CCCCCyx/Qwen3-8B-onpolicy-profiling-adam-20260403_091551
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
Qwen3-8B-onpolicy-profiling…/eval_20260409163705/run.log
ModelHub XC 1a22d07535 初始化项目，由ModelHub XC社区提供模型 
Model: CCCCCyx/Qwen3-8B-onpolicy-profiling-adam-20260403_091551
Source: Original Platform
2026-05-17 11:49:30 +08:00

131 lines
1003 KiB
Plaintext
Raw Permalink Blame History

[2026-04-09 16:37:05 nanoeval] run.py:42 - Step 0/1 system_prompt: None
[2026-04-09 16:37:05 nanoeval] task.py:107 - prepare_eval_input: system_prompt=None
[2026-04-09 16:37:08 nanoeval] run.py:52 - Step 0/1 completed: {'task_count': 6, 'instance_count': 5920, 'pass_k_by_task': {'gpqa_diamond': 4, 'hmmt2025': 4, 'aime2024': 32, 'aime2025': 32, 'math500': 4, 'minerva': 4}, 'task_sizes': {'gpqa_diamond': 792, 'hmmt2025': 120, 'aime2024': 960, 'aime2025': 960, 'math500': 2000, 'minerva': 1088}, 'output_path': '/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-20260403_091551/iter_0001023-hf/eval_20260409163705/step01_prepared.jsonl'}
[2026-04-09 16:37:08 nanoeval] base.py:107 - Initializing Engine (TP=1)...
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
[2026-04-09 16:37:14 nanoeval] server_args.py:1837 - Attention backend not specified. Use fa3 backend by default.
[2026-04-09 16:37:14 nanoeval] engine.py:157 - server_args=ServerArgs(model_path='/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-20260403_091551/iter_0001023-hf', tokenizer_path='/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-20260403_091551/iter_0001023-hf', tokenizer_mode='auto', tokenizer_worker_num=1, skip_tokenizer_init=False, load_format='auto', model_loader_extra_config='{}', trust_remote_code=True, context_length=None, is_embedding=False, enable_multimodal=None, revision=None, model_impl='auto', host='127.0.0.1', port=30000, fastapi_root_path='', grpc_mode=False, skip_server_warmup=False, warmups=None, nccl_port=None, checkpoint_engine_wait_weights_before_ready=False, dtype='auto', quantization=None, quantization_param_path=None, kv_cache_dtype='auto', enable_fp32_lm_head=False, modelopt_quant=None, modelopt_checkpoint_restore_path=None, modelopt_checkpoint_save_path=None, modelopt_export_path=None, quantize_and_serve=False, rl_quant_profile=None, mem_fraction_static=0.9, max_running_requests=None, max_queued_requests=None, max_total_tokens=None, chunked_prefill_size=8192, enable_dynamic_chunking=False, max_prefill_tokens=16384, prefill_max_requests=None, schedule_policy='fcfs', enable_priority_scheduling=False, abort_on_priority_when_disabled=False, schedule_low_priority_values_first=False, priority_scheduling_preemption_threshold=10, schedule_conservativeness=1.0, page_size=1, swa_full_tokens_ratio=0.8, disable_hybrid_swa_memory=False, radix_eviction_policy='lru', enable_prefill_delayer=False, prefill_delayer_max_delay_passes=30, prefill_delayer_token_usage_low_watermark=None, prefill_delayer_forward_passes_buckets=None, prefill_delayer_wait_seconds_buckets=None, device='cuda', tp_size=1, pp_size=1, pp_max_micro_batch_size=None, pp_async_batch_depth=0, stream_interval=1, stream_output=False, random_seed=199674123, constrained_json_whitespace_pattern=None, constrained_json_disable_any_whitespace=False, watchdog_timeout=300, soft_watchdog_timeout=None, dist_timeout=None, download_dir=None, model_checksum=None, base_gpu_id=0, gpu_id_step=1, sleep_on_idle=False, custom_sigquit_handler=None, log_level='error', log_level_http=None, log_requests=False, log_requests_level=2, log_requests_format='text', log_requests_target=None, uvicorn_access_log_exclude_prefixes=[], crash_dump_folder=None, show_time_cost=False, enable_metrics=False, enable_metrics_for_all_schedulers=False, tokenizer_metrics_custom_labels_header='x-custom-labels', tokenizer_metrics_allowed_custom_labels=None, extra_metric_labels=None, bucket_time_to_first_token=None, bucket_inter_token_latency=None, bucket_e2e_request_latency=None, collect_tokens_histogram=False, prompt_tokens_buckets=None, generation_tokens_buckets=None, gc_warning_threshold_secs=0.0, decode_log_interval=40, enable_request_time_stats_logging=False, kv_events_config=None, enable_trace=False, otlp_traces_endpoint='localhost:4317', export_metrics_to_file=False, export_metrics_to_file_dir=None, api_key=None, admin_api_key=None, served_model_name='/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-20260403_091551/iter_0001023-hf', weight_version='default', chat_template=None, hf_chat_template_name=None, completion_template=None, file_storage_path='sglang_storage', enable_cache_report=False, reasoning_parser=None, tool_call_parser=None, tool_server=None, sampling_defaults='model', dp_size=8, load_balance_method='round_robin', attn_cp_size=1, moe_dp_size=1, dist_init_addr=None, nnodes=1, node_rank=0, json_model_override_args='{}', preferred_sampling_params=None, enable_lora=None, enable_lora_overlap_loading=None, max_lora_rank=None, lora_target_modules=None, lora_paths=None, max_loaded_loras=None, max_loras_per_batch=8, lora_eviction_policy='lru', lora_backend='csgmv', max_lora_chunk_size=16, attention_backend='fa3', decode_attention_backend=None, prefill_attention_backend=None, sampling_backend='flashinfer', grammar_backend='xgrammar', mm_attention_backend=None, fp8_gemm_runner_backend='auto', fp4_gemm_runner_backend='flashinfer_cutlass', nsa_prefill_backend=None, nsa_decode_backend=None, disable_flashinfer_autotune=False, mamba_backend='triton', speculative_algorithm=None, speculative_draft_model_path=None, speculative_draft_model_revision=None, speculative_draft_load_format=None, speculative_num_steps=None, speculative_eagle_topk=None, speculative_num_draft_tokens=None, speculative_accept_threshold_single=1.0, speculative_accept_threshold_acc=1.0, speculative_token_map=None, speculative_attention_mode='prefill', speculative_draft_attention_backend=None, speculative_moe_runner_backend='auto', speculative_moe_a2a_backend=None, speculative_draft_model_quantization=None, speculative_ngram_min_match_window_size=1, speculative_ngram_max_match_window_size=12, speculative_ngram_min_bfs_breadth=1, speculative_ngram_max_bfs_breadth=10, speculative_ngram_match_type='BFS', speculative_ngram_branch_length=18, speculative_ngram_capacity=10000000, enable_multi_layer_eagle=False, ep_size=1, moe_a2a_backend='none', moe_runner_backend='auto', flashinfer_mxfp4_moe_precision='default', enable_flashinfer_allreduce_fusion=False, deepep_mode='auto', ep_num_redundant_experts=0, ep_dispatch_algorithm=None, init_expert_location='trivial', enable_eplb=False, eplb_algorithm='auto', eplb_rebalance_num_iterations=1000, eplb_rebalance_layers_per_chunk=None, eplb_min_rebalancing_utilization_threshold=1.0, expert_distribution_recorder_mode=None, expert_distribution_recorder_buffer_size=1000, enable_expert_distribution_metrics=False, deepep_config=None, moe_dense_tp_size=None, elastic_ep_backend=None, mooncake_ib_device=None, max_mamba_cache_size=None, mamba_ssm_dtype=None, mamba_full_memory_ratio=0.9, mamba_scheduler_strategy='no_buffer', mamba_track_interval=256, enable_hierarchical_cache=False, hicache_ratio=2.0, hicache_size=0, hicache_write_policy='write_through', hicache_io_backend='kernel', hicache_mem_layout='layer_first', disable_hicache_numa_detect=False, hicache_storage_backend=None, hicache_storage_prefetch_policy='best_effort', hicache_storage_backend_extra_config=None, hierarchical_sparse_attention_extra_config=None, enable_lmcache=False, kt_weight_path=None, kt_method=None, kt_cpuinfer=None, kt_threadpool_count=None, kt_num_gpu_experts=None, kt_max_deferred_experts_per_token=None, dllm_algorithm=None, dllm_algorithm_config=None, enable_double_sparsity=False, ds_channel_config_path=None, ds_heavy_channel_num=32, ds_heavy_token_num=256, ds_heavy_channel_type='qk', ds_sparse_decode_threshold=4096, cpu_offload_gb=0, offload_group_size=-1, offload_num_in_group=1, offload_prefetch_step=1, offload_mode='cpu', multi_item_scoring_delimiter=None, disable_radix_cache=True, cuda_graph_max_bs=256, cuda_graph_bs=[1, 2, 4, 8, 12, 16, 24, 32, 40, 48, 56, 64, 72, 80, 88, 96, 104, 112, 120, 128, 136, 144, 152, 160, 168, 176, 184, 192, 200, 208, 216, 224, 232, 240, 248, 256], disable_cuda_graph=False, disable_draft_cuda_graph=False, disable_cuda_graph_padding=False, enable_profile_cuda_graph=False, enable_cudagraph_gc=False, enable_layerwise_nvtx_marker=False, enable_nccl_nvls=False, enable_symm_mem=False, disable_flashinfer_cutlass_moe_fp4_allgather=False, enable_tokenizer_batch_encode=False, disable_tokenizer_batch_decode=False, disable_outlines_disk_cache=False, disable_custom_all_reduce=False, enable_mscclpp=False, enable_torch_symm_mem=False, disable_overlap_schedule=False, enable_mixed_chunk=False, enable_dp_attention=False, enable_dp_lm_head=False, enable_two_batch_overlap=False, enable_single_batch_overlap=False, tbo_token_distribution_threshold=0.48, enable_torch_compile=False, enable_piecewise_cuda_graph=False, enable_torch_compile_debug_mode=False, torch_compile_max_bs=32, piecewise_cuda_graph_max_tokens=8192, piecewise_cuda_graph_tokens=[4, 8, 12, 16, 20, 24, 28, 32, 48, 64, 80, 96, 112, 128, 144, 160, 176, 192, 208, 224, 240, 256, 288, 320, 352, 384, 416, 448, 480, 512, 576, 640, 704, 768, 832, 896, 960, 1024, 1280, 1536, 1792, 2048, 2304, 2560, 2816, 3072, 3328, 3584, 3840, 4096, 4608, 5120, 5632, 6144, 6656, 7168, 7680, 8192], piecewise_cuda_graph_compiler='eager', torchao_config='', enable_nan_detection=False, enable_p2p_check=False, triton_attention_reduce_in_fp32=False, triton_attention_num_kv_splits=8, triton_attention_split_tile_size=None, num_continuous_decode_steps=1, delete_ckpt_after_loading=False, enable_memory_saver=False, enable_weights_cpu_backup=False, enable_draft_weights_cpu_backup=False, allow_auto_truncate=False, enable_custom_logit_processor=False, flashinfer_mla_disable_ragged=False, disable_shared_experts_fusion=False, disable_chunked_prefix_cache=False, disable_fast_image_processor=False, keep_mm_feature_on_device=False, enable_return_hidden_states=False, enable_return_routed_experts=False, scheduler_recv_interval=1, numa_node=None, enable_deterministic_inference=False, rl_on_policy_target=None, enable_attn_tp_input_scattered=False, enable_nsa_prefill_context_parallel=False, nsa_prefill_cp_mode='round-robin-split', disable_indexer_rope_neox_style=False, enable_fused_qk_norm_rope=False, enable_precise_embedding_interpolation=False, enable_dynamic_batch_tokenizer=False, dynamic_batch_tokenizer_batch_size=32, dynamic_batch_tokenizer_batch_timeout=0.002, debug_tensor_dump_output_folder=None, debug_tensor_dump_layers=None, debug_tensor_dump_input_file=None, debug_tensor_dump_inject=False, disaggregation_mode='null', disaggregation_transfer_backend='mooncake', disaggregation_bootstrap_port=8998, disaggregation_decode_tp=None, disaggregation_decode_dp=None, disaggregation_prefill_pp=1, disaggregation_ib_device=None, disaggregation_decode_enable_offload_kvcache=False, num_reserved_decode_tokens=512, disaggregation_decode_polling_interval=1, encoder_only=False, language_only=False, encoder_transfer_backend='zmq_to_scheduler', encoder_urls=[], custom_weight_loader=[], weight_loader_disable_mmap=False, remote_instance_weight_loader_seed_instance_ip=None, remote_instance_weight_loader_seed_instance_service_port=None, remote_instance_weight_loader_send_weights_group_ports=None, remote_instance_weight_loader_backend='nccl', remote_instance_weight_loader_start_seed_via_transfer_engine=False, enable_pdmux=False, pdmux_config_path=None, sm_group_num=8, mm_max_concurrent_calls=32, mm_per_request_timeout=10.0, enable_broadcast_mm_inputs_process=False, enable_prefix_mm_cache=False, mm_enable_dp_encoder=False, mm_process_config={}, limit_mm_data_per_request=None, decrypted_config_file=None, decrypted_draft_config_file=None, forward_hooks=None)
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.cudart module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.runtime module instead.
<frozen importlib._bootstrap_external>:1297: FutureWarning: The cuda.nvrtc module is deprecated and will be removed in a future release, please switch to use the cuda.bindings.nvrtc module instead.
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
[Gloo] Rank 0 is connected to 0 peer ranks. Expected number of connected peer ranks is : 0
Loading safetensors checkpoint shards: 0% Completed | 0/4 [00:00<?, ?it/s]
Loading safetensors checkpoint shards: 0% Completed | 0/4 [00:00<?, ?it/s]
Loading safetensors checkpoint shards: 0% Completed | 0/4 [00:00<?, ?it/s]
Loading safetensors checkpoint shards: 0% Completed | 0/4 [00:00<?, ?it/s]
Loading safetensors checkpoint shards: 0% Completed | 0/4 [00:00<?, ?it/s]
Loading safetensors checkpoint shards: 0% Completed | 0/4 [00:00<?, ?it/s]
Loading safetensors checkpoint shards: 0% Completed | 0/4 [00:00<?, ?it/s]
Loading safetensors checkpoint shards: 0% Completed | 0/4 [00:00<?, ?it/s]
Loading safetensors checkpoint shards: 25% Completed | 1/4 [00:04<00:13, 4.60s/it]
Loading safetensors checkpoint shards: 25% Completed | 1/4 [00:04<00:12, 4.33s/it]
Loading safetensors checkpoint shards: 25% Completed | 1/4 [00:04<00:13, 4.45s/it]
Loading safetensors checkpoint shards: 25% Completed | 1/4 [00:04<00:13, 4.58s/it]
Loading safetensors checkpoint shards: 25% Completed | 1/4 [00:04<00:13, 4.59s/it]
Loading safetensors checkpoint shards: 25% Completed | 1/4 [00:04<00:13, 4.63s/it]
Loading safetensors checkpoint shards: 25% Completed | 1/4 [00:04<00:13, 4.59s/it]
Loading safetensors checkpoint shards: 25% Completed | 1/4 [00:04<00:13, 4.63s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:08<00:08, 4.44s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:08<00:08, 4.44s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:08<00:08, 4.46s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:08<00:08, 4.45s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:08<00:08, 4.38s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:08<00:08, 4.44s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:08<00:08, 4.33s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:08<00:08, 4.46s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:12<00:04, 4.06s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:12<00:04, 4.09s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:12<00:04, 4.10s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:12<00:04, 4.09s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:12<00:04, 4.03s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:12<00:04, 4.09s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:12<00:04, 4.09s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:12<00:04, 4.10s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 2.77s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 2.77s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 2.75s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 2.77s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 2.77s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 2.77s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 2.77s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 2.73s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 3.34s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 3.30s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 3.34s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 3.34s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 3.33s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 3.33s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 3.27s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:13<00:00, 3.35s/it]
0%| | 0/36 [00:00<?, ?it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 0%| | 0/36 [00:00<?, ?it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 3%|▎ | 1/36 [00:00<00:07, 4.88it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:07, 4.88it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:07, 4.88it/s]
0%| | 0/36 [00:00<?, ?it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 0%| | 0/36 [00:00<?, ?it/s]
0%| | 0/36 [00:00<?, ?it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 0%| | 0/36 [00:00<?, ?it/s]
0%| | 0/36 [00:00<?, ?it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 0%| | 0/36 [00:00<?, ?it/s]
0%| | 0/36 [00:00<?, ?it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 0%| | 0/36 [00:00<?, ?it/s]
0%| | 0/36 [00:00<?, ?it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 0%| | 0/36 [00:00<?, ?it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:07, 4.88it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 11%|█ | 4/36 [00:00<00:02, 13.17it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:02, 13.17it/s]
Capturing batches (bs=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.17it/s]
0%| | 0/36 [00:00<?, ?it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 0%| | 0/36 [00:00<?, ?it/s]
0%| | 0/36 [00:00<?, ?it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 0%| | 0/36 [00:00<?, ?it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.17it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 3%|▎ | 1/36 [00:00<00:06, 5.10it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:06, 5.10it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 3%|▎ | 1/36 [00:00<00:06, 5.09it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:06, 5.09it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 3%|▎ | 1/36 [00:00<00:06, 5.21it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:06, 5.21it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.35it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.35it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.10it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.09it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.21it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.35it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.10it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.09it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 3%|▎ | 1/36 [00:00<00:10, 3.40it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:10, 3.40it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.21it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 3%|▎ | 1/36 [00:00<00:10, 3.34it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:10, 3.34it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.35it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 3%|▎ | 1/36 [00:00<00:06, 5.27it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:06, 5.27it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 11%|█ | 4/36 [00:00<00:02, 13.45it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:02, 13.45it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:10, 3.40it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 11%|█ | 4/36 [00:00<00:02, 13.45it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:02, 13.45it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 11%|█ | 4/36 [00:00<00:02, 13.63it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:02, 13.63it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:10, 3.34it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.25it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.25it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.27it/s]
Capturing batches (bs=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.45it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 3%|▎ | 1/36 [00:00<00:07, 4.94it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:07, 4.94it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:10, 3.40it/s]
Capturing batches (bs=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.45it/s]
Capturing batches (bs=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.63it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:10, 3.34it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.25it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.27it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:07, 4.94it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.45it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 11%|█ | 4/36 [00:00<00:02, 10.81it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:02, 10.81it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.45it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.63it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 11%|█ | 4/36 [00:00<00:03, 10.65it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:03, 10.65it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.25it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 11%|█ | 4/36 [00:00<00:02, 13.76it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:02, 13.76it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:07, 4.94it/s]
Capturing batches (bs=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 10.81it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.54it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.54it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.56it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.56it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.66it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.66it/s]
Capturing batches (bs=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:03, 10.65it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.48it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.48it/s]
Capturing batches (bs=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.76it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 11%|█ | 4/36 [00:00<00:02, 13.25it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:02, 13.25it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.54it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 10.81it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.56it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.66it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:03, 10.65it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.48it/s]
Capturing batches (bs=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.25it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.76it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.54it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:01, 14.55it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:01, 14.55it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.56it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.66it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:02, 14.35it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:02, 14.35it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.48it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.83it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.83it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.25it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.31it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.31it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 14.55it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.37it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.37it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.37it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.37it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:02, 14.35it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.02it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.02it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.83it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.41it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.41it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.31it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 14.55it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.37it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.37it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:02, 14.35it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.02it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.83it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.31it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.41it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 16.87it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 16.87it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.37it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.37it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 16.68it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 16.68it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:01<00:00, 20.02it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.61it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.61it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.45it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.45it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.41it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 16.87it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.49it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.49it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.47it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.47it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 16.68it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.61it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.34it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.34it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.28it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.28it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.45it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 16.87it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.49it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.47it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 16.68it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.61it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.34it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.28it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 18.45it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 18.45it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.45it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.49it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.47it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 18.25it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 18.25it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.74it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.74it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.28it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.34it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 18.45it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:01, 19.90it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:01, 19.90it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:01, 19.96it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:01, 19.96it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:01, 19.90it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:01, 19.90it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 18.25it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.74it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.46it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.46it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 18.45it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 20.06it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.06it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:00<00:01, 19.90it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:00<00:01, 19.96it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:00<00:01, 19.90it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 18.25it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.74it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.46it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:01, 19.27it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:01, 19.27it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:01<00:01, 19.90it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.06it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:00<00:01, 19.96it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:00<00:01, 19.90it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:01<00:01, 19.05it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:01<00:01, 19.05it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.17it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.17it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.46it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:01<00:01, 19.27it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.15it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.15it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 61%|██████ | 22/36 [00:01<00:00, 20.06it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.23it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.23it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.15it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.15it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:01<00:01, 19.05it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.17it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.00it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.00it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:01<00:01, 19.27it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.15it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.00it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.00it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.23it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.15it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:01<00:01, 19.05it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.17it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.00it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 19.77it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 19.77it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.00it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.15it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.23it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.15it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 19.52it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 19.52it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.37it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.37it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:01<00:00, 20.00it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 19.77it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.00it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 19.82it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 19.82it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 19.90it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 19.90it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.37it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 19.52it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 19.76it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 19.76it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.24it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.24it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 19.77it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.92it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 19.82it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.92it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 19.90it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 19.76it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.37it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 19.52it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.24it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 19.62it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 19.62it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 67%|██████▋ | 24/36 [00:01<00:00, 19.82it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.92it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 67%|██████▋ | 24/36 [00:01<00:00, 19.82it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 61%|██████ | 22/36 [00:01<00:00, 19.90it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 67%|██████▋ | 24/36 [00:01<00:00, 19.77it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 67%|██████▋ | 24/36 [00:01<00:00, 19.77it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 20.04it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.04it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 19.40it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 19.40it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.24it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 19.62it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.92it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 67%|██████▋ | 24/36 [00:01<00:00, 19.82it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 69%|██████▉ | 25/36 [00:01<00:00, 19.88it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 69%|██████▉ | 25/36 [00:01<00:00, 19.88it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 67%|██████▋ | 24/36 [00:01<00:00, 19.77it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.04it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 19.40it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 19.93it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 19.93it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 61%|██████ | 22/36 [00:01<00:00, 19.62it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 72%|███████▏ | 26/36 [00:01<00:00, 19.82it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 72%|███████▏ | 26/36 [00:01<00:00, 19.82it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 19.88it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 72%|███████▏ | 26/36 [00:01<00:00, 19.75it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 72%|███████▏ | 26/36 [00:01<00:00, 19.75it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 61%|██████ | 22/36 [00:01<00:00, 20.04it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 67%|██████▋ | 24/36 [00:01<00:00, 19.50it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 67%|██████▋ | 24/36 [00:01<00:00, 19.50it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 19.93it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.39it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.39it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 69%|██████▉ | 25/36 [00:01<00:00, 19.69it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 69%|██████▉ | 25/36 [00:01<00:00, 19.69it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 72%|███████▏ | 26/36 [00:01<00:00, 19.82it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 75%|███████▌ | 27/36 [00:01<00:00, 19.82it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 75%|███████▌ | 27/36 [00:01<00:00, 19.82it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.01it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.01it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 72%|███████▏ | 26/36 [00:01<00:00, 19.75it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 67%|██████▋ | 24/36 [00:01<00:00, 19.50it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.39it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 61%|██████ | 22/36 [00:01<00:00, 19.93it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 19.69it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.69it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.69it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 75%|███████▌ | 27/36 [00:01<00:00, 19.82it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.01it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 72%|███████▏ | 26/36 [00:01<00:00, 19.58it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.64it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 72%|███████▏ | 26/36 [00:01<00:00, 19.58it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.64it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.39it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 69%|██████▉ | 25/36 [00:01<00:00, 19.91it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 69%|██████▉ | 25/36 [00:01<00:00, 19.91it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 19.69it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.69it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 81%|████████ | 29/36 [00:01<00:00, 19.75it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 81%|████████ | 29/36 [00:01<00:00, 19.75it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 94%|█████████▍| 34/36 [00:01<00:00, 19.52it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 19.52it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.01it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.64it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 19.91it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 72%|███████▏ | 26/36 [00:01<00:00, 19.58it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.71it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.71it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.64it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.64it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 81%|████████ | 29/36 [00:01<00:00, 19.75it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 19.52it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.88it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.88it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.59it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.59it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.50it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.50it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 19.91it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.71it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:01<00:00, 18.89it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.88it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.50it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.81it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.81it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.71it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.71it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.64it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 17.88it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 17.88it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.59it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.86it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.86it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.53it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.53it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 78%|███████▊ | 28/36 [00:01<00:00, 19.81it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.14it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.14it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 86%|████████▌ | 31/36 [00:01<00:00, 17.88it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.14it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.14it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.78it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.71it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.78it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.14it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 86%|████████▌ | 31/36 [00:01<00:00, 17.88it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.14it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.86it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.53it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.44it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.44it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.14it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 94%|█████████▍| 34/36 [00:01<00:00, 19.30it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 19.30it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.14it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.54it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.54it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.11it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.11it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 97%|█████████▋| 35/36 [00:01<00:00, 19.54it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 97%|█████████▋| 35/36 [00:01<00:00, 19.54it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 19.30it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.44it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 83%|████████▎ | 30/36 [00:01<00:00, 19.78it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 97%|█████████▋| 35/36 [00:01<00:00, 19.55it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 97%|█████████▋| 35/36 [00:01<00:00, 19.55it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.54it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:01<00:00, 18.79it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:01<00:00, 18.85it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.11it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.44it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.44it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.44it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:01<00:00, 18.81it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.54it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 97%|█████████▋| 35/36 [00:01<00:00, 19.72it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 97%|█████████▋| 35/36 [00:01<00:00, 19.72it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.11it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.44it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 97%|█████████▋| 35/36 [00:01<00:00, 19.80it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 97%|█████████▋| 35/36 [00:01<00:00, 19.80it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:02<00:00, 18.00it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 97%|█████████▋| 35/36 [00:01<00:00, 19.54it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 97%|█████████▋| 35/36 [00:01<00:00, 19.54it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 89%|████████▉ | 32/36 [00:01<00:00, 18.44it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:01<00:00, 19.03it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:02<00:00, 17.80it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 97%|█████████▋| 35/36 [00:01<00:00, 19.70it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 97%|█████████▋| 35/36 [00:01<00:00, 19.70it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:01<00:00, 18.83it/s]
Inference: 0%| | 0/5920 [00:00<?, ?it/s][2026-04-09 16:37:52] Worker Error: Requested token count exceeds the model's maximum context length of 32768 tokens. You requested a total of 32806 tokens: 2806 tokens from the input messages and 30000 tokens for the completion. Please reduce the number of tokens in the input messages or the completion to fit within the limit.
[2026-04-09 16:37:52] Worker Error: Requested token count exceeds the model's maximum context length of 32768 tokens. You requested a total of 32806 tokens: 2806 tokens from the input messages and 30000 tokens for the completion. Please reduce the number of tokens in the input messages or the completion to fit within the limit.
[2026-04-09 16:37:52] Worker Error: Requested token count exceeds the model's maximum context length of 32768 tokens. You requested a total of 32806 tokens: 2806 tokens from the input messages and 30000 tokens for the completion. Please reduce the number of tokens in the input messages or the completion to fit within the limit.
[2026-04-09 16:37:52] Worker Error: Requested token count exceeds the model's maximum context length of 32768 tokens. You requested a total of 32806 tokens: 2806 tokens from the input messages and 30000 tokens for the completion. Please reduce the number of tokens in the input messages or the completion to fit within the limit.
Inference: 0%| | 1/5920 [00:07<12:31:32, 7.62s/it]
Inference: 0%| | 1/5920 [00:07<12:31:32, 7.62s/it, tok/s=132.3]
Inference: 0%| | 2/5920 [00:07<12:31:24, 7.62s/it, tok/s=263.2]
Inference: 0%| | 3/5920 [00:07<12:31:16, 7.62s/it, tok/s=394.6]
Inference: 0%| | 4/5920 [00:07<12:31:09, 7.62s/it, tok/s=524.7]
Inference: 0%| | 5/5920 [00:08<2:13:37, 1.36s/it, tok/s=524.7]
Inference: 0%| | 5/5920 [00:08<2:13:37, 1.36s/it, tok/s=590.0]
Inference: 0%| | 6/5920 [00:08<2:13:36, 1.36s/it, tok/s=707.7]
Inference: 0%| | 7/5920 [00:08<1:26:13, 1.14it/s, tok/s=707.7]
Inference: 0%| | 7/5920 [00:08<1:26:13, 1.14it/s, tok/s=825.7]
Inference: 0%| | 8/5920 [00:09<1:14:29, 1.32it/s, tok/s=825.7]
Inference: 0%| | 8/5920 [00:09<1:14:29, 1.32it/s, tok/s=921.2]
Inference: 0%| | 9/5920 [00:10<1:22:09, 1.20it/s, tok/s=921.2]
Inference: 0%| | 9/5920 [00:10<1:22:09, 1.20it/s, tok/s=947.6]
Inference: 0%| | 10/5920 [00:11<1:38:14, 1.00it/s, tok/s=947.6]
Inference: 0%| | 10/5920 [00:11<1:38:14, 1.00it/s, tok/s=949.0]
Inference: 0%| | 11/5920 [00:11<1:38:13, 1.00it/s, tok/s=1064.2]
Inference: 0%| | 12/5920 [00:12<1:11:52, 1.37it/s, tok/s=1064.2]
Inference: 0%| | 12/5920 [00:12<1:11:52, 1.37it/s, tok/s=1130.3]
Inference: 0%| | 13/5920 [00:13<1:08:59, 1.43it/s, tok/s=1130.3]
Inference: 0%| | 13/5920 [00:13<1:08:59, 1.43it/s, tok/s=1194.9]
Inference: 0%| | 14/5920 [00:13<1:04:24, 1.53it/s, tok/s=1194.9]
Inference: 0%| | 14/5920 [00:13<1:04:24, 1.53it/s, tok/s=1265.9]
Inference: 0%| | 15/5920 [00:13<50:17, 1.96it/s, tok/s=1265.9]
Inference: 0%| | 15/5920 [00:13<50:17, 1.96it/s, tok/s=1380.5]
Inference: 0%| | 16/5920 [00:14<50:03, 1.97it/s, tok/s=1380.5]
Inference: 0%| | 16/5920 [00:14<50:03, 1.97it/s, tok/s=1449.4]
Inference: 0%| | 17/5920 [00:14<45:13, 2.18it/s, tok/s=1449.4]
Inference: 0%| | 17/5920 [00:14<45:13, 2.18it/s, tok/s=1547.4]
Inference: 0%| | 18/5920 [00:16<1:34:17, 1.04it/s, tok/s=1547.4]
Inference: 0%| | 18/5920 [00:16<1:34:17, 1.04it/s, tok/s=1461.7]
Inference: 0%| | 19/5920 [00:16<1:11:38, 1.37it/s, tok/s=1461.7]
Inference: 0%| | 19/5920 [00:16<1:11:38, 1.37it/s, tok/s=1573.3]
Inference: 0%| | 20/5920 [00:19<1:55:05, 1.17s/it, tok/s=1573.3]
Inference: 0%| | 20/5920 [00:19<1:55:05, 1.17s/it, tok/s=1505.7]
Inference: 0%| | 21/5920 [00:19<1:34:20, 1.04it/s, tok/s=1505.7]
Inference: 0%| | 21/5920 [00:19<1:34:20, 1.04it/s, tok/s=1591.1]
Inference: 0%| | 22/5920 [00:19<1:34:19, 1.04it/s, tok/s=1710.0]
Inference: 0%| | 23/5920 [00:20<1:08:46, 1.43it/s, tok/s=1710.0]
Inference: 0%| | 23/5920 [00:20<1:08:46, 1.43it/s, tok/s=1758.5]
Inference: 0%| | 24/5920 [00:21<1:11:46, 1.37it/s, tok/s=1758.5]
Inference: 0%| | 24/5920 [00:21<1:11:46, 1.37it/s, tok/s=1800.6]
Inference: 0%| | 25/5920 [00:22<1:22:44, 1.19it/s, tok/s=1800.6]
Inference: 0%| | 25/5920 [00:22<1:22:44, 1.19it/s, tok/s=1812.3]
Inference: 0%| | 26/5920 [00:22<1:06:57, 1.47it/s, tok/s=1812.3]
Inference: 0%| | 26/5920 [00:22<1:06:57, 1.47it/s, tok/s=1897.3]
Inference: 0%| | 27/5920 [00:24<1:42:04, 1.04s/it, tok/s=1897.3]
Inference: 0%| | 27/5920 [00:24<1:42:04, 1.04s/it, tok/s=1846.8]
Inference: 0%| | 28/5920 [00:26<2:15:53, 1.38s/it, tok/s=1846.8]
Inference: 0%| | 28/5920 [00:26<2:15:53, 1.38s/it, tok/s=1793.1]
Inference: 0%| | 29/5920 [00:27<1:57:21, 1.20s/it, tok/s=1793.1]
Inference: 0%| | 29/5920 [00:27<1:57:21, 1.20s/it, tok/s=1848.7]
Inference: 1%| | 30/5920 [00:28<1:57:58, 1.20s/it, tok/s=1848.7]
Inference: 1%| | 30/5920 [00:28<1:57:58, 1.20s/it, tok/s=1866.8]
Inference: 1%| | 31/5920 [00:29<1:54:01, 1.16s/it, tok/s=1866.8]
Inference: 1%| | 31/5920 [00:29<1:54:01, 1.16s/it, tok/s=1899.6]
Inference: 1%| | 32/5920 [00:29<1:54:00, 1.16s/it, tok/s=2000.1]
Inference: 1%| | 33/5920 [00:30<1:18:25, 1.25it/s, tok/s=2000.1]
Inference: 1%| | 33/5920 [00:30<1:18:25, 1.25it/s, tok/s=2053.5]
Inference: 1%| | 34/5920 [00:30<1:06:30, 1.47it/s, tok/s=2053.5]
Inference: 1%| | 34/5920 [00:30<1:06:30, 1.47it/s, tok/s=2134.1]
Inference: 1%| | 35/5920 [00:30<1:06:30, 1.47it/s, tok/s=2228.8]
Inference: 1%| | 36/5920 [00:32<1:03:03, 1.56it/s, tok/s=2228.8]
Inference: 1%| | 36/5920 [00:32<1:03:03, 1.56it/s, tok/s=2249.9]
Inference: 1%| | 37/5920 [00:32<51:45, 1.89it/s, tok/s=2249.9]
Inference: 1%| | 37/5920 [00:32<51:45, 1.89it/s, tok/s=2333.3]
Inference: 1%| | 38/5920 [00:32<51:44, 1.89it/s, tok/s=2428.6]
Inference: 1%| | 39/5920 [00:32<37:59, 2.58it/s, tok/s=2428.6]
Inference: 1%| | 39/5920 [00:32<37:59, 2.58it/s, tok/s=2507.1]
Inference: 1%| | 40/5920 [00:32<35:22, 2.77it/s, tok/s=2507.1]
Inference: 1%| | 40/5920 [00:32<35:22, 2.77it/s, tok/s=2580.7]
Inference: 1%| | 41/5920 [00:33<44:29, 2.20it/s, tok/s=2580.7]
Inference: 1%| | 41/5920 [00:33<44:29, 2.20it/s, tok/s=2617.2]
Inference: 1%| | 42/5920 [00:35<1:14:50, 1.31it/s, tok/s=2617.2]
Inference: 1%| | 42/5920 [00:35<1:14:50, 1.31it/s, tok/s=2588.6]
Inference: 1%| | 43/5920 [00:36<1:30:31, 1.08it/s, tok/s=2588.6]
Inference: 1%| | 43/5920 [00:36<1:30:31, 1.08it/s, tok/s=2585.9]
Inference: 1%| | 44/5920 [00:40<2:45:54, 1.69s/it, tok/s=2585.9]
Inference: 1%| | 44/5920 [00:40<2:45:54, 1.69s/it, tok/s=2443.5]
Inference: 1%| | 45/5920 [00:41<2:27:20, 1.50s/it, tok/s=2443.5]
Inference: 1%| | 45/5920 [00:41<2:27:20, 1.50s/it, tok/s=2473.9]
Inference: 1%| | 46/5920 [00:42<2:13:48, 1.37s/it, tok/s=2473.9]
Inference: 1%| | 46/5920 [00:42<2:13:48, 1.37s/it, tok/s=2505.2]
Inference: 1%| | 47/5920 [00:45<3:05:56, 1.90s/it, tok/s=2505.2]
Inference: 1%| | 47/5920 [00:45<3:05:56, 1.90s/it, tok/s=2416.0]
Inference: 1%| | 48/5920 [00:46<2:38:29, 1.62s/it, tok/s=2416.0]
Inference: 1%| | 48/5920 [00:46<2:38:29, 1.62s/it, tok/s=2450.7]
Inference: 1%| | 49/5920 [00:46<1:54:55, 1.17s/it, tok/s=2450.7]
Inference: 1%| | 49/5920 [00:46<1:54:55, 1.17s/it, tok/s=2530.6]
Inference: 1%| | 50/5920 [00:46<1:30:41, 1.08it/s, tok/s=2530.6]
Inference: 1%| | 50/5920 [00:46<1:30:41, 1.08it/s, tok/s=2600.2]
Inference: 1%| | 51/5920 [00:49<2:21:52, 1.45s/it, tok/s=2600.2]
Inference: 1%| | 51/5920 [00:49<2:21:52, 1.45s/it, tok/s=2543.9]
Inference: 1%| | 52/5920 [00:50<1:54:47, 1.17s/it, tok/s=2543.9]
Inference: 1%| | 52/5920 [00:50<1:54:47, 1.17s/it, tok/s=2604.3]
Inference: 1%| | 53/5920 [00:50<1:24:37, 1.16it/s, tok/s=2604.3]
Inference: 1%| | 53/5920 [00:50<1:24:37, 1.16it/s, tok/s=2680.9]
Inference: 1%| | 54/5920 [00:52<2:12:53, 1.36s/it, tok/s=2680.9]
Inference: 1%| | 54/5920 [00:52<2:12:53, 1.36s/it, tok/s=2635.3]
Inference: 1%| | 55/5920 [00:54<2:19:13, 1.42s/it, tok/s=2635.3]
Inference: 1%| | 55/5920 [00:54<2:19:13, 1.42s/it, tok/s=2641.3]
Inference: 1%| | 56/5920 [00:55<2:12:15, 1.35s/it, tok/s=2641.3]
Inference: 1%| | 56/5920 [00:55<2:12:15, 1.35s/it, tok/s=2665.9]
Inference: 1%| | 57/5920 [00:55<1:43:59, 1.06s/it, tok/s=2665.9]
Inference: 1%| | 57/5920 [00:55<1:43:59, 1.06s/it, tok/s=2730.9]
Inference: 1%| | 58/5920 [00:56<1:37:30, 1.00it/s, tok/s=2730.9]
Inference: 1%| | 58/5920 [00:56<1:37:30, 1.00it/s, tok/s=2772.6]
Inference: 1%| | 59/5920 [00:57<1:40:21, 1.03s/it, tok/s=2772.6]
Inference: 1%| | 59/5920 [00:57<1:40:21, 1.03s/it, tok/s=2801.3]
Inference: 1%| | 60/5920 [00:58<1:19:39, 1.23it/s, tok/s=2801.3]
Inference: 1%| | 60/5920 [00:58<1:19:39, 1.23it/s, tok/s=2866.2]
Inference: 1%| | 61/5920 [00:58<1:07:18, 1.45it/s, tok/s=2866.2]
Inference: 1%| | 61/5920 [00:58<1:07:18, 1.45it/s, tok/s=2927.4]
Inference: 1%| | 62/5920 [00:58<58:20, 1.67it/s, tok/s=2927.4]
Inference: 1%| | 62/5920 [00:58<58:20, 1.67it/s, tok/s=2987.2]
Inference: 1%| | 63/5920 [00:59<46:06, 2.12it/s, tok/s=2987.2]
Inference: 1%| | 63/5920 [00:59<46:06, 2.12it/s, tok/s=3059.2]
Inference: 1%| | 64/5920 [01:00<1:11:14, 1.37it/s, tok/s=3059.2]
Inference: 1%| | 64/5920 [01:00<1:11:14, 1.37it/s, tok/s=3071.1]
Inference: 1%| | 65/5920 [01:01<1:12:21, 1.35it/s, tok/s=3071.1]
Inference: 1%| | 65/5920 [01:01<1:12:21, 1.35it/s, tok/s=3114.0]
Inference: 1%| | 66/5920 [01:01<59:18, 1.64it/s, tok/s=3114.0]
Inference: 1%| | 66/5920 [01:01<59:18, 1.64it/s, tok/s=3177.2]
Inference: 1%| | 67/5920 [01:01<51:31, 1.89it/s, tok/s=3177.2]
Inference: 1%| | 67/5920 [01:01<51:31, 1.89it/s, tok/s=3239.4]
Inference: 1%| | 68/5920 [01:03<1:12:21, 1.35it/s, tok/s=3239.4]
Inference: 1%| | 68/5920 [01:03<1:12:21, 1.35it/s, tok/s=3254.2]
Inference: 1%| | 69/5920 [01:04<1:18:12, 1.25it/s, tok/s=3254.2]
Inference: 1%| | 69/5920 [01:04<1:18:12, 1.25it/s, tok/s=3282.7]
Inference: 1%| | 70/5920 [01:04<1:21:13, 1.20it/s, tok/s=3282.7]
Inference: 1%| | 70/5920 [01:04<1:21:13, 1.20it/s, tok/s=3313.2]
Inference: 1%| | 71/5920 [01:05<1:01:28, 1.59it/s, tok/s=3313.2]
Inference: 1%| | 71/5920 [01:05<1:01:28, 1.59it/s, tok/s=3382.7]
Inference: 1%| | 72/5920 [01:05<58:02, 1.68it/s, tok/s=3382.7]
Inference: 1%| | 72/5920 [01:05<58:02, 1.68it/s, tok/s=3435.5]
Inference: 1%| | 73/5920 [01:05<44:30, 2.19it/s, tok/s=3435.5]
Inference: 1%| | 73/5920 [01:05<44:30, 2.19it/s, tok/s=3504.7]
Inference: 1%|▏ | 74/5920 [01:06<41:33, 2.34it/s, tok/s=3504.7]
Inference: 1%|▏ | 74/5920 [01:06<41:33, 2.34it/s, tok/s=3564.0]
Inference: 1%|▏ | 75/5920 [01:06<40:51, 2.38it/s, tok/s=3564.0]
Inference: 1%|▏ | 75/5920 [01:06<40:51, 2.38it/s, tok/s=3620.4]
Inference: 1%|▏ | 76/5920 [01:07<58:40, 1.66it/s, tok/s=3620.4]
Inference: 1%|▏ | 76/5920 [01:07<58:40, 1.66it/s, tok/s=3642.6]
Inference: 1%|▏ | 77/5920 [01:10<2:01:57, 1.25s/it, tok/s=3642.6]
Inference: 1%|▏ | 77/5920 [01:10<2:01:57, 1.25s/it, tok/s=3573.9]
Inference: 1%|▏ | 78/5920 [01:10<1:31:57, 1.06it/s, tok/s=3573.9]
Inference: 1%|▏ | 78/5920 [01:10<1:31:57, 1.06it/s, tok/s=3636.5]
Inference: 1%|▏ | 79/5920 [01:10<1:09:45, 1.40it/s, tok/s=3636.5]
Inference: 1%|▏ | 79/5920 [01:10<1:09:45, 1.40it/s, tok/s=3701.7]
Inference: 1%|▏ | 80/5920 [01:10<54:12, 1.80it/s, tok/s=3701.7]
Inference: 1%|▏ | 80/5920 [01:10<54:12, 1.80it/s, tok/s=3766.6]
Inference: 1%|▏ | 81/5920 [01:13<1:41:18, 1.04s/it, tok/s=3766.6]
Inference: 1%|▏ | 81/5920 [01:13<1:41:18, 1.04s/it, tok/s=3729.0]
Inference: 1%|▏ | 82/5920 [01:14<1:48:10, 1.11s/it, tok/s=3729.0]
Inference: 1%|▏ | 82/5920 [01:14<1:48:10, 1.11s/it, tok/s=3738.7]
Inference: 1%|▏ | 83/5920 [01:14<1:18:57, 1.23it/s, tok/s=3738.7]
Inference: 1%|▏ | 83/5920 [01:14<1:18:57, 1.23it/s, tok/s=3809.5]
Inference: 1%|▏ | 84/5920 [01:14<1:05:44, 1.48it/s, tok/s=3809.5]
Inference: 1%|▏ | 84/5920 [01:14<1:05:44, 1.48it/s, tok/s=3865.8]
Inference: 1%|▏ | 85/5920 [01:15<1:03:25, 1.53it/s, tok/s=3865.8]
Inference: 1%|▏ | 85/5920 [01:15<1:03:25, 1.53it/s, tok/s=3909.2]
Inference: 1%|▏ | 86/5920 [01:17<1:34:33, 1.03it/s, tok/s=3909.2]
Inference: 1%|▏ | 86/5920 [01:17<1:34:33, 1.03it/s, tok/s=3894.7]
Inference: 1%|▏ | 87/5920 [01:19<2:15:03, 1.39s/it, tok/s=3894.7]
Inference: 1%|▏ | 87/5920 [01:19<2:15:03, 1.39s/it, tok/s=3851.4]
Inference: 1%|▏ | 88/5920 [01:21<2:36:59, 1.62s/it, tok/s=3851.4]
Inference: 1%|▏ | 88/5920 [01:21<2:36:59, 1.62s/it, tok/s=3823.6]
Inference: 2%|▏ | 89/5920 [01:22<2:08:58, 1.33s/it, tok/s=3823.6]
Inference: 2%|▏ | 89/5920 [01:22<2:08:58, 1.33s/it, tok/s=3864.6]
Inference: 2%|▏ | 90/5920 [01:22<1:41:02, 1.04s/it, tok/s=3864.6]
Inference: 2%|▏ | 90/5920 [01:22<1:41:02, 1.04s/it, tok/s=3920.7]
Inference: 2%|▏ | 91/5920 [01:22<1:14:42, 1.30it/s, tok/s=3920.7]
Inference: 2%|▏ | 91/5920 [01:22<1:14:42, 1.30it/s, tok/s=3986.8]
Inference: 2%|▏ | 92/5920 [01:23<56:57, 1.71it/s, tok/s=3986.8]
Inference: 2%|▏ | 92/5920 [01:23<56:57, 1.71it/s, tok/s=4057.6]
Inference: 2%|▏ | 93/5920 [01:24<1:26:44, 1.12it/s, tok/s=4057.6]
Inference: 2%|▏ | 93/5920 [01:24<1:26:44, 1.12it/s, tok/s=4051.9]
Inference: 2%|▏ | 94/5920 [01:27<2:38:17, 1.63s/it, tok/s=4051.9]
Inference: 2%|▏ | 94/5920 [01:27<2:38:17, 1.63s/it, tok/s=3969.1]
Inference: 2%|▏ | 95/5920 [01:32<4:03:59, 2.51s/it, tok/s=3969.1]
Inference: 2%|▏ | 95/5920 [01:32<4:03:59, 2.51s/it, tok/s=3841.6]
Inference: 2%|▏ | 96/5920 [01:32<4:03:56, 2.51s/it, tok/s=3911.1]
Inference: 2%|▏ | 97/5920 [01:33<2:22:45, 1.47s/it, tok/s=3911.1]
Inference: 2%|▏ | 97/5920 [01:33<2:22:45, 1.47s/it, tok/s=3964.0]
Inference: 2%|▏ | 98/5920 [01:33<1:52:48, 1.16s/it, tok/s=3964.0]
Inference: 2%|▏ | 98/5920 [01:33<1:52:48, 1.16s/it, tok/s=4022.9]
Inference: 2%|▏ | 99/5920 [01:33<1:37:18, 1.00s/it, tok/s=4022.9]
Inference: 2%|▏ | 99/5920 [01:33<1:37:18, 1.00s/it, tok/s=4067.3]
Inference: 2%|▏ | 100/5920 [01:35<2:01:55, 1.26s/it, tok/s=4067.3]
Inference: 2%|▏ | 100/5920 [01:35<2:01:55, 1.26s/it, tok/s=4053.1]
Inference: 2%|▏ | 101/5920 [01:36<1:44:30, 1.08s/it, tok/s=4053.1]
Inference: 2%|▏ | 101/5920 [01:36<1:44:30, 1.08s/it, tok/s=4096.8]
Inference: 2%|▏ | 102/5920 [01:36<1:25:06, 1.14it/s, tok/s=4096.8]
Inference: 2%|▏ | 102/5920 [01:36<1:25:06, 1.14it/s, tok/s=4146.8]
Inference: 2%|▏ | 103/5920 [01:36<1:05:39, 1.48it/s, tok/s=4146.8]
Inference: 2%|▏ | 103/5920 [01:36<1:05:39, 1.48it/s, tok/s=4206.6]
Inference: 2%|▏ | 104/5920 [01:37<55:04, 1.76it/s, tok/s=4206.6]
Inference: 2%|▏ | 104/5920 [01:37<55:04, 1.76it/s, tok/s=4260.5]
Inference: 2%|▏ | 105/5920 [01:37<51:12, 1.89it/s, tok/s=4260.5]
Inference: 2%|▏ | 105/5920 [01:37<51:12, 1.89it/s, tok/s=4308.9]
Inference: 2%|▏ | 106/5920 [01:40<1:46:15, 1.10s/it, tok/s=4308.9]
Inference: 2%|▏ | 106/5920 [01:40<1:46:15, 1.10s/it, tok/s=4270.5]
Inference: 2%|▏ | 107/5920 [01:42<2:14:16, 1.39s/it, tok/s=4270.5]
Inference: 2%|▏ | 107/5920 [01:42<2:14:16, 1.39s/it, tok/s=4253.1]
Inference: 2%|▏ | 108/5920 [01:42<1:37:48, 1.01s/it, tok/s=4253.1]
Inference: 2%|▏ | 108/5920 [01:42<1:37:48, 1.01s/it, tok/s=4314.2]
Inference: 2%|▏ | 109/5920 [01:42<1:19:30, 1.22it/s, tok/s=4314.2]
Inference: 2%|▏ | 109/5920 [01:42<1:19:30, 1.22it/s, tok/s=4364.8]
Inference: 2%|▏ | 110/5920 [01:45<2:27:02, 1.52s/it, tok/s=4364.8]
Inference: 2%|▏ | 110/5920 [01:45<2:27:02, 1.52s/it, tok/s=4300.3]
Inference: 2%|▏ | 111/5920 [01:48<3:08:40, 1.95s/it, tok/s=4300.3]
Inference: 2%|▏ | 111/5920 [01:48<3:08:40, 1.95s/it, tok/s=4248.6]
Inference: 2%|▏ | 112/5920 [01:53<4:19:29, 2.68s/it, tok/s=4248.6]
Inference: 2%|▏ | 112/5920 [01:53<4:19:29, 2.68s/it, tok/s=4148.4]
Inference: 2%|▏ | 113/5920 [01:54<3:50:35, 2.38s/it, tok/s=4148.4]
Inference: 2%|▏ | 113/5920 [01:54<3:50:35, 2.38s/it, tok/s=4151.8]
Inference: 2%|▏ | 114/5920 [01:55<3:07:13, 1.93s/it, tok/s=4151.8]
Inference: 2%|▏ | 114/5920 [01:55<3:07:13, 1.93s/it, tok/s=4184.6]
Inference: 2%|▏ | 115/5920 [01:56<2:43:41, 1.69s/it, tok/s=4184.6]
Inference: 2%|▏ | 115/5920 [01:56<2:43:41, 1.69s/it, tok/s=4207.3]
Inference: 2%|▏ | 116/5920 [01:57<2:21:09, 1.46s/it, tok/s=4207.3]
Inference: 2%|▏ | 116/5920 [01:57<2:21:09, 1.46s/it, tok/s=4236.7]
Inference: 2%|▏ | 117/5920 [01:58<2:10:29, 1.35s/it, tok/s=4236.7]
Inference: 2%|▏ | 117/5920 [01:58<2:10:29, 1.35s/it, tok/s=4262.7]
Inference: 2%|▏ | 118/5920 [01:59<1:34:30, 1.02it/s, tok/s=4262.7]
Inference: 2%|▏ | 118/5920 [01:59<1:34:30, 1.02it/s, tok/s=4323.0]
Inference: 2%|▏ | 119/5920 [01:59<1:09:40, 1.39it/s, tok/s=4323.0]
Inference: 2%|▏ | 119/5920 [01:59<1:09:40, 1.39it/s, tok/s=4383.7]
Inference: 2%|▏ | 120/5920 [01:59<57:01, 1.70it/s, tok/s=4383.7]
Inference: 2%|▏ | 120/5920 [01:59<57:01, 1.70it/s, tok/s=4435.2]
Inference: 2%|▏ | 121/5920 [01:59<55:50, 1.73it/s, tok/s=4435.2]
Inference: 2%|▏ | 121/5920 [01:59<55:50, 1.73it/s, tok/s=4476.7]
Inference: 2%|▏ | 122/5920 [02:00<44:59, 2.15it/s, tok/s=4476.7]
Inference: 2%|▏ | 122/5920 [02:00<44:59, 2.15it/s, tok/s=4479.4]
Inference: 2%|▏ | 123/5920 [02:00<54:48, 1.76it/s, tok/s=4479.4]
Inference: 2%|▏ | 123/5920 [02:00<54:48, 1.76it/s, tok/s=4514.3]
Inference: 2%|▏ | 124/5920 [02:02<1:36:21, 1.00it/s, tok/s=4514.3]
Inference: 2%|▏ | 124/5920 [02:02<1:36:21, 1.00it/s, tok/s=4504.3]
Inference: 2%|▏ | 125/5920 [02:04<1:40:38, 1.04s/it, tok/s=4504.3]
Inference: 2%|▏ | 125/5920 [02:04<1:40:38, 1.04s/it, tok/s=4524.0]
Inference: 2%|▏ | 126/5920 [02:04<1:15:51, 1.27it/s, tok/s=4524.0]
Inference: 2%|▏ | 126/5920 [02:04<1:15:51, 1.27it/s, tok/s=4580.4]
Inference: 2%|▏ | 127/5920 [02:04<57:11, 1.69it/s, tok/s=4580.4]
Inference: 2%|▏ | 127/5920 [02:04<57:11, 1.69it/s, tok/s=4637.4]
Inference: 2%|▏ | 128/5920 [02:04<57:10, 1.69it/s, tok/s=4705.9]
Inference: 2%|▏ | 129/5920 [02:05<1:01:49, 1.56it/s, tok/s=4705.9]
Inference: 2%|▏ | 129/5920 [02:05<1:01:49, 1.56it/s, tok/s=4665.1]
Inference: 2%|▏ | 130/5920 [02:06<58:07, 1.66it/s, tok/s=4665.1]
Inference: 2%|▏ | 130/5920 [02:06<58:07, 1.66it/s, tok/s=4714.7]
Inference: 2%|▏ | 131/5920 [02:06<50:19, 1.92it/s, tok/s=4714.7]
Inference: 2%|▏ | 131/5920 [02:06<50:19, 1.92it/s, tok/s=4766.0]
Inference: 2%|▏ | 132/5920 [02:07<56:38, 1.70it/s, tok/s=4766.0]
Inference: 2%|▏ | 132/5920 [02:07<56:38, 1.70it/s, tok/s=4800.1]
Inference: 2%|▏ | 133/5920 [02:07<51:16, 1.88it/s, tok/s=4800.1]
Inference: 2%|▏ | 133/5920 [02:07<51:16, 1.88it/s, tok/s=4846.7]
Inference: 2%|▏ | 134/5920 [02:08<46:53, 2.06it/s, tok/s=4846.7]
Inference: 2%|▏ | 134/5920 [02:08<46:53, 2.06it/s, tok/s=4894.6]
Inference: 2%|▏ | 135/5920 [02:08<45:48, 2.10it/s, tok/s=4894.6]
Inference: 2%|▏ | 135/5920 [02:08<45:48, 2.10it/s, tok/s=4888.7]
Inference: 2%|▏ | 136/5920 [02:08<35:25, 2.72it/s, tok/s=4888.7]
Inference: 2%|▏ | 136/5920 [02:08<35:25, 2.72it/s, tok/s=4945.5]
Inference: 2%|▏ | 137/5920 [02:09<49:33, 1.94it/s, tok/s=4945.5]
Inference: 2%|▏ | 137/5920 [02:09<49:33, 1.94it/s, tok/s=4925.2]
Inference: 2%|▏ | 138/5920 [02:11<1:41:54, 1.06s/it, tok/s=4925.2]
Inference: 2%|▏ | 138/5920 [02:11<1:41:54, 1.06s/it, tok/s=4897.2]
Inference: 2%|▏ | 139/5920 [02:13<1:57:20, 1.22s/it, tok/s=4897.2]
Inference: 2%|▏ | 139/5920 [02:13<1:57:20, 1.22s/it, tok/s=4902.6]
Inference: 2%|▏ | 140/5920 [02:13<1:31:32, 1.05it/s, tok/s=4902.6]
Inference: 2%|▏ | 140/5920 [02:13<1:31:32, 1.05it/s, tok/s=4950.8]
Inference: 2%|▏ | 141/5920 [02:14<1:13:13, 1.32it/s, tok/s=4950.8]
Inference: 2%|▏ | 141/5920 [02:14<1:13:13, 1.32it/s, tok/s=5000.5]
Inference: 2%|▏ | 142/5920 [02:17<2:38:25, 1.65s/it, tok/s=5000.5]
Inference: 2%|▏ | 142/5920 [02:17<2:38:25, 1.65s/it, tok/s=4926.7]
Inference: 2%|▏ | 143/5920 [02:18<1:57:32, 1.22s/it, tok/s=4926.7]
Inference: 2%|▏ | 143/5920 [02:18<1:57:32, 1.22s/it, tok/s=4978.6]
Inference: 2%|▏ | 144/5920 [02:19<2:13:56, 1.39s/it, tok/s=4978.6]
Inference: 2%|▏ | 144/5920 [02:19<2:13:56, 1.39s/it, tok/s=4969.6]
Inference: 2%|▏ | 145/5920 [02:23<3:27:40, 2.16s/it, tok/s=4969.6]
Inference: 2%|▏ | 145/5920 [02:23<3:27:40, 2.16s/it, tok/s=4891.4]
Inference: 2%|▏ | 146/5920 [02:25<3:24:19, 2.12s/it, tok/s=4891.4]
Inference: 2%|▏ | 146/5920 [02:25<3:24:19, 2.12s/it, tok/s=4881.4]
Inference: 2%|▏ | 147/5920 [02:26<2:42:53, 1.69s/it, tok/s=4881.4]
Inference: 2%|▏ | 147/5920 [02:26<2:42:53, 1.69s/it, tok/s=4918.0]
Inference: 2%|▎ | 148/5920 [02:27<2:19:49, 1.45s/it, tok/s=4918.0]
Inference: 2%|▎ | 148/5920 [02:27<2:19:49, 1.45s/it, tok/s=4946.1]
Inference: 3%|▎ | 149/5920 [02:27<1:45:08, 1.09s/it, tok/s=4946.1]
Inference: 3%|▎ | 149/5920 [02:27<1:45:08, 1.09s/it, tok/s=4998.7]
Inference: 3%|▎ | 150/5920 [02:28<1:35:42, 1.00it/s, tok/s=4998.7]
Inference: 3%|▎ | 150/5920 [02:28<1:35:42, 1.00it/s, tok/s=5033.4]
Inference: 3%|▎ | 151/5920 [02:28<1:11:35, 1.34it/s, tok/s=5033.4]
Inference: 3%|▎ | 151/5920 [02:28<1:11:35, 1.34it/s, tok/s=5076.2]
Inference: 3%|▎ | 152/5920 [02:29<1:12:37, 1.32it/s, tok/s=5076.2]
Inference: 3%|▎ | 152/5920 [02:29<1:12:37, 1.32it/s, tok/s=5111.5]
Inference: 3%|▎ | 153/5920 [02:29<1:03:04, 1.52it/s, tok/s=5111.5]
Inference: 3%|▎ | 153/5920 [02:29<1:03:04, 1.52it/s, tok/s=5159.5]
Inference: 3%|▎ | 154/5920 [02:30<57:07, 1.68it/s, tok/s=5159.5]
Inference: 3%|▎ | 154/5920 [02:30<57:07, 1.68it/s, tok/s=5201.2]
Inference: 3%|▎ | 155/5920 [02:31<1:12:26, 1.33it/s, tok/s=5201.2]
Inference: 3%|▎ | 155/5920 [02:31<1:12:26, 1.33it/s, tok/s=5221.2]
Inference: 3%|▎ | 156/5920 [02:31<1:12:25, 1.33it/s, tok/s=5279.1]
Inference: 3%|▎ | 157/5920 [02:32<1:09:39, 1.38it/s, tok/s=5279.1]
Inference: 3%|▎ | 157/5920 [02:32<1:09:39, 1.38it/s, tok/s=5290.0]
Inference: 3%|▎ | 158/5920 [02:33<1:04:30, 1.49it/s, tok/s=5290.0]
Inference: 3%|▎ | 158/5920 [02:33<1:04:30, 1.49it/s, tok/s=5329.1]
Inference: 3%|▎ | 159/5920 [02:33<1:04:30, 1.49it/s, tok/s=5388.3]
Inference: 3%|▎ | 160/5920 [02:33<1:04:29, 1.49it/s, tok/s=5444.2]
Inference: 3%|▎ | 161/5920 [02:34<52:17, 1.84it/s, tok/s=5444.2]
Inference: 3%|▎ | 161/5920 [02:34<52:17, 1.84it/s, tok/s=5460.1]
Inference: 3%|▎ | 162/5920 [02:35<1:01:01, 1.57it/s, tok/s=5460.1]
Inference: 3%|▎ | 162/5920 [02:35<1:01:01, 1.57it/s, tok/s=5482.1]
Inference: 3%|▎ | 163/5920 [02:36<1:01:26, 1.56it/s, tok/s=5482.1]
Inference: 3%|▎ | 163/5920 [02:36<1:01:26, 1.56it/s, tok/s=5516.0]
Inference: 3%|▎ | 164/5920 [02:39<2:07:13, 1.33s/it, tok/s=5516.0]
Inference: 3%|▎ | 164/5920 [02:39<2:07:13, 1.33s/it, tok/s=5451.7]
Inference: 3%|▎ | 165/5920 [02:41<2:16:47, 1.43s/it, tok/s=5451.7]
Inference: 3%|▎ | 165/5920 [02:41<2:16:47, 1.43s/it, tok/s=5448.9]
Inference: 3%|▎ | 166/5920 [02:46<3:58:12, 2.48s/it, tok/s=5448.9]
Inference: 3%|▎ | 166/5920 [02:46<3:58:12, 2.48s/it, tok/s=5328.6]
Inference: 3%|▎ | 167/5920 [02:50<4:15:37, 2.67s/it, tok/s=5328.6]
Inference: 3%|▎ | 167/5920 [02:50<4:15:37, 2.67s/it, tok/s=5285.9]
Inference: 3%|▎ | 168/5920 [02:50<3:06:24, 1.94s/it, tok/s=5285.9]
Inference: 3%|▎ | 168/5920 [02:50<3:06:24, 1.94s/it, tok/s=5338.0]
Inference: 3%|▎ | 169/5920 [02:51<2:49:01, 1.76s/it, tok/s=5338.0]
Inference: 3%|▎ | 169/5920 [02:51<2:49:01, 1.76s/it, tok/s=5351.7]
Inference: 3%|▎ | 170/5920 [02:54<3:38:36, 2.28s/it, tok/s=5351.7]
Inference: 3%|▎ | 170/5920 [02:54<3:38:36, 2.28s/it, tok/s=5298.2]
Inference: 3%|▎ | 171/5920 [02:55<2:43:17, 1.70s/it, tok/s=5298.2]
Inference: 3%|▎ | 171/5920 [02:55<2:43:17, 1.70s/it, tok/s=5343.3]
Inference: 3%|▎ | 172/5920 [02:56<2:28:48, 1.55s/it, tok/s=5343.3]
Inference: 3%|▎ | 172/5920 [02:56<2:28:48, 1.55s/it, tok/s=5362.3]
Inference: 3%|▎ | 173/5920 [02:56<1:57:43, 1.23s/it, tok/s=5362.3]
Inference: 3%|▎ | 173/5920 [02:56<1:57:43, 1.23s/it, tok/s=5396.2]
Inference: 3%|▎ | 174/5920 [02:57<1:25:37, 1.12it/s, tok/s=5396.2]
Inference: 3%|▎ | 174/5920 [02:57<1:25:37, 1.12it/s, tok/s=5448.0]
Inference: 3%|▎ | 175/5920 [02:58<1:53:15, 1.18s/it, tok/s=5448.0]
Inference: 3%|▎ | 175/5920 [02:58<1:53:15, 1.18s/it, tok/s=5451.1]
Inference: 3%|▎ | 176/5920 [02:59<1:33:36, 1.02it/s, tok/s=5451.1]
Inference: 3%|▎ | 176/5920 [02:59<1:33:36, 1.02it/s, tok/s=5492.4]
Inference: 3%|▎ | 177/5920 [02:59<1:13:46, 1.30it/s, tok/s=5492.4]
Inference: 3%|▎ | 177/5920 [02:59<1:13:46, 1.30it/s, tok/s=5523.7]
Inference: 3%|▎ | 178/5920 [02:59<57:04, 1.68it/s, tok/s=5523.7]
Inference: 3%|▎ | 178/5920 [02:59<57:04, 1.68it/s, tok/s=5571.8]
Inference: 3%|▎ | 179/5920 [03:01<1:13:43, 1.30it/s, tok/s=5571.8]
Inference: 3%|▎ | 179/5920 [03:01<1:13:43, 1.30it/s, tok/s=5589.1]
Inference: 3%|▎ | 180/5920 [03:01<1:13:42, 1.30it/s, tok/s=5641.8]
Inference: 3%|▎ | 181/5920 [03:02<1:00:43, 1.58it/s, tok/s=5641.8]
Inference: 3%|▎ | 181/5920 [03:02<1:00:43, 1.58it/s, tok/s=5669.9]
Inference: 3%|▎ | 182/5920 [03:03<1:22:34, 1.16it/s, tok/s=5669.9]
Inference: 3%|▎ | 182/5920 [03:03<1:22:34, 1.16it/s, tok/s=5676.4]
Inference: 3%|▎ | 183/5920 [03:05<1:43:14, 1.08s/it, tok/s=5676.4]
Inference: 3%|▎ | 183/5920 [03:05<1:43:14, 1.08s/it, tok/s=5678.2]
Inference: 3%|▎ | 184/5920 [03:05<1:19:08, 1.21it/s, tok/s=5678.2]
Inference: 3%|▎ | 184/5920 [03:05<1:19:08, 1.21it/s, tok/s=5727.3]
Inference: 3%|▎ | 185/5920 [03:05<1:00:42, 1.57it/s, tok/s=5727.3]
Inference: 3%|▎ | 185/5920 [03:05<1:00:42, 1.57it/s, tok/s=5737.5]
Inference: 3%|▎ | 186/5920 [03:05<46:44, 2.04it/s, tok/s=5737.5]
Inference: 3%|▎ | 186/5920 [03:05<46:44, 2.04it/s, tok/s=5789.3]
Inference: 3%|▎ | 187/5920 [03:05<36:23, 2.63it/s, tok/s=5789.3]
Inference: 3%|▎ | 187/5920 [03:05<36:23, 2.63it/s, tok/s=5844.4]
Inference: 3%|▎ | 188/5920 [03:07<59:48, 1.60it/s, tok/s=5844.4]
Inference: 3%|▎ | 188/5920 [03:07<59:48, 1.60it/s, tok/s=5859.6]
Inference: 3%|▎ | 189/5920 [03:07<1:01:51, 1.54it/s, tok/s=5859.6]
Inference: 3%|▎ | 189/5920 [03:07<1:01:51, 1.54it/s, tok/s=5864.7]
Inference: 3%|▎ | 190/5920 [03:11<2:33:16, 1.60s/it, tok/s=5864.7]
Inference: 3%|▎ | 190/5920 [03:11<2:33:16, 1.60s/it, tok/s=5798.3]
Inference: 3%|▎ | 191/5920 [03:12<2:03:24, 1.29s/it, tok/s=5798.3]
Inference: 3%|▎ | 191/5920 [03:12<2:03:24, 1.29s/it, tok/s=5837.7]
Inference: 3%|▎ | 192/5920 [03:15<2:53:00, 1.81s/it, tok/s=5837.7]
Inference: 3%|▎ | 192/5920 [03:15<2:53:00, 1.81s/it, tok/s=5802.7]
Inference: 3%|▎ | 193/5920 [03:15<2:04:27, 1.30s/it, tok/s=5802.7]
Inference: 3%|▎ | 193/5920 [03:15<2:04:27, 1.30s/it, tok/s=5854.4]
Inference: 3%|▎ | 194/5920 [03:15<1:40:06, 1.05s/it, tok/s=5854.4]
Inference: 3%|▎ | 194/5920 [03:15<1:40:06, 1.05s/it, tok/s=5863.6]
Inference: 3%|▎ | 195/5920 [03:17<2:11:05, 1.37s/it, tok/s=5863.6]
Inference: 3%|▎ | 195/5920 [03:17<2:11:05, 1.37s/it, tok/s=5854.0]
Inference: 3%|▎ | 196/5920 [03:18<1:37:04, 1.02s/it, tok/s=5854.0]
Inference: 3%|▎ | 196/5920 [03:18<1:37:04, 1.02s/it, tok/s=5901.9]
Inference: 3%|▎ | 197/5920 [03:21<2:37:01, 1.65s/it, tok/s=5901.9]
Inference: 3%|▎ | 197/5920 [03:21<2:37:01, 1.65s/it, tok/s=5863.0]
Inference: 3%|▎ | 198/5920 [03:23<2:45:39, 1.74s/it, tok/s=5863.0]
Inference: 3%|▎ | 198/5920 [03:23<2:45:39, 1.74s/it, tok/s=5830.4]
Inference: 3%|▎ | 199/5920 [03:23<2:06:33, 1.33s/it, tok/s=5830.4]
Inference: 3%|▎ | 199/5920 [03:23<2:06:33, 1.33s/it, tok/s=5873.5]
Inference: 3%|▎ | 200/5920 [03:24<1:54:34, 1.20s/it, tok/s=5873.5]
Inference: 3%|▎ | 200/5920 [03:24<1:54:34, 1.20s/it, tok/s=5899.3]
Inference: 3%|▎ | 201/5920 [03:24<1:54:33, 1.20s/it, tok/s=5955.1]
Inference: 3%|▎ | 202/5920 [03:25<1:28:43, 1.07it/s, tok/s=5955.1]
Inference: 3%|▎ | 202/5920 [03:25<1:28:43, 1.07it/s, tok/s=5974.6]
Inference: 3%|▎ | 203/5920 [03:25<1:28:42, 1.07it/s, tok/s=6025.4]
Inference: 3%|▎ | 204/5920 [03:26<1:14:26, 1.28it/s, tok/s=6025.4]
Inference: 3%|▎ | 204/5920 [03:26<1:14:26, 1.28it/s, tok/s=6048.6]
Inference: 3%|▎ | 205/5920 [03:26<1:14:25, 1.28it/s, tok/s=6097.7]
Inference: 3%|▎ | 206/5920 [03:28<1:21:01, 1.18it/s, tok/s=6097.7]
Inference: 3%|▎ | 206/5920 [03:28<1:21:01, 1.18it/s, tok/s=6094.4]
Inference: 3%|▎ | 207/5920 [03:29<1:12:05, 1.32it/s, tok/s=6094.4]
Inference: 3%|▎ | 207/5920 [03:29<1:12:05, 1.32it/s, tok/s=6134.9]
Inference: 4%|▎ | 208/5920 [03:30<1:22:49, 1.15it/s, tok/s=6134.9]
Inference: 4%|▎ | 208/5920 [03:30<1:22:49, 1.15it/s, tok/s=6123.4]
Inference: 4%|▎ | 209/5920 [03:30<1:06:45, 1.43it/s, tok/s=6123.4]
Inference: 4%|▎ | 209/5920 [03:30<1:06:45, 1.43it/s, tok/s=6172.1]
Inference: 4%|▎ | 210/5920 [03:30<56:24, 1.69it/s, tok/s=6172.1]
Inference: 4%|▎ | 210/5920 [03:30<56:24, 1.69it/s, tok/s=6214.2]
Inference: 4%|▎ | 211/5920 [03:31<1:06:17, 1.44it/s, tok/s=6214.2]
Inference: 4%|▎ | 211/5920 [03:31<1:06:17, 1.44it/s, tok/s=6215.8]
Inference: 4%|▎ | 212/5920 [03:32<1:01:18, 1.55it/s, tok/s=6215.8]
Inference: 4%|▎ | 212/5920 [03:32<1:01:18, 1.55it/s, tok/s=6252.0]
Inference: 4%|▎ | 213/5920 [03:32<1:01:18, 1.55it/s, tok/s=6264.9]
Inference: 4%|▎ | 214/5920 [03:33<49:58, 1.90it/s, tok/s=6264.9]
Inference: 4%|▎ | 214/5920 [03:33<49:58, 1.90it/s, tok/s=6298.3]
Inference: 4%|▎ | 215/5920 [03:33<42:47, 2.22it/s, tok/s=6298.3]
Inference: 4%|▎ | 215/5920 [03:33<42:47, 2.22it/s, tok/s=6345.3]
Inference: 4%|▎ | 216/5920 [03:33<38:32, 2.47it/s, tok/s=6345.3]
Inference: 4%|▎ | 216/5920 [03:33<38:32, 2.47it/s, tok/s=6360.5]
Inference: 4%|▎ | 217/5920 [03:34<49:56, 1.90it/s, tok/s=6360.5]
Inference: 4%|▎ | 217/5920 [03:34<49:56, 1.90it/s, tok/s=6364.3]
Inference: 4%|▎ | 218/5920 [03:34<43:17, 2.20it/s, tok/s=6364.3]
Inference: 4%|▎ | 218/5920 [03:34<43:17, 2.20it/s, tok/s=6410.2]
Inference: 4%|▎ | 219/5920 [03:38<2:04:29, 1.31s/it, tok/s=6410.2]
Inference: 4%|▎ | 219/5920 [03:38<2:04:29, 1.31s/it, tok/s=6364.0]
Inference: 4%|▎ | 220/5920 [03:38<1:50:29, 1.16s/it, tok/s=6364.0]
Inference: 4%|▎ | 220/5920 [03:38<1:50:29, 1.16s/it, tok/s=6394.6]
Inference: 4%|▎ | 221/5920 [03:38<1:50:28, 1.16s/it, tok/s=6444.2]
Inference: 4%|▍ | 222/5920 [03:39<1:11:55, 1.32it/s, tok/s=6444.2]
Inference: 4%|▍ | 222/5920 [03:39<1:11:55, 1.32it/s, tok/s=6478.8]
Inference: 4%|▍ | 223/5920 [03:39<1:06:18, 1.43it/s, tok/s=6478.8]
Inference: 4%|▍ | 223/5920 [03:39<1:06:18, 1.43it/s, tok/s=6514.4]
Inference: 4%|▍ | 224/5920 [03:41<1:18:34, 1.21it/s, tok/s=6514.4]
Inference: 4%|▍ | 224/5920 [03:41<1:18:34, 1.21it/s, tok/s=6529.5]
Inference: 4%|▍ | 225/5920 [03:42<1:42:15, 1.08s/it, tok/s=6529.5]
Inference: 4%|▍ | 225/5920 [03:42<1:42:15, 1.08s/it, tok/s=6530.3]
Inference: 4%|▍ | 226/5920 [03:43<1:31:09, 1.04it/s, tok/s=6530.3]
Inference: 4%|▍ | 226/5920 [03:43<1:31:09, 1.04it/s, tok/s=6535.5]
Inference: 4%|▍ | 227/5920 [03:44<1:33:35, 1.01it/s, tok/s=6535.5]
Inference: 4%|▍ | 227/5920 [03:44<1:33:35, 1.01it/s, tok/s=6545.5]
Inference: 4%|▍ | 228/5920 [03:44<1:13:58, 1.28it/s, tok/s=6545.5]
Inference: 4%|▍ | 228/5920 [03:44<1:13:58, 1.28it/s, tok/s=6590.2]
Inference: 4%|▍ | 229/5920 [03:46<1:30:01, 1.05it/s, tok/s=6590.2]
Inference: 4%|▍ | 229/5920 [03:46<1:30:01, 1.05it/s, tok/s=6600.9]
Inference: 4%|▍ | 230/5920 [03:46<1:13:21, 1.29it/s, tok/s=6600.9]
Inference: 4%|▍ | 230/5920 [03:46<1:13:21, 1.29it/s, tok/s=6644.3]
Inference: 4%|▍ | 231/5920 [03:47<1:14:59, 1.26it/s, tok/s=6644.3]
Inference: 4%|▍ | 231/5920 [03:47<1:14:59, 1.26it/s, tok/s=6630.8]
Inference: 4%|▍ | 232/5920 [03:47<1:00:40, 1.56it/s, tok/s=6630.8]
Inference: 4%|▍ | 232/5920 [03:47<1:00:40, 1.56it/s, tok/s=6675.1]
Inference: 4%|▍ | 233/5920 [03:48<51:10, 1.85it/s, tok/s=6675.1]
Inference: 4%|▍ | 233/5920 [03:48<51:10, 1.85it/s, tok/s=6716.5]
Inference: 4%|▍ | 234/5920 [03:48<44:02, 2.15it/s, tok/s=6716.5]
Inference: 4%|▍ | 234/5920 [03:48<44:02, 2.15it/s, tok/s=6759.4]
Inference: 4%|▍ | 235/5920 [03:49<52:25, 1.81it/s, tok/s=6759.4]
Inference: 4%|▍ | 235/5920 [03:49<52:25, 1.81it/s, tok/s=6787.6]
Inference: 4%|▍ | 236/5920 [03:49<57:37, 1.64it/s, tok/s=6787.6]
Inference: 4%|▍ | 236/5920 [03:49<57:37, 1.64it/s, tok/s=6815.7]
Inference: 4%|▍ | 237/5920 [03:50<56:55, 1.66it/s, tok/s=6815.7]
Inference: 4%|▍ | 237/5920 [03:50<56:55, 1.66it/s, tok/s=6814.1]
Inference: 4%|▍ | 238/5920 [03:51<58:27, 1.62it/s, tok/s=6814.1]
Inference: 4%|▍ | 238/5920 [03:51<58:27, 1.62it/s, tok/s=6846.4]
Inference: 4%|▍ | 239/5920 [03:51<58:17, 1.62it/s, tok/s=6846.4]
Inference: 4%|▍ | 239/5920 [03:51<58:17, 1.62it/s, tok/s=6880.9]
Inference: 4%|▍ | 240/5920 [03:51<58:16, 1.62it/s, tok/s=6928.1]
Inference: 4%|▍ | 241/5920 [03:51<34:25, 2.75it/s, tok/s=6928.1]
Inference: 4%|▍ | 241/5920 [03:51<34:25, 2.75it/s, tok/s=6943.6]
Inference: 4%|▍ | 242/5920 [03:52<33:19, 2.84it/s, tok/s=6943.6]
Inference: 4%|▍ | 242/5920 [03:52<33:19, 2.84it/s, tok/s=6945.5]
Inference: 4%|▍ | 243/5920 [03:52<31:48, 2.97it/s, tok/s=6945.5]
Inference: 4%|▍ | 243/5920 [03:52<31:48, 2.97it/s, tok/s=6977.2]
Inference: 4%|▍ | 244/5920 [03:52<31:48, 2.97it/s, tok/s=7025.6]
Inference: 4%|▍ | 245/5920 [03:52<22:20, 4.23it/s, tok/s=7025.6]
Inference: 4%|▍ | 245/5920 [03:52<22:20, 4.23it/s, tok/s=7072.4]
Inference: 4%|▍ | 246/5920 [03:53<38:16, 2.47it/s, tok/s=7072.4]
Inference: 4%|▍ | 246/5920 [03:53<38:16, 2.47it/s, tok/s=7082.7]
Inference: 4%|▍ | 247/5920 [03:54<55:58, 1.69it/s, tok/s=7082.7]
Inference: 4%|▍ | 247/5920 [03:54<55:58, 1.69it/s, tok/s=7077.7]
Inference: 4%|▍ | 248/5920 [03:55<54:16, 1.74it/s, tok/s=7077.7]
Inference: 4%|▍ | 248/5920 [03:55<54:16, 1.74it/s, tok/s=7113.7]
Inference: 4%|▍ | 249/5920 [03:55<54:16, 1.74it/s, tok/s=7162.0]
Inference: 4%|▍ | 250/5920 [03:59<1:56:18, 1.23s/it, tok/s=7162.0]
Inference: 4%|▍ | 250/5920 [03:59<1:56:18, 1.23s/it, tok/s=7089.2]
Inference: 4%|▍ | 251/5920 [04:00<1:53:19, 1.20s/it, tok/s=7089.2]
Inference: 4%|▍ | 251/5920 [04:00<1:53:19, 1.20s/it, tok/s=7107.6]
Inference: 4%|▍ | 252/5920 [04:00<1:28:35, 1.07it/s, tok/s=7107.6]
Inference: 4%|▍ | 252/5920 [04:00<1:28:35, 1.07it/s, tok/s=7151.2]
Inference: 4%|▍ | 253/5920 [04:00<1:28:34, 1.07it/s, tok/s=7166.3]
Inference: 4%|▍ | 254/5920 [04:01<1:05:28, 1.44it/s, tok/s=7166.3]
Inference: 4%|▍ | 254/5920 [04:01<1:05:28, 1.44it/s, tok/s=7196.1]
Inference: 4%|▍ | 255/5920 [04:01<58:36, 1.61it/s, tok/s=7196.1]
Inference: 4%|▍ | 255/5920 [04:01<58:36, 1.61it/s, tok/s=7233.3]
Inference: 4%|▍ | 256/5920 [04:02<49:48, 1.90it/s, tok/s=7233.3]
Inference: 4%|▍ | 256/5920 [04:02<49:48, 1.90it/s, tok/s=7277.4]
Inference: 4%|▍ | 257/5920 [04:02<49:32, 1.90it/s, tok/s=7277.4]
Inference: 4%|▍ | 257/5920 [04:02<49:32, 1.90it/s, tok/s=7290.9]
Inference: 4%|▍ | 258/5920 [04:03<49:15, 1.92it/s, tok/s=7290.9]
Inference: 4%|▍ | 258/5920 [04:03<49:15, 1.92it/s, tok/s=7326.3]
Inference: 4%|▍ | 259/5920 [04:03<51:48, 1.82it/s, tok/s=7326.3]
Inference: 4%|▍ | 259/5920 [04:03<51:48, 1.82it/s, tok/s=7356.8]
Inference: 4%|▍ | 260/5920 [04:04<1:05:16, 1.45it/s, tok/s=7356.8]
Inference: 4%|▍ | 260/5920 [04:04<1:05:16, 1.45it/s, tok/s=7344.3]
Inference: 4%|▍ | 261/5920 [04:05<55:02, 1.71it/s, tok/s=7344.3]
Inference: 4%|▍ | 261/5920 [04:05<55:02, 1.71it/s, tok/s=7362.6]
Inference: 4%|▍ | 262/5920 [04:06<1:24:32, 1.12it/s, tok/s=7362.6]
Inference: 4%|▍ | 262/5920 [04:06<1:24:32, 1.12it/s, tok/s=7364.3]
Inference: 4%|▍ | 263/5920 [04:06<1:02:46, 1.50it/s, tok/s=7364.3]
Inference: 4%|▍ | 263/5920 [04:06<1:02:46, 1.50it/s, tok/s=7390.9]
Inference: 4%|▍ | 264/5920 [04:08<1:43:06, 1.09s/it, tok/s=7390.9]
Inference: 4%|▍ | 264/5920 [04:08<1:43:06, 1.09s/it, tok/s=7340.4]
Inference: 4%|▍ | 265/5920 [04:10<2:02:07, 1.30s/it, tok/s=7340.4]
Inference: 4%|▍ | 265/5920 [04:10<2:02:07, 1.30s/it, tok/s=7300.7]
Inference: 4%|▍ | 266/5920 [04:11<1:44:09, 1.11s/it, tok/s=7300.7]
Inference: 4%|▍ | 266/5920 [04:11<1:44:09, 1.11s/it, tok/s=7320.1]
Inference: 5%|▍ | 267/5920 [04:11<1:17:39, 1.21it/s, tok/s=7320.1]
Inference: 5%|▍ | 267/5920 [04:11<1:17:39, 1.21it/s, tok/s=7364.8]
Inference: 5%|▍ | 268/5920 [04:13<1:58:38, 1.26s/it, tok/s=7364.8]
Inference: 5%|▍ | 268/5920 [04:13<1:58:38, 1.26s/it, tok/s=7316.8]
Inference: 5%|▍ | 269/5920 [04:13<1:58:37, 1.26s/it, tok/s=7363.3]
Inference: 5%|▍ | 270/5920 [04:14<1:16:01, 1.24it/s, tok/s=7363.3]
Inference: 5%|▍ | 270/5920 [04:14<1:16:01, 1.24it/s, tok/s=7362.5]
Inference: 5%|▍ | 271/5920 [04:14<1:16:00, 1.24it/s, tok/s=7379.4]
Inference: 5%|▍ | 272/5920 [04:14<54:53, 1.71it/s, tok/s=7379.4]
Inference: 5%|▍ | 272/5920 [04:14<54:53, 1.71it/s, tok/s=7417.9]
Inference: 5%|▍ | 273/5920 [04:15<1:00:41, 1.55it/s, tok/s=7417.9]
Inference: 5%|▍ | 273/5920 [04:15<1:00:41, 1.55it/s, tok/s=7441.0]
Inference: 5%|▍ | 274/5920 [04:16<56:37, 1.66it/s, tok/s=7441.0]
Inference: 5%|▍ | 274/5920 [04:16<56:37, 1.66it/s, tok/s=7446.5]
Inference: 5%|▍ | 275/5920 [04:18<1:36:48, 1.03s/it, tok/s=7446.5]
Inference: 5%|▍ | 275/5920 [04:18<1:36:48, 1.03s/it, tok/s=7431.2]
Inference: 5%|▍ | 276/5920 [04:18<1:36:47, 1.03s/it, tok/s=7478.5]
Inference: 5%|▍ | 277/5920 [04:19<1:13:48, 1.27it/s, tok/s=7478.5]
Inference: 5%|▍ | 277/5920 [04:19<1:13:48, 1.27it/s, tok/s=7500.9]
Inference: 5%|▍ | 278/5920 [04:20<1:16:45, 1.23it/s, tok/s=7500.9]
Inference: 5%|▍ | 278/5920 [04:20<1:16:45, 1.23it/s, tok/s=7521.6]
Inference: 5%|▍ | 279/5920 [04:22<1:46:49, 1.14s/it, tok/s=7521.6]
Inference: 5%|▍ | 279/5920 [04:22<1:46:49, 1.14s/it, tok/s=7511.4]
Inference: 5%|▍ | 280/5920 [04:22<1:46:48, 1.14s/it, tok/s=7559.1]
Inference: 5%|▍ | 281/5920 [04:22<1:10:06, 1.34it/s, tok/s=7559.1]
Inference: 5%|▍ | 281/5920 [04:22<1:10:06, 1.34it/s, tok/s=7597.4]
Inference: 5%|▍ | 282/5920 [04:23<1:07:44, 1.39it/s, tok/s=7597.4]
Inference: 5%|▍ | 282/5920 [04:23<1:07:44, 1.39it/s, tok/s=7595.9]
Inference: 5%|▍ | 283/5920 [04:23<1:00:40, 1.55it/s, tok/s=7595.9]
Inference: 5%|▍ | 283/5920 [04:23<1:00:40, 1.55it/s, tok/s=7633.7]
Inference: 5%|▍ | 284/5920 [04:24<55:47, 1.68it/s, tok/s=7633.7]
Inference: 5%|▍ | 284/5920 [04:24<55:47, 1.68it/s, tok/s=7668.6]
Inference: 5%|▍ | 285/5920 [04:25<1:01:45, 1.52it/s, tok/s=7668.6]
Inference: 5%|▍ | 285/5920 [04:25<1:01:45, 1.52it/s, tok/s=7673.7]
Inference: 5%|▍ | 286/5920 [04:25<1:06:58, 1.40it/s, tok/s=7673.7]
Inference: 5%|▍ | 286/5920 [04:25<1:06:58, 1.40it/s, tok/s=7695.4]
Inference: 5%|▍ | 287/5920 [04:26<1:11:01, 1.32it/s, tok/s=7695.4]
Inference: 5%|▍ | 287/5920 [04:26<1:11:01, 1.32it/s, tok/s=7717.1]
Inference: 5%|▍ | 288/5920 [04:27<56:56, 1.65it/s, tok/s=7717.1]
Inference: 5%|▍ | 288/5920 [04:27<56:56, 1.65it/s, tok/s=7756.8]
Inference: 5%|▍ | 289/5920 [04:27<52:42, 1.78it/s, tok/s=7756.8]
Inference: 5%|▍ | 289/5920 [04:27<52:42, 1.78it/s, tok/s=7793.3]
Inference: 5%|▍ | 290/5920 [04:27<52:42, 1.78it/s, tok/s=7832.2]
Inference: 5%|▍ | 291/5920 [04:27<37:56, 2.47it/s, tok/s=7832.2]
Inference: 5%|▍ | 291/5920 [04:27<37:56, 2.47it/s, tok/s=7866.3]
Inference: 5%|▍ | 292/5920 [04:28<36:09, 2.59it/s, tok/s=7866.3]
Inference: 5%|▍ | 292/5920 [04:28<36:09, 2.59it/s, tok/s=7904.1]
Inference: 5%|▍ | 293/5920 [04:29<55:02, 1.70it/s, tok/s=7904.1]
Inference: 5%|▍ | 293/5920 [04:29<55:02, 1.70it/s, tok/s=7919.7]
Inference: 5%|▍ | 294/5920 [04:29<52:03, 1.80it/s, tok/s=7919.7]
Inference: 5%|▍ | 294/5920 [04:29<52:03, 1.80it/s, tok/s=7952.1]
Inference: 5%|▍ | 295/5920 [04:30<58:53, 1.59it/s, tok/s=7952.1]
Inference: 5%|▍ | 295/5920 [04:30<58:53, 1.59it/s, tok/s=7977.4]
Inference: 5%|▌ | 296/5920 [04:31<54:52, 1.71it/s, tok/s=7977.4]
Inference: 5%|▌ | 296/5920 [04:31<54:52, 1.71it/s, tok/s=8011.4]
Inference: 5%|▌ | 297/5920 [04:31<42:55, 2.18it/s, tok/s=8011.4]
Inference: 5%|▌ | 297/5920 [04:31<42:55, 2.18it/s, tok/s=8053.9]
Inference: 5%|▌ | 298/5920 [04:31<33:52, 2.77it/s, tok/s=8053.9]
Inference: 5%|▌ | 298/5920 [04:31<33:52, 2.77it/s, tok/s=8096.2]
Inference: 5%|▌ | 299/5920 [04:32<38:30, 2.43it/s, tok/s=8096.2]
Inference: 5%|▌ | 299/5920 [04:32<38:30, 2.43it/s, tok/s=8111.5]
Inference: 5%|▌ | 300/5920 [04:32<35:02, 2.67it/s, tok/s=8111.5]
Inference: 5%|▌ | 300/5920 [04:32<35:02, 2.67it/s, tok/s=8149.0]
Inference: 5%|▌ | 301/5920 [04:32<35:02, 2.67it/s, tok/s=8197.9]
Inference: 5%|▌ | 302/5920 [04:32<32:22, 2.89it/s, tok/s=8197.9]
Inference: 5%|▌ | 302/5920 [04:32<32:22, 2.89it/s, tok/s=8226.5]
Inference: 5%|▌ | 303/5920 [04:33<45:39, 2.05it/s, tok/s=8226.5]
Inference: 5%|▌ | 303/5920 [04:33<45:39, 2.05it/s, tok/s=8245.4]
Inference: 5%|▌ | 304/5920 [04:35<1:11:54, 1.30it/s, tok/s=8245.4]
Inference: 5%|▌ | 304/5920 [04:35<1:11:54, 1.30it/s, tok/s=8214.4]
Inference: 5%|▌ | 305/5920 [04:35<1:11:53, 1.30it/s, tok/s=8259.0]
Inference: 5%|▌ | 306/5920 [04:35<52:14, 1.79it/s, tok/s=8259.0]
Inference: 5%|▌ | 306/5920 [04:35<52:14, 1.79it/s, tok/s=8293.8]
Inference: 5%|▌ | 307/5920 [04:36<50:35, 1.85it/s, tok/s=8293.8]
Inference: 5%|▌ | 307/5920 [04:36<50:35, 1.85it/s, tok/s=8325.8]
Inference: 5%|▌ | 308/5920 [04:37<1:06:01, 1.42it/s, tok/s=8325.8]
Inference: 5%|▌ | 308/5920 [04:37<1:06:01, 1.42it/s, tok/s=8308.2]
Inference: 5%|▌ | 309/5920 [04:37<53:06, 1.76it/s, tok/s=8308.2]
Inference: 5%|▌ | 309/5920 [04:37<53:06, 1.76it/s, tok/s=8320.7]
Inference: 5%|▌ | 310/5920 [04:39<1:14:44, 1.25it/s, tok/s=8320.7]
Inference: 5%|▌ | 310/5920 [04:39<1:14:44, 1.25it/s, tok/s=8325.4]
Inference: 5%|▌ | 311/5920 [04:39<1:14:43, 1.25it/s, tok/s=8353.0]
Inference: 5%|▌ | 312/5920 [04:39<48:31, 1.93it/s, tok/s=8353.0]
Inference: 5%|▌ | 312/5920 [04:39<48:31, 1.93it/s, tok/s=8368.3]
Inference: 5%|▌ | 313/5920 [04:39<45:02, 2.07it/s, tok/s=8368.3]
Inference: 5%|▌ | 313/5920 [04:39<45:02, 2.07it/s, tok/s=8379.7]
Inference: 5%|▌ | 314/5920 [04:40<57:25, 1.63it/s, tok/s=8379.7]
Inference: 5%|▌ | 314/5920 [04:40<57:25, 1.63it/s, tok/s=8377.4]
Inference: 5%|▌ | 315/5920 [04:41<44:45, 2.09it/s, tok/s=8377.4]
Inference: 5%|▌ | 315/5920 [04:41<44:45, 2.09it/s, tok/s=8423.1]
Inference: 5%|▌ | 316/5920 [04:41<44:44, 2.09it/s, tok/s=8469.4]
Inference: 5%|▌ | 317/5920 [04:41<30:51, 3.03it/s, tok/s=8469.4]
Inference: 5%|▌ | 317/5920 [04:41<30:51, 3.03it/s, tok/s=8510.9]
Inference: 5%|▌ | 318/5920 [04:42<49:12, 1.90it/s, tok/s=8510.9]
Inference: 5%|▌ | 318/5920 [04:42<49:12, 1.90it/s, tok/s=8507.0]
Inference: 5%|▌ | 319/5920 [04:42<47:11, 1.98it/s, tok/s=8507.0]
Inference: 5%|▌ | 319/5920 [04:42<47:11, 1.98it/s, tok/s=8542.1]
Inference: 5%|▌ | 320/5920 [04:44<1:13:51, 1.26it/s, tok/s=8542.1]
Inference: 5%|▌ | 320/5920 [04:44<1:13:51, 1.26it/s, tok/s=8542.0]
Inference: 5%|▌ | 321/5920 [04:44<1:04:22, 1.45it/s, tok/s=8542.0]
Inference: 5%|▌ | 321/5920 [04:44<1:04:22, 1.45it/s, tok/s=8540.0]
Inference: 5%|▌ | 322/5920 [04:45<1:00:59, 1.53it/s, tok/s=8540.0]
Inference: 5%|▌ | 322/5920 [04:45<1:00:59, 1.53it/s, tok/s=8569.5]
Inference: 5%|▌ | 323/5920 [04:45<53:45, 1.74it/s, tok/s=8569.5]
Inference: 5%|▌ | 323/5920 [04:45<53:45, 1.74it/s, tok/s=8602.2]
Inference: 5%|▌ | 324/5920 [04:47<1:18:42, 1.19it/s, tok/s=8602.2]
Inference: 5%|▌ | 324/5920 [04:47<1:18:42, 1.19it/s, tok/s=8603.3]
Inference: 5%|▌ | 325/5920 [04:47<1:02:30, 1.49it/s, tok/s=8603.3]
Inference: 5%|▌ | 325/5920 [04:47<1:02:30, 1.49it/s, tok/s=8611.3]
Inference: 6%|▌ | 326/5920 [04:50<2:16:24, 1.46s/it, tok/s=8611.3]
Inference: 6%|▌ | 326/5920 [04:50<2:16:24, 1.46s/it, tok/s=8558.8]
Inference: 6%|▌ | 327/5920 [04:51<1:40:14, 1.08s/it, tok/s=8558.8]
Inference: 6%|▌ | 327/5920 [04:51<1:40:14, 1.08s/it, tok/s=8599.8]
Inference: 6%|▌ | 328/5920 [04:51<1:27:25, 1.07it/s, tok/s=8599.8]
Inference: 6%|▌ | 328/5920 [04:51<1:27:25, 1.07it/s, tok/s=8627.1]
Inference: 6%|▌ | 329/5920 [04:52<1:17:35, 1.20it/s, tok/s=8627.1]
Inference: 6%|▌ | 329/5920 [04:52<1:17:35, 1.20it/s, tok/s=8651.5]
Inference: 6%|▌ | 330/5920 [04:53<1:36:00, 1.03s/it, tok/s=8651.5]
Inference: 6%|▌ | 330/5920 [04:53<1:36:00, 1.03s/it, tok/s=8654.2]
Inference: 6%|▌ | 331/5920 [04:53<1:11:38, 1.30it/s, tok/s=8654.2]
Inference: 6%|▌ | 331/5920 [04:53<1:11:38, 1.30it/s, tok/s=8698.4]
Inference: 6%|▌ | 332/5920 [04:54<58:51, 1.58it/s, tok/s=8698.4]
Inference: 6%|▌ | 332/5920 [04:54<58:51, 1.58it/s, tok/s=8736.2]
Inference: 6%|▌ | 333/5920 [04:54<58:51, 1.58it/s, tok/s=8747.4]
Inference: 6%|▌ | 334/5920 [04:55<49:04, 1.90it/s, tok/s=8747.4]
Inference: 6%|▌ | 334/5920 [04:55<49:04, 1.90it/s, tok/s=8734.4]
Inference: 6%|▌ | 335/5920 [04:57<1:22:42, 1.13it/s, tok/s=8734.4]
Inference: 6%|▌ | 335/5920 [04:57<1:22:42, 1.13it/s, tok/s=8694.5]
Inference: 6%|▌ | 336/5920 [04:57<1:10:33, 1.32it/s, tok/s=8694.5]
Inference: 6%|▌ | 336/5920 [04:57<1:10:33, 1.32it/s, tok/s=8731.8]
Inference: 6%|▌ | 337/5920 [04:58<1:15:27, 1.23it/s, tok/s=8731.8]
Inference: 6%|▌ | 337/5920 [04:58<1:15:27, 1.23it/s, tok/s=8752.4]
Inference: 6%|▌ | 338/5920 [04:58<57:00, 1.63it/s, tok/s=8752.4]
Inference: 6%|▌ | 338/5920 [04:58<57:00, 1.63it/s, tok/s=8796.0]
Inference: 6%|▌ | 339/5920 [04:58<56:59, 1.63it/s, tok/s=8819.6]
Inference: 6%|▌ | 340/5920 [05:00<1:12:27, 1.28it/s, tok/s=8819.6]
Inference: 6%|▌ | 340/5920 [05:00<1:12:27, 1.28it/s, tok/s=8808.9]
Inference: 6%|▌ | 341/5920 [05:02<1:31:28, 1.02it/s, tok/s=8808.9]
Inference: 6%|▌ | 341/5920 [05:02<1:31:28, 1.02it/s, tok/s=8807.4]
Inference: 6%|▌ | 342/5920 [05:02<1:10:50, 1.31it/s, tok/s=8807.4]
Inference: 6%|▌ | 342/5920 [05:02<1:10:50, 1.31it/s, tok/s=8828.8]
Inference: 6%|▌ | 343/5920 [05:02<1:06:07, 1.41it/s, tok/s=8828.8]
Inference: 6%|▌ | 343/5920 [05:02<1:06:07, 1.41it/s, tok/s=8822.5]
Inference: 6%|▌ | 344/5920 [05:03<59:11, 1.57it/s, tok/s=8822.5]
Inference: 6%|▌ | 344/5920 [05:03<59:11, 1.57it/s, tok/s=8855.7]
Inference: 6%|▌ | 345/5920 [05:03<54:18, 1.71it/s, tok/s=8855.7]
Inference: 6%|▌ | 345/5920 [05:03<54:18, 1.71it/s, tok/s=8888.1]
Inference: 6%|▌ | 346/5920 [05:04<49:10, 1.89it/s, tok/s=8888.1]
Inference: 6%|▌ | 346/5920 [05:04<49:10, 1.89it/s, tok/s=8922.9]
Inference: 6%|▌ | 347/5920 [05:04<53:30, 1.74it/s, tok/s=8922.9]
Inference: 6%|▌ | 347/5920 [05:04<53:30, 1.74it/s, tok/s=8949.4]
Inference: 6%|▌ | 348/5920 [05:05<48:34, 1.91it/s, tok/s=8949.4]
Inference: 6%|▌ | 348/5920 [05:05<48:34, 1.91it/s, tok/s=8983.9]
Inference: 6%|▌ | 349/5920 [05:05<55:31, 1.67it/s, tok/s=8983.9]
Inference: 6%|▌ | 349/5920 [05:05<55:31, 1.67it/s, tok/s=9009.5]
Inference: 6%|▌ | 350/5920 [05:07<1:09:53, 1.33it/s, tok/s=9009.5]
Inference: 6%|▌ | 350/5920 [05:07<1:09:53, 1.33it/s, tok/s=8987.1]
Inference: 6%|▌ | 351/5920 [05:07<54:03, 1.72it/s, tok/s=8987.1]
Inference: 6%|▌ | 351/5920 [05:07<54:03, 1.72it/s, tok/s=8994.3]
Inference: 6%|▌ | 352/5920 [05:09<1:28:48, 1.04it/s, tok/s=8994.3]
Inference: 6%|▌ | 352/5920 [05:09<1:28:48, 1.04it/s, tok/s=8988.0]
Inference: 6%|▌ | 353/5920 [05:09<1:11:16, 1.30it/s, tok/s=8988.0]
Inference: 6%|▌ | 353/5920 [05:09<1:11:16, 1.30it/s, tok/s=9010.3]
Inference: 6%|▌ | 354/5920 [05:10<1:06:41, 1.39it/s, tok/s=9010.3]
Inference: 6%|▌ | 354/5920 [05:10<1:06:41, 1.39it/s, tok/s=9038.6]
Inference: 6%|▌ | 355/5920 [05:10<1:09:35, 1.33it/s, tok/s=9038.6]
Inference: 6%|▌ | 355/5920 [05:10<1:09:35, 1.33it/s, tok/s=9059.2]
Inference: 6%|▌ | 356/5920 [05:11<1:04:37, 1.43it/s, tok/s=9059.2]
Inference: 6%|▌ | 356/5920 [05:11<1:04:37, 1.43it/s, tok/s=9089.4]
Inference: 6%|▌ | 357/5920 [05:11<49:59, 1.85it/s, tok/s=9089.4]
Inference: 6%|▌ | 357/5920 [05:11<49:59, 1.85it/s, tok/s=9130.3]
Inference: 6%|▌ | 358/5920 [05:12<48:42, 1.90it/s, tok/s=9130.3]
Inference: 6%|▌ | 358/5920 [05:12<48:42, 1.90it/s, tok/s=9139.9]
Inference: 6%|▌ | 359/5920 [05:12<48:42, 1.90it/s, tok/s=9183.8]
Inference: 6%|▌ | 360/5920 [05:12<30:05, 3.08it/s, tok/s=9183.8]
Inference: 6%|▌ | 360/5920 [05:12<30:05, 3.08it/s, tok/s=9226.5]
Inference: 6%|▌ | 361/5920 [05:12<38:19, 2.42it/s, tok/s=9226.5]
Inference: 6%|▌ | 361/5920 [05:12<38:19, 2.42it/s, tok/s=9221.2]
Inference: 6%|▌ | 362/5920 [05:13<38:34, 2.40it/s, tok/s=9221.2]
Inference: 6%|▌ | 362/5920 [05:13<38:34, 2.40it/s, tok/s=9253.1]
Inference: 6%|▌ | 363/5920 [05:13<40:53, 2.26it/s, tok/s=9253.1]
Inference: 6%|▌ | 363/5920 [05:13<40:53, 2.26it/s, tok/s=9283.6]
Inference: 6%|▌ | 364/5920 [05:14<47:51, 1.94it/s, tok/s=9283.6]
Inference: 6%|▌ | 364/5920 [05:14<47:51, 1.94it/s, tok/s=9299.0]
Inference: 6%|▌ | 365/5920 [05:14<47:50, 1.94it/s, tok/s=9347.1]
Inference: 6%|▌ | 366/5920 [05:15<36:09, 2.56it/s, tok/s=9347.1]
Inference: 6%|▌ | 366/5920 [05:15<36:09, 2.56it/s, tok/s=9353.0]
Inference: 6%|▌ | 367/5920 [05:15<30:01, 3.08it/s, tok/s=9353.0]
Inference: 6%|▌ | 367/5920 [05:15<30:01, 3.08it/s, tok/s=9372.0]
Inference: 6%|▌ | 368/5920 [05:15<25:51, 3.58it/s, tok/s=9372.0]
Inference: 6%|▌ | 368/5920 [05:15<25:51, 3.58it/s, tok/s=9414.6]
Inference: 6%|▌ | 369/5920 [05:15<25:46, 3.59it/s, tok/s=9414.6]
Inference: 6%|▌ | 369/5920 [05:15<25:46, 3.59it/s, tok/s=9452.9]
Inference: 6%|▋ | 370/5920 [05:16<30:28, 3.04it/s, tok/s=9452.9]
Inference: 6%|▋ | 370/5920 [05:16<30:28, 3.04it/s, tok/s=9486.1]
Inference: 6%|▋ | 371/5920 [05:17<49:55, 1.85it/s, tok/s=9486.1]
Inference: 6%|▋ | 371/5920 [05:17<49:55, 1.85it/s, tok/s=9465.2]
Inference: 6%|▋ | 372/5920 [05:19<1:47:58, 1.17s/it, tok/s=9465.2]
Inference: 6%|▋ | 372/5920 [05:19<1:47:58, 1.17s/it, tok/s=9432.0]
Inference: 6%|▋ | 373/5920 [05:21<2:07:20, 1.38s/it, tok/s=9432.0]
Inference: 6%|▋ | 373/5920 [05:21<2:07:20, 1.38s/it, tok/s=9420.5]
Inference: 6%|▋ | 374/5920 [05:22<1:45:52, 1.15s/it, tok/s=9420.5]
Inference: 6%|▋ | 374/5920 [05:22<1:45:52, 1.15s/it, tok/s=9446.1]
Inference: 6%|▋ | 375/5920 [05:22<1:45:51, 1.15s/it, tok/s=9454.3]
Inference: 6%|▋ | 376/5920 [05:23<1:27:28, 1.06it/s, tok/s=9454.3]
Inference: 6%|▋ | 376/5920 [05:23<1:27:28, 1.06it/s, tok/s=9460.3]
Inference: 6%|▋ | 377/5920 [05:24<1:16:30, 1.21it/s, tok/s=9460.3]
Inference: 6%|▋ | 377/5920 [05:24<1:16:30, 1.21it/s, tok/s=9494.8]
Inference: 6%|▋ | 378/5920 [05:24<1:00:10, 1.54it/s, tok/s=9494.8]
Inference: 6%|▋ | 378/5920 [05:24<1:00:10, 1.54it/s, tok/s=9536.0]
Inference: 6%|▋ | 379/5920 [05:25<1:04:25, 1.43it/s, tok/s=9536.0]
Inference: 6%|▋ | 379/5920 [05:25<1:04:25, 1.43it/s, tok/s=9555.7]
Inference: 6%|▋ | 380/5920 [05:25<1:02:53, 1.47it/s, tok/s=9555.7]
Inference: 6%|▋ | 380/5920 [05:25<1:02:53, 1.47it/s, tok/s=9581.8]
Inference: 6%|▋ | 381/5920 [05:26<56:53, 1.62it/s, tok/s=9581.8]
Inference: 6%|▋ | 381/5920 [05:26<56:53, 1.62it/s, tok/s=9612.8]
Inference: 6%|▋ | 382/5920 [05:26<44:41, 2.07it/s, tok/s=9612.8]
Inference: 6%|▋ | 382/5920 [05:26<44:41, 2.07it/s, tok/s=9634.7]
Inference: 6%|▋ | 383/5920 [05:27<1:02:30, 1.48it/s, tok/s=9634.7]
Inference: 6%|▋ | 383/5920 [05:27<1:02:30, 1.48it/s, tok/s=9646.1]
Inference: 6%|▋ | 384/5920 [05:28<1:01:06, 1.51it/s, tok/s=9646.1]
Inference: 6%|▋ | 384/5920 [05:28<1:01:06, 1.51it/s, tok/s=9653.7]
Inference: 7%|▋ | 385/5920 [05:28<55:37, 1.66it/s, tok/s=9653.7]
Inference: 7%|▋ | 385/5920 [05:28<55:37, 1.66it/s, tok/s=9687.4]
Inference: 7%|▋ | 386/5920 [05:28<43:28, 2.12it/s, tok/s=9687.4]
Inference: 7%|▋ | 386/5920 [05:28<43:28, 2.12it/s, tok/s=9695.3]
Inference: 7%|▋ | 387/5920 [05:29<39:11, 2.35it/s, tok/s=9695.3]
Inference: 7%|▋ | 387/5920 [05:29<39:11, 2.35it/s, tok/s=9719.1]
Inference: 7%|▋ | 388/5920 [05:29<42:12, 2.18it/s, tok/s=9719.1]
Inference: 7%|▋ | 388/5920 [05:29<42:12, 2.18it/s, tok/s=9749.6]
Inference: 7%|▋ | 389/5920 [05:31<1:19:36, 1.16it/s, tok/s=9749.6]
Inference: 7%|▋ | 389/5920 [05:31<1:19:36, 1.16it/s, tok/s=9740.4]
Inference: 7%|▋ | 390/5920 [05:31<59:37, 1.55it/s, tok/s=9740.4]
Inference: 7%|▋ | 390/5920 [05:31<59:37, 1.55it/s, tok/s=9781.1]
Inference: 7%|▋ | 391/5920 [05:32<53:37, 1.72it/s, tok/s=9781.1]
Inference: 7%|▋ | 391/5920 [05:32<53:37, 1.72it/s, tok/s=9791.1]
Inference: 7%|▋ | 392/5920 [05:33<1:12:09, 1.28it/s, tok/s=9791.1]
Inference: 7%|▋ | 392/5920 [05:33<1:12:09, 1.28it/s, tok/s=9783.2]
Inference: 7%|▋ | 393/5920 [05:33<1:12:08, 1.28it/s, tok/s=9826.7]
Inference: 7%|▋ | 394/5920 [05:33<47:42, 1.93it/s, tok/s=9826.7]
Inference: 7%|▋ | 394/5920 [05:33<47:42, 1.93it/s, tok/s=9860.0]
Inference: 7%|▋ | 395/5920 [05:34<44:26, 2.07it/s, tok/s=9860.0]
Inference: 7%|▋ | 395/5920 [05:34<44:26, 2.07it/s, tok/s=9892.7]
Inference: 7%|▋ | 396/5920 [05:35<1:03:11, 1.46it/s, tok/s=9892.7]
Inference: 7%|▋ | 396/5920 [05:35<1:03:11, 1.46it/s, tok/s=9898.9]
Inference: 7%|▋ | 397/5920 [05:35<54:45, 1.68it/s, tok/s=9898.9]
Inference: 7%|▋ | 397/5920 [05:35<54:45, 1.68it/s, tok/s=9936.1]
Inference: 7%|▋ | 398/5920 [05:35<54:44, 1.68it/s, tok/s=9978.3]
Inference: 7%|▋ | 399/5920 [05:36<39:54, 2.31it/s, tok/s=9978.3]
Inference: 7%|▋ | 399/5920 [05:36<39:54, 2.31it/s, tok/s=9984.9]
Inference: 7%|▋ | 400/5920 [05:36<39:16, 2.34it/s, tok/s=9984.9]
Inference: 7%|▋ | 400/5920 [05:36<39:16, 2.34it/s, tok/s=10016.8]
Inference: 7%|▋ | 401/5920 [05:37<45:29, 2.02it/s, tok/s=10016.8]
Inference: 7%|▋ | 401/5920 [05:37<45:29, 2.02it/s, tok/s=10039.7]
Inference: 7%|▋ | 402/5920 [05:37<37:23, 2.46it/s, tok/s=10039.7]
Inference: 7%|▋ | 402/5920 [05:37<37:23, 2.46it/s, tok/s=10079.9]
Inference: 7%|▋ | 403/5920 [05:38<44:42, 2.06it/s, tok/s=10079.9]
Inference: 7%|▋ | 403/5920 [05:38<44:42, 2.06it/s, tok/s=10102.8]
Inference: 7%|▋ | 404/5920 [05:38<44:12, 2.08it/s, tok/s=10102.8]
Inference: 7%|▋ | 404/5920 [05:38<44:12, 2.08it/s, tok/s=10132.4]
Inference: 7%|▋ | 405/5920 [05:39<47:12, 1.95it/s, tok/s=10132.4]
Inference: 7%|▋ | 405/5920 [05:39<47:12, 1.95it/s, tok/s=10158.1]
Inference: 7%|▋ | 406/5920 [05:39<38:49, 2.37it/s, tok/s=10158.1]
Inference: 7%|▋ | 406/5920 [05:39<38:49, 2.37it/s, tok/s=10195.7]
Inference: 7%|▋ | 407/5920 [05:39<35:39, 2.58it/s, tok/s=10195.7]
Inference: 7%|▋ | 407/5920 [05:39<35:39, 2.58it/s, tok/s=10233.3]
Inference: 7%|▋ | 408/5920 [05:40<33:55, 2.71it/s, tok/s=10233.3]
Inference: 7%|▋ | 408/5920 [05:40<33:55, 2.71it/s, tok/s=10267.3]
Inference: 7%|▋ | 409/5920 [05:40<32:27, 2.83it/s, tok/s=10267.3]
Inference: 7%|▋ | 409/5920 [05:40<32:27, 2.83it/s, tok/s=10301.7]
Inference: 7%|▋ | 410/5920 [05:41<43:08, 2.13it/s, tok/s=10301.7]
Inference: 7%|▋ | 410/5920 [05:41<43:08, 2.13it/s, tok/s=10322.7]
Inference: 7%|▋ | 411/5920 [05:41<33:24, 2.75it/s, tok/s=10322.7]
Inference: 7%|▋ | 411/5920 [05:41<33:24, 2.75it/s, tok/s=10365.7]
Inference: 7%|▋ | 412/5920 [05:41<30:44, 2.99it/s, tok/s=10365.7]
Inference: 7%|▋ | 412/5920 [05:41<30:44, 2.99it/s, tok/s=10377.0]
Inference: 7%|▋ | 413/5920 [05:42<43:57, 2.09it/s, tok/s=10377.0]
Inference: 7%|▋ | 413/5920 [05:42<43:57, 2.09it/s, tok/s=10396.5]
Inference: 7%|▋ | 414/5920 [05:43<1:06:43, 1.38it/s, tok/s=10396.5]
Inference: 7%|▋ | 414/5920 [05:43<1:06:43, 1.38it/s, tok/s=10388.9]
Inference: 7%|▋ | 415/5920 [05:44<1:01:40, 1.49it/s, tok/s=10388.9]
Inference: 7%|▋ | 415/5920 [05:44<1:01:40, 1.49it/s, tok/s=10385.3]
Inference: 7%|▋ | 416/5920 [05:44<45:57, 2.00it/s, tok/s=10385.3]
Inference: 7%|▋ | 416/5920 [05:44<45:57, 2.00it/s, tok/s=10409.5]
Inference: 7%|▋ | 417/5920 [05:45<56:08, 1.63it/s, tok/s=10409.5]
Inference: 7%|▋ | 417/5920 [05:45<56:08, 1.63it/s, tok/s=10430.2]
Inference: 7%|▋ | 418/5920 [05:45<42:11, 2.17it/s, tok/s=10430.2]
Inference: 7%|▋ | 418/5920 [05:45<42:11, 2.17it/s, tok/s=10438.6]
Inference: 7%|▋ | 419/5920 [05:46<1:03:31, 1.44it/s, tok/s=10438.6]
Inference: 7%|▋ | 419/5920 [05:46<1:03:31, 1.44it/s, tok/s=10414.4]
Inference: 7%|▋ | 420/5920 [05:47<1:10:46, 1.30it/s, tok/s=10414.4]
Inference: 7%|▋ | 420/5920 [05:47<1:10:46, 1.30it/s, tok/s=10428.8]
Inference: 7%|▋ | 421/5920 [05:47<54:12, 1.69it/s, tok/s=10428.8]
Inference: 7%|▋ | 421/5920 [05:47<54:12, 1.69it/s, tok/s=10469.2]
Inference: 7%|▋ | 422/5920 [05:47<54:11, 1.69it/s, tok/s=10476.7]
Inference: 7%|▋ | 423/5920 [05:48<40:50, 2.24it/s, tok/s=10476.7]
Inference: 7%|▋ | 423/5920 [05:48<40:50, 2.24it/s, tok/s=10489.9]
Inference: 7%|▋ | 424/5920 [05:48<33:35, 2.73it/s, tok/s=10489.9]
Inference: 7%|▋ | 424/5920 [05:48<33:35, 2.73it/s, tok/s=10532.6]
Inference: 7%|▋ | 425/5920 [05:48<29:03, 3.15it/s, tok/s=10532.6]
Inference: 7%|▋ | 425/5920 [05:48<29:03, 3.15it/s, tok/s=10538.5]
Inference: 7%|▋ | 426/5920 [05:48<23:58, 3.82it/s, tok/s=10538.5]
Inference: 7%|▋ | 426/5920 [05:48<23:58, 3.82it/s, tok/s=10579.3]
Inference: 7%|▋ | 427/5920 [05:48<23:58, 3.82it/s, tok/s=10622.2]
Inference: 7%|▋ | 428/5920 [05:49<24:12, 3.78it/s, tok/s=10622.2]
Inference: 7%|▋ | 428/5920 [05:49<24:12, 3.78it/s, tok/s=10653.7]
Inference: 7%|▋ | 429/5920 [05:49<24:15, 3.77it/s, tok/s=10653.7]
Inference: 7%|▋ | 429/5920 [05:49<24:15, 3.77it/s, tok/s=10684.4]
Inference: 7%|▋ | 430/5920 [05:49<24:14, 3.77it/s, tok/s=10726.2]
Inference: 7%|▋ | 431/5920 [05:49<22:46, 4.02it/s, tok/s=10726.2]
Inference: 7%|▋ | 431/5920 [05:49<22:46, 4.02it/s, tok/s=10761.4]
Inference: 7%|▋ | 432/5920 [05:50<22:19, 4.10it/s, tok/s=10761.4]
Inference: 7%|▋ | 432/5920 [05:50<22:19, 4.10it/s, tok/s=10792.9]
Inference: 7%|▋ | 433/5920 [05:50<20:21, 4.49it/s, tok/s=10792.9]
Inference: 7%|▋ | 433/5920 [05:50<20:21, 4.49it/s, tok/s=10812.7]
Inference: 7%|▋ | 434/5920 [05:51<35:16, 2.59it/s, tok/s=10812.7]
Inference: 7%|▋ | 434/5920 [05:51<35:16, 2.59it/s, tok/s=10830.8]
Inference: 7%|▋ | 435/5920 [05:51<35:16, 2.59it/s, tok/s=10865.2]
Inference: 7%|▋ | 436/5920 [05:51<29:01, 3.15it/s, tok/s=10865.2]
Inference: 7%|▋ | 436/5920 [05:51<29:01, 3.15it/s, tok/s=10895.7]
Inference: 7%|▋ | 437/5920 [05:51<29:00, 3.15it/s, tok/s=10910.8]
Inference: 7%|▋ | 438/5920 [05:51<22:09, 4.12it/s, tok/s=10910.8]
Inference: 7%|▋ | 438/5920 [05:51<22:09, 4.12it/s, tok/s=10947.4]
Inference: 7%|▋ | 439/5920 [05:51<22:09, 4.12it/s, tok/s=10993.2]
Inference: 7%|▋ | 440/5920 [05:52<19:48, 4.61it/s, tok/s=10993.2]
Inference: 7%|▋ | 440/5920 [05:52<19:48, 4.61it/s, tok/s=10997.7]
Inference: 7%|▋ | 441/5920 [05:52<23:45, 3.84it/s, tok/s=10997.7]
Inference: 7%|▋ | 441/5920 [05:52<23:45, 3.84it/s, tok/s=11030.4]
Inference: 7%|▋ | 442/5920 [05:53<43:49, 2.08it/s, tok/s=11030.4]
Inference: 7%|▋ | 442/5920 [05:53<43:49, 2.08it/s, tok/s=11036.0]
Inference: 7%|▋ | 443/5920 [05:53<37:51, 2.41it/s, tok/s=11036.0]
Inference: 7%|▋ | 443/5920 [05:53<37:51, 2.41it/s, tok/s=11049.5]
Inference: 8%|▊ | 444/5920 [05:54<37:51, 2.41it/s, tok/s=11075.4]
Inference: 8%|▊ | 445/5920 [05:54<28:20, 3.22it/s, tok/s=11075.4]
Inference: 8%|▊ | 445/5920 [05:54<28:20, 3.22it/s, tok/s=11081.5]
Inference: 8%|▊ | 446/5920 [05:54<26:55, 3.39it/s, tok/s=11081.5]
Inference: 8%|▊ | 446/5920 [05:54<26:55, 3.39it/s, tok/s=11089.4]
Inference: 8%|▊ | 447/5920 [05:54<28:08, 3.24it/s, tok/s=11089.4]
Inference: 8%|▊ | 447/5920 [05:54<28:08, 3.24it/s, tok/s=11117.3]
Inference: 8%|▊ | 448/5920 [05:55<42:58, 2.12it/s, tok/s=11117.3]
Inference: 8%|▊ | 448/5920 [05:55<42:58, 2.12it/s, tok/s=11132.5]
Inference: 8%|▊ | 449/5920 [05:56<37:42, 2.42it/s, tok/s=11132.5]
Inference: 8%|▊ | 449/5920 [05:56<37:42, 2.42it/s, tok/s=11137.0]
Inference: 8%|▊ | 450/5920 [05:56<47:46, 1.91it/s, tok/s=11137.0]
Inference: 8%|▊ | 450/5920 [05:56<47:46, 1.91it/s, tok/s=11157.6]
Inference: 8%|▊ | 451/5920 [05:57<42:48, 2.13it/s, tok/s=11157.6]
Inference: 8%|▊ | 451/5920 [05:57<42:48, 2.13it/s, tok/s=11161.6]
Inference: 8%|▊ | 452/5920 [05:57<47:13, 1.93it/s, tok/s=11161.6]
Inference: 8%|▊ | 452/5920 [05:57<47:13, 1.93it/s, tok/s=11185.9]
Inference: 8%|▊ | 453/5920 [05:58<46:00, 1.98it/s, tok/s=11185.9]
Inference: 8%|▊ | 453/5920 [05:58<46:00, 1.98it/s, tok/s=11184.9]
Inference: 8%|▊ | 454/5920 [05:59<53:57, 1.69it/s, tok/s=11184.9]
Inference: 8%|▊ | 454/5920 [05:59<53:57, 1.69it/s, tok/s=11169.1]
Inference: 8%|▊ | 455/5920 [06:00<1:13:22, 1.24it/s, tok/s=11169.1]
Inference: 8%|▊ | 455/5920 [06:00<1:13:22, 1.24it/s, tok/s=11168.0]
Inference: 8%|▊ | 456/5920 [06:01<1:26:55, 1.05it/s, tok/s=11168.0]
Inference: 8%|▊ | 456/5920 [06:01<1:26:55, 1.05it/s, tok/s=11157.6]
Inference: 8%|▊ | 457/5920 [06:01<1:26:55, 1.05it/s, tok/s=11164.7]
Inference: 8%|▊ | 458/5920 [06:01<50:46, 1.79it/s, tok/s=11164.7]
Inference: 8%|▊ | 458/5920 [06:01<50:46, 1.79it/s, tok/s=11177.6]
Inference: 8%|▊ | 459/5920 [06:03<1:02:02, 1.47it/s, tok/s=11177.6]
Inference: 8%|▊ | 459/5920 [06:03<1:02:02, 1.47it/s, tok/s=11189.1]
Inference: 8%|▊ | 460/5920 [06:03<1:08:06, 1.34it/s, tok/s=11189.1]
Inference: 8%|▊ | 460/5920 [06:03<1:08:06, 1.34it/s, tok/s=11183.1]
Inference: 8%|▊ | 461/5920 [06:04<56:55, 1.60it/s, tok/s=11183.1]
Inference: 8%|▊ | 461/5920 [06:04<56:55, 1.60it/s, tok/s=11219.1]
Inference: 8%|▊ | 462/5920 [06:06<1:26:07, 1.06it/s, tok/s=11219.1]
Inference: 8%|▊ | 462/5920 [06:06<1:26:07, 1.06it/s, tok/s=11185.1]
Inference: 8%|▊ | 463/5920 [06:06<1:14:43, 1.22it/s, tok/s=11185.1]
Inference: 8%|▊ | 463/5920 [06:06<1:14:43, 1.22it/s, tok/s=11212.9]
Inference: 8%|▊ | 464/5920 [06:07<1:12:05, 1.26it/s, tok/s=11212.9]
Inference: 8%|▊ | 464/5920 [06:07<1:12:05, 1.26it/s, tok/s=11234.8]
Inference: 8%|▊ | 465/5920 [06:07<56:31, 1.61it/s, tok/s=11234.8]
Inference: 8%|▊ | 465/5920 [06:07<56:31, 1.61it/s, tok/s=11272.2]
Inference: 8%|▊ | 466/5920 [06:07<56:30, 1.61it/s, tok/s=11288.4]
Inference: 8%|▊ | 467/5920 [06:07<35:42, 2.54it/s, tok/s=11288.4]
Inference: 8%|▊ | 467/5920 [06:07<35:42, 2.54it/s, tok/s=11295.6]
Inference: 8%|▊ | 468/5920 [06:07<30:01, 3.03it/s, tok/s=11295.6]
Inference: 8%|▊ | 468/5920 [06:07<30:01, 3.03it/s, tok/s=11334.3]
Inference: 8%|▊ | 469/5920 [06:07<30:01, 3.03it/s, tok/s=11376.4]
Inference: 8%|▊ | 470/5920 [06:08<27:08, 3.35it/s, tok/s=11376.4]
Inference: 8%|▊ | 470/5920 [06:08<27:08, 3.35it/s, tok/s=11408.5]
Inference: 8%|▊ | 471/5920 [06:09<36:15, 2.50it/s, tok/s=11408.5]
Inference: 8%|▊ | 471/5920 [06:09<36:15, 2.50it/s, tok/s=11428.6]
Inference: 8%|▊ | 472/5920 [06:09<44:17, 2.05it/s, tok/s=11428.6]
Inference: 8%|▊ | 472/5920 [06:09<44:17, 2.05it/s, tok/s=11450.1]
Inference: 8%|▊ | 473/5920 [06:11<1:09:02, 1.31it/s, tok/s=11450.1]
Inference: 8%|▊ | 473/5920 [06:11<1:09:02, 1.31it/s, tok/s=11428.0]
Inference: 8%|▊ | 474/5920 [06:11<56:43, 1.60it/s, tok/s=11428.0]
Inference: 8%|▊ | 474/5920 [06:11<56:43, 1.60it/s, tok/s=11445.3]
Inference: 8%|▊ | 475/5920 [06:12<58:32, 1.55it/s, tok/s=11445.3]
Inference: 8%|▊ | 475/5920 [06:12<58:32, 1.55it/s, tok/s=11441.2]
Inference: 8%|▊ | 476/5920 [06:13<1:05:27, 1.39it/s, tok/s=11441.2]
Inference: 8%|▊ | 476/5920 [06:13<1:05:27, 1.39it/s, tok/s=11437.2]
Inference: 8%|▊ | 477/5920 [06:14<1:26:43, 1.05it/s, tok/s=11437.2]
Inference: 8%|▊ | 477/5920 [06:14<1:26:43, 1.05it/s, tok/s=11433.5]
Inference: 8%|▊ | 478/5920 [06:15<1:09:59, 1.30it/s, tok/s=11433.5]
Inference: 8%|▊ | 478/5920 [06:15<1:09:59, 1.30it/s, tok/s=11467.3]
Inference: 8%|▊ | 479/5920 [06:15<1:01:36, 1.47it/s, tok/s=11467.3]
Inference: 8%|▊ | 479/5920 [06:15<1:01:36, 1.47it/s, tok/s=11491.1]
Inference: 8%|▊ | 480/5920 [06:16<1:06:47, 1.36it/s, tok/s=11491.1]
Inference: 8%|▊ | 480/5920 [06:16<1:06:47, 1.36it/s, tok/s=11506.9]
Inference: 8%|▊ | 481/5920 [06:17<1:12:34, 1.25it/s, tok/s=11506.9]
Inference: 8%|▊ | 481/5920 [06:17<1:12:34, 1.25it/s, tok/s=11509.8]
Inference: 8%|▊ | 482/5920 [06:18<1:25:10, 1.06it/s, tok/s=11509.8]
Inference: 8%|▊ | 482/5920 [06:18<1:25:10, 1.06it/s, tok/s=11491.7]
Inference: 8%|▊ | 483/5920 [06:19<1:14:40, 1.21it/s, tok/s=11491.7]
Inference: 8%|▊ | 483/5920 [06:19<1:14:40, 1.21it/s, tok/s=11518.8]
Inference: 8%|▊ | 484/5920 [06:19<1:00:22, 1.50it/s, tok/s=11518.8]
Inference: 8%|▊ | 484/5920 [06:19<1:00:22, 1.50it/s, tok/s=11553.2]
Inference: 8%|▊ | 485/5920 [06:19<53:37, 1.69it/s, tok/s=11553.2]
Inference: 8%|▊ | 485/5920 [06:19<53:37, 1.69it/s, tok/s=11569.3]
Inference: 8%|▊ | 486/5920 [06:20<41:42, 2.17it/s, tok/s=11569.3]
Inference: 8%|▊ | 486/5920 [06:20<41:42, 2.17it/s, tok/s=11608.2]
Inference: 8%|▊ | 487/5920 [06:20<41:42, 2.17it/s, tok/s=11640.8]
Inference: 8%|▊ | 488/5920 [06:20<27:27, 3.30it/s, tok/s=11640.8]
Inference: 8%|▊ | 488/5920 [06:20<27:27, 3.30it/s, tok/s=11681.0]
Inference: 8%|▊ | 489/5920 [06:21<51:41, 1.75it/s, tok/s=11681.0]
Inference: 8%|▊ | 489/5920 [06:21<51:41, 1.75it/s, tok/s=11650.2]
Inference: 8%|▊ | 490/5920 [06:21<41:25, 2.19it/s, tok/s=11650.2]
Inference: 8%|▊ | 490/5920 [06:21<41:25, 2.19it/s, tok/s=11659.5]
Inference: 8%|▊ | 491/5920 [06:22<46:26, 1.95it/s, tok/s=11659.5]
Inference: 8%|▊ | 491/5920 [06:22<46:26, 1.95it/s, tok/s=11682.3]
Inference: 8%|▊ | 492/5920 [06:22<38:22, 2.36it/s, tok/s=11682.3]
Inference: 8%|▊ | 492/5920 [06:22<38:22, 2.36it/s, tok/s=11691.5]
Inference: 8%|▊ | 493/5920 [06:22<33:14, 2.72it/s, tok/s=11691.5]
Inference: 8%|▊ | 493/5920 [06:22<33:14, 2.72it/s, tok/s=11726.4]
Inference: 8%|▊ | 494/5920 [06:23<30:22, 2.98it/s, tok/s=11726.4]
Inference: 8%|▊ | 494/5920 [06:23<30:22, 2.98it/s, tok/s=11760.6]
Inference: 8%|▊ | 495/5920 [06:23<33:06, 2.73it/s, tok/s=11760.6]
Inference: 8%|▊ | 495/5920 [06:23<33:06, 2.73it/s, tok/s=11759.9]
Inference: 8%|▊ | 496/5920 [06:24<48:07, 1.88it/s, tok/s=11759.9]
Inference: 8%|▊ | 496/5920 [06:24<48:07, 1.88it/s, tok/s=11777.2]
Inference: 8%|▊ | 497/5920 [06:24<36:35, 2.47it/s, tok/s=11777.2]
Inference: 8%|▊ | 497/5920 [06:24<36:35, 2.47it/s, tok/s=11817.1]
Inference: 8%|▊ | 498/5920 [06:24<30:07, 3.00it/s, tok/s=11817.1]
Inference: 8%|▊ | 498/5920 [06:24<30:07, 3.00it/s, tok/s=11824.5]
Inference: 8%|▊ | 499/5920 [06:24<25:29, 3.54it/s, tok/s=11824.5]
Inference: 8%|▊ | 499/5920 [06:24<25:29, 3.54it/s, tok/s=11861.7]
Inference: 8%|▊ | 500/5920 [06:26<53:09, 1.70it/s, tok/s=11861.7]
Inference: 8%|▊ | 500/5920 [06:26<53:09, 1.70it/s, tok/s=11864.6]
Inference: 8%|▊ | 501/5920 [06:26<49:57, 1.81it/s, tok/s=11864.6]
Inference: 8%|▊ | 501/5920 [06:26<49:57, 1.81it/s, tok/s=11871.5]
Inference: 8%|▊ | 502/5920 [06:27<53:08, 1.70it/s, tok/s=11871.5]
Inference: 8%|▊ | 502/5920 [06:27<53:08, 1.70it/s, tok/s=11895.7]
Inference: 8%|▊ | 503/5920 [06:27<53:07, 1.70it/s, tok/s=11910.9]
Inference: 9%|▊ | 504/5920 [06:27<53:07, 1.70it/s, tok/s=11955.9]
Inference: 9%|▊ | 505/5920 [06:28<33:08, 2.72it/s, tok/s=11955.9]
Inference: 9%|▊ | 505/5920 [06:28<33:08, 2.72it/s, tok/s=11981.4]
Inference: 9%|▊ | 506/5920 [06:28<35:54, 2.51it/s, tok/s=11981.4]
Inference: 9%|▊ | 506/5920 [06:28<35:54, 2.51it/s, tok/s=12008.7]
Inference: 9%|▊ | 507/5920 [06:29<43:56, 2.05it/s, tok/s=12008.7]
Inference: 9%|▊ | 507/5920 [06:29<43:56, 2.05it/s, tok/s=12027.5]
Inference: 9%|▊ | 508/5920 [06:30<57:10, 1.58it/s, tok/s=12027.5]
Inference: 9%|▊ | 508/5920 [06:30<57:10, 1.58it/s, tok/s=12017.9]
Inference: 9%|▊ | 509/5920 [06:30<48:18, 1.87it/s, tok/s=12017.9]
Inference: 9%|▊ | 509/5920 [06:30<48:18, 1.87it/s, tok/s=12022.0]
Inference: 9%|▊ | 510/5920 [06:30<38:16, 2.36it/s, tok/s=12022.0]
Inference: 9%|▊ | 510/5920 [06:30<38:16, 2.36it/s, tok/s=12033.8]
Inference: 9%|▊ | 511/5920 [06:31<48:50, 1.85it/s, tok/s=12033.8]
Inference: 9%|▊ | 511/5920 [06:31<48:50, 1.85it/s, tok/s=12053.2]
Inference: 9%|▊ | 512/5920 [06:32<48:56, 1.84it/s, tok/s=12053.2]
Inference: 9%|▊ | 512/5920 [06:32<48:56, 1.84it/s, tok/s=12059.8]
Inference: 9%|▊ | 513/5920 [06:32<39:36, 2.28it/s, tok/s=12059.8]
Inference: 9%|▊ | 513/5920 [06:32<39:36, 2.28it/s, tok/s=12097.9]
Inference: 9%|▊ | 514/5920 [06:33<1:09:10, 1.30it/s, tok/s=12097.9]
Inference: 9%|▊ | 514/5920 [06:33<1:09:10, 1.30it/s, tok/s=12063.5]
Inference: 9%|▊ | 515/5920 [06:36<1:48:54, 1.21s/it, tok/s=12063.5]
Inference: 9%|▊ | 515/5920 [06:36<1:48:54, 1.21s/it, tok/s=12040.0]
Inference: 9%|▊ | 516/5920 [06:38<2:09:31, 1.44s/it, tok/s=12040.0]
Inference: 9%|▊ | 516/5920 [06:38<2:09:31, 1.44s/it, tok/s=12022.1]
Inference: 9%|▊ | 517/5920 [06:39<2:01:27, 1.35s/it, tok/s=12022.1]
Inference: 9%|▊ | 517/5920 [06:39<2:01:27, 1.35s/it, tok/s=12033.2]
Inference: 9%|▉ | 518/5920 [06:45<4:18:07, 2.87s/it, tok/s=12033.2]
Inference: 9%|▉ | 518/5920 [06:45<4:18:07, 2.87s/it, tok/s=11883.9]
Inference: 9%|▉ | 519/5920 [06:45<4:18:04, 2.87s/it, tok/s=11924.2]
Inference: 9%|▉ | 520/5920 [06:46<2:25:51, 1.62s/it, tok/s=11924.2]
Inference: 9%|▉ | 520/5920 [06:46<2:25:51, 1.62s/it, tok/s=11953.0]
Inference: 9%|▉ | 521/5920 [06:46<2:06:04, 1.40s/it, tok/s=11953.0]
Inference: 9%|▉ | 521/5920 [06:46<2:06:04, 1.40s/it, tok/s=11971.9]
Inference: 9%|▉ | 522/5920 [06:46<2:06:03, 1.40s/it, tok/s=12012.8]
Inference: 9%|▉ | 523/5920 [06:47<1:26:03, 1.05it/s, tok/s=12012.8]
Inference: 9%|▉ | 523/5920 [06:47<1:26:03, 1.05it/s, tok/s=12007.1]
Inference: 9%|▉ | 524/5920 [06:47<1:13:24, 1.23it/s, tok/s=12007.1]
Inference: 9%|▉ | 524/5920 [06:47<1:13:24, 1.23it/s, tok/s=12041.9]
Inference: 9%|▉ | 525/5920 [06:47<1:00:10, 1.49it/s, tok/s=12041.9]
Inference: 9%|▉ | 525/5920 [06:47<1:00:10, 1.49it/s, tok/s=12042.2]
Inference: 9%|▉ | 526/5920 [06:48<48:12, 1.86it/s, tok/s=12042.2]
Inference: 9%|▉ | 526/5920 [06:48<48:12, 1.86it/s, tok/s=12050.7]
Inference: 9%|▉ | 527/5920 [06:48<48:12, 1.86it/s, tok/s=12063.9]
Inference: 9%|▉ | 528/5920 [06:48<30:28, 2.95it/s, tok/s=12063.9]
Inference: 9%|▉ | 528/5920 [06:48<30:28, 2.95it/s, tok/s=12104.9]
Inference: 9%|▉ | 529/5920 [06:49<41:14, 2.18it/s, tok/s=12104.9]
Inference: 9%|▉ | 529/5920 [06:49<41:14, 2.18it/s, tok/s=12124.4]
Inference: 9%|▉ | 530/5920 [06:49<41:23, 2.17it/s, tok/s=12124.4]
Inference: 9%|▉ | 530/5920 [06:49<41:23, 2.17it/s, tok/s=12132.8]
Inference: 9%|▉ | 531/5920 [06:49<41:23, 2.17it/s, tok/s=12170.6]
Inference: 9%|▉ | 532/5920 [06:49<30:27, 2.95it/s, tok/s=12170.6]
Inference: 9%|▉ | 532/5920 [06:49<30:27, 2.95it/s, tok/s=12205.7]
Inference: 9%|▉ | 533/5920 [06:50<26:28, 3.39it/s, tok/s=12205.7]
Inference: 9%|▉ | 533/5920 [06:50<26:28, 3.39it/s, tok/s=12218.4]
Inference: 9%|▉ | 534/5920 [06:50<23:55, 3.75it/s, tok/s=12218.4]
Inference: 9%|▉ | 534/5920 [06:50<23:55, 3.75it/s, tok/s=12232.2]
Inference: 9%|▉ | 535/5920 [06:50<21:29, 4.18it/s, tok/s=12232.2]
Inference: 9%|▉ | 535/5920 [06:50<21:29, 4.18it/s, tok/s=12269.1]
Inference: 9%|▉ | 536/5920 [06:53<1:19:46, 1.12it/s, tok/s=12269.1]
Inference: 9%|▉ | 536/5920 [06:53<1:19:46, 1.12it/s, tok/s=12208.4]
Inference: 9%|▉ | 537/5920 [06:53<1:04:54, 1.38it/s, tok/s=12208.4]
Inference: 9%|▉ | 537/5920 [06:53<1:04:54, 1.38it/s, tok/s=12240.5]
Inference: 9%|▉ | 538/5920 [06:54<1:07:30, 1.33it/s, tok/s=12240.5]
Inference: 9%|▉ | 538/5920 [06:54<1:07:30, 1.33it/s, tok/s=12258.7]
Inference: 9%|▉ | 539/5920 [06:54<55:14, 1.62it/s, tok/s=12258.7]
Inference: 9%|▉ | 539/5920 [06:54<55:14, 1.62it/s, tok/s=12291.9]
Inference: 9%|▉ | 540/5920 [06:54<55:14, 1.62it/s, tok/s=12334.9]
Inference: 9%|▉ | 541/5920 [06:55<47:06, 1.90it/s, tok/s=12334.9]
Inference: 9%|▉ | 541/5920 [06:55<47:06, 1.90it/s, tok/s=12340.7]
Inference: 9%|▉ | 542/5920 [06:56<54:16, 1.65it/s, tok/s=12340.7]
Inference: 9%|▉ | 542/5920 [06:56<54:16, 1.65it/s, tok/s=12321.8]
Inference: 9%|▉ | 543/5920 [06:56<43:42, 2.05it/s, tok/s=12321.8]
Inference: 9%|▉ | 543/5920 [06:56<43:42, 2.05it/s, tok/s=12323.8]
Inference: 9%|▉ | 544/5920 [06:56<36:35, 2.45it/s, tok/s=12323.8]
Inference: 9%|▉ | 544/5920 [06:56<36:35, 2.45it/s, tok/s=12362.1]
Inference: 9%|▉ | 545/5920 [06:56<36:35, 2.45it/s, tok/s=12366.1]
Inference: 9%|▉ | 546/5920 [06:57<36:34, 2.45it/s, tok/s=12366.1]
Inference: 9%|▉ | 546/5920 [06:57<36:34, 2.45it/s, tok/s=12351.9]
Inference: 9%|▉ | 547/5920 [06:57<37:14, 2.40it/s, tok/s=12351.9]
Inference: 9%|▉ | 547/5920 [06:57<37:14, 2.40it/s, tok/s=12378.9]
Inference: 9%|▉ | 548/5920 [06:57<37:14, 2.40it/s, tok/s=12420.8]
Inference: 9%|▉ | 549/5920 [06:58<39:41, 2.26it/s, tok/s=12420.8]
Inference: 9%|▉ | 549/5920 [06:58<39:41, 2.26it/s, tok/s=12424.6]
Inference: 9%|▉ | 550/5920 [06:59<36:45, 2.43it/s, tok/s=12424.6]
Inference: 9%|▉ | 550/5920 [06:59<36:45, 2.43it/s, tok/s=12458.1]
Inference: 9%|▉ | 551/5920 [07:00<1:08:27, 1.31it/s, tok/s=12458.1]
Inference: 9%|▉ | 551/5920 [07:00<1:08:27, 1.31it/s, tok/s=12432.3]
Inference: 9%|▉ | 552/5920 [07:01<1:09:35, 1.29it/s, tok/s=12432.3]
Inference: 9%|▉ | 552/5920 [07:01<1:09:35, 1.29it/s, tok/s=12451.5]
Inference: 9%|▉ | 553/5920 [07:02<1:08:49, 1.30it/s, tok/s=12451.5]
Inference: 9%|▉ | 553/5920 [07:02<1:08:49, 1.30it/s, tok/s=12436.7]
Inference: 9%|▉ | 554/5920 [07:02<1:02:07, 1.44it/s, tok/s=12436.7]
Inference: 9%|▉ | 554/5920 [07:02<1:02:07, 1.44it/s, tok/s=12465.9]
Inference: 9%|▉ | 555/5920 [07:03<47:27, 1.88it/s, tok/s=12465.9]
Inference: 9%|▉ | 555/5920 [07:03<47:27, 1.88it/s, tok/s=12504.0]
Inference: 9%|▉ | 556/5920 [07:03<47:08, 1.90it/s, tok/s=12504.0]
Inference: 9%|▉ | 556/5920 [07:03<47:08, 1.90it/s, tok/s=12508.8]
Inference: 9%|▉ | 557/5920 [07:03<42:18, 2.11it/s, tok/s=12508.8]
Inference: 9%|▉ | 557/5920 [07:03<42:18, 2.11it/s, tok/s=12506.3]
Inference: 9%|▉ | 558/5920 [07:04<40:06, 2.23it/s, tok/s=12506.3]
Inference: 9%|▉ | 558/5920 [07:04<40:06, 2.23it/s, tok/s=12508.8]
Inference: 9%|▉ | 559/5920 [07:04<45:07, 1.98it/s, tok/s=12508.8]
Inference: 9%|▉ | 559/5920 [07:04<45:07, 1.98it/s, tok/s=12534.4]
Inference: 9%|▉ | 560/5920 [07:05<35:39, 2.51it/s, tok/s=12534.4]
Inference: 9%|▉ | 560/5920 [07:05<35:39, 2.51it/s, tok/s=12536.7]
Inference: 9%|▉ | 561/5920 [07:05<44:22, 2.01it/s, tok/s=12536.7]
Inference: 9%|▉ | 561/5920 [07:05<44:22, 2.01it/s, tok/s=12559.7]
Inference: 9%|▉ | 562/5920 [07:07<1:03:09, 1.41it/s, tok/s=12559.7]
Inference: 9%|▉ | 562/5920 [07:07<1:03:09, 1.41it/s, tok/s=12567.6]
Inference: 10%|▉ | 563/5920 [07:08<1:27:37, 1.02it/s, tok/s=12567.6]
Inference: 10%|▉ | 563/5920 [07:08<1:27:37, 1.02it/s, tok/s=12543.9]
Inference: 10%|▉ | 564/5920 [07:08<1:08:16, 1.31it/s, tok/s=12543.9]
Inference: 10%|▉ | 564/5920 [07:08<1:08:16, 1.31it/s, tok/s=12543.9]
Inference: 10%|▉ | 565/5920 [07:09<59:34, 1.50it/s, tok/s=12543.9]
Inference: 10%|▉ | 565/5920 [07:09<59:34, 1.50it/s, tok/s=12575.0]
Inference: 10%|▉ | 566/5920 [07:09<59:33, 1.50it/s, tok/s=12615.6]
Inference: 10%|▉ | 567/5920 [07:09<40:48, 2.19it/s, tok/s=12615.6]
Inference: 10%|▉ | 567/5920 [07:09<40:48, 2.19it/s, tok/s=12647.0]
Inference: 10%|▉ | 568/5920 [07:10<37:16, 2.39it/s, tok/s=12647.0]
Inference: 10%|▉ | 568/5920 [07:10<37:16, 2.39it/s, tok/s=12682.1]
Inference: 10%|▉ | 569/5920 [07:11<54:29, 1.64it/s, tok/s=12682.1]
Inference: 10%|▉ | 569/5920 [07:11<54:29, 1.64it/s, tok/s=12655.2]
Inference: 10%|▉ | 570/5920 [07:11<42:10, 2.11it/s, tok/s=12655.2]
Inference: 10%|▉ | 570/5920 [07:11<42:10, 2.11it/s, tok/s=12694.3]
Inference: 10%|▉ | 571/5920 [07:12<52:21, 1.70it/s, tok/s=12694.3]
Inference: 10%|▉ | 571/5920 [07:12<52:21, 1.70it/s, tok/s=12676.2]
Inference: 10%|▉ | 572/5920 [07:12<52:20, 1.70it/s, tok/s=12711.0]
Inference: 10%|▉ | 573/5920 [07:12<36:23, 2.45it/s, tok/s=12711.0]
Inference: 10%|▉ | 573/5920 [07:12<36:23, 2.45it/s, tok/s=12708.3]
Inference: 10%|▉ | 574/5920 [07:12<36:17, 2.46it/s, tok/s=12708.3]
Inference: 10%|▉ | 574/5920 [07:12<36:17, 2.46it/s, tok/s=12703.6]
Inference: 10%|▉ | 575/5920 [07:13<33:42, 2.64it/s, tok/s=12703.6]
Inference: 10%|▉ | 575/5920 [07:13<33:42, 2.64it/s, tok/s=12709.3]
Inference: 10%|▉ | 576/5920 [07:13<34:13, 2.60it/s, tok/s=12709.3]
Inference: 10%|▉ | 576/5920 [07:13<34:13, 2.60it/s, tok/s=12705.4]
Inference: 10%|▉ | 577/5920 [07:14<36:12, 2.46it/s, tok/s=12705.4]
Inference: 10%|▉ | 577/5920 [07:14<36:12, 2.46it/s, tok/s=12698.7]
Inference: 10%|▉ | 578/5920 [07:14<28:41, 3.10it/s, tok/s=12698.7]
Inference: 10%|▉ | 578/5920 [07:14<28:41, 3.10it/s, tok/s=12722.1]
Inference: 10%|▉ | 579/5920 [07:15<45:56, 1.94it/s, tok/s=12722.1]
Inference: 10%|▉ | 579/5920 [07:15<45:56, 1.94it/s, tok/s=12700.3]
Inference: 10%|▉ | 580/5920 [07:15<40:40, 2.19it/s, tok/s=12700.3]
Inference: 10%|▉ | 580/5920 [07:15<40:40, 2.19it/s, tok/s=12700.1]
Inference: 10%|▉ | 581/5920 [07:15<36:41, 2.43it/s, tok/s=12700.1]
Inference: 10%|▉ | 581/5920 [07:15<36:41, 2.43it/s, tok/s=12733.4]
Inference: 10%|▉ | 582/5920 [07:15<28:54, 3.08it/s, tok/s=12733.4]
Inference: 10%|▉ | 582/5920 [07:15<28:54, 3.08it/s, tok/s=12737.9]
Inference: 10%|▉ | 583/5920 [07:16<25:17, 3.52it/s, tok/s=12737.9]
Inference: 10%|▉ | 583/5920 [07:16<25:17, 3.52it/s, tok/s=12740.9]
Inference: 10%|▉ | 584/5920 [07:17<53:49, 1.65it/s, tok/s=12740.9]
Inference: 10%|▉ | 584/5920 [07:17<53:49, 1.65it/s, tok/s=12709.0]
Inference: 10%|▉ | 585/5920 [07:17<53:48, 1.65it/s, tok/s=12715.7]
Inference: 10%|▉ | 586/5920 [07:17<35:42, 2.49it/s, tok/s=12715.7]
Inference: 10%|▉ | 586/5920 [07:17<35:42, 2.49it/s, tok/s=12720.5]
Inference: 10%|▉ | 587/5920 [07:18<34:37, 2.57it/s, tok/s=12720.5]
Inference: 10%|▉ | 587/5920 [07:18<34:37, 2.57it/s, tok/s=12726.1]
Inference: 10%|▉ | 588/5920 [07:19<44:17, 2.01it/s, tok/s=12726.1]
Inference: 10%|▉ | 588/5920 [07:19<44:17, 2.01it/s, tok/s=12711.2]
Inference: 10%|▉ | 589/5920 [07:19<44:16, 2.01it/s, tok/s=12753.8]
Inference: 10%|▉ | 590/5920 [07:19<44:16, 2.01it/s, tok/s=12794.7]
Inference: 10%|▉ | 591/5920 [07:19<25:48, 3.44it/s, tok/s=12794.7]
Inference: 10%|▉ | 591/5920 [07:19<25:48, 3.44it/s, tok/s=12828.5]
Inference: 10%|█ | 592/5920 [07:20<42:18, 2.10it/s, tok/s=12828.5]
Inference: 10%|█ | 592/5920 [07:20<42:18, 2.10it/s, tok/s=12802.2]
Inference: 10%|█ | 593/5920 [07:20<41:16, 2.15it/s, tok/s=12802.2]
Inference: 10%|█ | 593/5920 [07:20<41:16, 2.15it/s, tok/s=12831.3]
Inference: 10%|█ | 594/5920 [07:23<1:15:30, 1.18it/s, tok/s=12831.3]
Inference: 10%|█ | 594/5920 [07:23<1:15:30, 1.18it/s, tok/s=12780.3]
Inference: 10%|█ | 595/5920 [07:23<1:10:50, 1.25it/s, tok/s=12780.3]
Inference: 10%|█ | 595/5920 [07:23<1:10:50, 1.25it/s, tok/s=12797.6]
Inference: 10%|█ | 596/5920 [07:24<59:42, 1.49it/s, tok/s=12797.6]
Inference: 10%|█ | 596/5920 [07:24<59:42, 1.49it/s, tok/s=12829.7]
Inference: 10%|█ | 597/5920 [07:25<1:14:40, 1.19it/s, tok/s=12829.7]
Inference: 10%|█ | 597/5920 [07:25<1:14:40, 1.19it/s, tok/s=12811.5]
Inference: 10%|█ | 598/5920 [07:25<58:44, 1.51it/s, tok/s=12811.5]
Inference: 10%|█ | 598/5920 [07:25<58:44, 1.51it/s, tok/s=12848.6]
Inference: 10%|█ | 599/5920 [07:25<51:27, 1.72it/s, tok/s=12848.6]
Inference: 10%|█ | 599/5920 [07:25<51:27, 1.72it/s, tok/s=12853.3]
Inference: 10%|█ | 600/5920 [07:26<1:03:11, 1.40it/s, tok/s=12853.3]
Inference: 10%|█ | 600/5920 [07:26<1:03:11, 1.40it/s, tok/s=12842.3]
Inference: 10%|█ | 601/5920 [07:27<57:41, 1.54it/s, tok/s=12842.3]
Inference: 10%|█ | 601/5920 [07:27<57:41, 1.54it/s, tok/s=12838.8]
Inference: 10%|█ | 602/5920 [07:27<57:40, 1.54it/s, tok/s=12870.9]
Inference: 10%|█ | 603/5920 [07:27<41:08, 2.15it/s, tok/s=12870.9]
Inference: 10%|█ | 603/5920 [07:27<41:08, 2.15it/s, tok/s=12900.8]
Inference: 10%|█ | 604/5920 [07:28<33:51, 2.62it/s, tok/s=12900.8]
Inference: 10%|█ | 604/5920 [07:28<33:51, 2.62it/s, tok/s=12941.7]
Inference: 10%|█ | 605/5920 [07:28<33:51, 2.62it/s, tok/s=12948.3]
Inference: 10%|█ | 606/5920 [07:28<24:59, 3.54it/s, tok/s=12948.3]
Inference: 10%|█ | 606/5920 [07:28<24:59, 3.54it/s, tok/s=12951.2]
Inference: 10%|█ | 607/5920 [07:29<37:09, 2.38it/s, tok/s=12951.2]
Inference: 10%|█ | 607/5920 [07:29<37:09, 2.38it/s, tok/s=12945.4]
Inference: 10%|█ | 608/5920 [07:29<37:09, 2.38it/s, tok/s=12967.6]
Inference: 10%|█ | 609/5920 [07:30<42:15, 2.09it/s, tok/s=12967.6]
Inference: 10%|█ | 609/5920 [07:30<42:15, 2.09it/s, tok/s=12972.0]
Inference: 10%|█ | 610/5920 [07:31<49:01, 1.81it/s, tok/s=12972.0]
Inference: 10%|█ | 610/5920 [07:31<49:01, 1.81it/s, tok/s=12988.8]
Inference: 10%|█ | 611/5920 [07:31<45:25, 1.95it/s, tok/s=12988.8]
Inference: 10%|█ | 611/5920 [07:31<45:25, 1.95it/s, tok/s=13019.9]
Inference: 10%|█ | 612/5920 [07:31<39:59, 2.21it/s, tok/s=13019.9]
Inference: 10%|█ | 612/5920 [07:31<39:59, 2.21it/s, tok/s=13054.2]
Inference: 10%|█ | 613/5920 [07:32<44:40, 1.98it/s, tok/s=13054.2]
Inference: 10%|█ | 613/5920 [07:32<44:40, 1.98it/s, tok/s=13071.7]
Inference: 10%|█ | 614/5920 [07:32<36:52, 2.40it/s, tok/s=13071.7]
Inference: 10%|█ | 614/5920 [07:32<36:52, 2.40it/s, tok/s=13110.2]
Inference: 10%|█ | 615/5920 [07:33<48:40, 1.82it/s, tok/s=13110.2]
Inference: 10%|█ | 615/5920 [07:33<48:40, 1.82it/s, tok/s=13101.6]
Inference: 10%|█ | 616/5920 [07:34<48:53, 1.81it/s, tok/s=13101.6]
Inference: 10%|█ | 616/5920 [07:34<48:53, 1.81it/s, tok/s=13094.6]
Inference: 10%|█ | 617/5920 [07:34<45:38, 1.94it/s, tok/s=13094.6]
Inference: 10%|█ | 617/5920 [07:34<45:38, 1.94it/s, tok/s=13104.8]
Inference: 10%|█ | 618/5920 [07:34<35:00, 2.52it/s, tok/s=13104.8]
Inference: 10%|█ | 618/5920 [07:34<35:00, 2.52it/s, tok/s=13132.1]
Inference: 10%|█ | 619/5920 [07:35<36:55, 2.39it/s, tok/s=13132.1]
Inference: 10%|█ | 619/5920 [07:35<36:55, 2.39it/s, tok/s=13160.0]
Inference: 10%|█ | 620/5920 [07:36<49:12, 1.79it/s, tok/s=13160.0]
Inference: 10%|█ | 620/5920 [07:36<49:12, 1.79it/s, tok/s=13144.2]
Inference: 10%|█ | 621/5920 [07:36<49:02, 1.80it/s, tok/s=13144.2]
Inference: 10%|█ | 621/5920 [07:36<49:02, 1.80it/s, tok/s=13147.9]
Inference: 11%|█ | 622/5920 [07:37<1:01:31, 1.44it/s, tok/s=13147.9]
Inference: 11%|█ | 622/5920 [07:37<1:01:31, 1.44it/s, tok/s=13162.3]
Inference: 11%|█ | 623/5920 [07:38<59:48, 1.48it/s, tok/s=13162.3]
Inference: 11%|█ | 623/5920 [07:38<59:48, 1.48it/s, tok/s=13187.3]
Inference: 11%|█ | 624/5920 [07:39<1:04:02, 1.38it/s, tok/s=13187.3]
Inference: 11%|█ | 624/5920 [07:39<1:04:02, 1.38it/s, tok/s=13183.9]
Inference: 11%|█ | 625/5920 [07:39<59:57, 1.47it/s, tok/s=13183.9]
Inference: 11%|█ | 625/5920 [07:39<59:57, 1.47it/s, tok/s=13187.6]
Inference: 11%|█ | 626/5920 [07:41<1:24:25, 1.05it/s, tok/s=13187.6]
Inference: 11%|█ | 626/5920 [07:41<1:24:25, 1.05it/s, tok/s=13151.5]
Inference: 11%|█ | 627/5920 [07:41<1:18:16, 1.13it/s, tok/s=13151.5]
Inference: 11%|█ | 627/5920 [07:41<1:18:16, 1.13it/s, tok/s=13173.0]
Inference: 11%|█ | 628/5920 [07:43<1:23:51, 1.05it/s, tok/s=13173.0]
Inference: 11%|█ | 628/5920 [07:43<1:23:51, 1.05it/s, tok/s=13175.6]
Inference: 11%|█ | 629/5920 [07:43<1:08:47, 1.28it/s, tok/s=13175.6]
Inference: 11%|█ | 629/5920 [07:43<1:08:47, 1.28it/s, tok/s=13208.4]
Inference: 11%|█ | 630/5920 [07:44<1:20:39, 1.09it/s, tok/s=13208.4]
Inference: 11%|█ | 630/5920 [07:44<1:20:39, 1.09it/s, tok/s=13208.7]
Inference: 11%|█ | 631/5920 [07:47<2:04:13, 1.41s/it, tok/s=13208.7]
Inference: 11%|█ | 631/5920 [07:47<2:04:13, 1.41s/it, tok/s=13177.3]
Inference: 11%|█ | 632/5920 [07:47<1:39:17, 1.13s/it, tok/s=13177.3]
Inference: 11%|█ | 632/5920 [07:47<1:39:17, 1.13s/it, tok/s=13187.9]
Inference: 11%|█ | 633/5920 [07:49<1:59:58, 1.36s/it, tok/s=13187.9]
Inference: 11%|█ | 633/5920 [07:49<1:59:58, 1.36s/it, tok/s=13177.2]
Inference: 11%|█ | 634/5920 [07:49<1:28:22, 1.00s/it, tok/s=13177.2]
Inference: 11%|█ | 634/5920 [07:49<1:28:22, 1.00s/it, tok/s=13214.6]
Inference: 11%|█ | 635/5920 [07:50<1:21:57, 1.07it/s, tok/s=13214.6]
Inference: 11%|█ | 635/5920 [07:50<1:21:57, 1.07it/s, tok/s=13234.7]
Inference: 11%|█ | 636/5920 [07:50<1:02:47, 1.40it/s, tok/s=13234.7]
Inference: 11%|█ | 636/5920 [07:50<1:02:47, 1.40it/s, tok/s=13240.0]
Inference: 11%|█ | 637/5920 [07:51<55:50, 1.58it/s, tok/s=13240.0]
Inference: 11%|█ | 637/5920 [07:51<55:50, 1.58it/s, tok/s=13245.9]
Inference: 11%|█ | 638/5920 [07:52<1:06:39, 1.32it/s, tok/s=13245.9]
Inference: 11%|█ | 638/5920 [07:52<1:06:39, 1.32it/s, tok/s=13228.5]
Inference: 11%|█ | 639/5920 [07:53<1:11:25, 1.23it/s, tok/s=13228.5]
Inference: 11%|█ | 639/5920 [07:53<1:11:25, 1.23it/s, tok/s=13246.1]
Inference: 11%|█ | 640/5920 [07:53<53:56, 1.63it/s, tok/s=13246.1]
Inference: 11%|█ | 640/5920 [07:53<53:56, 1.63it/s, tok/s=13279.9]
Inference: 11%|█ | 641/5920 [07:53<41:33, 2.12it/s, tok/s=13279.9]
Inference: 11%|█ | 641/5920 [07:53<41:33, 2.12it/s, tok/s=13312.2]
Inference: 11%|█ | 642/5920 [07:53<41:33, 2.12it/s, tok/s=13347.4]
Inference: 11%|█ | 643/5920 [07:53<27:07, 3.24it/s, tok/s=13347.4]
Inference: 11%|█ | 643/5920 [07:53<27:07, 3.24it/s, tok/s=13363.6]
Inference: 11%|█ | 644/5920 [07:54<41:26, 2.12it/s, tok/s=13363.6]
Inference: 11%|█ | 644/5920 [07:54<41:26, 2.12it/s, tok/s=13348.7]
Inference: 11%|█ | 645/5920 [07:54<33:24, 2.63it/s, tok/s=13348.7]
Inference: 11%|█ | 645/5920 [07:54<33:24, 2.63it/s, tok/s=13387.9]
Inference: 11%|█ | 646/5920 [07:54<26:45, 3.28it/s, tok/s=13387.9]
Inference: 11%|█ | 646/5920 [07:54<26:45, 3.28it/s, tok/s=13427.0]
Inference: 11%|█ | 647/5920 [07:55<26:53, 3.27it/s, tok/s=13427.0]
Inference: 11%|█ | 647/5920 [07:55<26:53, 3.27it/s, tok/s=13453.7]
Inference: 11%|█ | 648/5920 [07:55<26:53, 3.27it/s, tok/s=13495.8]
Inference: 11%|█ | 649/5920 [07:55<27:00, 3.25it/s, tok/s=13495.8]
Inference: 11%|█ | 649/5920 [07:55<27:00, 3.25it/s, tok/s=13500.7]
Inference: 11%|█ | 650/5920 [07:56<29:05, 3.02it/s, tok/s=13500.7]
Inference: 11%|█ | 650/5920 [07:56<29:05, 3.02it/s, tok/s=13530.4]
Inference: 11%|█ | 651/5920 [07:56<29:04, 3.02it/s, tok/s=13567.5]
Inference: 11%|█ | 652/5920 [07:58<50:27, 1.74it/s, tok/s=13567.5]
Inference: 11%|█ | 652/5920 [07:58<50:27, 1.74it/s, tok/s=13559.9]
Inference: 11%|█ | 653/5920 [07:58<45:33, 1.93it/s, tok/s=13559.9]
Inference: 11%|█ | 653/5920 [07:58<45:33, 1.93it/s, tok/s=13570.8]
Inference: 11%|█ | 654/5920 [08:00<1:10:06, 1.25it/s, tok/s=13570.8]
Inference: 11%|█ | 654/5920 [08:00<1:10:06, 1.25it/s, tok/s=13566.3]
Inference: 11%|█ | 655/5920 [08:00<1:10:05, 1.25it/s, tok/s=13608.7]
Inference: 11%|█ | 656/5920 [08:01<1:01:54, 1.42it/s, tok/s=13608.7]
Inference: 11%|█ | 656/5920 [08:01<1:01:54, 1.42it/s, tok/s=13588.5]
Inference: 11%|█ | 657/5920 [08:01<1:01:54, 1.42it/s, tok/s=13628.9]
Inference: 11%|█ | 658/5920 [08:01<45:05, 1.95it/s, tok/s=13628.9]
Inference: 11%|█ | 658/5920 [08:01<45:05, 1.95it/s, tok/s=13663.1]
Inference: 11%|█ | 659/5920 [08:03<1:17:43, 1.13it/s, tok/s=13663.1]
Inference: 11%|█ | 659/5920 [08:03<1:17:43, 1.13it/s, tok/s=13625.9]
Inference: 11%|█ | 660/5920 [08:05<1:36:11, 1.10s/it, tok/s=13625.9]
Inference: 11%|█ | 660/5920 [08:05<1:36:11, 1.10s/it, tok/s=13595.5]
Inference: 11%|█ | 661/5920 [08:05<1:36:10, 1.10s/it, tok/s=13636.8]
Inference: 11%|█ | 662/5920 [08:06<1:04:32, 1.36it/s, tok/s=13636.8]
Inference: 11%|█ | 662/5920 [08:06<1:04:32, 1.36it/s, tok/s=13642.8]
Inference: 11%|█ | 663/5920 [08:06<1:05:02, 1.35it/s, tok/s=13642.8]
Inference: 11%|█ | 663/5920 [08:06<1:05:02, 1.35it/s, tok/s=13662.4]
Inference: 11%|█ | 664/5920 [08:06<52:00, 1.68it/s, tok/s=13662.4]
Inference: 11%|█ | 664/5920 [08:06<52:00, 1.68it/s, tok/s=13700.9]
Inference: 11%|█ | 665/5920 [08:06<51:59, 1.68it/s, tok/s=13740.9]
Inference: 11%|█▏ | 666/5920 [08:09<1:15:15, 1.16it/s, tok/s=13740.9]
Inference: 11%|█▏ | 666/5920 [08:09<1:15:15, 1.16it/s, tok/s=13715.2]
Inference: 11%|█▏ | 667/5920 [08:10<1:11:47, 1.22it/s, tok/s=13715.2]
Inference: 11%|█▏ | 667/5920 [08:10<1:11:47, 1.22it/s, tok/s=13728.5]
Inference: 11%|█▏ | 668/5920 [08:12<1:38:11, 1.12s/it, tok/s=13728.5]
Inference: 11%|█▏ | 668/5920 [08:12<1:38:11, 1.12s/it, tok/s=13713.8]
Inference: 11%|█▏ | 669/5920 [08:13<1:48:47, 1.24s/it, tok/s=13713.8]
Inference: 11%|█▏ | 669/5920 [08:13<1:48:47, 1.24s/it, tok/s=13710.2]
Inference: 11%|█▏ | 670/5920 [08:14<1:40:05, 1.14s/it, tok/s=13710.2]
Inference: 11%|█▏ | 670/5920 [08:14<1:40:05, 1.14s/it, tok/s=13729.1]
Inference: 11%|█▏ | 671/5920 [08:14<1:40:04, 1.14s/it, tok/s=13770.5]
Inference: 11%|█▏ | 672/5920 [08:14<59:58, 1.46it/s, tok/s=13770.5]
Inference: 11%|█▏ | 672/5920 [08:14<59:58, 1.46it/s, tok/s=13783.6]
Inference: 11%|█▏ | 673/5920 [08:15<1:02:11, 1.41it/s, tok/s=13783.6]
Inference: 11%|█▏ | 673/5920 [08:15<1:02:11, 1.41it/s, tok/s=13781.2]
Inference: 11%|█▏ | 674/5920 [08:16<1:16:01, 1.15it/s, tok/s=13781.2]
Inference: 11%|█▏ | 674/5920 [08:16<1:16:01, 1.15it/s, tok/s=13786.0]
Inference: 11%|█▏ | 675/5920 [08:17<1:01:31, 1.42it/s, tok/s=13786.0]
Inference: 11%|█▏ | 675/5920 [08:17<1:01:31, 1.42it/s, tok/s=13798.3]
Inference: 11%|█▏ | 676/5920 [08:19<1:40:40, 1.15s/it, tok/s=13798.3]
Inference: 11%|█▏ | 676/5920 [08:19<1:40:40, 1.15s/it, tok/s=13745.2]
Inference: 11%|█▏ | 677/5920 [08:23<2:40:38, 1.84s/it, tok/s=13745.2]
Inference: 11%|█▏ | 677/5920 [08:23<2:40:38, 1.84s/it, tok/s=13689.4]
Inference: 11%|█▏ | 678/5920 [08:23<2:40:36, 1.84s/it, tok/s=13706.4]
Inference: 11%|█▏ | 679/5920 [08:23<1:41:07, 1.16s/it, tok/s=13706.4]
Inference: 11%|█▏ | 679/5920 [08:23<1:41:07, 1.16s/it, tok/s=13732.6]
Inference: 11%|█▏ | 680/5920 [08:24<1:39:07, 1.14s/it, tok/s=13732.6]
Inference: 11%|█▏ | 680/5920 [08:24<1:39:07, 1.14s/it, tok/s=13745.5]
Inference: 12%|█▏ | 681/5920 [08:26<1:47:10, 1.23s/it, tok/s=13745.5]
Inference: 12%|█▏ | 681/5920 [08:26<1:47:10, 1.23s/it, tok/s=13747.1]
Inference: 12%|█▏ | 682/5920 [08:26<1:22:40, 1.06it/s, tok/s=13747.1]
Inference: 12%|█▏ | 682/5920 [08:26<1:22:40, 1.06it/s, tok/s=13777.5]
Inference: 12%|█▏ | 683/5920 [08:27<1:23:46, 1.04it/s, tok/s=13777.5]
Inference: 12%|█▏ | 683/5920 [08:27<1:23:46, 1.04it/s, tok/s=13787.7]
Inference: 12%|█▏ | 684/5920 [08:28<1:35:09, 1.09s/it, tok/s=13787.7]
Inference: 12%|█▏ | 684/5920 [08:28<1:35:09, 1.09s/it, tok/s=13792.2]
Inference: 12%|█▏ | 685/5920 [08:31<2:12:37, 1.52s/it, tok/s=13792.2]
Inference: 12%|█▏ | 685/5920 [08:31<2:12:37, 1.52s/it, tok/s=13726.3]
Inference: 12%|█▏ | 686/5920 [08:32<2:02:36, 1.41s/it, tok/s=13726.3]
Inference: 12%|█▏ | 686/5920 [08:32<2:02:36, 1.41s/it, tok/s=13723.3]
Inference: 12%|█▏ | 687/5920 [08:33<1:44:56, 1.20s/it, tok/s=13723.3]
Inference: 12%|█▏ | 687/5920 [08:33<1:44:56, 1.20s/it, tok/s=13718.7]
Inference: 12%|█▏ | 688/5920 [08:33<1:26:31, 1.01it/s, tok/s=13718.7]
Inference: 12%|█▏ | 688/5920 [08:33<1:26:31, 1.01it/s, tok/s=13724.8]
Inference: 12%|█▏ | 689/5920 [08:33<1:04:17, 1.36it/s, tok/s=13724.8]
Inference: 12%|█▏ | 689/5920 [08:33<1:04:17, 1.36it/s, tok/s=13764.0]
Inference: 12%|█▏ | 690/5920 [08:34<51:28, 1.69it/s, tok/s=13764.0]
Inference: 12%|█▏ | 690/5920 [08:34<51:28, 1.69it/s, tok/s=13798.9]
Inference: 12%|█▏ | 691/5920 [08:34<51:22, 1.70it/s, tok/s=13798.9]
Inference: 12%|█▏ | 691/5920 [08:34<51:22, 1.70it/s, tok/s=13803.3]
Inference: 12%|█▏ | 692/5920 [08:35<50:04, 1.74it/s, tok/s=13803.3]
Inference: 12%|█▏ | 692/5920 [08:35<50:04, 1.74it/s, tok/s=13831.4]
Inference: 12%|█▏ | 693/5920 [08:39<2:17:21, 1.58s/it, tok/s=13831.4]
Inference: 12%|█▏ | 693/5920 [08:39<2:17:21, 1.58s/it, tok/s=13738.5]
Inference: 12%|█▏ | 694/5920 [08:41<2:21:46, 1.63s/it, tok/s=13738.5]
Inference: 12%|█▏ | 694/5920 [08:41<2:21:46, 1.63s/it, tok/s=13696.2]
Inference: 12%|█▏ | 695/5920 [08:41<2:21:45, 1.63s/it, tok/s=13713.5]
Inference: 12%|█▏ | 696/5920 [08:41<1:21:14, 1.07it/s, tok/s=13713.5]
Inference: 12%|█▏ | 696/5920 [08:41<1:21:14, 1.07it/s, tok/s=13751.8]
Inference: 12%|█▏ | 697/5920 [08:41<1:08:24, 1.27it/s, tok/s=13751.8]
Inference: 12%|█▏ | 697/5920 [08:41<1:08:24, 1.27it/s, tok/s=13760.7]
Inference: 12%|█▏ | 698/5920 [08:41<56:17, 1.55it/s, tok/s=13760.7]
Inference: 12%|█▏ | 698/5920 [08:41<56:17, 1.55it/s, tok/s=13795.9]
Inference: 12%|█▏ | 699/5920 [08:41<56:16, 1.55it/s, tok/s=13834.1]
Inference: 12%|█▏ | 700/5920 [08:44<1:23:33, 1.04it/s, tok/s=13834.1]
Inference: 12%|█▏ | 700/5920 [08:44<1:23:33, 1.04it/s, tok/s=13780.7]
Inference: 12%|█▏ | 701/5920 [08:44<1:08:57, 1.26it/s, tok/s=13780.7]
Inference: 12%|█▏ | 701/5920 [08:44<1:08:57, 1.26it/s, tok/s=13778.8]
Inference: 12%|█▏ | 702/5920 [08:46<1:18:09, 1.11it/s, tok/s=13778.8]
Inference: 12%|█▏ | 702/5920 [08:46<1:18:09, 1.11it/s, tok/s=13784.6]
Inference: 12%|█▏ | 703/5920 [08:48<2:00:37, 1.39s/it, tok/s=13784.6]
Inference: 12%|█▏ | 703/5920 [08:48<2:00:37, 1.39s/it, tok/s=13753.2]
Inference: 12%|█▏ | 704/5920 [08:49<1:47:10, 1.23s/it, tok/s=13753.2]
Inference: 12%|█▏ | 704/5920 [08:49<1:47:10, 1.23s/it, tok/s=13773.6]
Inference: 12%|█▏ | 705/5920 [08:51<1:59:38, 1.38s/it, tok/s=13773.6]
Inference: 12%|█▏ | 705/5920 [08:51<1:59:38, 1.38s/it, tok/s=13752.7]
Inference: 12%|█▏ | 706/5920 [08:53<2:10:53, 1.51s/it, tok/s=13752.7]
Inference: 12%|█▏ | 706/5920 [08:53<2:10:53, 1.51s/it, tok/s=13731.4]
Inference: 12%|█▏ | 707/5920 [08:55<2:21:56, 1.63s/it, tok/s=13731.4]
Inference: 12%|█▏ | 707/5920 [08:55<2:21:56, 1.63s/it, tok/s=13724.0]
Inference: 12%|█▏ | 708/5920 [08:55<1:43:57, 1.20s/it, tok/s=13724.0]
Inference: 12%|█▏ | 708/5920 [08:55<1:43:57, 1.20s/it, tok/s=13724.0]
Inference: 12%|█▏ | 709/5920 [08:55<1:19:11, 1.10it/s, tok/s=13724.0]
Inference: 12%|█▏ | 709/5920 [08:55<1:19:11, 1.10it/s, tok/s=13739.1]
Inference: 12%|█▏ | 710/5920 [08:55<1:19:10, 1.10it/s, tok/s=13780.6]
Inference: 12%|█▏ | 711/5920 [08:55<46:09, 1.88it/s, tok/s=13780.6]
Inference: 12%|█▏ | 711/5920 [08:55<46:09, 1.88it/s, tok/s=13814.1]
Inference: 12%|█▏ | 712/5920 [08:57<1:20:11, 1.08it/s, tok/s=13814.1]
Inference: 12%|█▏ | 712/5920 [08:57<1:20:11, 1.08it/s, tok/s=13775.9]
Inference: 12%|█▏ | 713/5920 [08:58<1:08:45, 1.26it/s, tok/s=13775.9]
Inference: 12%|█▏ | 713/5920 [08:58<1:08:45, 1.26it/s, tok/s=13788.2]
Inference: 12%|█▏ | 714/5920 [08:58<55:10, 1.57it/s, tok/s=13788.2]
Inference: 12%|█▏ | 714/5920 [08:58<55:10, 1.57it/s, tok/s=13825.1]
Inference: 12%|█▏ | 715/5920 [08:58<44:10, 1.96it/s, tok/s=13825.1]
Inference: 12%|█▏ | 715/5920 [08:58<44:10, 1.96it/s, tok/s=13848.4]
Inference: 12%|█▏ | 716/5920 [08:59<1:03:06, 1.37it/s, tok/s=13848.4]
Inference: 12%|█▏ | 716/5920 [08:59<1:03:06, 1.37it/s, tok/s=13837.4]
Inference: 12%|█▏ | 717/5920 [09:02<1:43:46, 1.20s/it, tok/s=13837.4]
Inference: 12%|█▏ | 717/5920 [09:02<1:43:46, 1.20s/it, tok/s=13798.9]
Inference: 12%|█▏ | 718/5920 [09:02<1:43:44, 1.20s/it, tok/s=13824.2]
Inference: 12%|█▏ | 719/5920 [09:05<2:02:43, 1.42s/it, tok/s=13824.2]
Inference: 12%|█▏ | 719/5920 [09:05<2:02:43, 1.42s/it, tok/s=13782.3]
Inference: 12%|█▏ | 720/5920 [09:06<1:51:02, 1.28s/it, tok/s=13782.3]
Inference: 12%|█▏ | 720/5920 [09:06<1:51:02, 1.28s/it, tok/s=13784.9]
Inference: 12%|█▏ | 721/5920 [09:06<1:29:32, 1.03s/it, tok/s=13784.9]
Inference: 12%|█▏ | 721/5920 [09:06<1:29:32, 1.03s/it, tok/s=13795.7]
Inference: 12%|█▏ | 722/5920 [09:06<1:29:31, 1.03s/it, tok/s=13828.0]
Inference: 12%|█▏ | 723/5920 [09:07<58:10, 1.49it/s, tok/s=13828.0]
Inference: 12%|█▏ | 723/5920 [09:07<58:10, 1.49it/s, tok/s=13861.7]
Inference: 12%|█▏ | 724/5920 [09:10<1:58:18, 1.37s/it, tok/s=13861.7]
Inference: 12%|█▏ | 724/5920 [09:10<1:58:18, 1.37s/it, tok/s=13811.3]
Inference: 12%|█▏ | 725/5920 [09:12<2:13:09, 1.54s/it, tok/s=13811.3]
Inference: 12%|█▏ | 725/5920 [09:12<2:13:09, 1.54s/it, tok/s=13802.2]
Inference: 12%|█▏ | 726/5920 [09:13<1:48:02, 1.25s/it, tok/s=13802.2]
Inference: 12%|█▏ | 726/5920 [09:13<1:48:02, 1.25s/it, tok/s=13811.2]
Inference: 12%|█▏ | 727/5920 [09:14<1:37:31, 1.13s/it, tok/s=13811.2]
Inference: 12%|█▏ | 727/5920 [09:14<1:37:31, 1.13s/it, tok/s=13833.1]
Inference: 12%|█▏ | 728/5920 [09:15<1:32:45, 1.07s/it, tok/s=13833.1]
Inference: 12%|█▏ | 728/5920 [09:15<1:32:45, 1.07s/it, tok/s=13815.8]
Inference: 12%|█▏ | 729/5920 [09:15<1:32:44, 1.07s/it, tok/s=13854.6]
Inference: 12%|█▏ | 730/5920 [09:17<1:29:18, 1.03s/it, tok/s=13854.6]
Inference: 12%|█▏ | 730/5920 [09:17<1:29:18, 1.03s/it, tok/s=13849.6]
Inference: 12%|█▏ | 731/5920 [09:18<1:36:59, 1.12s/it, tok/s=13849.6]
Inference: 12%|█▏ | 731/5920 [09:18<1:36:59, 1.12s/it, tok/s=13833.9]
Inference: 12%|█▏ | 732/5920 [09:19<1:25:39, 1.01it/s, tok/s=13833.9]
Inference: 12%|█▏ | 732/5920 [09:19<1:25:39, 1.01it/s, tok/s=13825.1]
Inference: 12%|█▏ | 733/5920 [09:19<1:25:38, 1.01it/s, tok/s=13865.9]
Inference: 12%|█▏ | 734/5920 [09:20<1:21:48, 1.06it/s, tok/s=13865.9]
Inference: 12%|█▏ | 734/5920 [09:20<1:21:48, 1.06it/s, tok/s=13864.4]
Inference: 12%|█▏ | 735/5920 [09:24<2:14:07, 1.55s/it, tok/s=13864.4]
Inference: 12%|█▏ | 735/5920 [09:24<2:14:07, 1.55s/it, tok/s=13784.2]
Inference: 12%|█▏ | 736/5920 [09:24<1:52:44, 1.30s/it, tok/s=13784.2]
Inference: 12%|█▏ | 736/5920 [09:24<1:52:44, 1.30s/it, tok/s=13776.4]
Inference: 12%|█▏ | 737/5920 [09:25<1:35:39, 1.11s/it, tok/s=13776.4]
Inference: 12%|█▏ | 737/5920 [09:25<1:35:39, 1.11s/it, tok/s=13802.7]
Inference: 12%|█▏ | 738/5920 [09:25<1:35:38, 1.11s/it, tok/s=13842.8]
Inference: 12%|█▏ | 739/5920 [09:26<1:10:03, 1.23it/s, tok/s=13842.8]
Inference: 12%|█▏ | 739/5920 [09:26<1:10:03, 1.23it/s, tok/s=13829.8]
Inference: 12%|█▎ | 740/5920 [09:26<58:11, 1.48it/s, tok/s=13829.8]
Inference: 12%|█▎ | 740/5920 [09:26<58:11, 1.48it/s, tok/s=13863.8]
Inference: 13%|█▎ | 741/5920 [09:26<58:10, 1.48it/s, tok/s=13867.7]
Inference: 13%|█▎ | 742/5920 [09:26<42:15, 2.04it/s, tok/s=13867.7]
Inference: 13%|█▎ | 742/5920 [09:26<42:15, 2.04it/s, tok/s=13880.7]
Inference: 13%|█▎ | 743/5920 [09:28<1:08:31, 1.26it/s, tok/s=13880.7]
Inference: 13%|█▎ | 743/5920 [09:28<1:08:31, 1.26it/s, tok/s=13876.6]
Inference: 13%|█▎ | 744/5920 [09:29<57:16, 1.51it/s, tok/s=13876.6]
Inference: 13%|█▎ | 744/5920 [09:29<57:16, 1.51it/s, tok/s=13876.0]
Inference: 13%|█▎ | 745/5920 [09:29<48:37, 1.77it/s, tok/s=13876.0]
Inference: 13%|█▎ | 745/5920 [09:29<48:37, 1.77it/s, tok/s=13911.3]
Inference: 13%|█▎ | 746/5920 [09:29<48:30, 1.78it/s, tok/s=13911.3]
Inference: 13%|█▎ | 746/5920 [09:29<48:30, 1.78it/s, tok/s=13938.7]
Inference: 13%|█▎ | 747/5920 [09:30<42:10, 2.04it/s, tok/s=13938.7]
Inference: 13%|█▎ | 747/5920 [09:30<42:10, 2.04it/s, tok/s=13949.4]
Inference: 13%|█▎ | 748/5920 [09:30<42:10, 2.04it/s, tok/s=13968.6]
Inference: 13%|█▎ | 749/5920 [09:30<35:28, 2.43it/s, tok/s=13968.6]
Inference: 13%|█▎ | 749/5920 [09:30<35:28, 2.43it/s, tok/s=13997.9]
Inference: 13%|█▎ | 750/5920 [09:32<1:11:45, 1.20it/s, tok/s=13997.9]
Inference: 13%|█▎ | 750/5920 [09:32<1:11:45, 1.20it/s, tok/s=13973.2]
Inference: 13%|█▎ | 751/5920 [09:33<55:38, 1.55it/s, tok/s=13973.2]
Inference: 13%|█▎ | 751/5920 [09:33<55:38, 1.55it/s, tok/s=14012.0]
Inference: 13%|█▎ | 752/5920 [09:33<54:21, 1.58it/s, tok/s=14012.0]
Inference: 13%|█▎ | 752/5920 [09:33<54:21, 1.58it/s, tok/s=14014.6]
Inference: 13%|█▎ | 753/5920 [09:34<1:04:29, 1.34it/s, tok/s=14014.6]
Inference: 13%|█▎ | 753/5920 [09:34<1:04:29, 1.34it/s, tok/s=14007.7]
Inference: 13%|█▎ | 754/5920 [09:36<1:18:05, 1.10it/s, tok/s=14007.7]
Inference: 13%|█▎ | 754/5920 [09:36<1:18:05, 1.10it/s, tok/s=14018.0]
Inference: 13%|█▎ | 755/5920 [09:36<1:06:53, 1.29it/s, tok/s=14018.0]
Inference: 13%|█▎ | 755/5920 [09:36<1:06:53, 1.29it/s, tok/s=14046.4]
Inference: 13%|█▎ | 756/5920 [09:37<1:01:47, 1.39it/s, tok/s=14046.4]
Inference: 13%|█▎ | 756/5920 [09:37<1:01:47, 1.39it/s, tok/s=14065.9]
Inference: 13%|█▎ | 757/5920 [09:39<1:40:04, 1.16s/it, tok/s=14065.9]
Inference: 13%|█▎ | 757/5920 [09:39<1:40:04, 1.16s/it, tok/s=14052.9]
Inference: 13%|█▎ | 758/5920 [09:39<1:19:46, 1.08it/s, tok/s=14052.9]
Inference: 13%|█▎ | 758/5920 [09:39<1:19:46, 1.08it/s, tok/s=14050.7]
Inference: 13%|█▎ | 759/5920 [09:41<1:41:33, 1.18s/it, tok/s=14050.7]
Inference: 13%|█▎ | 759/5920 [09:41<1:41:33, 1.18s/it, tok/s=14047.1]
Inference: 13%|█▎ | 760/5920 [09:42<1:32:50, 1.08s/it, tok/s=14047.1]
Inference: 13%|█▎ | 760/5920 [09:42<1:32:50, 1.08s/it, tok/s=14047.5]
Inference: 13%|█▎ | 761/5920 [09:42<1:11:50, 1.20it/s, tok/s=14047.5]
Inference: 13%|█▎ | 761/5920 [09:42<1:11:50, 1.20it/s, tok/s=14051.5]
Inference: 13%|█▎ | 762/5920 [09:43<1:11:15, 1.21it/s, tok/s=14051.5]
Inference: 13%|█▎ | 762/5920 [09:43<1:11:15, 1.21it/s, tok/s=14049.6]
Inference: 13%|█▎ | 763/5920 [09:44<1:21:35, 1.05it/s, tok/s=14049.6]
Inference: 13%|█▎ | 763/5920 [09:44<1:21:35, 1.05it/s, tok/s=14061.2]
Inference: 13%|█▎ | 764/5920 [09:49<2:53:27, 2.02s/it, tok/s=14061.2]
Inference: 13%|█▎ | 764/5920 [09:49<2:53:27, 2.02s/it, tok/s=13994.9]
Inference: 13%|█▎ | 765/5920 [09:49<2:23:17, 1.67s/it, tok/s=13994.9]
Inference: 13%|█▎ | 765/5920 [09:49<2:23:17, 1.67s/it, tok/s=14016.5]
Inference: 13%|█▎ | 766/5920 [09:51<2:15:29, 1.58s/it, tok/s=14016.5]
Inference: 13%|█▎ | 766/5920 [09:51<2:15:29, 1.58s/it, tok/s=14026.1]
Inference: 13%|█▎ | 767/5920 [09:51<1:46:14, 1.24s/it, tok/s=14026.1]
Inference: 13%|█▎ | 767/5920 [09:51<1:46:14, 1.24s/it, tok/s=14057.1]
Inference: 13%|█▎ | 768/5920 [09:52<1:23:33, 1.03it/s, tok/s=14057.1]
Inference: 13%|█▎ | 768/5920 [09:52<1:23:33, 1.03it/s, tok/s=14079.7]
Inference: 13%|█▎ | 769/5920 [09:54<2:06:59, 1.48s/it, tok/s=14079.7]
Inference: 13%|█▎ | 769/5920 [09:54<2:06:59, 1.48s/it, tok/s=14042.6]
Inference: 13%|█▎ | 770/5920 [09:54<2:06:57, 1.48s/it, tok/s=14048.9]
Inference: 13%|█▎ | 771/5920 [09:54<2:06:56, 1.48s/it, tok/s=14081.5]
Inference: 13%|█▎ | 772/5920 [09:58<1:49:27, 1.28s/it, tok/s=14081.5]
Inference: 13%|█▎ | 772/5920 [09:58<1:49:27, 1.28s/it, tok/s=14025.9]
Inference: 13%|█▎ | 773/5920 [09:58<1:38:20, 1.15s/it, tok/s=14025.9]
Inference: 13%|█▎ | 773/5920 [09:58<1:38:20, 1.15s/it, tok/s=14018.5]
Inference: 13%|█▎ | 774/5920 [09:59<1:28:14, 1.03s/it, tok/s=14018.5]
Inference: 13%|█▎ | 774/5920 [09:59<1:28:14, 1.03s/it, tok/s=14011.4]
Inference: 13%|█▎ | 775/5920 [10:00<1:30:20, 1.05s/it, tok/s=14011.4]
Inference: 13%|█▎ | 775/5920 [10:00<1:30:20, 1.05s/it, tok/s=13992.7]
Inference: 13%|█▎ | 776/5920 [10:01<1:25:08, 1.01it/s, tok/s=13992.7]
Inference: 13%|█▎ | 776/5920 [10:01<1:25:08, 1.01it/s, tok/s=13981.6]
Inference: 13%|█▎ | 777/5920 [10:03<1:54:34, 1.34s/it, tok/s=13981.6]
Inference: 13%|█▎ | 777/5920 [10:03<1:54:34, 1.34s/it, tok/s=13970.7]
Inference: 13%|█▎ | 778/5920 [10:05<2:07:19, 1.49s/it, tok/s=13970.7]
Inference: 13%|█▎ | 778/5920 [10:05<2:07:19, 1.49s/it, tok/s=13969.0]
Inference: 13%|█▎ | 779/5920 [10:08<2:31:34, 1.77s/it, tok/s=13969.0]
Inference: 13%|█▎ | 779/5920 [10:08<2:31:34, 1.77s/it, tok/s=13935.3]
Inference: 13%|█▎ | 780/5920 [10:08<2:10:19, 1.52s/it, tok/s=13935.3]
Inference: 13%|█▎ | 780/5920 [10:08<2:10:19, 1.52s/it, tok/s=13932.7]
Inference: 13%|█▎ | 781/5920 [10:09<1:36:25, 1.13s/it, tok/s=13932.7]
Inference: 13%|█▎ | 781/5920 [10:09<1:36:25, 1.13s/it, tok/s=13940.9]
Inference: 13%|█▎ | 782/5920 [10:10<1:45:05, 1.23s/it, tok/s=13940.9]
Inference: 13%|█▎ | 782/5920 [10:10<1:45:05, 1.23s/it, tok/s=13948.7]
Inference: 13%|█▎ | 783/5920 [10:10<1:17:36, 1.10it/s, tok/s=13948.7]
Inference: 13%|█▎ | 783/5920 [10:10<1:17:36, 1.10it/s, tok/s=13986.6]
Inference: 13%|█▎ | 784/5920 [10:10<59:41, 1.43it/s, tok/s=13986.6]
Inference: 13%|█▎ | 784/5920 [10:10<59:41, 1.43it/s, tok/s=14003.4]
Inference: 13%|█▎ | 785/5920 [10:12<1:12:59, 1.17it/s, tok/s=14003.4]
Inference: 13%|█▎ | 785/5920 [10:12<1:12:59, 1.17it/s, tok/s=14017.4]
Inference: 13%|█▎ | 786/5920 [10:12<1:00:26, 1.42it/s, tok/s=14017.4]
Inference: 13%|█▎ | 786/5920 [10:12<1:00:26, 1.42it/s, tok/s=14019.2]
Inference: 13%|█▎ | 787/5920 [10:12<53:44, 1.59it/s, tok/s=14019.2]
Inference: 13%|█▎ | 787/5920 [10:12<53:44, 1.59it/s, tok/s=14043.1]
Inference: 13%|█▎ | 788/5920 [10:13<44:04, 1.94it/s, tok/s=14043.1]
Inference: 13%|█▎ | 788/5920 [10:13<44:04, 1.94it/s, tok/s=14054.3]
Inference: 13%|█▎ | 789/5920 [10:13<44:43, 1.91it/s, tok/s=14054.3]
Inference: 13%|█▎ | 789/5920 [10:13<44:43, 1.91it/s, tok/s=14060.0]
Inference: 13%|█▎ | 790/5920 [10:15<1:15:33, 1.13it/s, tok/s=14060.0]
Inference: 13%|█▎ | 790/5920 [10:15<1:15:33, 1.13it/s, tok/s=14041.8]
Inference: 13%|█▎ | 791/5920 [10:15<1:03:10, 1.35it/s, tok/s=14041.8]
Inference: 13%|█▎ | 791/5920 [10:15<1:03:10, 1.35it/s, tok/s=14066.0]
Inference: 13%|█▎ | 792/5920 [10:16<1:07:25, 1.27it/s, tok/s=14066.0]
Inference: 13%|█▎ | 792/5920 [10:16<1:07:25, 1.27it/s, tok/s=14056.3]
Inference: 13%|█▎ | 793/5920 [10:23<3:50:15, 2.69s/it, tok/s=14056.3]
Inference: 13%|█▎ | 793/5920 [10:23<3:50:15, 2.69s/it, tok/s=13913.4]
Inference: 13%|█▎ | 794/5920 [10:24<2:44:06, 1.92s/it, tok/s=13913.4]
Inference: 13%|█▎ | 794/5920 [10:24<2:44:06, 1.92s/it, tok/s=13921.1]
Inference: 13%|█▎ | 795/5920 [10:24<1:58:18, 1.39s/it, tok/s=13921.1]
Inference: 13%|█▎ | 795/5920 [10:24<1:58:18, 1.39s/it, tok/s=13931.9]
Inference: 13%|█▎ | 796/5920 [10:24<1:41:47, 1.19s/it, tok/s=13931.9]
Inference: 13%|█▎ | 796/5920 [10:24<1:41:47, 1.19s/it, tok/s=13932.7]
Inference: 13%|█▎ | 797/5920 [10:25<1:23:15, 1.03it/s, tok/s=13932.7]
Inference: 13%|█▎ | 797/5920 [10:25<1:23:15, 1.03it/s, tok/s=13932.2]
Inference: 13%|█▎ | 798/5920 [10:30<3:13:47, 2.27s/it, tok/s=13932.2]
Inference: 13%|█▎ | 798/5920 [10:30<3:13:47, 2.27s/it, tok/s=13832.5]
Inference: 13%|█▎ | 799/5920 [10:30<2:21:21, 1.66s/it, tok/s=13832.5]
Inference: 13%|█▎ | 799/5920 [10:30<2:21:21, 1.66s/it, tok/s=13845.0]
Inference: 14%|█▎ | 800/5920 [10:32<2:17:37, 1.61s/it, tok/s=13845.0]
Inference: 14%|█▎ | 800/5920 [10:32<2:17:37, 1.61s/it, tok/s=13844.3]
Inference: 14%|█▎ | 801/5920 [10:35<3:00:07, 2.11s/it, tok/s=13844.3]
Inference: 14%|█▎ | 801/5920 [10:35<3:00:07, 2.11s/it, tok/s=13798.2]
Inference: 14%|█▎ | 802/5920 [10:37<2:56:02, 2.06s/it, tok/s=13798.2]
Inference: 14%|█▎ | 802/5920 [10:37<2:56:02, 2.06s/it, tok/s=13774.3]
Inference: 14%|█▎ | 803/5920 [10:38<2:15:37, 1.59s/it, tok/s=13774.3]
Inference: 14%|█▎ | 803/5920 [10:38<2:15:37, 1.59s/it, tok/s=13802.6]
Inference: 14%|█▎ | 804/5920 [10:38<1:54:47, 1.35s/it, tok/s=13802.6]
Inference: 14%|█▎ | 804/5920 [10:38<1:54:47, 1.35s/it, tok/s=13792.7]
Inference: 14%|█▎ | 805/5920 [10:39<1:37:57, 1.15s/it, tok/s=13792.7]
Inference: 14%|█▎ | 805/5920 [10:39<1:37:57, 1.15s/it, tok/s=13790.9]
Inference: 14%|█▎ | 806/5920 [10:39<1:37:56, 1.15s/it, tok/s=13812.0]
Inference: 14%|█▎ | 807/5920 [10:39<1:37:55, 1.15s/it, tok/s=13818.5]
Inference: 14%|█▎ | 808/5920 [10:41<1:09:04, 1.23it/s, tok/s=13818.5]
Inference: 14%|█▎ | 808/5920 [10:41<1:09:04, 1.23it/s, tok/s=13820.3]
Inference: 14%|█▎ | 809/5920 [10:42<1:17:11, 1.10it/s, tok/s=13820.3]
Inference: 14%|█▎ | 809/5920 [10:42<1:17:11, 1.10it/s, tok/s=13822.1]
Inference: 14%|█▎ | 810/5920 [10:47<2:46:25, 1.95s/it, tok/s=13822.1]
Inference: 14%|█▎ | 810/5920 [10:47<2:46:25, 1.95s/it, tok/s=13726.5]
Inference: 14%|█▎ | 811/5920 [10:48<2:10:26, 1.53s/it, tok/s=13726.5]
Inference: 14%|█▎ | 811/5920 [10:48<2:10:26, 1.53s/it, tok/s=13746.5]
Inference: 14%|█▎ | 812/5920 [10:50<2:20:46, 1.65s/it, tok/s=13746.5]
Inference: 14%|█▎ | 812/5920 [10:50<2:20:46, 1.65s/it, tok/s=13712.8]
Inference: 14%|█▎ | 813/5920 [10:52<2:43:36, 1.92s/it, tok/s=13712.8]
Inference: 14%|█▎ | 813/5920 [10:52<2:43:36, 1.92s/it, tok/s=13697.8]
Inference: 14%|█▍ | 814/5920 [10:54<2:29:48, 1.76s/it, tok/s=13697.8]
Inference: 14%|█▍ | 814/5920 [10:54<2:29:48, 1.76s/it, tok/s=13692.2]
Inference: 14%|█▍ | 815/5920 [10:56<2:49:03, 1.99s/it, tok/s=13692.2]
Inference: 14%|█▍ | 815/5920 [10:56<2:49:03, 1.99s/it, tok/s=13679.7]
Inference: 14%|█▍ | 816/5920 [10:56<2:49:01, 1.99s/it, tok/s=13694.9]
Inference: 14%|█▍ | 817/5920 [10:57<1:39:17, 1.17s/it, tok/s=13694.9]
Inference: 14%|█▍ | 817/5920 [10:57<1:39:17, 1.17s/it, tok/s=13694.8]
Inference: 14%|█▍ | 818/5920 [10:57<1:39:16, 1.17s/it, tok/s=13701.5]
Inference: 14%|█▍ | 819/5920 [10:58<1:22:55, 1.03it/s, tok/s=13701.5]
Inference: 14%|█▍ | 819/5920 [10:58<1:22:55, 1.03it/s, tok/s=13685.8]
Inference: 14%|█▍ | 820/5920 [10:58<1:09:29, 1.22it/s, tok/s=13685.8]
Inference: 14%|█▍ | 820/5920 [10:58<1:09:29, 1.22it/s, tok/s=13687.9]
Inference: 14%|█▍ | 821/5920 [10:58<1:09:29, 1.22it/s, tok/s=13725.0]
Inference: 14%|█▍ | 822/5920 [11:00<1:14:27, 1.14it/s, tok/s=13725.0]
Inference: 14%|█▍ | 822/5920 [11:00<1:14:27, 1.14it/s, tok/s=13719.4]
Inference: 14%|█▍ | 823/5920 [11:03<1:50:25, 1.30s/it, tok/s=13719.4]
Inference: 14%|█▍ | 823/5920 [11:03<1:50:25, 1.30s/it, tok/s=13683.1]
Inference: 14%|█▍ | 824/5920 [11:03<1:28:14, 1.04s/it, tok/s=13683.1]
Inference: 14%|█▍ | 824/5920 [11:03<1:28:14, 1.04s/it, tok/s=13687.0]
Inference: 14%|█▍ | 825/5920 [11:03<1:08:50, 1.23it/s, tok/s=13687.0]
Inference: 14%|█▍ | 825/5920 [11:03<1:08:50, 1.23it/s, tok/s=13692.2]
Inference: 14%|█▍ | 826/5920 [11:04<1:01:41, 1.38it/s, tok/s=13692.2]
Inference: 14%|█▍ | 826/5920 [11:04<1:01:41, 1.38it/s, tok/s=13692.1]
Inference: 14%|█▍ | 827/5920 [11:04<48:04, 1.77it/s, tok/s=13692.1]
Inference: 14%|█▍ | 827/5920 [11:04<48:04, 1.77it/s, tok/s=13727.7]
Inference: 14%|█▍ | 828/5920 [11:05<51:06, 1.66it/s, tok/s=13727.7]
Inference: 14%|█▍ | 828/5920 [11:05<51:06, 1.66it/s, tok/s=13746.2]
Inference: 14%|█▍ | 829/5920 [11:05<38:53, 2.18it/s, tok/s=13746.2]
Inference: 14%|█▍ | 829/5920 [11:05<38:53, 2.18it/s, tok/s=13785.0]
Inference: 14%|█▍ | 830/5920 [11:07<1:20:03, 1.06it/s, tok/s=13785.0]
Inference: 14%|█▍ | 830/5920 [11:07<1:20:03, 1.06it/s, tok/s=13753.6]
Inference: 14%|█▍ | 831/5920 [11:07<1:00:07, 1.41it/s, tok/s=13753.6]
Inference: 14%|█▍ | 831/5920 [11:07<1:00:07, 1.41it/s, tok/s=13769.5]
Inference: 14%|█▍ | 832/5920 [11:08<1:20:17, 1.06it/s, tok/s=13769.5]
Inference: 14%|█▍ | 832/5920 [11:08<1:20:17, 1.06it/s, tok/s=13750.2]
Inference: 14%|█▍ | 833/5920 [11:09<1:20:16, 1.06it/s, tok/s=13759.7]
Inference: 14%|█▍ | 834/5920 [11:09<45:55, 1.85it/s, tok/s=13759.7]
Inference: 14%|█▍ | 834/5920 [11:09<45:55, 1.85it/s, tok/s=13768.1]
Inference: 14%|█▍ | 835/5920 [11:09<49:22, 1.72it/s, tok/s=13768.1]
Inference: 14%|█▍ | 835/5920 [11:09<49:22, 1.72it/s, tok/s=13766.8]
Inference: 14%|█▍ | 836/5920 [11:10<51:13, 1.65it/s, tok/s=13766.8]
Inference: 14%|█▍ | 836/5920 [11:10<51:13, 1.65it/s, tok/s=13767.9]
Inference: 14%|█▍ | 837/5920 [11:11<55:27, 1.53it/s, tok/s=13767.9]
Inference: 14%|█▍ | 837/5920 [11:11<55:27, 1.53it/s, tok/s=13790.1]
Inference: 14%|█▍ | 838/5920 [11:11<54:28, 1.55it/s, tok/s=13790.1]
Inference: 14%|█▍ | 838/5920 [11:11<54:28, 1.55it/s, tok/s=13811.1]
Inference: 14%|█▍ | 839/5920 [11:13<1:09:10, 1.22it/s, tok/s=13811.1]
Inference: 14%|█▍ | 839/5920 [11:13<1:09:10, 1.22it/s, tok/s=13807.1]
Inference: 14%|█▍ | 840/5920 [11:13<55:55, 1.51it/s, tok/s=13807.1]
Inference: 14%|█▍ | 840/5920 [11:13<55:55, 1.51it/s, tok/s=13814.5]
Inference: 14%|█▍ | 841/5920 [11:13<47:21, 1.79it/s, tok/s=13814.5]
Inference: 14%|█▍ | 841/5920 [11:13<47:21, 1.79it/s, tok/s=13821.6]
Inference: 14%|█▍ | 842/5920 [11:14<56:38, 1.49it/s, tok/s=13821.6]
Inference: 14%|█▍ | 842/5920 [11:14<56:38, 1.49it/s, tok/s=13842.7]
Inference: 14%|█▍ | 843/5920 [11:14<44:39, 1.89it/s, tok/s=13842.7]
Inference: 14%|█▍ | 843/5920 [11:14<44:39, 1.89it/s, tok/s=13846.2]
Inference: 14%|█▍ | 844/5920 [11:15<54:26, 1.55it/s, tok/s=13846.2]
Inference: 14%|█▍ | 844/5920 [11:15<54:26, 1.55it/s, tok/s=13835.8]
Inference: 14%|█▍ | 845/5920 [11:15<54:25, 1.55it/s, tok/s=13844.6]
Inference: 14%|█▍ | 846/5920 [11:16<41:10, 2.05it/s, tok/s=13844.6]
Inference: 14%|█▍ | 846/5920 [11:16<41:10, 2.05it/s, tok/s=13866.4]
Inference: 14%|█▍ | 847/5920 [11:16<36:42, 2.30it/s, tok/s=13866.4]
Inference: 14%|█▍ | 847/5920 [11:16<36:42, 2.30it/s, tok/s=13869.6]
Inference: 14%|█▍ | 848/5920 [11:16<31:03, 2.72it/s, tok/s=13869.6]
Inference: 14%|█▍ | 848/5920 [11:16<31:03, 2.72it/s, tok/s=13883.2]
Inference: 14%|█▍ | 849/5920 [11:20<1:46:20, 1.26s/it, tok/s=13883.2]
Inference: 14%|█▍ | 849/5920 [11:20<1:46:20, 1.26s/it, tok/s=13832.0]
Inference: 14%|█▍ | 850/5920 [11:20<1:24:33, 1.00s/it, tok/s=13832.0]
Inference: 14%|█▍ | 850/5920 [11:20<1:24:33, 1.00s/it, tok/s=13833.0]
Inference: 14%|█▍ | 851/5920 [11:21<1:17:03, 1.10it/s, tok/s=13833.0]
Inference: 14%|█▍ | 851/5920 [11:21<1:17:03, 1.10it/s, tok/s=13859.1]
Inference: 14%|█▍ | 852/5920 [11:21<1:07:43, 1.25it/s, tok/s=13859.1]
Inference: 14%|█▍ | 852/5920 [11:21<1:07:43, 1.25it/s, tok/s=13861.8]
Inference: 14%|█▍ | 853/5920 [11:22<1:04:03, 1.32it/s, tok/s=13861.8]
Inference: 14%|█▍ | 853/5920 [11:22<1:04:03, 1.32it/s, tok/s=13860.2]
Inference: 14%|█▍ | 854/5920 [11:23<1:06:02, 1.28it/s, tok/s=13860.2]
Inference: 14%|█▍ | 854/5920 [11:23<1:06:02, 1.28it/s, tok/s=13851.1]
Inference: 14%|█▍ | 855/5920 [11:24<1:03:09, 1.34it/s, tok/s=13851.1]
Inference: 14%|█▍ | 855/5920 [11:24<1:03:09, 1.34it/s, tok/s=13846.1]
Inference: 14%|█▍ | 856/5920 [11:26<1:45:52, 1.25s/it, tok/s=13846.1]
Inference: 14%|█▍ | 856/5920 [11:26<1:45:52, 1.25s/it, tok/s=13808.8]
Inference: 14%|█▍ | 857/5920 [11:27<1:25:50, 1.02s/it, tok/s=13808.8]
Inference: 14%|█▍ | 857/5920 [11:27<1:25:50, 1.02s/it, tok/s=13807.0]
Inference: 14%|█▍ | 858/5920 [11:28<1:42:44, 1.22s/it, tok/s=13807.0]
Inference: 14%|█▍ | 858/5920 [11:28<1:42:44, 1.22s/it, tok/s=13782.1]
Inference: 15%|█▍ | 859/5920 [11:29<1:31:55, 1.09s/it, tok/s=13782.1]
Inference: 15%|█▍ | 859/5920 [11:29<1:31:55, 1.09s/it, tok/s=13781.3]
Inference: 15%|█▍ | 860/5920 [11:30<1:16:48, 1.10it/s, tok/s=13781.3]
Inference: 15%|█▍ | 860/5920 [11:30<1:16:48, 1.10it/s, tok/s=13793.5]
Inference: 15%|█▍ | 861/5920 [11:30<1:06:19, 1.27it/s, tok/s=13793.5]
Inference: 15%|█▍ | 861/5920 [11:30<1:06:19, 1.27it/s, tok/s=13818.0]
Inference: 15%|█▍ | 862/5920 [11:30<49:48, 1.69it/s, tok/s=13818.0]
Inference: 15%|█▍ | 862/5920 [11:30<49:48, 1.69it/s, tok/s=13852.6]
Inference: 15%|█▍ | 863/5920 [11:32<1:12:04, 1.17it/s, tok/s=13852.6]
Inference: 15%|█▍ | 863/5920 [11:32<1:12:04, 1.17it/s, tok/s=13834.0]
Inference: 15%|█▍ | 864/5920 [11:32<1:12:03, 1.17it/s, tok/s=13847.5]
Inference: 15%|█▍ | 865/5920 [11:33<59:34, 1.41it/s, tok/s=13847.5]
Inference: 15%|█▍ | 865/5920 [11:33<59:34, 1.41it/s, tok/s=13835.0]
Inference: 15%|█▍ | 866/5920 [11:34<1:17:24, 1.09it/s, tok/s=13835.0]
Inference: 15%|█▍ | 866/5920 [11:34<1:17:24, 1.09it/s, tok/s=13823.5]
Inference: 15%|█▍ | 867/5920 [11:34<1:01:52, 1.36it/s, tok/s=13823.5]
Inference: 15%|█▍ | 867/5920 [11:34<1:01:52, 1.36it/s, tok/s=13830.5]
Inference: 15%|█▍ | 868/5920 [11:35<1:02:10, 1.35it/s, tok/s=13830.5]
Inference: 15%|█▍ | 868/5920 [11:35<1:02:10, 1.35it/s, tok/s=13833.7]
Inference: 15%|█▍ | 869/5920 [11:35<47:40, 1.77it/s, tok/s=13833.7]
Inference: 15%|█▍ | 869/5920 [11:35<47:40, 1.77it/s, tok/s=13841.1]
Inference: 15%|█▍ | 870/5920 [11:35<47:39, 1.77it/s, tok/s=13852.7]
Inference: 15%|█▍ | 871/5920 [11:37<48:11, 1.75it/s, tok/s=13852.7]
Inference: 15%|█▍ | 871/5920 [11:37<48:11, 1.75it/s, tok/s=13844.4]
Inference: 15%|█▍ | 872/5920 [11:37<48:10, 1.75it/s, tok/s=13856.7]
Inference: 15%|█▍ | 873/5920 [11:37<34:45, 2.42it/s, tok/s=13856.7]
Inference: 15%|█▍ | 873/5920 [11:37<34:45, 2.42it/s, tok/s=13871.5]
Inference: 15%|█▍ | 874/5920 [11:38<43:27, 1.94it/s, tok/s=13871.5]
Inference: 15%|█▍ | 874/5920 [11:38<43:27, 1.94it/s, tok/s=13869.0]
Inference: 15%|█▍ | 875/5920 [11:42<2:02:38, 1.46s/it, tok/s=13869.0]
Inference: 15%|█▍ | 875/5920 [11:42<2:02:38, 1.46s/it, tok/s=13795.2]
Inference: 15%|█▍ | 876/5920 [11:44<2:05:30, 1.49s/it, tok/s=13795.2]
Inference: 15%|█▍ | 876/5920 [11:44<2:05:30, 1.49s/it, tok/s=13772.3]
Inference: 15%|█▍ | 877/5920 [11:44<1:44:54, 1.25s/it, tok/s=13772.3]
Inference: 15%|█▍ | 877/5920 [11:44<1:44:54, 1.25s/it, tok/s=13769.5]
Inference: 15%|█▍ | 878/5920 [11:45<1:22:15, 1.02it/s, tok/s=13769.5]
Inference: 15%|█▍ | 878/5920 [11:45<1:22:15, 1.02it/s, tok/s=13772.2]
Inference: 15%|█▍ | 879/5920 [11:45<1:07:12, 1.25it/s, tok/s=13772.2]
Inference: 15%|█▍ | 879/5920 [11:45<1:07:12, 1.25it/s, tok/s=13774.0]
Inference: 15%|█▍ | 880/5920 [11:46<1:17:22, 1.09it/s, tok/s=13774.0]
Inference: 15%|█▍ | 880/5920 [11:46<1:17:22, 1.09it/s, tok/s=13758.9]
Inference: 15%|█▍ | 881/5920 [11:47<1:24:21, 1.00s/it, tok/s=13758.9]
Inference: 15%|█▍ | 881/5920 [11:47<1:24:21, 1.00s/it, tok/s=13752.5]
Inference: 15%|█▍ | 882/5920 [11:48<1:18:02, 1.08it/s, tok/s=13752.5]
Inference: 15%|█▍ | 882/5920 [11:48<1:18:02, 1.08it/s, tok/s=13746.6]
Inference: 15%|█▍ | 883/5920 [11:49<1:18:28, 1.07it/s, tok/s=13746.6]
Inference: 15%|█▍ | 883/5920 [11:49<1:18:28, 1.07it/s, tok/s=13747.9]
Inference: 15%|█▍ | 884/5920 [11:50<1:25:21, 1.02s/it, tok/s=13747.9]
Inference: 15%|█▍ | 884/5920 [11:50<1:25:21, 1.02s/it, tok/s=13733.7]
Inference: 15%|█▍ | 885/5920 [11:52<1:46:57, 1.27s/it, tok/s=13733.7]
Inference: 15%|█▍ | 885/5920 [11:52<1:46:57, 1.27s/it, tok/s=13710.6]
Inference: 15%|█▍ | 886/5920 [11:52<1:19:28, 1.06it/s, tok/s=13710.6]
Inference: 15%|█▍ | 886/5920 [11:52<1:19:28, 1.06it/s, tok/s=13747.4]
Inference: 15%|█▍ | 887/5920 [11:53<1:05:22, 1.28it/s, tok/s=13747.4]
Inference: 15%|█▍ | 887/5920 [11:53<1:05:22, 1.28it/s, tok/s=13758.2]
Inference: 15%|█▌ | 888/5920 [11:54<1:11:54, 1.17it/s, tok/s=13758.2]
Inference: 15%|█▌ | 888/5920 [11:54<1:11:54, 1.17it/s, tok/s=13773.3]
Inference: 15%|█▌ | 889/5920 [11:54<56:47, 1.48it/s, tok/s=13773.3]
Inference: 15%|█▌ | 889/5920 [11:54<56:47, 1.48it/s, tok/s=13780.8]
Inference: 15%|█▌ | 890/5920 [11:55<56:49, 1.48it/s, tok/s=13780.8]
Inference: 15%|█▌ | 890/5920 [11:55<56:49, 1.48it/s, tok/s=13776.8]
Inference: 15%|█▌ | 891/5920 [11:55<43:17, 1.94it/s, tok/s=13776.8]
Inference: 15%|█▌ | 891/5920 [11:55<43:17, 1.94it/s, tok/s=13788.6]
Inference: 15%|█▌ | 892/5920 [11:57<1:13:57, 1.13it/s, tok/s=13788.6]
Inference: 15%|█▌ | 892/5920 [11:57<1:13:57, 1.13it/s, tok/s=13777.9]
Inference: 15%|█▌ | 893/5920 [11:59<1:44:17, 1.24s/it, tok/s=13777.9]
Inference: 15%|█▌ | 893/5920 [11:59<1:44:17, 1.24s/it, tok/s=13752.0]
Inference: 15%|█▌ | 894/5920 [12:00<1:47:15, 1.28s/it, tok/s=13752.0]
Inference: 15%|█▌ | 894/5920 [12:00<1:47:15, 1.28s/it, tok/s=13757.7]
Inference: 15%|█▌ | 895/5920 [12:00<1:24:04, 1.00s/it, tok/s=13757.7]
Inference: 15%|█▌ | 895/5920 [12:00<1:24:04, 1.00s/it, tok/s=13761.6]
Inference: 15%|█▌ | 896/5920 [12:01<1:09:39, 1.20it/s, tok/s=13761.6]
Inference: 15%|█▌ | 896/5920 [12:01<1:09:39, 1.20it/s, tok/s=13762.6]
Inference: 15%|█▌ | 897/5920 [12:02<1:05:08, 1.29it/s, tok/s=13762.6]
Inference: 15%|█▌ | 897/5920 [12:02<1:05:08, 1.29it/s, tok/s=13783.1]
Inference: 15%|█▌ | 898/5920 [12:02<56:19, 1.49it/s, tok/s=13783.1]
Inference: 15%|█▌ | 898/5920 [12:02<56:19, 1.49it/s, tok/s=13798.5]
Inference: 15%|█▌ | 899/5920 [12:02<45:35, 1.84it/s, tok/s=13798.5]
Inference: 15%|█▌ | 899/5920 [12:02<45:35, 1.84it/s, tok/s=13807.3]
Inference: 15%|█▌ | 900/5920 [12:04<1:07:43, 1.24it/s, tok/s=13807.3]
Inference: 15%|█▌ | 900/5920 [12:04<1:07:43, 1.24it/s, tok/s=13796.8]
Inference: 15%|█▌ | 901/5920 [12:05<1:11:32, 1.17it/s, tok/s=13796.8]
Inference: 15%|█▌ | 901/5920 [12:05<1:11:32, 1.17it/s, tok/s=13805.8]
Inference: 15%|█▌ | 902/5920 [12:05<1:11:05, 1.18it/s, tok/s=13805.8]
Inference: 15%|█▌ | 902/5920 [12:05<1:11:05, 1.18it/s, tok/s=13810.0]
Inference: 15%|█▌ | 903/5920 [12:08<1:42:58, 1.23s/it, tok/s=13810.0]
Inference: 15%|█▌ | 903/5920 [12:08<1:42:58, 1.23s/it, tok/s=13782.7]
Inference: 15%|█▌ | 904/5920 [12:08<1:27:20, 1.04s/it, tok/s=13782.7]
Inference: 15%|█▌ | 904/5920 [12:08<1:27:20, 1.04s/it, tok/s=13782.5]
Inference: 15%|█▌ | 905/5920 [12:09<1:27:08, 1.04s/it, tok/s=13782.5]
Inference: 15%|█▌ | 905/5920 [12:09<1:27:08, 1.04s/it, tok/s=13772.7]
Inference: 15%|█▌ | 906/5920 [12:10<1:09:39, 1.20it/s, tok/s=13772.7]
Inference: 15%|█▌ | 906/5920 [12:10<1:09:39, 1.20it/s, tok/s=13787.5]
Inference: 15%|█▌ | 907/5920 [12:10<59:03, 1.41it/s, tok/s=13787.5]
Inference: 15%|█▌ | 907/5920 [12:10<59:03, 1.41it/s, tok/s=13792.9]
Inference: 15%|█▌ | 908/5920 [12:12<1:20:00, 1.04it/s, tok/s=13792.9]
Inference: 15%|█▌ | 908/5920 [12:12<1:20:00, 1.04it/s, tok/s=13782.8]
Inference: 15%|█▌ | 909/5920 [12:13<1:34:27, 1.13s/it, tok/s=13782.8]
Inference: 15%|█▌ | 909/5920 [12:13<1:34:27, 1.13s/it, tok/s=13768.1]
Inference: 15%|█▌ | 910/5920 [12:13<1:34:25, 1.13s/it, tok/s=13792.1]
Inference: 15%|█▌ | 911/5920 [12:15<1:24:06, 1.01s/it, tok/s=13792.1]
Inference: 15%|█▌ | 911/5920 [12:15<1:24:06, 1.01s/it, tok/s=13774.6]
Inference: 15%|█▌ | 912/5920 [12:16<1:35:44, 1.15s/it, tok/s=13774.6]
Inference: 15%|█▌ | 912/5920 [12:16<1:35:44, 1.15s/it, tok/s=13756.7]
Inference: 15%|█▌ | 913/5920 [12:17<1:25:20, 1.02s/it, tok/s=13756.7]
Inference: 15%|█▌ | 913/5920 [12:17<1:25:20, 1.02s/it, tok/s=13757.6]
Inference: 15%|█▌ | 914/5920 [12:19<1:37:11, 1.16s/it, tok/s=13757.6]
Inference: 15%|█▌ | 914/5920 [12:19<1:37:11, 1.16s/it, tok/s=13743.6]
Inference: 15%|█▌ | 915/5920 [12:20<1:46:39, 1.28s/it, tok/s=13743.6]
Inference: 15%|█▌ | 915/5920 [12:20<1:46:39, 1.28s/it, tok/s=13728.1]
Inference: 15%|█▌ | 916/5920 [12:23<2:33:06, 1.84s/it, tok/s=13728.1]
Inference: 15%|█▌ | 916/5920 [12:23<2:33:06, 1.84s/it, tok/s=13682.0]
Inference: 15%|█▌ | 917/5920 [12:23<2:33:05, 1.84s/it, tok/s=13694.8]
Inference: 16%|█▌ | 918/5920 [12:26<2:06:42, 1.52s/it, tok/s=13694.8]
Inference: 16%|█▌ | 918/5920 [12:26<2:06:42, 1.52s/it, tok/s=13670.6]
Inference: 16%|█▌ | 919/5920 [12:26<1:44:51, 1.26s/it, tok/s=13670.6]
Inference: 16%|█▌ | 919/5920 [12:26<1:44:51, 1.26s/it, tok/s=13676.1]
Inference: 16%|█▌ | 920/5920 [12:26<1:26:27, 1.04s/it, tok/s=13676.1]
Inference: 16%|█▌ | 920/5920 [12:26<1:26:27, 1.04s/it, tok/s=13679.6]
Inference: 16%|█▌ | 921/5920 [12:27<1:05:27, 1.27it/s, tok/s=13679.6]
Inference: 16%|█▌ | 921/5920 [12:27<1:05:27, 1.27it/s, tok/s=13702.0]
Inference: 16%|█▌ | 922/5920 [12:29<1:33:27, 1.12s/it, tok/s=13702.0]
Inference: 16%|█▌ | 922/5920 [12:29<1:33:27, 1.12s/it, tok/s=13685.6]
Inference: 16%|█▌ | 923/5920 [12:30<1:30:05, 1.08s/it, tok/s=13685.6]
Inference: 16%|█▌ | 923/5920 [12:30<1:30:05, 1.08s/it, tok/s=13679.7]
Inference: 16%|█▌ | 924/5920 [12:30<1:19:28, 1.05it/s, tok/s=13679.7]
Inference: 16%|█▌ | 924/5920 [12:30<1:19:28, 1.05it/s, tok/s=13681.2]
Inference: 16%|█▌ | 925/5920 [12:32<1:36:23, 1.16s/it, tok/s=13681.2]
Inference: 16%|█▌ | 925/5920 [12:32<1:36:23, 1.16s/it, tok/s=13665.1]
Inference: 16%|█▌ | 926/5920 [12:32<1:15:02, 1.11it/s, tok/s=13665.1]
Inference: 16%|█▌ | 926/5920 [12:32<1:15:02, 1.11it/s, tok/s=13680.1]
Inference: 16%|█▌ | 927/5920 [12:32<57:04, 1.46it/s, tok/s=13680.1]
Inference: 16%|█▌ | 927/5920 [12:32<57:04, 1.46it/s, tok/s=13684.6]
Inference: 16%|█▌ | 928/5920 [12:34<1:12:25, 1.15it/s, tok/s=13684.6]
Inference: 16%|█▌ | 928/5920 [12:34<1:12:25, 1.15it/s, tok/s=13684.1]
Inference: 16%|█▌ | 929/5920 [12:34<57:13, 1.45it/s, tok/s=13684.1]
Inference: 16%|█▌ | 929/5920 [12:34<57:13, 1.45it/s, tok/s=13708.3]
Inference: 16%|█▌ | 930/5920 [12:36<1:27:07, 1.05s/it, tok/s=13708.3]
Inference: 16%|█▌ | 930/5920 [12:36<1:27:07, 1.05s/it, tok/s=13690.6]
Inference: 16%|█▌ | 931/5920 [12:36<1:18:50, 1.05it/s, tok/s=13690.6]
Inference: 16%|█▌ | 931/5920 [12:36<1:18:50, 1.05it/s, tok/s=13685.9]
Inference: 16%|█▌ | 932/5920 [12:38<1:33:26, 1.12s/it, tok/s=13685.9]
Inference: 16%|█▌ | 932/5920 [12:38<1:33:26, 1.12s/it, tok/s=13667.6]
Inference: 16%|█▌ | 933/5920 [12:38<1:33:25, 1.12s/it, tok/s=13705.0]
Inference: 16%|█▌ | 934/5920 [12:39<1:08:49, 1.21it/s, tok/s=13705.0]
Inference: 16%|█▌ | 934/5920 [12:39<1:08:49, 1.21it/s, tok/s=13705.9]
Inference: 16%|█▌ | 935/5920 [12:40<1:05:43, 1.26it/s, tok/s=13705.9]
Inference: 16%|█▌ | 935/5920 [12:40<1:05:43, 1.26it/s, tok/s=13710.5]
Inference: 16%|█▌ | 936/5920 [12:40<59:09, 1.40it/s, tok/s=13710.5]
Inference: 16%|█▌ | 936/5920 [12:40<59:09, 1.40it/s, tok/s=13722.0]
Inference: 16%|█▌ | 937/5920 [12:41<1:07:37, 1.23it/s, tok/s=13722.0]
Inference: 16%|█▌ | 937/5920 [12:41<1:07:37, 1.23it/s, tok/s=13741.5]
Inference: 16%|█▌ | 938/5920 [12:42<58:56, 1.41it/s, tok/s=13741.5]
Inference: 16%|█▌ | 938/5920 [12:42<58:56, 1.41it/s, tok/s=13739.9]
Inference: 16%|█▌ | 939/5920 [12:42<56:57, 1.46it/s, tok/s=13739.9]
Inference: 16%|█▌ | 939/5920 [12:42<56:57, 1.46it/s, tok/s=13741.2]
Inference: 16%|█▌ | 940/5920 [12:42<56:56, 1.46it/s, tok/s=13754.4]
Inference: 16%|█▌ | 941/5920 [12:42<56:56, 1.46it/s, tok/s=13768.1]
Inference: 16%|█▌ | 942/5920 [12:42<28:34, 2.90it/s, tok/s=13768.1]
Inference: 16%|█▌ | 942/5920 [12:42<28:34, 2.90it/s, tok/s=13805.7]
Inference: 16%|█▌ | 943/5920 [12:43<27:13, 3.05it/s, tok/s=13805.7]
Inference: 16%|█▌ | 943/5920 [12:43<27:13, 3.05it/s, tok/s=13820.1]
Inference: 16%|█▌ | 944/5920 [12:43<33:52, 2.45it/s, tok/s=13820.1]
Inference: 16%|█▌ | 944/5920 [12:43<33:52, 2.45it/s, tok/s=13814.4]
Inference: 16%|█▌ | 945/5920 [12:46<1:16:05, 1.09it/s, tok/s=13814.4]
Inference: 16%|█▌ | 945/5920 [12:46<1:16:05, 1.09it/s, tok/s=13809.1]
Inference: 16%|█▌ | 946/5920 [12:46<59:28, 1.39it/s, tok/s=13809.1]
Inference: 16%|█▌ | 946/5920 [12:46<59:28, 1.39it/s, tok/s=13820.0]
Inference: 16%|█▌ | 947/5920 [12:47<55:38, 1.49it/s, tok/s=13820.0]
Inference: 16%|█▌ | 947/5920 [12:47<55:38, 1.49it/s, tok/s=13823.6]
Inference: 16%|█▌ | 948/5920 [12:48<1:04:47, 1.28it/s, tok/s=13823.6]
Inference: 16%|█▌ | 948/5920 [12:48<1:04:47, 1.28it/s, tok/s=13843.8]
Inference: 16%|█▌ | 949/5920 [12:48<52:22, 1.58it/s, tok/s=13843.8]
Inference: 16%|█▌ | 949/5920 [12:48<52:22, 1.58it/s, tok/s=13861.7]
Inference: 16%|█▌ | 950/5920 [12:49<1:01:12, 1.35it/s, tok/s=13861.7]
Inference: 16%|█▌ | 950/5920 [12:49<1:01:12, 1.35it/s, tok/s=13873.1]
Inference: 16%|█▌ | 951/5920 [12:50<1:06:55, 1.24it/s, tok/s=13873.1]
Inference: 16%|█▌ | 951/5920 [12:50<1:06:55, 1.24it/s, tok/s=13883.1]
Inference: 16%|█▌ | 952/5920 [12:50<1:00:44, 1.36it/s, tok/s=13883.1]
Inference: 16%|█▌ | 952/5920 [12:50<1:00:44, 1.36it/s, tok/s=13886.9]
Inference: 16%|█▌ | 953/5920 [12:51<50:30, 1.64it/s, tok/s=13886.9]
Inference: 16%|█▌ | 953/5920 [12:51<50:30, 1.64it/s, tok/s=13894.0]
Inference: 16%|█▌ | 954/5920 [12:53<1:19:18, 1.04it/s, tok/s=13894.0]
Inference: 16%|█▌ | 954/5920 [12:53<1:19:18, 1.04it/s, tok/s=13873.9]
Inference: 16%|█▌ | 955/5920 [12:53<1:19:17, 1.04it/s, tok/s=13879.6]
Inference: 16%|█▌ | 956/5920 [12:53<45:06, 1.83it/s, tok/s=13879.6]
Inference: 16%|█▌ | 956/5920 [12:53<45:06, 1.83it/s, tok/s=13892.9]
Inference: 16%|█▌ | 957/5920 [12:54<1:07:09, 1.23it/s, tok/s=13892.9]
Inference: 16%|█▌ | 957/5920 [12:54<1:07:09, 1.23it/s, tok/s=13870.1]
Inference: 16%|█▌ | 958/5920 [12:54<52:48, 1.57it/s, tok/s=13870.1]
Inference: 16%|█▌ | 958/5920 [12:54<52:48, 1.57it/s, tok/s=13881.3]
Inference: 16%|█▌ | 959/5920 [12:55<42:05, 1.96it/s, tok/s=13881.3]
Inference: 16%|█▌ | 959/5920 [12:55<42:05, 1.96it/s, tok/s=13917.4]
Inference: 16%|█▌ | 960/5920 [12:55<42:04, 1.96it/s, tok/s=13956.4]
Inference: 16%|█▌ | 961/5920 [12:56<49:39, 1.66it/s, tok/s=13956.4]
Inference: 16%|█▌ | 961/5920 [12:56<49:39, 1.66it/s, tok/s=13938.2]
Inference: 16%|█▋ | 962/5920 [12:57<48:03, 1.72it/s, tok/s=13938.2]
Inference: 16%|█▋ | 962/5920 [12:57<48:03, 1.72it/s, tok/s=13946.9]
Inference: 16%|█▋ | 963/5920 [12:57<51:37, 1.60it/s, tok/s=13946.9]
Inference: 16%|█▋ | 963/5920 [12:57<51:37, 1.60it/s, tok/s=13939.9]
Inference: 16%|█▋ | 964/5920 [12:58<46:04, 1.79it/s, tok/s=13939.9]
Inference: 16%|█▋ | 964/5920 [12:58<46:04, 1.79it/s, tok/s=13956.3]
Inference: 16%|█▋ | 965/5920 [12:59<1:02:58, 1.31it/s, tok/s=13956.3]
Inference: 16%|█▋ | 965/5920 [12:59<1:02:58, 1.31it/s, tok/s=13946.6]
Inference: 16%|█▋ | 966/5920 [13:01<1:25:35, 1.04s/it, tok/s=13946.6]
Inference: 16%|█▋ | 966/5920 [13:01<1:25:35, 1.04s/it, tok/s=13925.7]
Inference: 16%|█▋ | 967/5920 [13:01<1:05:15, 1.27it/s, tok/s=13925.7]
Inference: 16%|█▋ | 967/5920 [13:01<1:05:15, 1.27it/s, tok/s=13938.8]
Inference: 16%|█▋ | 968/5920 [13:02<1:07:21, 1.23it/s, tok/s=13938.8]
Inference: 16%|█▋ | 968/5920 [13:02<1:07:21, 1.23it/s, tok/s=13932.9]
Inference: 16%|█▋ | 969/5920 [13:02<52:17, 1.58it/s, tok/s=13932.9]
Inference: 16%|█▋ | 969/5920 [13:02<52:17, 1.58it/s, tok/s=13951.8]
Inference: 16%|█▋ | 970/5920 [13:03<50:11, 1.64it/s, tok/s=13951.8]
Inference: 16%|█▋ | 970/5920 [13:03<50:11, 1.64it/s, tok/s=13958.2]
Inference: 16%|█▋ | 971/5920 [13:03<42:32, 1.94it/s, tok/s=13958.2]
Inference: 16%|█▋ | 971/5920 [13:03<42:32, 1.94it/s, tok/s=13965.8]
Inference: 16%|█▋ | 972/5920 [13:04<1:00:14, 1.37it/s, tok/s=13965.8]
Inference: 16%|█▋ | 972/5920 [13:04<1:00:14, 1.37it/s, tok/s=13966.1]
Inference: 16%|█▋ | 973/5920 [13:04<48:43, 1.69it/s, tok/s=13966.1]
Inference: 16%|█▋ | 973/5920 [13:04<48:43, 1.69it/s, tok/s=13972.1]
Inference: 16%|█▋ | 974/5920 [13:06<1:05:11, 1.26it/s, tok/s=13972.1]
Inference: 16%|█▋ | 974/5920 [13:06<1:05:11, 1.26it/s, tok/s=13974.9]
Inference: 16%|█▋ | 975/5920 [13:07<1:18:46, 1.05it/s, tok/s=13974.9]
Inference: 16%|█▋ | 975/5920 [13:07<1:18:46, 1.05it/s, tok/s=13964.3]
Inference: 16%|█▋ | 976/5920 [13:07<1:18:45, 1.05it/s, tok/s=13976.9]
Inference: 17%|█▋ | 977/5920 [13:07<45:51, 1.80it/s, tok/s=13976.9]
Inference: 17%|█▋ | 977/5920 [13:07<45:51, 1.80it/s, tok/s=13981.4]
Inference: 17%|█▋ | 978/5920 [13:08<42:43, 1.93it/s, tok/s=13981.4]
Inference: 17%|█▋ | 978/5920 [13:08<42:43, 1.93it/s, tok/s=13982.8]
Inference: 17%|█▋ | 979/5920 [13:08<35:15, 2.34it/s, tok/s=13982.8]
Inference: 17%|█▋ | 979/5920 [13:08<35:15, 2.34it/s, tok/s=13987.3]
Inference: 17%|█▋ | 980/5920 [13:08<29:47, 2.76it/s, tok/s=13987.3]
Inference: 17%|█▋ | 980/5920 [13:08<29:47, 2.76it/s, tok/s=13992.8]
Inference: 17%|█▋ | 981/5920 [13:09<37:55, 2.17it/s, tok/s=13992.8]
Inference: 17%|█▋ | 981/5920 [13:09<37:55, 2.17it/s, tok/s=13993.7]
Inference: 17%|█▋ | 982/5920 [13:09<40:39, 2.02it/s, tok/s=13993.7]
Inference: 17%|█▋ | 982/5920 [13:09<40:39, 2.02it/s, tok/s=13991.0]
Inference: 17%|█▋ | 983/5920 [13:09<31:55, 2.58it/s, tok/s=13991.0]
Inference: 17%|█▋ | 983/5920 [13:09<31:55, 2.58it/s, tok/s=14002.3]
Inference: 17%|█▋ | 984/5920 [13:11<55:02, 1.49it/s, tok/s=14002.3]
Inference: 17%|█▋ | 984/5920 [13:11<55:02, 1.49it/s, tok/s=13986.0]
Inference: 17%|█▋ | 985/5920 [13:12<1:15:56, 1.08it/s, tok/s=13986.0]
Inference: 17%|█▋ | 985/5920 [13:12<1:15:56, 1.08it/s, tok/s=13965.6]
Inference: 17%|█▋ | 986/5920 [13:12<58:25, 1.41it/s, tok/s=13965.6]
Inference: 17%|█▋ | 986/5920 [13:12<58:25, 1.41it/s, tok/s=13973.2]
Inference: 17%|█▋ | 987/5920 [13:14<1:18:21, 1.05it/s, tok/s=13973.2]
Inference: 17%|█▋ | 987/5920 [13:14<1:18:21, 1.05it/s, tok/s=13952.5]
Inference: 17%|█▋ | 988/5920 [13:14<1:18:20, 1.05it/s, tok/s=13961.1]
Inference: 17%|█▋ | 989/5920 [13:14<52:26, 1.57it/s, tok/s=13961.1]
Inference: 17%|█▋ | 989/5920 [13:14<52:26, 1.57it/s, tok/s=13958.9]
Inference: 17%|█▋ | 990/5920 [13:15<51:19, 1.60it/s, tok/s=13958.9]
Inference: 17%|█▋ | 990/5920 [13:15<51:19, 1.60it/s, tok/s=13956.0]
Inference: 17%|█▋ | 991/5920 [13:16<59:51, 1.37it/s, tok/s=13956.0]
Inference: 17%|█▋ | 991/5920 [13:16<59:51, 1.37it/s, tok/s=13951.0]
Inference: 17%|█▋ | 992/5920 [13:16<45:49, 1.79it/s, tok/s=13951.0]
Inference: 17%|█▋ | 992/5920 [13:16<45:49, 1.79it/s, tok/s=13972.7]
Inference: 17%|█▋ | 993/5920 [13:17<43:59, 1.87it/s, tok/s=13972.7]
Inference: 17%|█▋ | 993/5920 [13:17<43:59, 1.87it/s, tok/s=13982.5]
Inference: 17%|█▋ | 994/5920 [13:17<36:31, 2.25it/s, tok/s=13982.5]
Inference: 17%|█▋ | 994/5920 [13:17<36:31, 2.25it/s, tok/s=13985.6]
Inference: 17%|█▋ | 995/5920 [13:17<36:31, 2.25it/s, tok/s=13990.8]
Inference: 17%|█▋ | 996/5920 [13:17<29:02, 2.83it/s, tok/s=13990.8]
Inference: 17%|█▋ | 996/5920 [13:17<29:02, 2.83it/s, tok/s=13992.7]
Inference: 17%|█▋ | 997/5920 [13:17<29:02, 2.83it/s, tok/s=13998.7]
Inference: 17%|█▋ | 998/5920 [13:18<26:43, 3.07it/s, tok/s=13998.7]
Inference: 17%|█▋ | 998/5920 [13:18<26:43, 3.07it/s, tok/s=14006.4]
Inference: 17%|█▋ | 999/5920 [13:18<27:06, 3.03it/s, tok/s=14006.4]
Inference: 17%|█▋ | 999/5920 [13:18<27:06, 3.03it/s, tok/s=14015.2]
Inference: 17%|█▋ | 1000/5920 [13:18<25:39, 3.20it/s, tok/s=14015.2]
Inference: 17%|█▋ | 1000/5920 [13:18<25:39, 3.20it/s, tok/s=14038.5]
Inference: 17%|█▋ | 1001/5920 [13:20<59:53, 1.37it/s, tok/s=14038.5]
Inference: 17%|█▋ | 1001/5920 [13:20<59:53, 1.37it/s, tok/s=14016.5]
Inference: 17%|█▋ | 1002/5920 [13:24<2:03:55, 1.51s/it, tok/s=14016.5]
Inference: 17%|█▋ | 1002/5920 [13:24<2:03:55, 1.51s/it, tok/s=13961.4]
Inference: 17%|█▋ | 1003/5920 [13:25<1:55:14, 1.41s/it, tok/s=13961.4]
Inference: 17%|█▋ | 1003/5920 [13:25<1:55:14, 1.41s/it, tok/s=13948.7]
Inference: 17%|█▋ | 1004/5920 [13:26<1:29:32, 1.09s/it, tok/s=13948.7]
Inference: 17%|█▋ | 1004/5920 [13:26<1:29:32, 1.09s/it, tok/s=13965.3]
Inference: 17%|█▋ | 1005/5920 [13:26<1:25:15, 1.04s/it, tok/s=13965.3]
Inference: 17%|█▋ | 1005/5920 [13:26<1:25:15, 1.04s/it, tok/s=13967.8]
Inference: 17%|█▋ | 1006/5920 [13:31<2:37:16, 1.92s/it, tok/s=13967.8]
Inference: 17%|█▋ | 1006/5920 [13:31<2:37:16, 1.92s/it, tok/s=13905.4]
Inference: 17%|█▋ | 1007/5920 [13:33<2:56:05, 2.15s/it, tok/s=13905.4]
Inference: 17%|█▋ | 1007/5920 [13:33<2:56:05, 2.15s/it, tok/s=13876.2]
Inference: 17%|█▋ | 1008/5920 [13:36<3:15:51, 2.39s/it, tok/s=13876.2]
Inference: 17%|█▋ | 1008/5920 [13:36<3:15:51, 2.39s/it, tok/s=13833.4]
Inference: 17%|█▋ | 1009/5920 [13:36<2:23:03, 1.75s/it, tok/s=13833.4]
Inference: 17%|█▋ | 1009/5920 [13:36<2:23:03, 1.75s/it, tok/s=13842.9]
Inference: 17%|█▋ | 1010/5920 [13:36<2:23:01, 1.75s/it, tok/s=13850.9]
Inference: 17%|█▋ | 1011/5920 [13:37<1:25:45, 1.05s/it, tok/s=13850.9]
Inference: 17%|█▋ | 1011/5920 [13:37<1:25:45, 1.05s/it, tok/s=13853.7]
Inference: 17%|█▋ | 1012/5920 [13:37<1:09:46, 1.17it/s, tok/s=13853.7]
Inference: 17%|█▋ | 1012/5920 [13:37<1:09:46, 1.17it/s, tok/s=13858.1]
Inference: 17%|█▋ | 1013/5920 [13:37<1:09:45, 1.17it/s, tok/s=13883.2]
Inference: 17%|█▋ | 1014/5920 [13:41<1:49:31, 1.34s/it, tok/s=13883.2]
Inference: 17%|█▋ | 1014/5920 [13:41<1:49:31, 1.34s/it, tok/s=13832.5]
Inference: 17%|█▋ | 1015/5920 [13:42<1:31:46, 1.12s/it, tok/s=13832.5]
Inference: 17%|█▋ | 1015/5920 [13:42<1:31:46, 1.12s/it, tok/s=13838.8]
Inference: 17%|█▋ | 1016/5920 [13:43<1:28:42, 1.09s/it, tok/s=13838.8]
Inference: 17%|█▋ | 1016/5920 [13:43<1:28:42, 1.09s/it, tok/s=13832.0]
Inference: 17%|█▋ | 1017/5920 [13:44<1:26:22, 1.06s/it, tok/s=13832.0]
Inference: 17%|█▋ | 1017/5920 [13:44<1:26:22, 1.06s/it, tok/s=13836.9]
Inference: 17%|█▋ | 1018/5920 [13:44<1:14:08, 1.10it/s, tok/s=13836.9]
Inference: 17%|█▋ | 1018/5920 [13:44<1:14:08, 1.10it/s, tok/s=13837.8]
Inference: 17%|█▋ | 1019/5920 [13:45<1:24:41, 1.04s/it, tok/s=13837.8]
Inference: 17%|█▋ | 1019/5920 [13:45<1:24:41, 1.04s/it, tok/s=13837.9]
Inference: 17%|█▋ | 1020/5920 [13:46<1:23:33, 1.02s/it, tok/s=13837.9]
Inference: 17%|█▋ | 1020/5920 [13:46<1:23:33, 1.02s/it, tok/s=13828.1]
Inference: 17%|█▋ | 1021/5920 [13:46<1:23:32, 1.02s/it, tok/s=13841.5]
Inference: 17%|█▋ | 1022/5920 [13:47<52:33, 1.55it/s, tok/s=13841.5]
Inference: 17%|█▋ | 1022/5920 [13:47<52:33, 1.55it/s, tok/s=13856.9]
Inference: 17%|█▋ | 1023/5920 [13:47<43:13, 1.89it/s, tok/s=13856.9]
Inference: 17%|█▋ | 1023/5920 [13:47<43:13, 1.89it/s, tok/s=13883.1]
Inference: 17%|█▋ | 1024/5920 [13:47<40:00, 2.04it/s, tok/s=13883.1]
Inference: 17%|█▋ | 1024/5920 [13:47<40:00, 2.04it/s, tok/s=13883.8]
Inference: 17%|█▋ | 1025/5920 [13:47<40:00, 2.04it/s, tok/s=13891.9]
Inference: 17%|█▋ | 1026/5920 [13:47<39:59, 2.04it/s, tok/s=13900.6]
Inference: 17%|█▋ | 1027/5920 [13:48<24:48, 3.29it/s, tok/s=13900.6]
Inference: 17%|█▋ | 1027/5920 [13:48<24:48, 3.29it/s, tok/s=13912.0]
Inference: 17%|█▋ | 1028/5920 [13:48<22:42, 3.59it/s, tok/s=13912.0]
Inference: 17%|█▋ | 1028/5920 [13:48<22:42, 3.59it/s, tok/s=13931.4]
Inference: 17%|█▋ | 1029/5920 [13:49<32:41, 2.49it/s, tok/s=13931.4]
Inference: 17%|█▋ | 1029/5920 [13:49<32:41, 2.49it/s, tok/s=13931.1]
Inference: 17%|█▋ | 1030/5920 [13:49<32:40, 2.49it/s, tok/s=13931.1]
Inference: 17%|█▋ | 1030/5920 [13:49<32:40, 2.49it/s, tok/s=13934.4]
Inference: 17%|█▋ | 1031/5920 [13:49<26:41, 3.05it/s, tok/s=13934.4]
Inference: 17%|█▋ | 1031/5920 [13:49<26:41, 3.05it/s, tok/s=13961.4]
Inference: 17%|█▋ | 1032/5920 [13:50<34:41, 2.35it/s, tok/s=13961.4]
Inference: 17%|█▋ | 1032/5920 [13:50<34:41, 2.35it/s, tok/s=13957.8]
Inference: 17%|█▋ | 1033/5920 [13:50<35:29, 2.30it/s, tok/s=13957.8]
Inference: 17%|█▋ | 1033/5920 [13:50<35:29, 2.30it/s, tok/s=13958.5]
Inference: 17%|█▋ | 1034/5920 [13:50<35:28, 2.30it/s, tok/s=13977.1]
Inference: 17%|█▋ | 1035/5920 [13:51<32:27, 2.51it/s, tok/s=13977.1]
Inference: 17%|█▋ | 1035/5920 [13:51<32:27, 2.51it/s, tok/s=13988.5]
Inference: 18%|█▊ | 1036/5920 [13:51<29:46, 2.73it/s, tok/s=13988.5]
Inference: 18%|█▊ | 1036/5920 [13:51<29:46, 2.73it/s, tok/s=14001.0]
Inference: 18%|█▊ | 1037/5920 [13:52<28:32, 2.85it/s, tok/s=14001.0]
Inference: 18%|█▊ | 1037/5920 [13:52<28:32, 2.85it/s, tok/s=14005.2]
Inference: 18%|█▊ | 1038/5920 [13:53<51:51, 1.57it/s, tok/s=14005.2]
Inference: 18%|█▊ | 1038/5920 [13:53<51:51, 1.57it/s, tok/s=14010.4]
Inference: 18%|█▊ | 1039/5920 [13:54<52:07, 1.56it/s, tok/s=14010.4]
Inference: 18%|█▊ | 1039/5920 [13:54<52:07, 1.56it/s, tok/s=14012.0]
Inference: 18%|█▊ | 1040/5920 [13:55<58:48, 1.38it/s, tok/s=14012.0]
Inference: 18%|█▊ | 1040/5920 [13:55<58:48, 1.38it/s, tok/s=14022.1]
Inference: 18%|█▊ | 1041/5920 [13:55<44:27, 1.83it/s, tok/s=14022.1]
Inference: 18%|█▊ | 1041/5920 [13:55<44:27, 1.83it/s, tok/s=14033.5]
Inference: 18%|█▊ | 1042/5920 [13:55<44:27, 1.83it/s, tok/s=14042.7]
Inference: 18%|█▊ | 1043/5920 [13:55<26:29, 3.07it/s, tok/s=14042.7]
Inference: 18%|█▊ | 1043/5920 [13:55<26:29, 3.07it/s, tok/s=14061.1]
Inference: 18%|█▊ | 1044/5920 [13:55<26:29, 3.07it/s, tok/s=14074.4]
Inference: 18%|█▊ | 1045/5920 [13:55<25:23, 3.20it/s, tok/s=14074.4]
Inference: 18%|█▊ | 1045/5920 [13:55<25:23, 3.20it/s, tok/s=14076.5]
Inference: 18%|█▊ | 1046/5920 [13:56<27:06, 3.00it/s, tok/s=14076.5]
Inference: 18%|█▊ | 1046/5920 [13:56<27:06, 3.00it/s, tok/s=14078.1]
Inference: 18%|█▊ | 1047/5920 [13:57<37:56, 2.14it/s, tok/s=14078.1]
Inference: 18%|█▊ | 1047/5920 [13:57<37:56, 2.14it/s, tok/s=14079.9]
Inference: 18%|█▊ | 1048/5920 [13:58<1:03:15, 1.28it/s, tok/s=14079.9]
Inference: 18%|█▊ | 1048/5920 [13:58<1:03:15, 1.28it/s, tok/s=14068.6]
Inference: 18%|█▊ | 1049/5920 [13:59<1:01:46, 1.31it/s, tok/s=14068.6]
Inference: 18%|█▊ | 1049/5920 [13:59<1:01:46, 1.31it/s, tok/s=14089.5]
Inference: 18%|█▊ | 1050/5920 [13:59<51:49, 1.57it/s, tok/s=14089.5]
Inference: 18%|█▊ | 1050/5920 [13:59<51:49, 1.57it/s, tok/s=14094.1]
Inference: 18%|█▊ | 1051/5920 [14:00<41:13, 1.97it/s, tok/s=14094.1]
Inference: 18%|█▊ | 1051/5920 [14:00<41:13, 1.97it/s, tok/s=14127.0]
Inference: 18%|█▊ | 1052/5920 [14:00<36:28, 2.22it/s, tok/s=14127.0]
Inference: 18%|█▊ | 1052/5920 [14:00<36:28, 2.22it/s, tok/s=14129.8]
Inference: 18%|█▊ | 1053/5920 [14:00<36:27, 2.22it/s, tok/s=14145.6]
Inference: 18%|█▊ | 1054/5920 [14:00<24:18, 3.34it/s, tok/s=14145.6]
Inference: 18%|█▊ | 1054/5920 [14:00<24:18, 3.34it/s, tok/s=14149.9]
Inference: 18%|█▊ | 1055/5920 [14:01<30:03, 2.70it/s, tok/s=14149.9]
Inference: 18%|█▊ | 1055/5920 [14:01<30:03, 2.70it/s, tok/s=14156.2]
Inference: 18%|█▊ | 1056/5920 [14:01<26:09, 3.10it/s, tok/s=14156.2]
Inference: 18%|█▊ | 1056/5920 [14:01<26:09, 3.10it/s, tok/s=14173.1]
Inference: 18%|█▊ | 1057/5920 [14:02<43:01, 1.88it/s, tok/s=14173.1]
Inference: 18%|█▊ | 1057/5920 [14:02<43:01, 1.88it/s, tok/s=14174.4]
Inference: 18%|█▊ | 1058/5920 [14:03<54:26, 1.49it/s, tok/s=14174.4]
Inference: 18%|█▊ | 1058/5920 [14:03<54:26, 1.49it/s, tok/s=14170.7]
Inference: 18%|█▊ | 1059/5920 [14:04<1:04:54, 1.25it/s, tok/s=14170.7]
Inference: 18%|█▊ | 1059/5920 [14:04<1:04:54, 1.25it/s, tok/s=14162.1]
Inference: 18%|█▊ | 1060/5920 [14:06<1:23:03, 1.03s/it, tok/s=14162.1]
Inference: 18%|█▊ | 1060/5920 [14:06<1:23:03, 1.03s/it, tok/s=14158.5]
Inference: 18%|█▊ | 1061/5920 [14:07<1:15:22, 1.07it/s, tok/s=14158.5]
Inference: 18%|█▊ | 1061/5920 [14:07<1:15:22, 1.07it/s, tok/s=14169.9]
Inference: 18%|█▊ | 1062/5920 [14:07<1:16:27, 1.06it/s, tok/s=14169.9]
Inference: 18%|█▊ | 1062/5920 [14:07<1:16:27, 1.06it/s, tok/s=14160.9]
Inference: 18%|█▊ | 1063/5920 [14:08<1:16:26, 1.06it/s, tok/s=14180.3]
Inference: 18%|█▊ | 1064/5920 [14:08<59:35, 1.36it/s, tok/s=14180.3]
Inference: 18%|█▊ | 1064/5920 [14:08<59:35, 1.36it/s, tok/s=14180.9]
Inference: 18%|█▊ | 1065/5920 [14:10<1:19:17, 1.02it/s, tok/s=14180.9]
Inference: 18%|█▊ | 1065/5920 [14:10<1:19:17, 1.02it/s, tok/s=14171.4]
Inference: 18%|█▊ | 1066/5920 [14:12<1:29:57, 1.11s/it, tok/s=14171.4]
Inference: 18%|█▊ | 1066/5920 [14:12<1:29:57, 1.11s/it, tok/s=14169.0]
Inference: 18%|█▊ | 1067/5920 [14:12<1:16:31, 1.06it/s, tok/s=14169.0]
Inference: 18%|█▊ | 1067/5920 [14:12<1:16:31, 1.06it/s, tok/s=14168.7]
Inference: 18%|█▊ | 1068/5920 [14:13<1:21:47, 1.01s/it, tok/s=14168.7]
Inference: 18%|█▊ | 1068/5920 [14:13<1:21:47, 1.01s/it, tok/s=14158.8]
Inference: 18%|█▊ | 1069/5920 [14:16<1:49:29, 1.35s/it, tok/s=14158.8]
Inference: 18%|█▊ | 1069/5920 [14:16<1:49:29, 1.35s/it, tok/s=14135.9]
Inference: 18%|█▊ | 1070/5920 [14:16<1:24:25, 1.04s/it, tok/s=14135.9]
Inference: 18%|█▊ | 1070/5920 [14:16<1:24:25, 1.04s/it, tok/s=14148.5]
Inference: 18%|█▊ | 1071/5920 [14:17<1:18:38, 1.03it/s, tok/s=14148.5]
Inference: 18%|█▊ | 1071/5920 [14:17<1:18:38, 1.03it/s, tok/s=14145.5]
Inference: 18%|█▊ | 1072/5920 [14:18<1:20:41, 1.00it/s, tok/s=14145.5]
Inference: 18%|█▊ | 1072/5920 [14:18<1:20:41, 1.00it/s, tok/s=14138.2]
Inference: 18%|█▊ | 1073/5920 [14:18<1:12:00, 1.12it/s, tok/s=14138.2]
Inference: 18%|█▊ | 1073/5920 [14:18<1:12:00, 1.12it/s, tok/s=14153.0]
Inference: 18%|█▊ | 1074/5920 [14:19<1:04:33, 1.25it/s, tok/s=14153.0]
Inference: 18%|█▊ | 1074/5920 [14:19<1:04:33, 1.25it/s, tok/s=14159.2]
Inference: 18%|█▊ | 1075/5920 [14:20<1:23:03, 1.03s/it, tok/s=14159.2]
Inference: 18%|█▊ | 1075/5920 [14:20<1:23:03, 1.03s/it, tok/s=14150.8]
Inference: 18%|█▊ | 1076/5920 [14:23<1:49:45, 1.36s/it, tok/s=14150.8]
Inference: 18%|█▊ | 1076/5920 [14:23<1:49:45, 1.36s/it, tok/s=14142.6]
Inference: 18%|█▊ | 1077/5920 [14:23<1:49:44, 1.36s/it, tok/s=14149.8]
Inference: 18%|█▊ | 1078/5920 [14:23<1:05:57, 1.22it/s, tok/s=14149.8]
Inference: 18%|█▊ | 1078/5920 [14:23<1:05:57, 1.22it/s, tok/s=14162.5]
Inference: 18%|█▊ | 1079/5920 [14:24<1:03:18, 1.27it/s, tok/s=14162.5]
Inference: 18%|█▊ | 1079/5920 [14:24<1:03:18, 1.27it/s, tok/s=14162.3]
Inference: 18%|█▊ | 1080/5920 [14:25<1:08:03, 1.19it/s, tok/s=14162.3]
Inference: 18%|█▊ | 1080/5920 [14:25<1:08:03, 1.19it/s, tok/s=14172.7]
Inference: 18%|█▊ | 1081/5920 [14:26<1:24:18, 1.05s/it, tok/s=14172.7]
Inference: 18%|█▊ | 1081/5920 [14:26<1:24:18, 1.05s/it, tok/s=14168.5]
Inference: 18%|█▊ | 1082/5920 [14:28<1:45:58, 1.31s/it, tok/s=14168.5]
Inference: 18%|█▊ | 1082/5920 [14:28<1:45:58, 1.31s/it, tok/s=14156.5]
Inference: 18%|█▊ | 1083/5920 [14:29<1:40:20, 1.24s/it, tok/s=14156.5]
Inference: 18%|█▊ | 1083/5920 [14:29<1:40:20, 1.24s/it, tok/s=14168.1]
Inference: 18%|█▊ | 1084/5920 [14:30<1:22:15, 1.02s/it, tok/s=14168.1]
Inference: 18%|█▊ | 1084/5920 [14:30<1:22:15, 1.02s/it, tok/s=14177.1]
Inference: 18%|█▊ | 1085/5920 [14:30<1:08:08, 1.18it/s, tok/s=14177.1]
Inference: 18%|█▊ | 1085/5920 [14:30<1:08:08, 1.18it/s, tok/s=14181.2]
Inference: 18%|█▊ | 1086/5920 [14:31<1:03:35, 1.27it/s, tok/s=14181.2]
Inference: 18%|█▊ | 1086/5920 [14:31<1:03:35, 1.27it/s, tok/s=14187.2]
Inference: 18%|█▊ | 1087/5920 [14:33<1:42:11, 1.27s/it, tok/s=14187.2]
Inference: 18%|█▊ | 1087/5920 [14:33<1:42:11, 1.27s/it, tok/s=14169.0]
Inference: 18%|█▊ | 1088/5920 [14:35<1:42:27, 1.27s/it, tok/s=14169.0]
Inference: 18%|█▊ | 1088/5920 [14:35<1:42:27, 1.27s/it, tok/s=14156.6]
Inference: 18%|█▊ | 1089/5920 [14:35<1:31:45, 1.14s/it, tok/s=14156.6]
Inference: 18%|█▊ | 1089/5920 [14:35<1:31:45, 1.14s/it, tok/s=14151.8]
Inference: 18%|█▊ | 1090/5920 [14:37<1:33:56, 1.17s/it, tok/s=14151.8]
Inference: 18%|█▊ | 1090/5920 [14:37<1:33:56, 1.17s/it, tok/s=14155.4]
Inference: 18%|█▊ | 1091/5920 [14:37<1:21:08, 1.01s/it, tok/s=14155.4]
Inference: 18%|█▊ | 1091/5920 [14:37<1:21:08, 1.01s/it, tok/s=14156.4]
Inference: 18%|█▊ | 1092/5920 [14:38<1:23:29, 1.04s/it, tok/s=14156.4]
Inference: 18%|█▊ | 1092/5920 [14:38<1:23:29, 1.04s/it, tok/s=14156.2]
Inference: 18%|█▊ | 1093/5920 [14:40<1:37:11, 1.21s/it, tok/s=14156.2]
Inference: 18%|█▊ | 1093/5920 [14:40<1:37:11, 1.21s/it, tok/s=14147.5]
Inference: 18%|█▊ | 1094/5920 [14:40<1:13:29, 1.09it/s, tok/s=14147.5]
Inference: 18%|█▊ | 1094/5920 [14:40<1:13:29, 1.09it/s, tok/s=14162.2]
Inference: 18%|█▊ | 1095/5920 [14:41<1:08:43, 1.17it/s, tok/s=14162.2]
Inference: 18%|█▊ | 1095/5920 [14:41<1:08:43, 1.17it/s, tok/s=14162.3]
Inference: 19%|█▊ | 1096/5920 [14:42<1:04:09, 1.25it/s, tok/s=14162.3]
Inference: 19%|█▊ | 1096/5920 [14:42<1:04:09, 1.25it/s, tok/s=14174.3]
Inference: 19%|█▊ | 1097/5920 [14:42<50:08, 1.60it/s, tok/s=14174.3]
Inference: 19%|█▊ | 1097/5920 [14:42<50:08, 1.60it/s, tok/s=14199.9]
Inference: 19%|█▊ | 1098/5920 [14:44<1:19:02, 1.02it/s, tok/s=14199.9]
Inference: 19%|█▊ | 1098/5920 [14:44<1:19:02, 1.02it/s, tok/s=14180.7]
Inference: 19%|█▊ | 1099/5920 [14:44<1:02:59, 1.28it/s, tok/s=14180.7]
Inference: 19%|█▊ | 1099/5920 [14:44<1:02:59, 1.28it/s, tok/s=14186.8]
Inference: 19%|█▊ | 1100/5920 [14:44<49:52, 1.61it/s, tok/s=14186.8]
Inference: 19%|█▊ | 1100/5920 [14:44<49:52, 1.61it/s, tok/s=14193.4]
Inference: 19%|█▊ | 1101/5920 [14:44<41:47, 1.92it/s, tok/s=14193.4]
Inference: 19%|█▊ | 1101/5920 [14:44<41:47, 1.92it/s, tok/s=14207.4]
Inference: 19%|█▊ | 1102/5920 [14:46<55:13, 1.45it/s, tok/s=14207.4]
Inference: 19%|█▊ | 1102/5920 [14:46<55:13, 1.45it/s, tok/s=14205.3]
Inference: 19%|█▊ | 1103/5920 [14:46<47:37, 1.69it/s, tok/s=14205.3]
Inference: 19%|█▊ | 1103/5920 [14:46<47:37, 1.69it/s, tok/s=14212.8]
Inference: 19%|█▊ | 1104/5920 [14:47<1:08:28, 1.17it/s, tok/s=14212.8]
Inference: 19%|█▊ | 1104/5920 [14:47<1:08:28, 1.17it/s, tok/s=14207.0]
Inference: 19%|█▊ | 1105/5920 [14:50<1:40:24, 1.25s/it, tok/s=14207.0]
Inference: 19%|█▊ | 1105/5920 [14:50<1:40:24, 1.25s/it, tok/s=14183.0]
Inference: 19%|█▊ | 1106/5920 [14:51<1:35:06, 1.19s/it, tok/s=14183.0]
Inference: 19%|█▊ | 1106/5920 [14:51<1:35:06, 1.19s/it, tok/s=14194.5]
Inference: 19%|█▊ | 1107/5920 [14:52<1:34:36, 1.18s/it, tok/s=14194.5]
Inference: 19%|█▊ | 1107/5920 [14:52<1:34:36, 1.18s/it, tok/s=14188.2]
Inference: 19%|█▊ | 1108/5920 [14:52<1:10:46, 1.13it/s, tok/s=14188.2]
Inference: 19%|█▊ | 1108/5920 [14:52<1:10:46, 1.13it/s, tok/s=14191.8]
Inference: 19%|█▊ | 1109/5920 [14:54<1:28:06, 1.10s/it, tok/s=14191.8]
Inference: 19%|█▊ | 1109/5920 [14:54<1:28:06, 1.10s/it, tok/s=14183.3]
Inference: 19%|█▉ | 1110/5920 [14:54<1:17:08, 1.04it/s, tok/s=14183.3]
Inference: 19%|█▉ | 1110/5920 [14:54<1:17:08, 1.04it/s, tok/s=14183.3]
Inference: 19%|█▉ | 1111/5920 [14:57<1:53:29, 1.42s/it, tok/s=14183.3]
Inference: 19%|█▉ | 1111/5920 [14:57<1:53:29, 1.42s/it, tok/s=14151.4]
Inference: 19%|█▉ | 1112/5920 [14:58<1:58:39, 1.48s/it, tok/s=14151.4]
Inference: 19%|█▉ | 1112/5920 [14:58<1:58:39, 1.48s/it, tok/s=14137.8]
Inference: 19%|█▉ | 1113/5920 [15:00<2:12:57, 1.66s/it, tok/s=14137.8]
Inference: 19%|█▉ | 1113/5920 [15:00<2:12:57, 1.66s/it, tok/s=14128.9]
Inference: 19%|█▉ | 1114/5920 [15:02<2:01:15, 1.51s/it, tok/s=14128.9]
Inference: 19%|█▉ | 1114/5920 [15:02<2:01:15, 1.51s/it, tok/s=14122.5]
Inference: 19%|█▉ | 1115/5920 [15:02<1:34:10, 1.18s/it, tok/s=14122.5]
Inference: 19%|█▉ | 1115/5920 [15:02<1:34:10, 1.18s/it, tok/s=14138.0]
Inference: 19%|█▉ | 1116/5920 [15:02<1:10:13, 1.14it/s, tok/s=14138.0]
Inference: 19%|█▉ | 1116/5920 [15:02<1:10:13, 1.14it/s, tok/s=14141.0]
Inference: 19%|█▉ | 1117/5920 [15:07<2:35:18, 1.94s/it, tok/s=14141.0]
Inference: 19%|█▉ | 1117/5920 [15:07<2:35:18, 1.94s/it, tok/s=14085.5]
Inference: 19%|█▉ | 1118/5920 [15:09<2:46:35, 2.08s/it, tok/s=14085.5]
Inference: 19%|█▉ | 1118/5920 [15:09<2:46:35, 2.08s/it, tok/s=14062.2]
Inference: 19%|█▉ | 1119/5920 [15:09<2:01:59, 1.52s/it, tok/s=14062.2]
Inference: 19%|█▉ | 1119/5920 [15:09<2:01:59, 1.52s/it, tok/s=14074.1]
Inference: 19%|█▉ | 1120/5920 [15:10<1:51:21, 1.39s/it, tok/s=14074.1]
Inference: 19%|█▉ | 1120/5920 [15:10<1:51:21, 1.39s/it, tok/s=14074.1]
Inference: 19%|█▉ | 1121/5920 [15:11<1:28:30, 1.11s/it, tok/s=14074.1]
Inference: 19%|█▉ | 1121/5920 [15:11<1:28:30, 1.11s/it, tok/s=14090.7]
Inference: 19%|█▉ | 1122/5920 [15:11<1:12:37, 1.10it/s, tok/s=14090.7]
Inference: 19%|█▉ | 1122/5920 [15:11<1:12:37, 1.10it/s, tok/s=14107.1]
Inference: 19%|█▉ | 1123/5920 [15:13<1:37:34, 1.22s/it, tok/s=14107.1]
Inference: 19%|█▉ | 1123/5920 [15:13<1:37:34, 1.22s/it, tok/s=14093.3]
Inference: 19%|█▉ | 1124/5920 [15:13<1:13:54, 1.08it/s, tok/s=14093.3]
Inference: 19%|█▉ | 1124/5920 [15:13<1:13:54, 1.08it/s, tok/s=14107.1]
Inference: 19%|█▉ | 1125/5920 [15:16<1:45:24, 1.32s/it, tok/s=14107.1]
Inference: 19%|█▉ | 1125/5920 [15:16<1:45:24, 1.32s/it, tok/s=14084.6]
Inference: 19%|█▉ | 1126/5920 [15:18<2:01:10, 1.52s/it, tok/s=14084.6]
Inference: 19%|█▉ | 1126/5920 [15:18<2:01:10, 1.52s/it, tok/s=14076.9]
Inference: 19%|█▉ | 1127/5920 [15:18<1:33:48, 1.17s/it, tok/s=14076.9]
Inference: 19%|█▉ | 1127/5920 [15:18<1:33:48, 1.17s/it, tok/s=14095.6]
Inference: 19%|█▉ | 1128/5920 [15:20<2:03:14, 1.54s/it, tok/s=14095.6]
Inference: 19%|█▉ | 1128/5920 [15:20<2:03:14, 1.54s/it, tok/s=14071.2]
Inference: 19%|█▉ | 1129/5920 [15:22<2:05:36, 1.57s/it, tok/s=14071.2]
Inference: 19%|█▉ | 1129/5920 [15:22<2:05:36, 1.57s/it, tok/s=14069.0]
Inference: 19%|█▉ | 1130/5920 [15:24<2:15:46, 1.70s/it, tok/s=14069.0]
Inference: 19%|█▉ | 1130/5920 [15:24<2:15:46, 1.70s/it, tok/s=14064.0]
Inference: 19%|█▉ | 1131/5920 [15:24<1:47:12, 1.34s/it, tok/s=14064.0]
Inference: 19%|█▉ | 1131/5920 [15:24<1:47:12, 1.34s/it, tok/s=14089.0]
Inference: 19%|█▉ | 1132/5920 [15:25<1:47:11, 1.34s/it, tok/s=14111.0]
Inference: 19%|█▉ | 1133/5920 [15:28<2:05:02, 1.57s/it, tok/s=14111.0]
Inference: 19%|█▉ | 1133/5920 [15:28<2:05:02, 1.57s/it, tok/s=14079.8]
Inference: 19%|█▉ | 1134/5920 [15:28<1:38:01, 1.23s/it, tok/s=14079.8]
Inference: 19%|█▉ | 1134/5920 [15:28<1:38:01, 1.23s/it, tok/s=14100.1]
Inference: 19%|█▉ | 1135/5920 [15:29<1:17:31, 1.03it/s, tok/s=14100.1]
Inference: 19%|█▉ | 1135/5920 [15:29<1:17:31, 1.03it/s, tok/s=14113.9]
Inference: 19%|█▉ | 1136/5920 [15:29<1:06:56, 1.19it/s, tok/s=14113.9]
Inference: 19%|█▉ | 1136/5920 [15:29<1:06:56, 1.19it/s, tok/s=14121.6]
Inference: 19%|█▉ | 1137/5920 [15:29<52:52, 1.51it/s, tok/s=14121.6]
Inference: 19%|█▉ | 1137/5920 [15:29<52:52, 1.51it/s, tok/s=14131.4]
Inference: 19%|█▉ | 1138/5920 [15:32<1:40:13, 1.26s/it, tok/s=14131.4]
Inference: 19%|█▉ | 1138/5920 [15:32<1:40:13, 1.26s/it, tok/s=14097.0]
Inference: 19%|█▉ | 1139/5920 [15:34<1:51:24, 1.40s/it, tok/s=14097.0]
Inference: 19%|█▉ | 1139/5920 [15:34<1:51:24, 1.40s/it, tok/s=14078.1]
Inference: 19%|█▉ | 1140/5920 [15:36<2:07:44, 1.60s/it, tok/s=14078.1]
Inference: 19%|█▉ | 1140/5920 [15:36<2:07:44, 1.60s/it, tok/s=14054.3]
Inference: 19%|█▉ | 1141/5920 [15:36<2:07:43, 1.60s/it, tok/s=14066.9]
Inference: 19%|█▉ | 1142/5920 [15:39<2:08:11, 1.61s/it, tok/s=14066.9]
Inference: 19%|█▉ | 1142/5920 [15:39<2:08:11, 1.61s/it, tok/s=14041.6]
Inference: 19%|█▉ | 1143/5920 [15:40<1:53:02, 1.42s/it, tok/s=14041.6]
Inference: 19%|█▉ | 1143/5920 [15:40<1:53:02, 1.42s/it, tok/s=14036.6]
Inference: 19%|█▉ | 1144/5920 [15:41<1:46:32, 1.34s/it, tok/s=14036.6]
Inference: 19%|█▉ | 1144/5920 [15:41<1:46:32, 1.34s/it, tok/s=14042.3]
Inference: 19%|█▉ | 1145/5920 [15:42<1:28:20, 1.11s/it, tok/s=14042.3]
Inference: 19%|█▉ | 1145/5920 [15:42<1:28:20, 1.11s/it, tok/s=14042.7]
Inference: 19%|█▉ | 1146/5920 [15:42<1:12:25, 1.10it/s, tok/s=14042.7]
Inference: 19%|█▉ | 1146/5920 [15:42<1:12:25, 1.10it/s, tok/s=14045.0]
Inference: 19%|█▉ | 1147/5920 [15:43<1:20:30, 1.01s/it, tok/s=14045.0]
Inference: 19%|█▉ | 1147/5920 [15:43<1:20:30, 1.01s/it, tok/s=14044.6]
Inference: 19%|█▉ | 1148/5920 [15:45<1:32:43, 1.17s/it, tok/s=14044.6]
Inference: 19%|█▉ | 1148/5920 [15:45<1:32:43, 1.17s/it, tok/s=14040.4]
Inference: 19%|█▉ | 1149/5920 [15:45<1:15:00, 1.06it/s, tok/s=14040.4]
Inference: 19%|█▉ | 1149/5920 [15:45<1:15:00, 1.06it/s, tok/s=14048.1]
Inference: 19%|█▉ | 1150/5920 [15:46<1:10:02, 1.14it/s, tok/s=14048.1]
Inference: 19%|█▉ | 1150/5920 [15:46<1:10:02, 1.14it/s, tok/s=14055.1]
Inference: 19%|█▉ | 1151/5920 [15:47<1:10:13, 1.13it/s, tok/s=14055.1]
Inference: 19%|█▉ | 1151/5920 [15:47<1:10:13, 1.13it/s, tok/s=14050.4]
Inference: 19%|█▉ | 1152/5920 [15:48<1:09:18, 1.15it/s, tok/s=14050.4]
Inference: 19%|█▉ | 1152/5920 [15:48<1:09:18, 1.15it/s, tok/s=14054.3]
Inference: 19%|█▉ | 1153/5920 [15:49<1:10:22, 1.13it/s, tok/s=14054.3]
Inference: 19%|█▉ | 1153/5920 [15:49<1:10:22, 1.13it/s, tok/s=14059.9]
Inference: 19%|█▉ | 1154/5920 [15:49<55:09, 1.44it/s, tok/s=14059.9]
Inference: 19%|█▉ | 1154/5920 [15:49<55:09, 1.44it/s, tok/s=14085.5]
Inference: 20%|█▉ | 1155/5920 [15:49<55:09, 1.44it/s, tok/s=14107.3]
Inference: 20%|█▉ | 1156/5920 [15:50<47:37, 1.67it/s, tok/s=14107.3]
Inference: 20%|█▉ | 1156/5920 [15:50<47:37, 1.67it/s, tok/s=14100.0]
Inference: 20%|█▉ | 1157/5920 [15:50<38:08, 2.08it/s, tok/s=14100.0]
Inference: 20%|█▉ | 1157/5920 [15:50<38:08, 2.08it/s, tok/s=14121.6]
Inference: 20%|█▉ | 1158/5920 [15:51<53:23, 1.49it/s, tok/s=14121.6]
Inference: 20%|█▉ | 1158/5920 [15:51<53:23, 1.49it/s, tok/s=14111.9]
Inference: 20%|█▉ | 1159/5920 [15:54<1:34:15, 1.19s/it, tok/s=14111.9]
Inference: 20%|█▉ | 1159/5920 [15:54<1:34:15, 1.19s/it, tok/s=14081.9]
Inference: 20%|█▉ | 1160/5920 [15:57<2:11:50, 1.66s/it, tok/s=14081.9]
Inference: 20%|█▉ | 1160/5920 [15:57<2:11:50, 1.66s/it, tok/s=14047.1]
Inference: 20%|█▉ | 1161/5920 [15:58<2:07:38, 1.61s/it, tok/s=14047.1]
Inference: 20%|█▉ | 1161/5920 [15:58<2:07:38, 1.61s/it, tok/s=14048.3]
Inference: 20%|█▉ | 1162/5920 [15:59<1:53:14, 1.43s/it, tok/s=14048.3]
Inference: 20%|█▉ | 1162/5920 [15:59<1:53:14, 1.43s/it, tok/s=14052.4]
Inference: 20%|█▉ | 1163/5920 [16:00<1:35:42, 1.21s/it, tok/s=14052.4]
Inference: 20%|█▉ | 1163/5920 [16:00<1:35:42, 1.21s/it, tok/s=14049.7]
Inference: 20%|█▉ | 1164/5920 [16:02<1:51:22, 1.41s/it, tok/s=14049.7]
Inference: 20%|█▉ | 1164/5920 [16:02<1:51:22, 1.41s/it, tok/s=14041.9]
Inference: 20%|█▉ | 1165/5920 [16:04<2:04:56, 1.58s/it, tok/s=14041.9]
Inference: 20%|█▉ | 1165/5920 [16:04<2:04:56, 1.58s/it, tok/s=14039.8]
Inference: 20%|█▉ | 1166/5920 [16:06<2:28:11, 1.87s/it, tok/s=14039.8]
Inference: 20%|█▉ | 1166/5920 [16:06<2:28:11, 1.87s/it, tok/s=14009.5]
Inference: 20%|█▉ | 1167/5920 [16:07<2:04:37, 1.57s/it, tok/s=14009.5]
Inference: 20%|█▉ | 1167/5920 [16:07<2:04:37, 1.57s/it, tok/s=14010.4]
Inference: 20%|█▉ | 1168/5920 [16:07<1:30:33, 1.14s/it, tok/s=14010.4]
Inference: 20%|█▉ | 1168/5920 [16:07<1:30:33, 1.14s/it, tok/s=14022.5]
Inference: 20%|█▉ | 1169/5920 [16:07<1:10:20, 1.13it/s, tok/s=14022.5]
Inference: 20%|█▉ | 1169/5920 [16:07<1:10:20, 1.13it/s, tok/s=14025.0]
Inference: 20%|█▉ | 1170/5920 [16:08<1:01:28, 1.29it/s, tok/s=14025.0]
Inference: 20%|█▉ | 1170/5920 [16:08<1:01:28, 1.29it/s, tok/s=14031.3]
Inference: 20%|█▉ | 1171/5920 [16:08<54:05, 1.46it/s, tok/s=14031.3]
Inference: 20%|█▉ | 1171/5920 [16:08<54:05, 1.46it/s, tok/s=14032.8]
Inference: 20%|█▉ | 1172/5920 [16:10<1:13:59, 1.07it/s, tok/s=14032.8]
Inference: 20%|█▉ | 1172/5920 [16:10<1:13:59, 1.07it/s, tok/s=14034.1]
Inference: 20%|█▉ | 1173/5920 [16:11<1:16:14, 1.04it/s, tok/s=14034.1]
Inference: 20%|█▉ | 1173/5920 [16:11<1:16:14, 1.04it/s, tok/s=14027.7]
Inference: 20%|█▉ | 1174/5920 [16:14<2:15:40, 1.72s/it, tok/s=14027.7]
Inference: 20%|█▉ | 1174/5920 [16:14<2:15:40, 1.72s/it, tok/s=13985.5]
Inference: 20%|█▉ | 1175/5920 [16:15<1:44:35, 1.32s/it, tok/s=13985.5]
Inference: 20%|█▉ | 1175/5920 [16:15<1:44:35, 1.32s/it, tok/s=13986.0]
Inference: 20%|█▉ | 1176/5920 [16:16<1:40:11, 1.27s/it, tok/s=13986.0]
Inference: 20%|█▉ | 1176/5920 [16:16<1:40:11, 1.27s/it, tok/s=13994.1]
Inference: 20%|█▉ | 1177/5920 [16:18<1:47:40, 1.36s/it, tok/s=13994.1]
Inference: 20%|█▉ | 1177/5920 [16:18<1:47:40, 1.36s/it, tok/s=13990.2]
Inference: 20%|█▉ | 1178/5920 [16:19<1:52:31, 1.42s/it, tok/s=13990.2]
Inference: 20%|█▉ | 1178/5920 [16:19<1:52:31, 1.42s/it, tok/s=13980.2]
Inference: 20%|█▉ | 1179/5920 [16:20<1:41:24, 1.28s/it, tok/s=13980.2]
Inference: 20%|█▉ | 1179/5920 [16:20<1:41:24, 1.28s/it, tok/s=13974.2]
Inference: 20%|█▉ | 1180/5920 [16:20<1:13:21, 1.08it/s, tok/s=13974.2]
Inference: 20%|█▉ | 1180/5920 [16:20<1:13:21, 1.08it/s, tok/s=13992.3]
Inference: 20%|█▉ | 1181/5920 [16:21<1:03:04, 1.25it/s, tok/s=13992.3]
Inference: 20%|█▉ | 1181/5920 [16:21<1:03:04, 1.25it/s, tok/s=13993.4]
Inference: 20%|█▉ | 1182/5920 [16:22<1:07:25, 1.17it/s, tok/s=13993.4]
Inference: 20%|█▉ | 1182/5920 [16:22<1:07:25, 1.17it/s, tok/s=13987.8]
Inference: 20%|█▉ | 1183/5920 [16:24<1:45:59, 1.34s/it, tok/s=13987.8]
Inference: 20%|█▉ | 1183/5920 [16:24<1:45:59, 1.34s/it, tok/s=13960.8]
Inference: 20%|██ | 1184/5920 [16:26<1:49:11, 1.38s/it, tok/s=13960.8]
Inference: 20%|██ | 1184/5920 [16:26<1:49:11, 1.38s/it, tok/s=13955.0]
Inference: 20%|██ | 1185/5920 [16:26<1:29:38, 1.14s/it, tok/s=13955.0]
Inference: 20%|██ | 1185/5920 [16:26<1:29:38, 1.14s/it, tok/s=13972.9]
Inference: 20%|██ | 1186/5920 [16:27<1:26:40, 1.10s/it, tok/s=13972.9]
Inference: 20%|██ | 1186/5920 [16:27<1:26:40, 1.10s/it, tok/s=13976.8]
Inference: 20%|██ | 1187/5920 [16:29<1:50:15, 1.40s/it, tok/s=13976.8]
Inference: 20%|██ | 1187/5920 [16:29<1:50:15, 1.40s/it, tok/s=13965.9]
Inference: 20%|██ | 1188/5920 [16:30<1:27:45, 1.11s/it, tok/s=13965.9]
Inference: 20%|██ | 1188/5920 [16:30<1:27:45, 1.11s/it, tok/s=13971.8]
Inference: 20%|██ | 1189/5920 [16:30<1:05:04, 1.21it/s, tok/s=13971.8]
Inference: 20%|██ | 1189/5920 [16:30<1:05:04, 1.21it/s, tok/s=13990.4]
Inference: 20%|██ | 1190/5920 [16:31<1:17:41, 1.01it/s, tok/s=13990.4]
Inference: 20%|██ | 1190/5920 [16:31<1:17:41, 1.01it/s, tok/s=13980.8]
Inference: 20%|██ | 1191/5920 [16:31<59:09, 1.33it/s, tok/s=13980.8]
Inference: 20%|██ | 1191/5920 [16:31<59:09, 1.33it/s, tok/s=13993.8]
Inference: 20%|██ | 1192/5920 [16:32<59:09, 1.33it/s, tok/s=14012.0]
Inference: 20%|██ | 1193/5920 [16:32<41:57, 1.88it/s, tok/s=14012.0]
Inference: 20%|██ | 1193/5920 [16:32<41:57, 1.88it/s, tok/s=14019.2]
Inference: 20%|██ | 1194/5920 [16:33<45:36, 1.73it/s, tok/s=14019.2]
Inference: 20%|██ | 1194/5920 [16:33<45:36, 1.73it/s, tok/s=14017.2]
Inference: 20%|██ | 1195/5920 [16:33<41:33, 1.90it/s, tok/s=14017.2]
Inference: 20%|██ | 1195/5920 [16:33<41:33, 1.90it/s, tok/s=14020.2]
Inference: 20%|██ | 1196/5920 [16:35<1:05:40, 1.20it/s, tok/s=14020.2]
Inference: 20%|██ | 1196/5920 [16:35<1:05:40, 1.20it/s, tok/s=14006.2]
Inference: 20%|██ | 1197/5920 [16:35<1:00:42, 1.30it/s, tok/s=14006.2]
Inference: 20%|██ | 1197/5920 [16:35<1:00:42, 1.30it/s, tok/s=14005.2]
Inference: 20%|██ | 1198/5920 [16:38<1:45:02, 1.33s/it, tok/s=14005.2]
Inference: 20%|██ | 1198/5920 [16:38<1:45:02, 1.33s/it, tok/s=13974.0]
Inference: 20%|██ | 1199/5920 [16:39<1:27:10, 1.11s/it, tok/s=13974.0]
Inference: 20%|██ | 1199/5920 [16:39<1:27:10, 1.11s/it, tok/s=13974.9]
Inference: 20%|██ | 1200/5920 [16:43<2:30:47, 1.92s/it, tok/s=13974.9]
Inference: 20%|██ | 1200/5920 [16:43<2:30:47, 1.92s/it, tok/s=13928.6]
Inference: 20%|██ | 1201/5920 [16:44<2:13:41, 1.70s/it, tok/s=13928.6]
Inference: 20%|██ | 1201/5920 [16:44<2:13:41, 1.70s/it, tok/s=13931.2]
Inference: 20%|██ | 1202/5920 [16:45<1:57:50, 1.50s/it, tok/s=13931.2]
Inference: 20%|██ | 1202/5920 [16:45<1:57:50, 1.50s/it, tok/s=13934.0]
Inference: 20%|██ | 1203/5920 [16:50<3:16:50, 2.50s/it, tok/s=13934.0]
Inference: 20%|██ | 1203/5920 [16:50<3:16:50, 2.50s/it, tok/s=13875.8]
Inference: 20%|██ | 1204/5920 [16:50<2:32:58, 1.95s/it, tok/s=13875.8]
Inference: 20%|██ | 1204/5920 [16:50<2:32:58, 1.95s/it, tok/s=13889.9]
Inference: 20%|██ | 1205/5920 [16:54<3:07:30, 2.39s/it, tok/s=13889.9]
Inference: 20%|██ | 1205/5920 [16:54<3:07:30, 2.39s/it, tok/s=13853.3]
Inference: 20%|██ | 1206/5920 [16:54<2:23:08, 1.82s/it, tok/s=13853.3]
Inference: 20%|██ | 1206/5920 [16:54<2:23:08, 1.82s/it, tok/s=13860.3]
Inference: 20%|██ | 1207/5920 [16:56<2:21:18, 1.80s/it, tok/s=13860.3]
Inference: 20%|██ | 1207/5920 [16:56<2:21:18, 1.80s/it, tok/s=13856.9]
Inference: 20%|██ | 1208/5920 [16:58<2:25:40, 1.85s/it, tok/s=13856.9]
Inference: 20%|██ | 1208/5920 [16:58<2:25:40, 1.85s/it, tok/s=13840.1]
Inference: 20%|██ | 1209/5920 [16:59<2:06:36, 1.61s/it, tok/s=13840.1]
Inference: 20%|██ | 1209/5920 [16:59<2:06:36, 1.61s/it, tok/s=13846.2]
Inference: 20%|██ | 1210/5920 [16:59<2:06:34, 1.61s/it, tok/s=13854.7]
Inference: 20%|██ | 1211/5920 [17:00<1:36:00, 1.22s/it, tok/s=13854.7]
Inference: 20%|██ | 1211/5920 [17:00<1:36:00, 1.22s/it, tok/s=13860.0]
Inference: 20%|██ | 1212/5920 [17:02<1:34:58, 1.21s/it, tok/s=13860.0]
Inference: 20%|██ | 1212/5920 [17:02<1:34:58, 1.21s/it, tok/s=13852.3]
Inference: 20%|██ | 1213/5920 [17:02<1:19:00, 1.01s/it, tok/s=13852.3]
Inference: 20%|██ | 1213/5920 [17:02<1:19:00, 1.01s/it, tok/s=13856.2]
Inference: 21%|██ | 1214/5920 [17:04<1:39:26, 1.27s/it, tok/s=13856.2]
Inference: 21%|██ | 1214/5920 [17:04<1:39:26, 1.27s/it, tok/s=13851.4]
Inference: 21%|██ | 1215/5920 [17:05<1:38:08, 1.25s/it, tok/s=13851.4]
Inference: 21%|██ | 1215/5920 [17:05<1:38:08, 1.25s/it, tok/s=13844.7]
Inference: 21%|██ | 1216/5920 [17:07<1:39:26, 1.27s/it, tok/s=13844.7]
Inference: 21%|██ | 1216/5920 [17:07<1:39:26, 1.27s/it, tok/s=13840.0]
Inference: 21%|██ | 1217/5920 [17:07<1:26:54, 1.11s/it, tok/s=13840.0]
Inference: 21%|██ | 1217/5920 [17:07<1:26:54, 1.11s/it, tok/s=13840.1]
Inference: 21%|██ | 1218/5920 [17:08<1:17:35, 1.01it/s, tok/s=13840.1]
Inference: 21%|██ | 1218/5920 [17:08<1:17:35, 1.01it/s, tok/s=13851.1]
Inference: 21%|██ | 1219/5920 [17:11<1:57:34, 1.50s/it, tok/s=13851.1]
Inference: 21%|██ | 1219/5920 [17:11<1:57:34, 1.50s/it, tok/s=13823.3]
Inference: 21%|██ | 1220/5920 [17:11<1:31:26, 1.17s/it, tok/s=13823.3]
Inference: 21%|██ | 1220/5920 [17:11<1:31:26, 1.17s/it, tok/s=13826.4]
Inference: 21%|██ | 1221/5920 [17:12<1:31:46, 1.17s/it, tok/s=13826.4]
Inference: 21%|██ | 1221/5920 [17:12<1:31:46, 1.17s/it, tok/s=13820.2]
Inference: 21%|██ | 1222/5920 [17:14<1:53:03, 1.44s/it, tok/s=13820.2]
Inference: 21%|██ | 1222/5920 [17:14<1:53:03, 1.44s/it, tok/s=13808.4]
Inference: 21%|██ | 1223/5920 [17:14<1:21:42, 1.04s/it, tok/s=13808.4]
Inference: 21%|██ | 1223/5920 [17:14<1:21:42, 1.04s/it, tok/s=13833.0]
Inference: 21%|██ | 1224/5920 [17:16<1:35:07, 1.22s/it, tok/s=13833.0]
Inference: 21%|██ | 1224/5920 [17:16<1:35:07, 1.22s/it, tok/s=13831.3]
Inference: 21%|██ | 1225/5920 [17:17<1:20:19, 1.03s/it, tok/s=13831.3]
Inference: 21%|██ | 1225/5920 [17:17<1:20:19, 1.03s/it, tok/s=13832.3]
Inference: 21%|██ | 1226/5920 [17:18<1:16:02, 1.03it/s, tok/s=13832.3]
Inference: 21%|██ | 1226/5920 [17:18<1:16:02, 1.03it/s, tok/s=13839.7]
Inference: 21%|██ | 1227/5920 [17:18<1:02:26, 1.25it/s, tok/s=13839.7]
Inference: 21%|██ | 1227/5920 [17:18<1:02:26, 1.25it/s, tok/s=13858.5]
Inference: 21%|██ | 1228/5920 [17:19<1:07:04, 1.17it/s, tok/s=13858.5]
Inference: 21%|██ | 1228/5920 [17:19<1:07:04, 1.17it/s, tok/s=13853.8]
Inference: 21%|██ | 1229/5920 [17:20<1:03:55, 1.22it/s, tok/s=13853.8]
Inference: 21%|██ | 1229/5920 [17:20<1:03:55, 1.22it/s, tok/s=13854.0]
Inference: 21%|██ | 1230/5920 [17:21<1:09:36, 1.12it/s, tok/s=13854.0]
Inference: 21%|██ | 1230/5920 [17:21<1:09:36, 1.12it/s, tok/s=13847.6]
Inference: 21%|██ | 1231/5920 [17:21<51:19, 1.52it/s, tok/s=13847.6]
Inference: 21%|██ | 1231/5920 [17:21<51:19, 1.52it/s, tok/s=13855.3]
Inference: 21%|██ | 1232/5920 [17:21<46:07, 1.69it/s, tok/s=13855.3]
Inference: 21%|██ | 1232/5920 [17:21<46:07, 1.69it/s, tok/s=13860.6]
Inference: 21%|██ | 1233/5920 [17:22<56:57, 1.37it/s, tok/s=13860.6]
Inference: 21%|██ | 1233/5920 [17:22<56:57, 1.37it/s, tok/s=13866.6]
Inference: 21%|██ | 1234/5920 [17:25<1:33:10, 1.19s/it, tok/s=13866.6]
Inference: 21%|██ | 1234/5920 [17:25<1:33:10, 1.19s/it, tok/s=13847.6]
Inference: 21%|██ | 1235/5920 [17:27<1:54:13, 1.46s/it, tok/s=13847.6]
Inference: 21%|██ | 1235/5920 [17:27<1:54:13, 1.46s/it, tok/s=13842.8]
Inference: 21%|██ | 1236/5920 [17:27<1:54:12, 1.46s/it, tok/s=13853.8]
Inference: 21%|██ | 1237/5920 [17:27<1:54:10, 1.46s/it, tok/s=13871.6]
Inference: 21%|██ | 1238/5920 [17:28<1:09:18, 1.13it/s, tok/s=13871.6]
Inference: 21%|██ | 1238/5920 [17:28<1:09:18, 1.13it/s, tok/s=13863.6]
Inference: 21%|██ | 1239/5920 [17:29<1:11:12, 1.10it/s, tok/s=13863.6]
Inference: 21%|██ | 1239/5920 [17:29<1:11:12, 1.10it/s, tok/s=13859.3]
Inference: 21%|██ | 1240/5920 [17:29<1:11:11, 1.10it/s, tok/s=13869.3]
Inference: 21%|██ | 1241/5920 [17:29<47:22, 1.65it/s, tok/s=13869.3]
Inference: 21%|██ | 1241/5920 [17:29<47:22, 1.65it/s, tok/s=13878.0]
Inference: 21%|██ | 1242/5920 [17:29<39:32, 1.97it/s, tok/s=13878.0]
Inference: 21%|██ | 1242/5920 [17:29<39:32, 1.97it/s, tok/s=13884.6]
Inference: 21%|██ | 1243/5920 [17:30<36:39, 2.13it/s, tok/s=13884.6]
Inference: 21%|██ | 1243/5920 [17:30<36:39, 2.13it/s, tok/s=13906.4]
Inference: 21%|██ | 1244/5920 [17:30<29:38, 2.63it/s, tok/s=13906.4]
Inference: 21%|██ | 1244/5920 [17:30<29:38, 2.63it/s, tok/s=13912.2]
Inference: 21%|██ | 1245/5920 [17:30<29:52, 2.61it/s, tok/s=13912.2]
Inference: 21%|██ | 1245/5920 [17:30<29:52, 2.61it/s, tok/s=13917.0]
Inference: 21%|██ | 1246/5920 [17:34<1:36:15, 1.24s/it, tok/s=13917.0]
Inference: 21%|██ | 1246/5920 [17:34<1:36:15, 1.24s/it, tok/s=13881.6]
Inference: 21%|██ | 1247/5920 [17:35<1:40:53, 1.30s/it, tok/s=13881.6]
Inference: 21%|██ | 1247/5920 [17:35<1:40:53, 1.30s/it, tok/s=13872.3]
Inference: 21%|██ | 1248/5920 [17:39<2:41:53, 2.08s/it, tok/s=13872.3]
Inference: 21%|██ | 1248/5920 [17:39<2:41:53, 2.08s/it, tok/s=13827.7]
Inference: 21%|██ | 1249/5920 [17:41<2:45:23, 2.12s/it, tok/s=13827.7]
Inference: 21%|██ | 1249/5920 [17:41<2:45:23, 2.12s/it, tok/s=13809.2]
Inference: 21%|██ | 1250/5920 [17:43<2:44:30, 2.11s/it, tok/s=13809.2]
Inference: 21%|██ | 1250/5920 [17:43<2:44:30, 2.11s/it, tok/s=13790.0]
Inference: 21%|██ | 1251/5920 [17:44<2:08:49, 1.66s/it, tok/s=13790.0]
Inference: 21%|██ | 1251/5920 [17:44<2:08:49, 1.66s/it, tok/s=13805.5]
Inference: 21%|██ | 1252/5920 [17:45<1:54:15, 1.47s/it, tok/s=13805.5]
Inference: 21%|██ | 1252/5920 [17:45<1:54:15, 1.47s/it, tok/s=13800.7]
Inference: 21%|██ | 1253/5920 [17:50<3:24:49, 2.63s/it, tok/s=13800.7]
Inference: 21%|██ | 1253/5920 [17:50<3:24:49, 2.63s/it, tok/s=13746.4]
Inference: 21%|██ | 1254/5920 [17:53<3:21:21, 2.59s/it, tok/s=13746.4]
Inference: 21%|██ | 1254/5920 [17:53<3:21:21, 2.59s/it, tok/s=13737.6]
Inference: 21%|██ | 1255/5920 [17:55<3:12:32, 2.48s/it, tok/s=13737.6]
Inference: 21%|██ | 1255/5920 [17:55<3:12:32, 2.48s/it, tok/s=13730.5]
Inference: 21%|██ | 1256/5920 [17:56<2:41:20, 2.08s/it, tok/s=13730.5]
Inference: 21%|██ | 1256/5920 [17:56<2:41:20, 2.08s/it, tok/s=13736.9]
Inference: 21%|██ | 1257/5920 [18:00<3:18:45, 2.56s/it, tok/s=13736.9]
Inference: 21%|██ | 1257/5920 [18:00<3:18:45, 2.56s/it, tok/s=13715.6]
Inference: 21%|██▏ | 1258/5920 [18:03<3:19:31, 2.57s/it, tok/s=13715.6]
Inference: 21%|██▏ | 1258/5920 [18:03<3:19:31, 2.57s/it, tok/s=13703.1]
Inference: 21%|██▏ | 1259/5920 [18:03<3:19:29, 2.57s/it, tok/s=13723.5]
Inference: 21%|██▏ | 1260/5920 [18:04<2:05:48, 1.62s/it, tok/s=13723.5]
Inference: 21%|██▏ | 1260/5920 [18:04<2:05:48, 1.62s/it, tok/s=13722.8]
Inference: 21%|██▏ | 1261/5920 [18:05<1:56:22, 1.50s/it, tok/s=13722.8]
Inference: 21%|██▏ | 1261/5920 [18:05<1:56:22, 1.50s/it, tok/s=13729.6]
Inference: 21%|██▏ | 1262/5920 [18:05<1:28:04, 1.13s/it, tok/s=13729.6]
Inference: 21%|██▏ | 1262/5920 [18:05<1:28:04, 1.13s/it, tok/s=13738.5]
Inference: 21%|██▏ | 1263/5920 [18:05<1:14:08, 1.05it/s, tok/s=13738.5]
Inference: 21%|██▏ | 1263/5920 [18:05<1:14:08, 1.05it/s, tok/s=13757.4]
Inference: 21%|██▏ | 1264/5920 [18:05<1:14:07, 1.05it/s, tok/s=13766.9]
Inference: 21%|██▏ | 1265/5920 [18:07<1:15:48, 1.02it/s, tok/s=13766.9]
Inference: 21%|██▏ | 1265/5920 [18:07<1:15:48, 1.02it/s, tok/s=13756.4]
Inference: 21%|██▏ | 1266/5920 [18:08<1:04:38, 1.20it/s, tok/s=13756.4]
Inference: 21%|██▏ | 1266/5920 [18:08<1:04:38, 1.20it/s, tok/s=13767.7]
Inference: 21%|██▏ | 1267/5920 [18:09<1:19:49, 1.03s/it, tok/s=13767.7]
Inference: 21%|██▏ | 1267/5920 [18:09<1:19:49, 1.03s/it, tok/s=13766.9]
Inference: 21%|██▏ | 1268/5920 [18:09<1:19:48, 1.03s/it, tok/s=13785.3]
Inference: 21%|██▏ | 1269/5920 [18:09<1:19:47, 1.03s/it, tok/s=13797.0]
Inference: 21%|██▏ | 1270/5920 [18:10<51:10, 1.51it/s, tok/s=13797.0]
Inference: 21%|██▏ | 1270/5920 [18:10<51:10, 1.51it/s, tok/s=13796.9]
Inference: 21%|██▏ | 1271/5920 [18:13<1:17:02, 1.01it/s, tok/s=13796.9]
Inference: 21%|██▏ | 1271/5920 [18:13<1:17:02, 1.01it/s, tok/s=13791.7]
Inference: 21%|██▏ | 1272/5920 [18:15<1:37:49, 1.26s/it, tok/s=13791.7]
Inference: 21%|██▏ | 1272/5920 [18:15<1:37:49, 1.26s/it, tok/s=13772.7]
Inference: 22%|██▏ | 1273/5920 [18:16<1:37:00, 1.25s/it, tok/s=13772.7]
Inference: 22%|██▏ | 1273/5920 [18:16<1:37:00, 1.25s/it, tok/s=13779.4]
Inference: 22%|██▏ | 1274/5920 [18:16<1:14:57, 1.03it/s, tok/s=13779.4]
Inference: 22%|██▏ | 1274/5920 [18:16<1:14:57, 1.03it/s, tok/s=13793.1]
Inference: 22%|██▏ | 1275/5920 [18:16<58:11, 1.33it/s, tok/s=13793.1]
Inference: 22%|██▏ | 1275/5920 [18:16<58:11, 1.33it/s, tok/s=13802.9]
Inference: 22%|██▏ | 1276/5920 [18:17<1:01:35, 1.26it/s, tok/s=13802.9]
Inference: 22%|██▏ | 1276/5920 [18:17<1:01:35, 1.26it/s, tok/s=13799.0]
Inference: 22%|██▏ | 1277/5920 [18:19<1:15:09, 1.03it/s, tok/s=13799.0]
Inference: 22%|██▏ | 1277/5920 [18:19<1:15:09, 1.03it/s, tok/s=13803.3]
Inference: 22%|██▏ | 1278/5920 [18:19<55:56, 1.38it/s, tok/s=13803.3]
Inference: 22%|██▏ | 1278/5920 [18:19<55:56, 1.38it/s, tok/s=13812.8]
Inference: 22%|██▏ | 1279/5920 [18:19<48:02, 1.61it/s, tok/s=13812.8]
Inference: 22%|██▏ | 1279/5920 [18:19<48:02, 1.61it/s, tok/s=13815.6]
Inference: 22%|██▏ | 1280/5920 [18:21<1:24:09, 1.09s/it, tok/s=13815.6]
Inference: 22%|██▏ | 1280/5920 [18:21<1:24:09, 1.09s/it, tok/s=13796.9]
Inference: 22%|██▏ | 1281/5920 [18:22<1:17:23, 1.00s/it, tok/s=13796.9]
Inference: 22%|██▏ | 1281/5920 [18:22<1:17:23, 1.00s/it, tok/s=13797.3]
Inference: 22%|██▏ | 1282/5920 [18:23<1:22:21, 1.07s/it, tok/s=13797.3]
Inference: 22%|██▏ | 1282/5920 [18:23<1:22:21, 1.07s/it, tok/s=13809.4]
Inference: 22%|██▏ | 1283/5920 [18:23<1:22:20, 1.07s/it, tok/s=13817.5]
Inference: 22%|██▏ | 1284/5920 [18:25<1:11:33, 1.08it/s, tok/s=13817.5]
Inference: 22%|██▏ | 1284/5920 [18:25<1:11:33, 1.08it/s, tok/s=13818.6]
Inference: 22%|██▏ | 1285/5920 [18:27<1:43:48, 1.34s/it, tok/s=13818.6]
Inference: 22%|██▏ | 1285/5920 [18:27<1:43:48, 1.34s/it, tok/s=13807.7]
Inference: 22%|██▏ | 1286/5920 [18:28<1:37:03, 1.26s/it, tok/s=13807.7]
Inference: 22%|██▏ | 1286/5920 [18:28<1:37:03, 1.26s/it, tok/s=13804.7]
Inference: 22%|██▏ | 1287/5920 [18:29<1:13:46, 1.05it/s, tok/s=13804.7]
Inference: 22%|██▏ | 1287/5920 [18:29<1:13:46, 1.05it/s, tok/s=13823.7]
Inference: 22%|██▏ | 1288/5920 [18:29<1:00:37, 1.27it/s, tok/s=13823.7]
Inference: 22%|██▏ | 1288/5920 [18:29<1:00:37, 1.27it/s, tok/s=13823.1]
Inference: 22%|██▏ | 1289/5920 [18:30<1:02:56, 1.23it/s, tok/s=13823.1]
Inference: 22%|██▏ | 1289/5920 [18:30<1:02:56, 1.23it/s, tok/s=13823.5]
Inference: 22%|██▏ | 1290/5920 [18:34<2:09:29, 1.68s/it, tok/s=13823.5]
Inference: 22%|██▏ | 1290/5920 [18:34<2:09:29, 1.68s/it, tok/s=13798.3]
Inference: 22%|██▏ | 1291/5920 [18:34<2:09:27, 1.68s/it, tok/s=13822.2]
Inference: 22%|██▏ | 1292/5920 [18:34<1:24:15, 1.09s/it, tok/s=13822.2]
Inference: 22%|██▏ | 1292/5920 [18:34<1:24:15, 1.09s/it, tok/s=13836.6]
Inference: 22%|██▏ | 1293/5920 [18:35<1:10:47, 1.09it/s, tok/s=13836.6]
Inference: 22%|██▏ | 1293/5920 [18:35<1:10:47, 1.09it/s, tok/s=13847.6]
Inference: 22%|██▏ | 1294/5920 [18:36<1:09:40, 1.11it/s, tok/s=13847.6]
Inference: 22%|██▏ | 1294/5920 [18:36<1:09:40, 1.11it/s, tok/s=13852.8]
Inference: 22%|██▏ | 1295/5920 [18:37<1:14:18, 1.04it/s, tok/s=13852.8]
Inference: 22%|██▏ | 1295/5920 [18:37<1:14:18, 1.04it/s, tok/s=13842.6]
Inference: 22%|██▏ | 1296/5920 [18:38<1:11:51, 1.07it/s, tok/s=13842.6]
Inference: 22%|██▏ | 1296/5920 [18:38<1:11:51, 1.07it/s, tok/s=13842.7]
Inference: 22%|██▏ | 1297/5920 [18:39<1:10:34, 1.09it/s, tok/s=13842.7]
Inference: 22%|██▏ | 1297/5920 [18:39<1:10:34, 1.09it/s, tok/s=13854.4]
Inference: 22%|██▏ | 1298/5920 [18:40<1:19:27, 1.03s/it, tok/s=13854.4]
Inference: 22%|██▏ | 1298/5920 [18:40<1:19:27, 1.03s/it, tok/s=13850.2]
Inference: 22%|██▏ | 1299/5920 [18:41<1:30:32, 1.18s/it, tok/s=13850.2]
Inference: 22%|██▏ | 1299/5920 [18:41<1:30:32, 1.18s/it, tok/s=13840.8]
Inference: 22%|██▏ | 1300/5920 [18:42<1:22:38, 1.07s/it, tok/s=13840.8]
Inference: 22%|██▏ | 1300/5920 [18:42<1:22:38, 1.07s/it, tok/s=13857.4]
Inference: 22%|██▏ | 1301/5920 [18:43<1:08:16, 1.13it/s, tok/s=13857.4]
Inference: 22%|██▏ | 1301/5920 [18:43<1:08:16, 1.13it/s, tok/s=13857.1]
Inference: 22%|██▏ | 1302/5920 [18:43<50:16, 1.53it/s, tok/s=13857.1]
Inference: 22%|██▏ | 1302/5920 [18:43<50:16, 1.53it/s, tok/s=13882.8]
Inference: 22%|██▏ | 1303/5920 [18:43<44:29, 1.73it/s, tok/s=13882.8]
Inference: 22%|██▏ | 1303/5920 [18:43<44:29, 1.73it/s, tok/s=13882.0]
Inference: 22%|██▏ | 1304/5920 [18:45<1:06:17, 1.16it/s, tok/s=13882.0]
Inference: 22%|██▏ | 1304/5920 [18:45<1:06:17, 1.16it/s, tok/s=13871.6]
Inference: 22%|██▏ | 1305/5920 [18:45<58:50, 1.31it/s, tok/s=13871.6]
Inference: 22%|██▏ | 1305/5920 [18:45<58:50, 1.31it/s, tok/s=13868.9]
Inference: 22%|██▏ | 1306/5920 [18:45<47:21, 1.62it/s, tok/s=13868.9]
Inference: 22%|██▏ | 1306/5920 [18:45<47:21, 1.62it/s, tok/s=13873.1]
Inference: 22%|██▏ | 1307/5920 [18:46<47:20, 1.62it/s, tok/s=13896.1]
Inference: 22%|██▏ | 1308/5920 [18:46<34:43, 2.21it/s, tok/s=13896.1]
Inference: 22%|██▏ | 1308/5920 [18:46<34:43, 2.21it/s, tok/s=13914.4]
Inference: 22%|██▏ | 1309/5920 [18:49<1:15:00, 1.02it/s, tok/s=13914.4]
Inference: 22%|██▏ | 1309/5920 [18:49<1:15:00, 1.02it/s, tok/s=13908.8]
Inference: 22%|██▏ | 1310/5920 [18:50<1:14:58, 1.02it/s, tok/s=13908.8]
Inference: 22%|██▏ | 1310/5920 [18:50<1:14:58, 1.02it/s, tok/s=13921.4]
Inference: 22%|██▏ | 1311/5920 [18:51<1:34:50, 1.23s/it, tok/s=13921.4]
Inference: 22%|██▏ | 1311/5920 [18:51<1:34:50, 1.23s/it, tok/s=13921.0]
Inference: 22%|██▏ | 1312/5920 [18:52<1:11:44, 1.07it/s, tok/s=13921.0]
Inference: 22%|██▏ | 1312/5920 [18:52<1:11:44, 1.07it/s, tok/s=13945.6]
Inference: 22%|██▏ | 1313/5920 [18:54<1:40:00, 1.30s/it, tok/s=13945.6]
Inference: 22%|██▏ | 1313/5920 [18:54<1:40:00, 1.30s/it, tok/s=13941.9]
Inference: 22%|██▏ | 1314/5920 [18:54<1:19:08, 1.03s/it, tok/s=13941.9]
Inference: 22%|██▏ | 1314/5920 [18:54<1:19:08, 1.03s/it, tok/s=13961.5]
Inference: 22%|██▏ | 1315/5920 [18:54<1:19:07, 1.03s/it, tok/s=13969.9]
Inference: 22%|██▏ | 1316/5920 [18:55<1:05:06, 1.18it/s, tok/s=13969.9]
Inference: 22%|██▏ | 1316/5920 [18:55<1:05:06, 1.18it/s, tok/s=13978.0]
Inference: 22%|██▏ | 1317/5920 [18:56<55:29, 1.38it/s, tok/s=13978.0]
Inference: 22%|██▏ | 1317/5920 [18:56<55:29, 1.38it/s, tok/s=13997.0]
Inference: 22%|██▏ | 1318/5920 [18:56<49:32, 1.55it/s, tok/s=13997.0]
Inference: 22%|██▏ | 1318/5920 [18:56<49:32, 1.55it/s, tok/s=13999.9]
Inference: 22%|██▏ | 1319/5920 [18:57<48:25, 1.58it/s, tok/s=13999.9]
Inference: 22%|██▏ | 1319/5920 [18:57<48:25, 1.58it/s, tok/s=14001.9]
Inference: 22%|██▏ | 1320/5920 [18:57<48:34, 1.58it/s, tok/s=14001.9]
Inference: 22%|██▏ | 1320/5920 [18:57<48:34, 1.58it/s, tok/s=14005.9]
Inference: 22%|██▏ | 1321/5920 [18:58<52:02, 1.47it/s, tok/s=14005.9]
Inference: 22%|██▏ | 1321/5920 [18:58<52:02, 1.47it/s, tok/s=14000.7]
Inference: 22%|██▏ | 1322/5920 [18:59<1:00:00, 1.28it/s, tok/s=14000.7]
Inference: 22%|██▏ | 1322/5920 [18:59<1:00:00, 1.28it/s, tok/s=13999.9]
Inference: 22%|██▏ | 1323/5920 [19:01<1:15:58, 1.01it/s, tok/s=13999.9]
Inference: 22%|██▏ | 1323/5920 [19:01<1:15:58, 1.01it/s, tok/s=13989.9]
Inference: 22%|██▏ | 1324/5920 [19:02<1:13:10, 1.05it/s, tok/s=13989.9]
Inference: 22%|██▏ | 1324/5920 [19:02<1:13:10, 1.05it/s, tok/s=13990.5]
Inference: 22%|██▏ | 1325/5920 [19:03<1:29:53, 1.17s/it, tok/s=13990.5]
Inference: 22%|██▏ | 1325/5920 [19:03<1:29:53, 1.17s/it, tok/s=13995.6]
Inference: 22%|██▏ | 1326/5920 [19:03<1:05:55, 1.16it/s, tok/s=13995.6]
Inference: 22%|██▏ | 1326/5920 [19:03<1:05:55, 1.16it/s, tok/s=14014.0]
Inference: 22%|██▏ | 1327/5920 [19:04<49:23, 1.55it/s, tok/s=14014.0]
Inference: 22%|██▏ | 1327/5920 [19:04<49:23, 1.55it/s, tok/s=14020.5]
Inference: 22%|██▏ | 1328/5920 [19:04<38:57, 1.96it/s, tok/s=14020.5]
Inference: 22%|██▏ | 1328/5920 [19:04<38:57, 1.96it/s, tok/s=14029.7]
Inference: 22%|██▏ | 1329/5920 [19:04<36:18, 2.11it/s, tok/s=14029.7]
Inference: 22%|██▏ | 1329/5920 [19:04<36:18, 2.11it/s, tok/s=14045.2]
Inference: 22%|██▏ | 1330/5920 [19:05<50:24, 1.52it/s, tok/s=14045.2]
Inference: 22%|██▏ | 1330/5920 [19:05<50:24, 1.52it/s, tok/s=14040.5]
Inference: 22%|██▏ | 1331/5920 [19:06<43:16, 1.77it/s, tok/s=14040.5]
Inference: 22%|██▏ | 1331/5920 [19:06<43:16, 1.77it/s, tok/s=14053.4]
Inference: 22%|██▎ | 1332/5920 [19:10<2:05:22, 1.64s/it, tok/s=14053.4]
Inference: 22%|██▎ | 1332/5920 [19:10<2:05:22, 1.64s/it, tok/s=14023.7]
Inference: 23%|██▎ | 1333/5920 [19:10<1:32:22, 1.21s/it, tok/s=14023.7]
Inference: 23%|██▎ | 1333/5920 [19:10<1:32:22, 1.21s/it, tok/s=14047.4]
Inference: 23%|██▎ | 1334/5920 [19:10<1:32:21, 1.21s/it, tok/s=14055.3]
Inference: 23%|██▎ | 1335/5920 [19:10<52:31, 1.45it/s, tok/s=14055.3]
Inference: 23%|██▎ | 1335/5920 [19:10<52:31, 1.45it/s, tok/s=14058.4]
Inference: 23%|██▎ | 1336/5920 [19:10<52:31, 1.45it/s, tok/s=14066.5]
Inference: 23%|██▎ | 1337/5920 [19:10<52:30, 1.45it/s, tok/s=14076.4]
Inference: 23%|██▎ | 1338/5920 [19:14<1:11:45, 1.06it/s, tok/s=14076.4]
Inference: 23%|██▎ | 1338/5920 [19:14<1:11:45, 1.06it/s, tok/s=14058.9]
Inference: 23%|██▎ | 1339/5920 [19:14<1:01:54, 1.23it/s, tok/s=14058.9]
Inference: 23%|██▎ | 1339/5920 [19:14<1:01:54, 1.23it/s, tok/s=14079.6]
Inference: 23%|██▎ | 1340/5920 [19:15<57:04, 1.34it/s, tok/s=14079.6]
Inference: 23%|██▎ | 1340/5920 [19:15<57:04, 1.34it/s, tok/s=14077.5]
Inference: 23%|██▎ | 1341/5920 [19:16<1:16:33, 1.00s/it, tok/s=14077.5]
Inference: 23%|██▎ | 1341/5920 [19:16<1:16:33, 1.00s/it, tok/s=14059.7]
Inference: 23%|██▎ | 1342/5920 [19:17<1:08:24, 1.12it/s, tok/s=14059.7]
Inference: 23%|██▎ | 1342/5920 [19:17<1:08:24, 1.12it/s, tok/s=14060.5]
Inference: 23%|██▎ | 1343/5920 [19:19<1:38:37, 1.29s/it, tok/s=14060.5]
Inference: 23%|██▎ | 1343/5920 [19:19<1:38:37, 1.29s/it, tok/s=14036.1]
Inference: 23%|██▎ | 1344/5920 [19:21<1:44:24, 1.37s/it, tok/s=14036.1]
Inference: 23%|██▎ | 1344/5920 [19:21<1:44:24, 1.37s/it, tok/s=14026.5]
Inference: 23%|██▎ | 1345/5920 [19:22<1:47:03, 1.40s/it, tok/s=14026.5]
Inference: 23%|██▎ | 1345/5920 [19:22<1:47:03, 1.40s/it, tok/s=14019.0]
Inference: 23%|██▎ | 1346/5920 [19:23<1:22:33, 1.08s/it, tok/s=14019.0]
Inference: 23%|██▎ | 1346/5920 [19:23<1:22:33, 1.08s/it, tok/s=14020.0]
Inference: 23%|██▎ | 1347/5920 [19:23<1:00:52, 1.25it/s, tok/s=14020.0]
Inference: 23%|██▎ | 1347/5920 [19:23<1:00:52, 1.25it/s, tok/s=14028.0]
Inference: 23%|██▎ | 1348/5920 [19:23<45:31, 1.67it/s, tok/s=14028.0]
Inference: 23%|██▎ | 1348/5920 [19:23<45:31, 1.67it/s, tok/s=14047.7]
Inference: 23%|██▎ | 1349/5920 [19:23<36:14, 2.10it/s, tok/s=14047.7]
Inference: 23%|██▎ | 1349/5920 [19:23<36:14, 2.10it/s, tok/s=14071.6]
Inference: 23%|██▎ | 1350/5920 [19:24<50:03, 1.52it/s, tok/s=14071.6]
Inference: 23%|██▎ | 1350/5920 [19:24<50:03, 1.52it/s, tok/s=14063.5]
Inference: 23%|██▎ | 1351/5920 [19:26<1:24:43, 1.11s/it, tok/s=14063.5]
Inference: 23%|██▎ | 1351/5920 [19:26<1:24:43, 1.11s/it, tok/s=14042.5]
Inference: 23%|██▎ | 1352/5920 [19:29<1:52:09, 1.47s/it, tok/s=14042.5]
Inference: 23%|██▎ | 1352/5920 [19:29<1:52:09, 1.47s/it, tok/s=14023.7]
Inference: 23%|██▎ | 1353/5920 [19:30<1:46:10, 1.39s/it, tok/s=14023.7]
Inference: 23%|██▎ | 1353/5920 [19:30<1:46:10, 1.39s/it, tok/s=14032.3]
Inference: 23%|██▎ | 1354/5920 [19:31<1:43:12, 1.36s/it, tok/s=14032.3]
Inference: 23%|██▎ | 1354/5920 [19:31<1:43:12, 1.36s/it, tok/s=14039.3]
Inference: 23%|██▎ | 1355/5920 [19:32<1:40:48, 1.33s/it, tok/s=14039.3]
Inference: 23%|██▎ | 1355/5920 [19:32<1:40:48, 1.33s/it, tok/s=14029.4]
Inference: 23%|██▎ | 1356/5920 [19:32<1:40:47, 1.33s/it, tok/s=14033.6]
Inference: 23%|██▎ | 1357/5920 [19:33<57:20, 1.33it/s, tok/s=14033.6]
Inference: 23%|██▎ | 1357/5920 [19:33<57:20, 1.33it/s, tok/s=14044.9]
Inference: 23%|██▎ | 1358/5920 [19:33<57:20, 1.33it/s, tok/s=14053.8]
Inference: 23%|██▎ | 1359/5920 [19:34<50:44, 1.50it/s, tok/s=14053.8]
Inference: 23%|██▎ | 1359/5920 [19:34<50:44, 1.50it/s, tok/s=14051.2]
Inference: 23%|██▎ | 1360/5920 [19:35<54:17, 1.40it/s, tok/s=14051.2]
Inference: 23%|██▎ | 1360/5920 [19:35<54:17, 1.40it/s, tok/s=14063.3]
Inference: 23%|██▎ | 1361/5920 [19:36<58:58, 1.29it/s, tok/s=14063.3]
Inference: 23%|██▎ | 1361/5920 [19:36<58:58, 1.29it/s, tok/s=14057.4]
Inference: 23%|██▎ | 1362/5920 [19:36<49:34, 1.53it/s, tok/s=14057.4]
Inference: 23%|██▎ | 1362/5920 [19:36<49:34, 1.53it/s, tok/s=14066.7]
Inference: 23%|██▎ | 1363/5920 [19:36<49:33, 1.53it/s, tok/s=14090.8]
Inference: 23%|██▎ | 1364/5920 [19:36<33:50, 2.24it/s, tok/s=14090.8]
Inference: 23%|██▎ | 1364/5920 [19:36<33:50, 2.24it/s, tok/s=14094.8]
Inference: 23%|██▎ | 1365/5920 [19:37<36:14, 2.09it/s, tok/s=14094.8]
Inference: 23%|██▎ | 1365/5920 [19:37<36:14, 2.09it/s, tok/s=14111.3]
Inference: 23%|██▎ | 1366/5920 [19:39<1:07:15, 1.13it/s, tok/s=14111.3]
Inference: 23%|██▎ | 1366/5920 [19:39<1:07:15, 1.13it/s, tok/s=14099.4]
Inference: 23%|██▎ | 1367/5920 [19:40<1:06:33, 1.14it/s, tok/s=14099.4]
Inference: 23%|██▎ | 1367/5920 [19:40<1:06:33, 1.14it/s, tok/s=14094.3]
Inference: 23%|██▎ | 1368/5920 [19:41<1:08:25, 1.11it/s, tok/s=14094.3]
Inference: 23%|██▎ | 1368/5920 [19:41<1:08:25, 1.11it/s, tok/s=14097.5]
Inference: 23%|██▎ | 1369/5920 [19:41<57:22, 1.32it/s, tok/s=14097.5]
Inference: 23%|██▎ | 1369/5920 [19:41<57:22, 1.32it/s, tok/s=14098.0]
Inference: 23%|██▎ | 1370/5920 [19:42<59:18, 1.28it/s, tok/s=14098.0]
Inference: 23%|██▎ | 1370/5920 [19:42<59:18, 1.28it/s, tok/s=14099.4]
Inference: 23%|██▎ | 1371/5920 [19:43<1:07:43, 1.12it/s, tok/s=14099.4]
Inference: 23%|██▎ | 1371/5920 [19:43<1:07:43, 1.12it/s, tok/s=14091.0]
Inference: 23%|██▎ | 1372/5920 [19:44<1:11:00, 1.07it/s, tok/s=14091.0]
Inference: 23%|██▎ | 1372/5920 [19:44<1:11:00, 1.07it/s, tok/s=14100.4]
Inference: 23%|██▎ | 1373/5920 [19:48<2:26:54, 1.94s/it, tok/s=14100.4]
Inference: 23%|██▎ | 1373/5920 [19:48<2:26:54, 1.94s/it, tok/s=14058.9]
Inference: 23%|██▎ | 1374/5920 [19:49<1:46:26, 1.40s/it, tok/s=14058.9]
Inference: 23%|██▎ | 1374/5920 [19:49<1:46:26, 1.40s/it, tok/s=14068.0]
Inference: 23%|██▎ | 1375/5920 [19:49<1:35:09, 1.26s/it, tok/s=14068.0]
Inference: 23%|██▎ | 1375/5920 [19:49<1:35:09, 1.26s/it, tok/s=14063.0]
Inference: 23%|██▎ | 1376/5920 [19:50<1:23:08, 1.10s/it, tok/s=14063.0]
Inference: 23%|██▎ | 1376/5920 [19:50<1:23:08, 1.10s/it, tok/s=14060.0]
Inference: 23%|██▎ | 1377/5920 [19:51<1:12:05, 1.05it/s, tok/s=14060.0]
Inference: 23%|██▎ | 1377/5920 [19:51<1:12:05, 1.05it/s, tok/s=14078.1]
Inference: 23%|██▎ | 1378/5920 [19:52<1:20:05, 1.06s/it, tok/s=14078.1]
Inference: 23%|██▎ | 1378/5920 [19:52<1:20:05, 1.06s/it, tok/s=14071.8]
Inference: 23%|██▎ | 1379/5920 [19:52<1:00:08, 1.26it/s, tok/s=14071.8]
Inference: 23%|██▎ | 1379/5920 [19:52<1:00:08, 1.26it/s, tok/s=14079.1]
Inference: 23%|██▎ | 1380/5920 [19:52<1:00:07, 1.26it/s, tok/s=14084.0]
Inference: 23%|██▎ | 1381/5920 [19:53<38:04, 1.99it/s, tok/s=14084.0]
Inference: 23%|██▎ | 1381/5920 [19:53<38:04, 1.99it/s, tok/s=14102.6]
Inference: 23%|██▎ | 1382/5920 [19:55<1:18:32, 1.04s/it, tok/s=14102.6]
Inference: 23%|██▎ | 1382/5920 [19:55<1:18:32, 1.04s/it, tok/s=14096.3]
Inference: 23%|██▎ | 1383/5920 [19:56<1:05:21, 1.16it/s, tok/s=14096.3]
Inference: 23%|██▎ | 1383/5920 [19:56<1:05:21, 1.16it/s, tok/s=14106.2]
Inference: 23%|██▎ | 1384/5920 [19:57<1:09:20, 1.09it/s, tok/s=14106.2]
Inference: 23%|██▎ | 1384/5920 [19:57<1:09:20, 1.09it/s, tok/s=14097.6]
Inference: 23%|██▎ | 1385/5920 [19:58<1:24:08, 1.11s/it, tok/s=14097.6]
Inference: 23%|██▎ | 1385/5920 [19:58<1:24:08, 1.11s/it, tok/s=14103.7]
Inference: 23%|██▎ | 1386/5920 [19:58<1:24:07, 1.11s/it, tok/s=14128.7]
Inference: 23%|██▎ | 1387/5920 [19:59<51:15, 1.47it/s, tok/s=14128.7]
Inference: 23%|██▎ | 1387/5920 [19:59<51:15, 1.47it/s, tok/s=14130.6]
Inference: 23%|██▎ | 1388/5920 [19:59<52:50, 1.43it/s, tok/s=14130.6]
Inference: 23%|██▎ | 1388/5920 [19:59<52:50, 1.43it/s, tok/s=14131.6]
Inference: 23%|██▎ | 1389/5920 [19:59<52:49, 1.43it/s, tok/s=14141.2]
Inference: 23%|██▎ | 1390/5920 [20:00<35:48, 2.11it/s, tok/s=14141.2]
Inference: 23%|██▎ | 1390/5920 [20:00<35:48, 2.11it/s, tok/s=14164.0]
Inference: 23%|██▎ | 1391/5920 [20:02<1:04:56, 1.16it/s, tok/s=14164.0]
Inference: 23%|██▎ | 1391/5920 [20:02<1:04:56, 1.16it/s, tok/s=14143.6]
Inference: 24%|██▎ | 1392/5920 [20:02<56:28, 1.34it/s, tok/s=14143.6]
Inference: 24%|██▎ | 1392/5920 [20:02<56:28, 1.34it/s, tok/s=14159.1]
Inference: 24%|██▎ | 1393/5920 [20:02<56:27, 1.34it/s, tok/s=14176.6]
Inference: 24%|██▎ | 1394/5920 [20:04<53:27, 1.41it/s, tok/s=14176.6]
Inference: 24%|██▎ | 1394/5920 [20:04<53:27, 1.41it/s, tok/s=14185.3]
Inference: 24%|██▎ | 1395/5920 [20:04<57:15, 1.32it/s, tok/s=14185.3]
Inference: 24%|██▎ | 1395/5920 [20:04<57:15, 1.32it/s, tok/s=14199.4]
Inference: 24%|██▎ | 1396/5920 [20:05<50:48, 1.48it/s, tok/s=14199.4]
Inference: 24%|██▎ | 1396/5920 [20:05<50:48, 1.48it/s, tok/s=14204.9]
Inference: 24%|██▎ | 1397/5920 [20:05<50:48, 1.48it/s, tok/s=14227.0]
Inference: 24%|██▎ | 1398/5920 [20:06<46:37, 1.62it/s, tok/s=14227.0]
Inference: 24%|██▎ | 1398/5920 [20:06<46:37, 1.62it/s, tok/s=14223.8]
Inference: 24%|██▎ | 1399/5920 [20:06<46:37, 1.62it/s, tok/s=14231.3]
Inference: 24%|██▎ | 1400/5920 [20:06<46:36, 1.62it/s, tok/s=14241.5]
Inference: 24%|██▎ | 1401/5920 [20:06<27:38, 2.73it/s, tok/s=14241.5]
Inference: 24%|██▎ | 1401/5920 [20:06<27:38, 2.73it/s, tok/s=14246.0]
Inference: 24%|██▎ | 1402/5920 [20:07<26:57, 2.79it/s, tok/s=14246.0]
Inference: 24%|██▎ | 1402/5920 [20:07<26:57, 2.79it/s, tok/s=14265.1]
Inference: 24%|██▎ | 1403/5920 [20:07<32:09, 2.34it/s, tok/s=14265.1]
Inference: 24%|██▎ | 1403/5920 [20:07<32:09, 2.34it/s, tok/s=14262.2]
Inference: 24%|██▎ | 1404/5920 [20:09<51:01, 1.48it/s, tok/s=14262.2]
Inference: 24%|██▎ | 1404/5920 [20:09<51:01, 1.48it/s, tok/s=14269.1]
Inference: 24%|██▎ | 1405/5920 [20:09<45:28, 1.65it/s, tok/s=14269.1]
Inference: 24%|██▎ | 1405/5920 [20:09<45:28, 1.65it/s, tok/s=14270.2]
Inference: 24%|██▍ | 1406/5920 [20:10<44:50, 1.68it/s, tok/s=14270.2]
Inference: 24%|██▍ | 1406/5920 [20:10<44:50, 1.68it/s, tok/s=14273.0]
Inference: 24%|██▍ | 1407/5920 [20:10<44:50, 1.68it/s, tok/s=14282.7]
Inference: 24%|██▍ | 1408/5920 [20:11<45:40, 1.65it/s, tok/s=14282.7]
Inference: 24%|██▍ | 1408/5920 [20:11<45:40, 1.65it/s, tok/s=14282.4]
Inference: 24%|██▍ | 1409/5920 [20:11<41:06, 1.83it/s, tok/s=14282.4]
Inference: 24%|██▍ | 1409/5920 [20:11<41:06, 1.83it/s, tok/s=14303.1]
Inference: 24%|██▍ | 1410/5920 [20:12<47:47, 1.57it/s, tok/s=14303.1]
Inference: 24%|██▍ | 1410/5920 [20:12<47:47, 1.57it/s, tok/s=14298.8]
Inference: 24%|██▍ | 1411/5920 [20:13<46:43, 1.61it/s, tok/s=14298.8]
Inference: 24%|██▍ | 1411/5920 [20:13<46:43, 1.61it/s, tok/s=14299.5]
Inference: 24%|██▍ | 1412/5920 [20:13<44:24, 1.69it/s, tok/s=14299.5]
Inference: 24%|██▍ | 1412/5920 [20:13<44:24, 1.69it/s, tok/s=14312.3]
Inference: 24%|██▍ | 1413/5920 [20:14<36:47, 2.04it/s, tok/s=14312.3]
Inference: 24%|██▍ | 1413/5920 [20:14<36:47, 2.04it/s, tok/s=14322.4]
Inference: 24%|██▍ | 1414/5920 [20:14<43:49, 1.71it/s, tok/s=14322.4]
Inference: 24%|██▍ | 1414/5920 [20:14<43:49, 1.71it/s, tok/s=14323.8]
Inference: 24%|██▍ | 1415/5920 [20:14<43:48, 1.71it/s, tok/s=14347.8]
Inference: 24%|██▍ | 1416/5920 [20:15<30:58, 2.42it/s, tok/s=14347.8]
Inference: 24%|██▍ | 1416/5920 [20:15<30:58, 2.42it/s, tok/s=14368.6]
Inference: 24%|██▍ | 1417/5920 [20:15<27:22, 2.74it/s, tok/s=14368.6]
Inference: 24%|██▍ | 1417/5920 [20:15<27:22, 2.74it/s, tok/s=14375.6]
Inference: 24%|██▍ | 1418/5920 [20:15<30:39, 2.45it/s, tok/s=14375.6]
Inference: 24%|██▍ | 1418/5920 [20:15<30:39, 2.45it/s, tok/s=14388.8]
Inference: 24%|██▍ | 1419/5920 [20:16<37:08, 2.02it/s, tok/s=14388.8]
Inference: 24%|██▍ | 1419/5920 [20:16<37:08, 2.02it/s, tok/s=14389.8]
Inference: 24%|██▍ | 1420/5920 [20:16<29:18, 2.56it/s, tok/s=14389.8]
Inference: 24%|██▍ | 1420/5920 [20:16<29:18, 2.56it/s, tok/s=14413.1]
Inference: 24%|██▍ | 1421/5920 [20:16<29:18, 2.56it/s, tok/s=14424.4]
Inference: 24%|██▍ | 1422/5920 [20:17<32:59, 2.27it/s, tok/s=14424.4]
Inference: 24%|██▍ | 1422/5920 [20:17<32:59, 2.27it/s, tok/s=14436.8]
Inference: 24%|██▍ | 1423/5920 [20:18<28:24, 2.64it/s, tok/s=14436.8]
Inference: 24%|██▍ | 1423/5920 [20:18<28:24, 2.64it/s, tok/s=14457.0]
Inference: 24%|██▍ | 1424/5920 [20:18<37:41, 1.99it/s, tok/s=14457.0]
Inference: 24%|██▍ | 1424/5920 [20:18<37:41, 1.99it/s, tok/s=14451.5]
Inference: 24%|██▍ | 1425/5920 [20:19<33:06, 2.26it/s, tok/s=14451.5]
Inference: 24%|██▍ | 1425/5920 [20:19<33:06, 2.26it/s, tok/s=14471.6]
Inference: 24%|██▍ | 1426/5920 [20:20<42:16, 1.77it/s, tok/s=14471.6]
Inference: 24%|██▍ | 1426/5920 [20:20<42:16, 1.77it/s, tok/s=14469.8]
Inference: 24%|██▍ | 1427/5920 [20:20<47:12, 1.59it/s, tok/s=14469.8]
Inference: 24%|██▍ | 1427/5920 [20:20<47:12, 1.59it/s, tok/s=14470.5]
Inference: 24%|██▍ | 1428/5920 [20:21<37:40, 1.99it/s, tok/s=14470.5]
Inference: 24%|██▍ | 1428/5920 [20:21<37:40, 1.99it/s, tok/s=14492.9]
Inference: 24%|██▍ | 1429/5920 [20:21<38:01, 1.97it/s, tok/s=14492.9]
Inference: 24%|██▍ | 1429/5920 [20:21<38:01, 1.97it/s, tok/s=14492.9]
Inference: 24%|██▍ | 1430/5920 [20:23<1:13:05, 1.02it/s, tok/s=14492.9]
Inference: 24%|██▍ | 1430/5920 [20:23<1:13:05, 1.02it/s, tok/s=14479.0]
Inference: 24%|██▍ | 1431/5920 [20:23<1:13:04, 1.02it/s, tok/s=14485.3]
Inference: 24%|██▍ | 1432/5920 [20:24<56:54, 1.31it/s, tok/s=14485.3]
Inference: 24%|██▍ | 1432/5920 [20:24<56:54, 1.31it/s, tok/s=14496.7]
Inference: 24%|██▍ | 1433/5920 [20:25<55:50, 1.34it/s, tok/s=14496.7]
Inference: 24%|██▍ | 1433/5920 [20:25<55:50, 1.34it/s, tok/s=14501.2]
Inference: 24%|██▍ | 1434/5920 [20:26<1:10:39, 1.06it/s, tok/s=14501.2]
Inference: 24%|██▍ | 1434/5920 [20:26<1:10:39, 1.06it/s, tok/s=14507.9]
Inference: 24%|██▍ | 1435/5920 [20:27<58:36, 1.28it/s, tok/s=14507.9]
Inference: 24%|██▍ | 1435/5920 [20:27<58:36, 1.28it/s, tok/s=14511.8]
Inference: 24%|██▍ | 1436/5920 [20:27<55:33, 1.35it/s, tok/s=14511.8]
Inference: 24%|██▍ | 1436/5920 [20:27<55:33, 1.35it/s, tok/s=14508.3]
Inference: 24%|██▍ | 1437/5920 [20:29<1:07:55, 1.10it/s, tok/s=14508.3]
Inference: 24%|██▍ | 1437/5920 [20:29<1:07:55, 1.10it/s, tok/s=14499.2]
Inference: 24%|██▍ | 1438/5920 [20:29<1:02:08, 1.20it/s, tok/s=14499.2]
Inference: 24%|██▍ | 1438/5920 [20:29<1:02:08, 1.20it/s, tok/s=14505.2]
Inference: 24%|██▍ | 1439/5920 [20:29<1:02:07, 1.20it/s, tok/s=14510.4]
Inference: 24%|██▍ | 1440/5920 [20:32<1:12:48, 1.03it/s, tok/s=14510.4]
Inference: 24%|██▍ | 1440/5920 [20:32<1:12:48, 1.03it/s, tok/s=14488.0]
Inference: 24%|██▍ | 1441/5920 [20:32<1:05:49, 1.13it/s, tok/s=14488.0]
Inference: 24%|██▍ | 1441/5920 [20:32<1:05:49, 1.13it/s, tok/s=14489.6]
Inference: 24%|██▍ | 1442/5920 [20:34<1:16:40, 1.03s/it, tok/s=14489.6]
Inference: 24%|██▍ | 1442/5920 [20:34<1:16:40, 1.03s/it, tok/s=14481.6]
Inference: 24%|██▍ | 1443/5920 [20:35<1:12:54, 1.02it/s, tok/s=14481.6]
Inference: 24%|██▍ | 1443/5920 [20:35<1:12:54, 1.02it/s, tok/s=14495.7]
Inference: 24%|██▍ | 1444/5920 [20:35<1:09:49, 1.07it/s, tok/s=14495.7]
Inference: 24%|██▍ | 1444/5920 [20:35<1:09:49, 1.07it/s, tok/s=14492.9]
Inference: 24%|██▍ | 1445/5920 [20:36<1:04:47, 1.15it/s, tok/s=14492.9]
Inference: 24%|██▍ | 1445/5920 [20:36<1:04:47, 1.15it/s, tok/s=14492.5]
Inference: 24%|██▍ | 1446/5920 [20:37<58:28, 1.28it/s, tok/s=14492.5]
Inference: 24%|██▍ | 1446/5920 [20:37<58:28, 1.28it/s, tok/s=14510.1]
Inference: 24%|██▍ | 1447/5920 [20:38<1:14:11, 1.00it/s, tok/s=14510.1]
Inference: 24%|██▍ | 1447/5920 [20:38<1:14:11, 1.00it/s, tok/s=14516.7]
Inference: 24%|██▍ | 1448/5920 [20:39<1:11:51, 1.04it/s, tok/s=14516.7]
Inference: 24%|██▍ | 1448/5920 [20:39<1:11:51, 1.04it/s, tok/s=14528.7]
Inference: 24%|██▍ | 1449/5920 [20:41<1:29:56, 1.21s/it, tok/s=14528.7]
Inference: 24%|██▍ | 1449/5920 [20:41<1:29:56, 1.21s/it, tok/s=14530.9]
Inference: 24%|██▍ | 1450/5920 [20:42<1:21:22, 1.09s/it, tok/s=14530.9]
Inference: 24%|██▍ | 1450/5920 [20:42<1:21:22, 1.09s/it, tok/s=14527.8]
Inference: 25%|██▍ | 1451/5920 [20:42<1:21:21, 1.09s/it, tok/s=14541.2]
Inference: 25%|██▍ | 1452/5920 [20:43<1:04:42, 1.15it/s, tok/s=14541.2]
Inference: 25%|██▍ | 1452/5920 [20:43<1:04:42, 1.15it/s, tok/s=14533.1]
Inference: 25%|██▍ | 1453/5920 [20:44<1:02:30, 1.19it/s, tok/s=14533.1]
Inference: 25%|██▍ | 1453/5920 [20:44<1:02:30, 1.19it/s, tok/s=14529.7]
Inference: 25%|██▍ | 1454/5920 [20:46<1:30:26, 1.22s/it, tok/s=14529.7]
Inference: 25%|██▍ | 1454/5920 [20:46<1:30:26, 1.22s/it, tok/s=14527.3]
Inference: 25%|██▍ | 1455/5920 [20:48<1:39:05, 1.33s/it, tok/s=14527.3]
Inference: 25%|██▍ | 1455/5920 [20:48<1:39:05, 1.33s/it, tok/s=14516.2]
Inference: 25%|██▍ | 1456/5920 [20:49<1:44:53, 1.41s/it, tok/s=14516.2]
Inference: 25%|██▍ | 1456/5920 [20:49<1:44:53, 1.41s/it, tok/s=14521.6]
Inference: 25%|██▍ | 1457/5920 [20:50<1:26:12, 1.16s/it, tok/s=14521.6]
Inference: 25%|██▍ | 1457/5920 [20:50<1:26:12, 1.16s/it, tok/s=14517.8]
Inference: 25%|██▍ | 1458/5920 [20:51<1:33:46, 1.26s/it, tok/s=14517.8]
Inference: 25%|██▍ | 1458/5920 [20:51<1:33:46, 1.26s/it, tok/s=14511.8]
Inference: 25%|██▍ | 1459/5920 [20:53<1:45:55, 1.42s/it, tok/s=14511.8]
Inference: 25%|██▍ | 1459/5920 [20:53<1:45:55, 1.42s/it, tok/s=14502.2]
Inference: 25%|██▍ | 1460/5920 [20:54<1:42:51, 1.38s/it, tok/s=14502.2]
Inference: 25%|██▍ | 1460/5920 [20:54<1:42:51, 1.38s/it, tok/s=14496.9]
Inference: 25%|██▍ | 1461/5920 [20:55<1:33:58, 1.26s/it, tok/s=14496.9]
Inference: 25%|██▍ | 1461/5920 [20:55<1:33:58, 1.26s/it, tok/s=14507.7]
Inference: 25%|██▍ | 1462/5920 [20:55<1:11:15, 1.04it/s, tok/s=14507.7]
Inference: 25%|██▍ | 1462/5920 [20:55<1:11:15, 1.04it/s, tok/s=14529.0]
Inference: 25%|██▍ | 1463/5920 [20:56<1:03:29, 1.17it/s, tok/s=14529.0]
Inference: 25%|██▍ | 1463/5920 [20:56<1:03:29, 1.17it/s, tok/s=14530.6]
Inference: 25%|██▍ | 1464/5920 [20:56<48:28, 1.53it/s, tok/s=14530.6]
Inference: 25%|██▍ | 1464/5920 [20:56<48:28, 1.53it/s, tok/s=14536.8]
Inference: 25%|██▍ | 1465/5920 [20:56<48:27, 1.53it/s, tok/s=14547.2]
Inference: 25%|██▍ | 1466/5920 [20:57<34:22, 2.16it/s, tok/s=14547.2]
Inference: 25%|██▍ | 1466/5920 [20:57<34:22, 2.16it/s, tok/s=14546.7]
Inference: 25%|██▍ | 1467/5920 [20:57<33:29, 2.22it/s, tok/s=14546.7]
Inference: 25%|██▍ | 1467/5920 [20:57<33:29, 2.22it/s, tok/s=14563.8]
Inference: 25%|██▍ | 1468/5920 [20:57<33:29, 2.22it/s, tok/s=14586.5]
Inference: 25%|██▍ | 1469/5920 [20:58<37:23, 1.98it/s, tok/s=14586.5]
Inference: 25%|██▍ | 1469/5920 [20:58<37:23, 1.98it/s, tok/s=14576.5]
Inference: 25%|██▍ | 1470/5920 [20:58<30:45, 2.41it/s, tok/s=14576.5]
Inference: 25%|██▍ | 1470/5920 [20:58<30:45, 2.41it/s, tok/s=14585.5]
Inference: 25%|██▍ | 1471/5920 [20:59<37:20, 1.99it/s, tok/s=14585.5]
Inference: 25%|██▍ | 1471/5920 [20:59<37:20, 1.99it/s, tok/s=14584.1]
Inference: 25%|██▍ | 1472/5920 [21:01<1:11:04, 1.04it/s, tok/s=14584.1]
Inference: 25%|██▍ | 1472/5920 [21:01<1:11:04, 1.04it/s, tok/s=14578.7]
Inference: 25%|██▍ | 1473/5920 [21:01<1:11:03, 1.04it/s, tok/s=14586.2]
Inference: 25%|██▍ | 1474/5920 [21:02<45:45, 1.62it/s, tok/s=14586.2]
Inference: 25%|██▍ | 1474/5920 [21:02<45:45, 1.62it/s, tok/s=14606.4]
Inference: 25%|██▍ | 1475/5920 [21:03<59:30, 1.24it/s, tok/s=14606.4]
Inference: 25%|██▍ | 1475/5920 [21:03<59:30, 1.24it/s, tok/s=14612.3]
Inference: 25%|██▍ | 1476/5920 [21:07<1:49:15, 1.48s/it, tok/s=14612.3]
Inference: 25%|██▍ | 1476/5920 [21:07<1:49:15, 1.48s/it, tok/s=14574.5]
Inference: 25%|██▍ | 1477/5920 [21:07<1:23:59, 1.13s/it, tok/s=14574.5]
Inference: 25%|██▍ | 1477/5920 [21:07<1:23:59, 1.13s/it, tok/s=14583.9]
Inference: 25%|██▍ | 1478/5920 [21:08<1:26:23, 1.17s/it, tok/s=14583.9]
Inference: 25%|██▍ | 1478/5920 [21:08<1:26:23, 1.17s/it, tok/s=14593.3]
Inference: 25%|██▍ | 1479/5920 [21:11<1:51:53, 1.51s/it, tok/s=14593.3]
Inference: 25%|██▍ | 1479/5920 [21:11<1:51:53, 1.51s/it, tok/s=14574.6]
Inference: 25%|██▌ | 1480/5920 [21:11<1:21:58, 1.11s/it, tok/s=14574.6]
Inference: 25%|██▌ | 1480/5920 [21:11<1:21:58, 1.11s/it, tok/s=14579.0]
Inference: 25%|██▌ | 1481/5920 [21:11<1:16:15, 1.03s/it, tok/s=14579.0]
Inference: 25%|██▌ | 1481/5920 [21:11<1:16:15, 1.03s/it, tok/s=14593.0]
Inference: 25%|██▌ | 1482/5920 [21:12<1:10:28, 1.05it/s, tok/s=14593.0]
Inference: 25%|██▌ | 1482/5920 [21:12<1:10:28, 1.05it/s, tok/s=14606.4]
Inference: 25%|██▌ | 1483/5920 [21:12<52:02, 1.42it/s, tok/s=14606.4]
Inference: 25%|██▌ | 1483/5920 [21:12<52:02, 1.42it/s, tok/s=14614.7]
Inference: 25%|██▌ | 1484/5920 [21:14<1:12:17, 1.02it/s, tok/s=14614.7]
Inference: 25%|██▌ | 1484/5920 [21:14<1:12:17, 1.02it/s, tok/s=14606.8]
Inference: 25%|██▌ | 1485/5920 [21:15<1:14:30, 1.01s/it, tok/s=14606.8]
Inference: 25%|██▌ | 1485/5920 [21:15<1:14:30, 1.01s/it, tok/s=14618.1]
Inference: 25%|██▌ | 1486/5920 [21:16<1:11:48, 1.03it/s, tok/s=14618.1]
Inference: 25%|██▌ | 1486/5920 [21:16<1:11:48, 1.03it/s, tok/s=14610.4]
Inference: 25%|██▌ | 1487/5920 [21:18<1:29:38, 1.21s/it, tok/s=14610.4]
Inference: 25%|██▌ | 1487/5920 [21:18<1:29:38, 1.21s/it, tok/s=14598.9]
Inference: 25%|██▌ | 1488/5920 [21:18<1:19:45, 1.08s/it, tok/s=14598.9]
Inference: 25%|██▌ | 1488/5920 [21:18<1:19:45, 1.08s/it, tok/s=14602.6]
Inference: 25%|██▌ | 1489/5920 [21:19<1:13:52, 1.00s/it, tok/s=14602.6]
Inference: 25%|██▌ | 1489/5920 [21:19<1:13:52, 1.00s/it, tok/s=14595.9]
Inference: 25%|██▌ | 1490/5920 [21:20<57:19, 1.29it/s, tok/s=14595.9]
Inference: 25%|██▌ | 1490/5920 [21:20<57:19, 1.29it/s, tok/s=14601.6]
Inference: 25%|██▌ | 1491/5920 [21:20<1:01:01, 1.21it/s, tok/s=14601.6]
Inference: 25%|██▌ | 1491/5920 [21:20<1:01:01, 1.21it/s, tok/s=14603.2]
Inference: 25%|██▌ | 1492/5920 [21:21<55:00, 1.34it/s, tok/s=14603.2]
Inference: 25%|██▌ | 1492/5920 [21:21<55:00, 1.34it/s, tok/s=14599.3]
Inference: 25%|██▌ | 1493/5920 [21:22<50:16, 1.47it/s, tok/s=14599.3]
Inference: 25%|██▌ | 1493/5920 [21:22<50:16, 1.47it/s, tok/s=14601.8]
Inference: 25%|██▌ | 1494/5920 [21:22<45:17, 1.63it/s, tok/s=14601.8]
Inference: 25%|██▌ | 1494/5920 [21:22<45:17, 1.63it/s, tok/s=14610.7]
Inference: 25%|██▌ | 1495/5920 [21:22<36:17, 2.03it/s, tok/s=14610.7]
Inference: 25%|██▌ | 1495/5920 [21:22<36:17, 2.03it/s, tok/s=14631.8]
Inference: 25%|██▌ | 1496/5920 [21:23<33:47, 2.18it/s, tok/s=14631.8]
Inference: 25%|██▌ | 1496/5920 [21:23<33:47, 2.18it/s, tok/s=14630.0]
Inference: 25%|██▌ | 1497/5920 [21:23<29:51, 2.47it/s, tok/s=14630.0]
Inference: 25%|██▌ | 1497/5920 [21:23<29:51, 2.47it/s, tok/s=14639.6]
Inference: 25%|██▌ | 1498/5920 [21:24<56:16, 1.31it/s, tok/s=14639.6]
Inference: 25%|██▌ | 1498/5920 [21:24<56:16, 1.31it/s, tok/s=14631.7]
Inference: 25%|██▌ | 1499/5920 [21:25<46:18, 1.59it/s, tok/s=14631.7]
Inference: 25%|██▌ | 1499/5920 [21:25<46:18, 1.59it/s, tok/s=14636.7]
Inference: 25%|██▌ | 1500/5920 [21:25<37:13, 1.98it/s, tok/s=14636.7]
Inference: 25%|██▌ | 1500/5920 [21:25<37:13, 1.98it/s, tok/s=14655.0]
Inference: 25%|██▌ | 1501/5920 [21:26<48:03, 1.53it/s, tok/s=14655.0]
Inference: 25%|██▌ | 1501/5920 [21:26<48:03, 1.53it/s, tok/s=14667.1]
Inference: 25%|██▌ | 1502/5920 [21:27<45:23, 1.62it/s, tok/s=14667.1]
Inference: 25%|██▌ | 1502/5920 [21:27<45:23, 1.62it/s, tok/s=14684.4]
Inference: 25%|██▌ | 1503/5920 [21:27<39:00, 1.89it/s, tok/s=14684.4]
Inference: 25%|██▌ | 1503/5920 [21:27<39:00, 1.89it/s, tok/s=14686.3]
Inference: 25%|██▌ | 1504/5920 [21:28<46:22, 1.59it/s, tok/s=14686.3]
Inference: 25%|██▌ | 1504/5920 [21:28<46:22, 1.59it/s, tok/s=14684.6]
Inference: 25%|██▌ | 1505/5920 [21:29<56:00, 1.31it/s, tok/s=14684.6]
Inference: 25%|██▌ | 1505/5920 [21:29<56:00, 1.31it/s, tok/s=14675.2]
Inference: 25%|██▌ | 1506/5920 [21:31<1:18:23, 1.07s/it, tok/s=14675.2]
Inference: 25%|██▌ | 1506/5920 [21:31<1:18:23, 1.07s/it, tok/s=14668.4]
Inference: 25%|██▌ | 1507/5920 [21:31<1:05:18, 1.13it/s, tok/s=14668.4]
Inference: 25%|██▌ | 1507/5920 [21:31<1:05:18, 1.13it/s, tok/s=14686.3]
Inference: 25%|██▌ | 1508/5920 [21:31<53:47, 1.37it/s, tok/s=14686.3]
Inference: 25%|██▌ | 1508/5920 [21:31<53:47, 1.37it/s, tok/s=14705.4]
Inference: 25%|██▌ | 1509/5920 [21:32<55:55, 1.31it/s, tok/s=14705.4]
Inference: 25%|██▌ | 1509/5920 [21:32<55:55, 1.31it/s, tok/s=14719.3]
Inference: 26%|██▌ | 1510/5920 [21:34<1:08:51, 1.07it/s, tok/s=14719.3]
Inference: 26%|██▌ | 1510/5920 [21:34<1:08:51, 1.07it/s, tok/s=14727.2]
Inference: 26%|██▌ | 1511/5920 [21:35<1:21:53, 1.11s/it, tok/s=14727.2]
Inference: 26%|██▌ | 1511/5920 [21:35<1:21:53, 1.11s/it, tok/s=14715.3]
Inference: 26%|██▌ | 1512/5920 [21:35<1:04:10, 1.14it/s, tok/s=14715.3]
Inference: 26%|██▌ | 1512/5920 [21:35<1:04:10, 1.14it/s, tok/s=14724.4]
Inference: 26%|██▌ | 1513/5920 [21:36<48:31, 1.51it/s, tok/s=14724.4]
Inference: 26%|██▌ | 1513/5920 [21:36<48:31, 1.51it/s, tok/s=14745.0]
Inference: 26%|██▌ | 1514/5920 [21:37<55:58, 1.31it/s, tok/s=14745.0]
Inference: 26%|██▌ | 1514/5920 [21:37<55:58, 1.31it/s, tok/s=14745.7]
Inference: 26%|██▌ | 1515/5920 [21:37<55:58, 1.31it/s, tok/s=14753.8]
Inference: 26%|██▌ | 1516/5920 [21:37<38:09, 1.92it/s, tok/s=14753.8]
Inference: 26%|██▌ | 1516/5920 [21:37<38:09, 1.92it/s, tok/s=14751.9]
Inference: 26%|██▌ | 1517/5920 [21:38<48:12, 1.52it/s, tok/s=14751.9]
Inference: 26%|██▌ | 1517/5920 [21:38<48:12, 1.52it/s, tok/s=14747.7]
Inference: 26%|██▌ | 1518/5920 [21:38<48:12, 1.52it/s, tok/s=14755.5]
Inference: 26%|██▌ | 1519/5920 [21:38<30:58, 2.37it/s, tok/s=14755.5]
Inference: 26%|██▌ | 1519/5920 [21:38<30:58, 2.37it/s, tok/s=14777.3]
Inference: 26%|██▌ | 1520/5920 [21:39<42:31, 1.72it/s, tok/s=14777.3]
Inference: 26%|██▌ | 1520/5920 [21:39<42:31, 1.72it/s, tok/s=14787.2]
Inference: 26%|██▌ | 1521/5920 [21:40<34:52, 2.10it/s, tok/s=14787.2]
Inference: 26%|██▌ | 1521/5920 [21:40<34:52, 2.10it/s, tok/s=14791.5]
Inference: 26%|██▌ | 1522/5920 [21:40<30:12, 2.43it/s, tok/s=14791.5]
Inference: 26%|██▌ | 1522/5920 [21:40<30:12, 2.43it/s, tok/s=14791.5]
Inference: 26%|██▌ | 1523/5920 [21:40<30:12, 2.43it/s, tok/s=14814.3]
Inference: 26%|██▌ | 1524/5920 [21:40<27:14, 2.69it/s, tok/s=14814.3]
Inference: 26%|██▌ | 1524/5920 [21:40<27:14, 2.69it/s, tok/s=14810.2]
Inference: 26%|██▌ | 1525/5920 [21:41<25:49, 2.84it/s, tok/s=14810.2]
Inference: 26%|██▌ | 1525/5920 [21:41<25:49, 2.84it/s, tok/s=14809.4]
Inference: 26%|██▌ | 1526/5920 [21:42<33:42, 2.17it/s, tok/s=14809.4]
Inference: 26%|██▌ | 1526/5920 [21:42<33:42, 2.17it/s, tok/s=14823.6]
Inference: 26%|██▌ | 1527/5920 [21:42<34:04, 2.15it/s, tok/s=14823.6]
Inference: 26%|██▌ | 1527/5920 [21:42<34:04, 2.15it/s, tok/s=14827.8]
Inference: 26%|██▌ | 1528/5920 [21:43<51:27, 1.42it/s, tok/s=14827.8]
Inference: 26%|██▌ | 1528/5920 [21:43<51:27, 1.42it/s, tok/s=14835.7]
Inference: 26%|██▌ | 1529/5920 [21:44<40:49, 1.79it/s, tok/s=14835.7]
Inference: 26%|██▌ | 1529/5920 [21:44<40:49, 1.79it/s, tok/s=14856.7]
Inference: 26%|██▌ | 1530/5920 [21:44<35:07, 2.08it/s, tok/s=14856.7]
Inference: 26%|██▌ | 1530/5920 [21:44<35:07, 2.08it/s, tok/s=14866.4]
Inference: 26%|██▌ | 1531/5920 [21:45<45:56, 1.59it/s, tok/s=14866.4]
Inference: 26%|██▌ | 1531/5920 [21:45<45:56, 1.59it/s, tok/s=14877.9]
Inference: 26%|██▌ | 1532/5920 [21:45<46:36, 1.57it/s, tok/s=14877.9]
Inference: 26%|██▌ | 1532/5920 [21:45<46:36, 1.57it/s, tok/s=14879.3]
Inference: 26%|██▌ | 1533/5920 [21:46<39:53, 1.83it/s, tok/s=14879.3]
Inference: 26%|██▌ | 1533/5920 [21:46<39:53, 1.83it/s, tok/s=14896.0]
Inference: 26%|██▌ | 1534/5920 [21:47<58:11, 1.26it/s, tok/s=14896.0]
Inference: 26%|██▌ | 1534/5920 [21:47<58:11, 1.26it/s, tok/s=14888.5]
Inference: 26%|██▌ | 1535/5920 [21:48<47:27, 1.54it/s, tok/s=14888.5]
Inference: 26%|██▌ | 1535/5920 [21:48<47:27, 1.54it/s, tok/s=14891.2]
Inference: 26%|██▌ | 1536/5920 [21:48<46:16, 1.58it/s, tok/s=14891.2]
Inference: 26%|██▌ | 1536/5920 [21:48<46:16, 1.58it/s, tok/s=14907.4]
Inference: 26%|██▌ | 1537/5920 [21:49<1:00:19, 1.21it/s, tok/s=14907.4]
Inference: 26%|██▌ | 1537/5920 [21:49<1:00:19, 1.21it/s, tok/s=14915.5]
Inference: 26%|██▌ | 1538/5920 [21:50<54:37, 1.34it/s, tok/s=14915.5]
Inference: 26%|██▌ | 1538/5920 [21:50<54:37, 1.34it/s, tok/s=14932.0]
Inference: 26%|██▌ | 1539/5920 [21:51<53:19, 1.37it/s, tok/s=14932.0]
Inference: 26%|██▌ | 1539/5920 [21:51<53:19, 1.37it/s, tok/s=14930.7]
Inference: 26%|██▌ | 1540/5920 [21:52<57:10, 1.28it/s, tok/s=14930.7]
Inference: 26%|██▌ | 1540/5920 [21:52<57:10, 1.28it/s, tok/s=14923.6]
Inference: 26%|██▌ | 1541/5920 [21:52<44:25, 1.64it/s, tok/s=14923.6]
Inference: 26%|██▌ | 1541/5920 [21:52<44:25, 1.64it/s, tok/s=14939.7]
Inference: 26%|██▌ | 1542/5920 [21:53<1:08:53, 1.06it/s, tok/s=14939.7]
Inference: 26%|██▌ | 1542/5920 [21:53<1:08:53, 1.06it/s, tok/s=14928.9]
Inference: 26%|██▌ | 1543/5920 [21:53<1:08:52, 1.06it/s, tok/s=14934.7]
Inference: 26%|██▌ | 1544/5920 [21:54<39:12, 1.86it/s, tok/s=14934.7]
Inference: 26%|██▌ | 1544/5920 [21:54<39:12, 1.86it/s, tok/s=14936.4]
Inference: 26%|██▌ | 1545/5920 [21:55<59:42, 1.22it/s, tok/s=14936.4]
Inference: 26%|██▌ | 1545/5920 [21:55<59:42, 1.22it/s, tok/s=14921.1]
Inference: 26%|██▌ | 1546/5920 [21:57<1:21:11, 1.11s/it, tok/s=14921.1]
Inference: 26%|██▌ | 1546/5920 [21:57<1:21:11, 1.11s/it, tok/s=14908.2]
Inference: 26%|██▌ | 1547/5920 [21:57<1:21:09, 1.11s/it, tok/s=14911.9]
Inference: 26%|██▌ | 1548/5920 [21:59<1:07:56, 1.07it/s, tok/s=14911.9]
Inference: 26%|██▌ | 1548/5920 [21:59<1:07:56, 1.07it/s, tok/s=14899.4]
Inference: 26%|██▌ | 1549/5920 [21:59<56:19, 1.29it/s, tok/s=14899.4]
Inference: 26%|██▌ | 1549/5920 [21:59<56:19, 1.29it/s, tok/s=14919.3]
Inference: 26%|██▌ | 1550/5920 [21:59<56:18, 1.29it/s, tok/s=14921.8]
Inference: 26%|██▌ | 1551/5920 [22:00<49:34, 1.47it/s, tok/s=14921.8]
Inference: 26%|██▌ | 1551/5920 [22:00<49:34, 1.47it/s, tok/s=14933.1]
Inference: 26%|██▌ | 1552/5920 [22:01<56:09, 1.30it/s, tok/s=14933.1]
Inference: 26%|██▌ | 1552/5920 [22:01<56:09, 1.30it/s, tok/s=14943.3]
Inference: 26%|██▌ | 1553/5920 [22:01<50:54, 1.43it/s, tok/s=14943.3]
Inference: 26%|██▌ | 1553/5920 [22:01<50:54, 1.43it/s, tok/s=14944.4]
Inference: 26%|██▋ | 1554/5920 [22:02<49:43, 1.46it/s, tok/s=14944.4]
Inference: 26%|██▋ | 1554/5920 [22:02<49:43, 1.46it/s, tok/s=14959.9]
Inference: 26%|██▋ | 1555/5920 [22:04<1:03:26, 1.15it/s, tok/s=14959.9]
Inference: 26%|██▋ | 1555/5920 [22:04<1:03:26, 1.15it/s, tok/s=14957.6]
Inference: 26%|██▋ | 1556/5920 [22:05<1:18:08, 1.07s/it, tok/s=14957.6]
Inference: 26%|██▋ | 1556/5920 [22:05<1:18:08, 1.07s/it, tok/s=14948.2]
Inference: 26%|██▋ | 1557/5920 [22:05<58:29, 1.24it/s, tok/s=14948.2]
Inference: 26%|██▋ | 1557/5920 [22:05<58:29, 1.24it/s, tok/s=14953.4]
Inference: 26%|██▋ | 1558/5920 [22:06<58:30, 1.24it/s, tok/s=14953.4]
Inference: 26%|██▋ | 1558/5920 [22:06<58:30, 1.24it/s, tok/s=14967.0]
Inference: 26%|██▋ | 1559/5920 [22:08<1:13:14, 1.01s/it, tok/s=14967.0]
Inference: 26%|██▋ | 1559/5920 [22:08<1:13:14, 1.01s/it, tok/s=14957.6]
Inference: 26%|██▋ | 1560/5920 [22:08<1:11:10, 1.02it/s, tok/s=14957.6]
Inference: 26%|██▋ | 1560/5920 [22:08<1:11:10, 1.02it/s, tok/s=14953.7]
Inference: 26%|██▋ | 1561/5920 [22:10<1:16:28, 1.05s/it, tok/s=14953.7]
Inference: 26%|██▋ | 1561/5920 [22:10<1:16:28, 1.05s/it, tok/s=14952.3]
Inference: 26%|██▋ | 1562/5920 [22:10<1:02:33, 1.16it/s, tok/s=14952.3]
Inference: 26%|██▋ | 1562/5920 [22:10<1:02:33, 1.16it/s, tok/s=14957.7]
Inference: 26%|██▋ | 1563/5920 [22:11<1:12:58, 1.00s/it, tok/s=14957.7]
Inference: 26%|██▋ | 1563/5920 [22:11<1:12:58, 1.00s/it, tok/s=14965.2]
Inference: 26%|██▋ | 1564/5920 [22:12<1:06:21, 1.09it/s, tok/s=14965.2]
Inference: 26%|██▋ | 1564/5920 [22:12<1:06:21, 1.09it/s, tok/s=14960.9]
Inference: 26%|██▋ | 1565/5920 [22:13<1:07:09, 1.08it/s, tok/s=14960.9]
Inference: 26%|██▋ | 1565/5920 [22:13<1:07:09, 1.08it/s, tok/s=14961.5]
Inference: 26%|██▋ | 1566/5920 [22:14<1:06:28, 1.09it/s, tok/s=14961.5]
Inference: 26%|██▋ | 1566/5920 [22:14<1:06:28, 1.09it/s, tok/s=14968.2]
Inference: 26%|██▋ | 1567/5920 [22:14<57:26, 1.26it/s, tok/s=14968.2]
Inference: 26%|██▋ | 1567/5920 [22:14<57:26, 1.26it/s, tok/s=14969.5]
Inference: 26%|██▋ | 1568/5920 [22:15<50:42, 1.43it/s, tok/s=14969.5]
Inference: 26%|██▋ | 1568/5920 [22:15<50:42, 1.43it/s, tok/s=14986.6]
Inference: 27%|██▋ | 1569/5920 [22:16<57:52, 1.25it/s, tok/s=14986.6]
Inference: 27%|██▋ | 1569/5920 [22:16<57:52, 1.25it/s, tok/s=14997.6]
Inference: 27%|██▋ | 1570/5920 [22:18<1:24:52, 1.17s/it, tok/s=14997.6]
Inference: 27%|██▋ | 1570/5920 [22:18<1:24:52, 1.17s/it, tok/s=14984.2]
Inference: 27%|██▋ | 1571/5920 [22:19<1:23:27, 1.15s/it, tok/s=14984.2]
Inference: 27%|██▋ | 1571/5920 [22:19<1:23:27, 1.15s/it, tok/s=14984.4]
Inference: 27%|██▋ | 1572/5920 [22:19<1:02:20, 1.16it/s, tok/s=14984.4]
Inference: 27%|██▋ | 1572/5920 [22:19<1:02:20, 1.16it/s, tok/s=14995.7]
Inference: 27%|██▋ | 1573/5920 [22:20<59:25, 1.22it/s, tok/s=14995.7]
Inference: 27%|██▋ | 1573/5920 [22:20<59:25, 1.22it/s, tok/s=14997.1]
Inference: 27%|██▋ | 1574/5920 [22:22<1:20:44, 1.11s/it, tok/s=14997.1]
Inference: 27%|██▋ | 1574/5920 [22:22<1:20:44, 1.11s/it, tok/s=14998.2]
Inference: 27%|██▋ | 1575/5920 [22:23<1:24:06, 1.16s/it, tok/s=14998.2]
Inference: 27%|██▋ | 1575/5920 [22:23<1:24:06, 1.16s/it, tok/s=15003.2]
Inference: 27%|██▋ | 1576/5920 [22:23<1:24:04, 1.16s/it, tok/s=15005.9]
Inference: 27%|██▋ | 1577/5920 [22:24<1:02:12, 1.16it/s, tok/s=15005.9]
Inference: 27%|██▋ | 1577/5920 [22:24<1:02:12, 1.16it/s, tok/s=14999.2]
Inference: 27%|██▋ | 1578/5920 [22:26<1:19:53, 1.10s/it, tok/s=14999.2]
Inference: 27%|██▋ | 1578/5920 [22:26<1:19:53, 1.10s/it, tok/s=14991.5]
Inference: 27%|██▋ | 1579/5920 [22:27<1:10:11, 1.03it/s, tok/s=14991.5]
Inference: 27%|██▋ | 1579/5920 [22:27<1:10:11, 1.03it/s, tok/s=14988.5]
Inference: 27%|██▋ | 1580/5920 [22:27<1:03:38, 1.14it/s, tok/s=14988.5]
Inference: 27%|██▋ | 1580/5920 [22:27<1:03:38, 1.14it/s, tok/s=14986.7]
Inference: 27%|██▋ | 1581/5920 [22:28<55:21, 1.31it/s, tok/s=14986.7]
Inference: 27%|██▋ | 1581/5920 [22:28<55:21, 1.31it/s, tok/s=14985.6]
Inference: 27%|██▋ | 1582/5920 [22:31<1:42:24, 1.42s/it, tok/s=14985.6]
Inference: 27%|██▋ | 1582/5920 [22:31<1:42:24, 1.42s/it, tok/s=14973.8]
Inference: 27%|██▋ | 1583/5920 [22:31<1:15:16, 1.04s/it, tok/s=14973.8]
Inference: 27%|██▋ | 1583/5920 [22:31<1:15:16, 1.04s/it, tok/s=14994.8]
Inference: 27%|██▋ | 1584/5920 [22:31<1:15:15, 1.04s/it, tok/s=14996.9]
Inference: 27%|██▋ | 1585/5920 [22:31<46:43, 1.55it/s, tok/s=14996.9]
Inference: 27%|██▋ | 1585/5920 [22:31<46:43, 1.55it/s, tok/s=14996.9]
Inference: 27%|██▋ | 1586/5920 [22:32<43:56, 1.64it/s, tok/s=14996.9]
Inference: 27%|██▋ | 1586/5920 [22:32<43:56, 1.64it/s, tok/s=15002.5]
Inference: 27%|██▋ | 1587/5920 [22:35<1:32:16, 1.28s/it, tok/s=15002.5]
Inference: 27%|██▋ | 1587/5920 [22:35<1:32:16, 1.28s/it, tok/s=14974.5]
Inference: 27%|██▋ | 1588/5920 [22:35<1:11:35, 1.01it/s, tok/s=14974.5]
Inference: 27%|██▋ | 1588/5920 [22:35<1:11:35, 1.01it/s, tok/s=14976.2]
Inference: 27%|██▋ | 1589/5920 [22:36<1:07:03, 1.08it/s, tok/s=14976.2]
Inference: 27%|██▋ | 1589/5920 [22:36<1:07:03, 1.08it/s, tok/s=14981.2]
Inference: 27%|██▋ | 1590/5920 [22:42<2:57:41, 2.46s/it, tok/s=14981.2]
Inference: 27%|██▋ | 1590/5920 [22:42<2:57:41, 2.46s/it, tok/s=14921.2]
Inference: 27%|██▋ | 1591/5920 [22:44<2:38:23, 2.20s/it, tok/s=14921.2]
Inference: 27%|██▋ | 1591/5920 [22:44<2:38:23, 2.20s/it, tok/s=14906.9]
Inference: 27%|██▋ | 1592/5920 [22:44<2:03:33, 1.71s/it, tok/s=14906.9]
Inference: 27%|██▋ | 1592/5920 [22:44<2:03:33, 1.71s/it, tok/s=14914.5]
Inference: 27%|██▋ | 1593/5920 [22:50<3:32:09, 2.94s/it, tok/s=14914.5]
Inference: 27%|██▋ | 1593/5920 [22:50<3:32:09, 2.94s/it, tok/s=14861.0]
Inference: 27%|██▋ | 1594/5920 [22:50<3:32:06, 2.94s/it, tok/s=14873.2]
Inference: 27%|██▋ | 1595/5920 [22:51<2:14:54, 1.87s/it, tok/s=14873.2]
Inference: 27%|██▋ | 1595/5920 [22:51<2:14:54, 1.87s/it, tok/s=14868.6]
Inference: 27%|██▋ | 1596/5920 [22:51<2:14:52, 1.87s/it, tok/s=14874.5]
Inference: 27%|██▋ | 1597/5920 [22:52<1:32:51, 1.29s/it, tok/s=14874.5]
Inference: 27%|██▋ | 1597/5920 [22:52<1:32:51, 1.29s/it, tok/s=14871.7]
Inference: 27%|██▋ | 1598/5920 [22:53<1:20:07, 1.11s/it, tok/s=14871.7]
Inference: 27%|██▋ | 1598/5920 [22:53<1:20:07, 1.11s/it, tok/s=14888.2]
Inference: 27%|██▋ | 1599/5920 [22:53<1:15:05, 1.04s/it, tok/s=14888.2]
Inference: 27%|██▋ | 1599/5920 [22:53<1:15:05, 1.04s/it, tok/s=14889.7]
Inference: 27%|██▋ | 1600/5920 [22:55<1:24:49, 1.18s/it, tok/s=14889.7]
Inference: 27%|██▋ | 1600/5920 [22:55<1:24:49, 1.18s/it, tok/s=14885.5]
Inference: 27%|██▋ | 1601/5920 [22:56<1:30:11, 1.25s/it, tok/s=14885.5]
Inference: 27%|██▋ | 1601/5920 [22:56<1:30:11, 1.25s/it, tok/s=14877.5]
Inference: 27%|██▋ | 1602/5920 [23:00<2:06:32, 1.76s/it, tok/s=14877.5]
Inference: 27%|██▋ | 1602/5920 [23:00<2:06:32, 1.76s/it, tok/s=14848.7]
Inference: 27%|██▋ | 1603/5920 [23:00<1:41:44, 1.41s/it, tok/s=14848.7]
Inference: 27%|██▋ | 1603/5920 [23:00<1:41:44, 1.41s/it, tok/s=14864.7]
Inference: 27%|██▋ | 1604/5920 [23:03<2:06:41, 1.76s/it, tok/s=14864.7]
Inference: 27%|██▋ | 1604/5920 [23:03<2:06:41, 1.76s/it, tok/s=14849.8]
Inference: 27%|██▋ | 1605/5920 [23:08<3:15:39, 2.72s/it, tok/s=14849.8]
Inference: 27%|██▋ | 1605/5920 [23:08<3:15:39, 2.72s/it, tok/s=14806.3]
Inference: 27%|██▋ | 1606/5920 [23:08<2:28:12, 2.06s/it, tok/s=14806.3]
Inference: 27%|██▋ | 1606/5920 [23:08<2:28:12, 2.06s/it, tok/s=14822.9]
Inference: 27%|██▋ | 1607/5920 [23:12<3:10:36, 2.65s/it, tok/s=14822.9]
Inference: 27%|██▋ | 1607/5920 [23:12<3:10:36, 2.65s/it, tok/s=14786.4]
Inference: 27%|██▋ | 1608/5920 [23:12<3:10:34, 2.65s/it, tok/s=14788.9]
Inference: 27%|██▋ | 1609/5920 [23:14<2:07:41, 1.78s/it, tok/s=14788.9]
Inference: 27%|██▋ | 1609/5920 [23:14<2:07:41, 1.78s/it, tok/s=14791.9]
Inference: 27%|██▋ | 1610/5920 [23:14<2:07:39, 1.78s/it, tok/s=14799.4]
Inference: 27%|██▋ | 1611/5920 [23:14<1:21:05, 1.13s/it, tok/s=14799.4]
Inference: 27%|██▋ | 1611/5920 [23:14<1:21:05, 1.13s/it, tok/s=14818.2]
Inference: 27%|██▋ | 1612/5920 [23:15<1:15:22, 1.05s/it, tok/s=14818.2]
Inference: 27%|██▋ | 1612/5920 [23:15<1:15:22, 1.05s/it, tok/s=14822.0]
Inference: 27%|██▋ | 1613/5920 [23:15<1:00:56, 1.18it/s, tok/s=14822.0]
Inference: 27%|██▋ | 1613/5920 [23:15<1:00:56, 1.18it/s, tok/s=14830.6]
Inference: 27%|██▋ | 1614/5920 [23:20<2:07:49, 1.78s/it, tok/s=14830.6]
Inference: 27%|██▋ | 1614/5920 [23:20<2:07:49, 1.78s/it, tok/s=14790.1]
Inference: 27%|██▋ | 1615/5920 [23:20<1:47:00, 1.49s/it, tok/s=14790.1]
Inference: 27%|██▋ | 1615/5920 [23:20<1:47:00, 1.49s/it, tok/s=14796.0]
Inference: 27%|██▋ | 1616/5920 [23:21<1:39:54, 1.39s/it, tok/s=14796.0]
Inference: 27%|██▋ | 1616/5920 [23:21<1:39:54, 1.39s/it, tok/s=14791.9]
Inference: 27%|██▋ | 1617/5920 [23:24<2:09:46, 1.81s/it, tok/s=14791.9]
Inference: 27%|██▋ | 1617/5920 [23:24<2:09:46, 1.81s/it, tok/s=14773.6]
Inference: 27%|██▋ | 1618/5920 [23:24<2:09:44, 1.81s/it, tok/s=14787.5]
Inference: 27%|██▋ | 1619/5920 [23:24<1:14:28, 1.04s/it, tok/s=14787.5]
Inference: 27%|██▋ | 1619/5920 [23:24<1:14:28, 1.04s/it, tok/s=14794.6]
Inference: 27%|██▋ | 1620/5920 [23:26<1:25:22, 1.19s/it, tok/s=14794.6]
Inference: 27%|██▋ | 1620/5920 [23:26<1:25:22, 1.19s/it, tok/s=14782.0]
Inference: 27%|██▋ | 1621/5920 [23:28<1:46:02, 1.48s/it, tok/s=14782.0]
Inference: 27%|██▋ | 1621/5920 [23:28<1:46:02, 1.48s/it, tok/s=14765.3]
Inference: 27%|██▋ | 1622/5920 [23:29<1:26:51, 1.21s/it, tok/s=14765.3]
Inference: 27%|██▋ | 1622/5920 [23:29<1:26:51, 1.21s/it, tok/s=14762.6]
Inference: 27%|██▋ | 1623/5920 [23:30<1:27:52, 1.23s/it, tok/s=14762.6]
Inference: 27%|██▋ | 1623/5920 [23:30<1:27:52, 1.23s/it, tok/s=14752.1]
Inference: 27%|██▋ | 1624/5920 [23:31<1:17:13, 1.08s/it, tok/s=14752.1]
Inference: 27%|██▋ | 1624/5920 [23:31<1:17:13, 1.08s/it, tok/s=14748.2]
Inference: 27%|██▋ | 1625/5920 [23:31<1:01:05, 1.17it/s, tok/s=14748.2]
Inference: 27%|██▋ | 1625/5920 [23:31<1:01:05, 1.17it/s, tok/s=14758.3]
Inference: 27%|██▋ | 1626/5920 [23:35<2:03:15, 1.72s/it, tok/s=14758.3]
Inference: 27%|██▋ | 1626/5920 [23:35<2:03:15, 1.72s/it, tok/s=14721.2]
Inference: 27%|██▋ | 1627/5920 [23:35<1:36:05, 1.34s/it, tok/s=14721.2]
Inference: 27%|██▋ | 1627/5920 [23:35<1:36:05, 1.34s/it, tok/s=14719.3]
Inference: 28%|██▊ | 1628/5920 [23:37<1:47:20, 1.50s/it, tok/s=14719.3]
Inference: 28%|██▊ | 1628/5920 [23:37<1:47:20, 1.50s/it, tok/s=14712.3]
Inference: 28%|██▊ | 1629/5920 [23:40<2:08:08, 1.79s/it, tok/s=14712.3]
Inference: 28%|██▊ | 1629/5920 [23:40<2:08:08, 1.79s/it, tok/s=14694.6]
Inference: 28%|██▊ | 1630/5920 [23:40<1:42:22, 1.43s/it, tok/s=14694.6]
Inference: 28%|██▊ | 1630/5920 [23:40<1:42:22, 1.43s/it, tok/s=14694.2]
Inference: 28%|██▊ | 1631/5920 [23:41<1:17:03, 1.08s/it, tok/s=14694.2]
Inference: 28%|██▊ | 1631/5920 [23:41<1:17:03, 1.08s/it, tok/s=14694.2]
Inference: 28%|██▊ | 1632/5920 [23:41<56:45, 1.26it/s, tok/s=14694.2]
Inference: 28%|██▊ | 1632/5920 [23:41<56:45, 1.26it/s, tok/s=14709.3]
Inference: 28%|██▊ | 1633/5920 [23:43<1:22:26, 1.15s/it, tok/s=14709.3]
Inference: 28%|██▊ | 1633/5920 [23:43<1:22:26, 1.15s/it, tok/s=14700.5]
Inference: 28%|██▊ | 1634/5920 [23:44<1:26:11, 1.21s/it, tok/s=14700.5]
Inference: 28%|██▊ | 1634/5920 [23:44<1:26:11, 1.21s/it, tok/s=14697.3]
Inference: 28%|██▊ | 1635/5920 [23:44<1:06:46, 1.07it/s, tok/s=14697.3]
Inference: 28%|██▊ | 1635/5920 [23:44<1:06:46, 1.07it/s, tok/s=14696.8]
Inference: 28%|██▊ | 1636/5920 [23:44<1:06:45, 1.07it/s, tok/s=14699.5]
Inference: 28%|██▊ | 1637/5920 [23:45<48:14, 1.48it/s, tok/s=14699.5]
Inference: 28%|██▊ | 1637/5920 [23:45<48:14, 1.48it/s, tok/s=14694.6]
Inference: 28%|██▊ | 1638/5920 [23:46<49:32, 1.44it/s, tok/s=14694.6]
Inference: 28%|██▊ | 1638/5920 [23:46<49:32, 1.44it/s, tok/s=14694.7]
Inference: 28%|██▊ | 1639/5920 [23:46<39:03, 1.83it/s, tok/s=14694.7]
Inference: 28%|██▊ | 1639/5920 [23:46<39:03, 1.83it/s, tok/s=14708.3]
Inference: 28%|██▊ | 1640/5920 [23:46<35:34, 2.01it/s, tok/s=14708.3]
Inference: 28%|██▊ | 1640/5920 [23:46<35:34, 2.01it/s, tok/s=14715.2]
Inference: 28%|██▊ | 1641/5920 [23:49<1:26:14, 1.21s/it, tok/s=14715.2]
Inference: 28%|██▊ | 1641/5920 [23:49<1:26:14, 1.21s/it, tok/s=14687.1]
Inference: 28%|██▊ | 1642/5920 [23:52<1:58:08, 1.66s/it, tok/s=14687.1]
Inference: 28%|██▊ | 1642/5920 [23:52<1:58:08, 1.66s/it, tok/s=14673.2]
Inference: 28%|██▊ | 1643/5920 [23:53<1:44:26, 1.47s/it, tok/s=14673.2]
Inference: 28%|██▊ | 1643/5920 [23:53<1:44:26, 1.47s/it, tok/s=14673.5]
Inference: 28%|██▊ | 1644/5920 [23:54<1:21:42, 1.15s/it, tok/s=14673.5]
Inference: 28%|██▊ | 1644/5920 [23:54<1:21:42, 1.15s/it, tok/s=14680.1]
Inference: 28%|██▊ | 1645/5920 [23:54<1:02:42, 1.14it/s, tok/s=14680.1]
Inference: 28%|██▊ | 1645/5920 [23:54<1:02:42, 1.14it/s, tok/s=14680.6]
Inference: 28%|██▊ | 1646/5920 [23:54<1:02:41, 1.14it/s, tok/s=14687.6]
Inference: 28%|██▊ | 1647/5920 [23:56<1:14:25, 1.05s/it, tok/s=14687.6]
Inference: 28%|██▊ | 1647/5920 [23:56<1:14:25, 1.05s/it, tok/s=14674.7]
Inference: 28%|██▊ | 1648/5920 [23:57<1:02:14, 1.14it/s, tok/s=14674.7]
Inference: 28%|██▊ | 1648/5920 [23:57<1:02:14, 1.14it/s, tok/s=14674.6]
Inference: 28%|██▊ | 1649/5920 [23:59<1:31:14, 1.28s/it, tok/s=14674.6]
Inference: 28%|██▊ | 1649/5920 [23:59<1:31:14, 1.28s/it, tok/s=14663.8]
Inference: 28%|██▊ | 1650/5920 [23:59<1:12:13, 1.01s/it, tok/s=14663.8]
Inference: 28%|██▊ | 1650/5920 [23:59<1:12:13, 1.01s/it, tok/s=14678.2]
Inference: 28%|██▊ | 1651/5920 [24:03<2:04:56, 1.76s/it, tok/s=14678.2]
Inference: 28%|██▊ | 1651/5920 [24:03<2:04:56, 1.76s/it, tok/s=14645.2]
Inference: 28%|██▊ | 1652/5920 [24:06<2:27:59, 2.08s/it, tok/s=14645.2]
Inference: 28%|██▊ | 1652/5920 [24:06<2:27:59, 2.08s/it, tok/s=14632.9]
Inference: 28%|██▊ | 1653/5920 [24:08<2:30:08, 2.11s/it, tok/s=14632.9]
Inference: 28%|██▊ | 1653/5920 [24:08<2:30:08, 2.11s/it, tok/s=14627.5]
Inference: 28%|██▊ | 1654/5920 [24:09<2:05:19, 1.76s/it, tok/s=14627.5]
Inference: 28%|██▊ | 1654/5920 [24:09<2:05:19, 1.76s/it, tok/s=14632.3]
Inference: 28%|██▊ | 1655/5920 [24:11<2:15:45, 1.91s/it, tok/s=14632.3]
Inference: 28%|██▊ | 1655/5920 [24:11<2:15:45, 1.91s/it, tok/s=14615.5]
Inference: 28%|██▊ | 1656/5920 [24:12<1:53:54, 1.60s/it, tok/s=14615.5]
Inference: 28%|██▊ | 1656/5920 [24:12<1:53:54, 1.60s/it, tok/s=14614.3]
Inference: 28%|██▊ | 1657/5920 [24:13<1:30:15, 1.27s/it, tok/s=14614.3]
Inference: 28%|██▊ | 1657/5920 [24:13<1:30:15, 1.27s/it, tok/s=14612.9]
Inference: 28%|██▊ | 1658/5920 [24:13<1:30:14, 1.27s/it, tok/s=14615.7]
Inference: 28%|██▊ | 1659/5920 [24:13<56:17, 1.26it/s, tok/s=14615.7]
Inference: 28%|██▊ | 1659/5920 [24:13<56:17, 1.26it/s, tok/s=14617.3]
Inference: 28%|██▊ | 1660/5920 [24:13<45:57, 1.54it/s, tok/s=14617.3]
Inference: 28%|██▊ | 1660/5920 [24:13<45:57, 1.54it/s, tok/s=14623.7]
Inference: 28%|██▊ | 1661/5920 [24:15<1:12:42, 1.02s/it, tok/s=14623.7]
Inference: 28%|██▊ | 1661/5920 [24:15<1:12:42, 1.02s/it, tok/s=14607.7]
Inference: 28%|██▊ | 1662/5920 [24:16<1:04:55, 1.09it/s, tok/s=14607.7]
Inference: 28%|██▊ | 1662/5920 [24:16<1:04:55, 1.09it/s, tok/s=14615.4]
Inference: 28%|██▊ | 1663/5920 [24:16<50:55, 1.39it/s, tok/s=14615.4]
Inference: 28%|██▊ | 1663/5920 [24:16<50:55, 1.39it/s, tok/s=14621.6]
Inference: 28%|██▊ | 1664/5920 [24:17<49:18, 1.44it/s, tok/s=14621.6]
Inference: 28%|██▊ | 1664/5920 [24:17<49:18, 1.44it/s, tok/s=14621.3]
Inference: 28%|██▊ | 1665/5920 [24:17<39:59, 1.77it/s, tok/s=14621.3]
Inference: 28%|██▊ | 1665/5920 [24:17<39:59, 1.77it/s, tok/s=14622.0]
Inference: 28%|██▊ | 1666/5920 [24:17<39:58, 1.77it/s, tok/s=14626.8]
Inference: 28%|██▊ | 1667/5920 [24:17<26:50, 2.64it/s, tok/s=14626.8]
Inference: 28%|██▊ | 1667/5920 [24:17<26:50, 2.64it/s, tok/s=14627.2]
Inference: 28%|██▊ | 1668/5920 [24:20<1:08:12, 1.04it/s, tok/s=14627.2]
Inference: 28%|██▊ | 1668/5920 [24:20<1:08:12, 1.04it/s, tok/s=14605.3]
Inference: 28%|██▊ | 1669/5920 [24:21<1:08:27, 1.03it/s, tok/s=14605.3]
Inference: 28%|██▊ | 1669/5920 [24:21<1:08:27, 1.03it/s, tok/s=14599.4]
Inference: 28%|██▊ | 1670/5920 [24:21<54:30, 1.30it/s, tok/s=14599.4]
Inference: 28%|██▊ | 1670/5920 [24:21<54:30, 1.30it/s, tok/s=14600.4]
Inference: 28%|██▊ | 1671/5920 [24:22<49:36, 1.43it/s, tok/s=14600.4]
Inference: 28%|██▊ | 1671/5920 [24:22<49:36, 1.43it/s, tok/s=14599.7]
Inference: 28%|██▊ | 1672/5920 [24:23<49:44, 1.42it/s, tok/s=14599.7]
Inference: 28%|██▊ | 1672/5920 [24:23<49:44, 1.42it/s, tok/s=14602.4]
Inference: 28%|██▊ | 1673/5920 [24:23<46:38, 1.52it/s, tok/s=14602.4]
Inference: 28%|██▊ | 1673/5920 [24:23<46:38, 1.52it/s, tok/s=14600.8]
Inference: 28%|██▊ | 1674/5920 [24:24<40:16, 1.76it/s, tok/s=14600.8]
Inference: 28%|██▊ | 1674/5920 [24:24<40:16, 1.76it/s, tok/s=14605.6]
Inference: 28%|██▊ | 1675/5920 [24:24<40:16, 1.76it/s, tok/s=14621.8]
Inference: 28%|██▊ | 1676/5920 [24:24<31:01, 2.28it/s, tok/s=14621.8]
Inference: 28%|██▊ | 1676/5920 [24:24<31:01, 2.28it/s, tok/s=14622.9]
Inference: 28%|██▊ | 1677/5920 [24:25<32:02, 2.21it/s, tok/s=14622.9]
Inference: 28%|██▊ | 1677/5920 [24:25<32:02, 2.21it/s, tok/s=14621.6]
Inference: 28%|██▊ | 1678/5920 [24:26<49:32, 1.43it/s, tok/s=14621.6]
Inference: 28%|██▊ | 1678/5920 [24:26<49:32, 1.43it/s, tok/s=14615.7]
Inference: 28%|██▊ | 1679/5920 [24:29<1:31:51, 1.30s/it, tok/s=14615.7]
Inference: 28%|██▊ | 1679/5920 [24:29<1:31:51, 1.30s/it, tok/s=14594.4]
Inference: 28%|██▊ | 1680/5920 [24:29<1:10:02, 1.01it/s, tok/s=14594.4]
Inference: 28%|██▊ | 1680/5920 [24:29<1:10:02, 1.01it/s, tok/s=14597.4]
Inference: 28%|██▊ | 1681/5920 [24:29<1:10:01, 1.01it/s, tok/s=14600.3]
Inference: 28%|██▊ | 1682/5920 [24:29<44:06, 1.60it/s, tok/s=14600.3]
Inference: 28%|██▊ | 1682/5920 [24:29<44:06, 1.60it/s, tok/s=14608.9]
Inference: 28%|██▊ | 1683/5920 [24:29<44:06, 1.60it/s, tok/s=14612.6]
Inference: 28%|██▊ | 1684/5920 [24:30<44:05, 1.60it/s, tok/s=14616.6]
Inference: 28%|██▊ | 1685/5920 [24:30<25:26, 2.77it/s, tok/s=14616.6]
Inference: 28%|██▊ | 1685/5920 [24:30<25:26, 2.77it/s, tok/s=14620.3]
Inference: 28%|██▊ | 1686/5920 [24:32<44:47, 1.58it/s, tok/s=14620.3]
Inference: 28%|██▊ | 1686/5920 [24:32<44:47, 1.58it/s, tok/s=14605.8]
Inference: 28%|██▊ | 1687/5920 [24:32<41:19, 1.71it/s, tok/s=14605.8]
Inference: 28%|██▊ | 1687/5920 [24:32<41:19, 1.71it/s, tok/s=14605.5]
Inference: 29%|██▊ | 1688/5920 [24:32<34:26, 2.05it/s, tok/s=14605.5]
Inference: 29%|██▊ | 1688/5920 [24:32<34:26, 2.05it/s, tok/s=14608.6]
Inference: 29%|██▊ | 1689/5920 [24:34<1:07:41, 1.04it/s, tok/s=14608.6]
Inference: 29%|██▊ | 1689/5920 [24:34<1:07:41, 1.04it/s, tok/s=14589.7]
Inference: 29%|██▊ | 1690/5920 [24:35<1:05:43, 1.07it/s, tok/s=14589.7]
Inference: 29%|██▊ | 1690/5920 [24:35<1:05:43, 1.07it/s, tok/s=14585.2]
Inference: 29%|██▊ | 1691/5920 [24:36<1:05:02, 1.08it/s, tok/s=14585.2]
Inference: 29%|██▊ | 1691/5920 [24:36<1:05:02, 1.08it/s, tok/s=14579.8]
Inference: 29%|██▊ | 1692/5920 [24:38<1:24:59, 1.21s/it, tok/s=14579.8]
Inference: 29%|██▊ | 1692/5920 [24:38<1:24:59, 1.21s/it, tok/s=14564.3]
Inference: 29%|██▊ | 1693/5920 [24:39<1:14:23, 1.06s/it, tok/s=14564.3]
Inference: 29%|██▊ | 1693/5920 [24:39<1:14:23, 1.06s/it, tok/s=14573.5]
Inference: 29%|██▊ | 1694/5920 [24:42<1:54:37, 1.63s/it, tok/s=14573.5]
Inference: 29%|██▊ | 1694/5920 [24:42<1:54:37, 1.63s/it, tok/s=14547.7]
Inference: 29%|██▊ | 1695/5920 [24:43<1:47:40, 1.53s/it, tok/s=14547.7]
Inference: 29%|██▊ | 1695/5920 [24:43<1:47:40, 1.53s/it, tok/s=14538.7]
Inference: 29%|██▊ | 1696/5920 [24:44<1:37:14, 1.38s/it, tok/s=14538.7]
Inference: 29%|██▊ | 1696/5920 [24:44<1:37:14, 1.38s/it, tok/s=14532.1]
Inference: 29%|██▊ | 1697/5920 [24:44<1:37:12, 1.38s/it, tok/s=14537.1]
Inference: 29%|██▊ | 1698/5920 [24:45<58:19, 1.21it/s, tok/s=14537.1]
Inference: 29%|██▊ | 1698/5920 [24:45<58:19, 1.21it/s, tok/s=14551.1]
Inference: 29%|██▊ | 1699/5920 [24:45<56:45, 1.24it/s, tok/s=14551.1]
Inference: 29%|██▊ | 1699/5920 [24:45<56:45, 1.24it/s, tok/s=14548.9]
Inference: 29%|██▊ | 1700/5920 [24:46<48:04, 1.46it/s, tok/s=14548.9]
Inference: 29%|██▊ | 1700/5920 [24:46<48:04, 1.46it/s, tok/s=14550.9]
Inference: 29%|██▊ | 1701/5920 [24:47<1:02:19, 1.13it/s, tok/s=14550.9]
Inference: 29%|██▊ | 1701/5920 [24:47<1:02:19, 1.13it/s, tok/s=14554.2]
Inference: 29%|██▉ | 1702/5920 [24:47<53:39, 1.31it/s, tok/s=14554.2]
Inference: 29%|██▉ | 1702/5920 [24:47<53:39, 1.31it/s, tok/s=14560.2]
Inference: 29%|██▉ | 1703/5920 [24:48<53:39, 1.31it/s, tok/s=14565.6]
Inference: 29%|██▉ | 1704/5920 [24:48<44:14, 1.59it/s, tok/s=14565.6]
Inference: 29%|██▉ | 1704/5920 [24:48<44:14, 1.59it/s, tok/s=14562.8]
Inference: 29%|██▉ | 1705/5920 [24:49<49:47, 1.41it/s, tok/s=14562.8]
Inference: 29%|██▉ | 1705/5920 [24:49<49:47, 1.41it/s, tok/s=14570.4]
Inference: 29%|██▉ | 1706/5920 [24:50<51:22, 1.37it/s, tok/s=14570.4]
Inference: 29%|██▉ | 1706/5920 [24:50<51:22, 1.37it/s, tok/s=14571.3]
Inference: 29%|██▉ | 1707/5920 [24:50<51:21, 1.37it/s, tok/s=14585.2]
Inference: 29%|██▉ | 1708/5920 [24:51<38:32, 1.82it/s, tok/s=14585.2]
Inference: 29%|██▉ | 1708/5920 [24:51<38:32, 1.82it/s, tok/s=14586.5]
Inference: 29%|██▉ | 1709/5920 [24:51<37:01, 1.90it/s, tok/s=14586.5]
Inference: 29%|██▉ | 1709/5920 [24:51<37:01, 1.90it/s, tok/s=14587.6]
Inference: 29%|██▉ | 1710/5920 [24:52<33:12, 2.11it/s, tok/s=14587.6]
Inference: 29%|██▉ | 1710/5920 [24:52<33:12, 2.11it/s, tok/s=14591.0]
Inference: 29%|██▉ | 1711/5920 [24:52<28:14, 2.48it/s, tok/s=14591.0]
Inference: 29%|██▉ | 1711/5920 [24:52<28:14, 2.48it/s, tok/s=14594.9]
Inference: 29%|██▉ | 1712/5920 [24:52<28:13, 2.48it/s, tok/s=14599.9]
Inference: 29%|██▉ | 1713/5920 [24:52<19:47, 3.54it/s, tok/s=14599.9]
Inference: 29%|██▉ | 1713/5920 [24:52<19:47, 3.54it/s, tok/s=14604.3]
Inference: 29%|██▉ | 1714/5920 [24:53<37:14, 1.88it/s, tok/s=14604.3]
Inference: 29%|██▉ | 1714/5920 [24:53<37:14, 1.88it/s, tok/s=14608.3]
Inference: 29%|██▉ | 1715/5920 [24:55<1:00:51, 1.15it/s, tok/s=14608.3]
Inference: 29%|██▉ | 1715/5920 [24:55<1:00:51, 1.15it/s, tok/s=14594.9]
Inference: 29%|██▉ | 1716/5920 [24:56<1:00:58, 1.15it/s, tok/s=14594.9]
Inference: 29%|██▉ | 1716/5920 [24:56<1:00:58, 1.15it/s, tok/s=14592.4]
Inference: 29%|██▉ | 1717/5920 [24:56<1:00:57, 1.15it/s, tok/s=14595.7]
Inference: 29%|██▉ | 1718/5920 [24:56<38:37, 1.81it/s, tok/s=14595.7]
Inference: 29%|██▉ | 1718/5920 [24:56<38:37, 1.81it/s, tok/s=14599.3]
Inference: 29%|██▉ | 1719/5920 [25:00<1:33:58, 1.34s/it, tok/s=14599.3]
Inference: 29%|██▉ | 1719/5920 [25:00<1:33:58, 1.34s/it, tok/s=14567.2]
Inference: 29%|██▉ | 1720/5920 [25:01<1:29:07, 1.27s/it, tok/s=14567.2]
Inference: 29%|██▉ | 1720/5920 [25:01<1:29:07, 1.27s/it, tok/s=14562.1]
Inference: 29%|██▉ | 1721/5920 [25:01<1:08:10, 1.03it/s, tok/s=14562.1]
Inference: 29%|██▉ | 1721/5920 [25:01<1:08:10, 1.03it/s, tok/s=14566.5]
Inference: 29%|██▉ | 1722/5920 [25:02<1:08:09, 1.03it/s, tok/s=14568.4]
Inference: 29%|██▉ | 1723/5920 [25:03<55:24, 1.26it/s, tok/s=14568.4]
Inference: 29%|██▉ | 1723/5920 [25:03<55:24, 1.26it/s, tok/s=14563.9]
Inference: 29%|██▉ | 1724/5920 [25:04<1:01:28, 1.14it/s, tok/s=14563.9]
Inference: 29%|██▉ | 1724/5920 [25:04<1:01:28, 1.14it/s, tok/s=14558.5]
Inference: 29%|██▉ | 1725/5920 [25:05<1:08:44, 1.02it/s, tok/s=14558.5]
Inference: 29%|██▉ | 1725/5920 [25:05<1:08:44, 1.02it/s, tok/s=14551.4]
Inference: 29%|██▉ | 1726/5920 [25:05<52:44, 1.33it/s, tok/s=14551.4]
Inference: 29%|██▉ | 1726/5920 [25:05<52:44, 1.33it/s, tok/s=14558.4]
Inference: 29%|██▉ | 1727/5920 [25:07<1:12:08, 1.03s/it, tok/s=14558.4]
Inference: 29%|██▉ | 1727/5920 [25:07<1:12:08, 1.03s/it, tok/s=14546.8]
Inference: 29%|██▉ | 1728/5920 [25:08<1:21:47, 1.17s/it, tok/s=14546.8]
Inference: 29%|██▉ | 1728/5920 [25:08<1:21:47, 1.17s/it, tok/s=14535.4]
Inference: 29%|██▉ | 1729/5920 [25:08<1:21:46, 1.17s/it, tok/s=14540.3]
Inference: 29%|██▉ | 1730/5920 [25:12<1:36:34, 1.38s/it, tok/s=14540.3]
Inference: 29%|██▉ | 1730/5920 [25:12<1:36:34, 1.38s/it, tok/s=14526.3]
Inference: 29%|██▉ | 1731/5920 [25:12<1:23:31, 1.20s/it, tok/s=14526.3]
Inference: 29%|██▉ | 1731/5920 [25:12<1:23:31, 1.20s/it, tok/s=14538.9]
Inference: 29%|██▉ | 1732/5920 [25:15<1:47:52, 1.55s/it, tok/s=14538.9]
Inference: 29%|██▉ | 1732/5920 [25:15<1:47:52, 1.55s/it, tok/s=14519.6]
Inference: 29%|██▉ | 1733/5920 [25:16<1:38:39, 1.41s/it, tok/s=14519.6]
Inference: 29%|██▉ | 1733/5920 [25:16<1:38:39, 1.41s/it, tok/s=14512.8]
Inference: 29%|██▉ | 1734/5920 [25:16<1:38:38, 1.41s/it, tok/s=14526.4]
Inference: 29%|██▉ | 1735/5920 [25:17<1:04:59, 1.07it/s, tok/s=14526.4]
Inference: 29%|██▉ | 1735/5920 [25:17<1:04:59, 1.07it/s, tok/s=14532.0]
Inference: 29%|██▉ | 1736/5920 [25:18<1:17:45, 1.12s/it, tok/s=14532.0]
Inference: 29%|██▉ | 1736/5920 [25:18<1:17:45, 1.12s/it, tok/s=14521.9]
Inference: 29%|██▉ | 1737/5920 [25:19<1:03:07, 1.10it/s, tok/s=14521.9]
Inference: 29%|██▉ | 1737/5920 [25:19<1:03:07, 1.10it/s, tok/s=14525.7]
Inference: 29%|██▉ | 1738/5920 [25:19<1:03:06, 1.10it/s, tok/s=14529.1]
Inference: 29%|██▉ | 1739/5920 [25:19<40:56, 1.70it/s, tok/s=14529.1]
Inference: 29%|██▉ | 1739/5920 [25:19<40:56, 1.70it/s, tok/s=14532.0]
Inference: 29%|██▉ | 1740/5920 [25:19<36:48, 1.89it/s, tok/s=14532.0]
Inference: 29%|██▉ | 1740/5920 [25:19<36:48, 1.89it/s, tok/s=14535.1]
Inference: 29%|██▉ | 1741/5920 [25:21<1:04:05, 1.09it/s, tok/s=14535.1]
Inference: 29%|██▉ | 1741/5920 [25:21<1:04:05, 1.09it/s, tok/s=14520.5]
Inference: 29%|██▉ | 1742/5920 [25:21<50:03, 1.39it/s, tok/s=14520.5]
Inference: 29%|██▉ | 1742/5920 [25:21<50:03, 1.39it/s, tok/s=14525.3]
Inference: 29%|██▉ | 1743/5920 [25:21<50:03, 1.39it/s, tok/s=14530.9]
Inference: 29%|██▉ | 1744/5920 [25:23<45:34, 1.53it/s, tok/s=14530.9]
Inference: 29%|██▉ | 1744/5920 [25:23<45:34, 1.53it/s, tok/s=14525.9]
Inference: 29%|██▉ | 1745/5920 [25:23<45:33, 1.53it/s, tok/s=14539.4]
Inference: 29%|██▉ | 1746/5920 [25:24<46:11, 1.51it/s, tok/s=14539.4]
Inference: 29%|██▉ | 1746/5920 [25:24<46:11, 1.51it/s, tok/s=14533.9]
Inference: 30%|██▉ | 1747/5920 [25:24<44:46, 1.55it/s, tok/s=14533.9]
Inference: 30%|██▉ | 1747/5920 [25:24<44:46, 1.55it/s, tok/s=14531.6]
Inference: 30%|██▉ | 1748/5920 [25:25<44:45, 1.55it/s, tok/s=14548.5]
Inference: 30%|██▉ | 1749/5920 [25:26<50:33, 1.37it/s, tok/s=14548.5]
Inference: 30%|██▉ | 1749/5920 [25:26<50:33, 1.37it/s, tok/s=14535.8]
Inference: 30%|██▉ | 1750/5920 [25:27<55:44, 1.25it/s, tok/s=14535.8]
Inference: 30%|██▉ | 1750/5920 [25:27<55:44, 1.25it/s, tok/s=14540.8]
Inference: 30%|██▉ | 1751/5920 [25:27<55:43, 1.25it/s, tok/s=14552.5]
Inference: 30%|██▉ | 1752/5920 [25:28<42:23, 1.64it/s, tok/s=14552.5]
Inference: 30%|██▉ | 1752/5920 [25:28<42:23, 1.64it/s, tok/s=14566.4]
Inference: 30%|██▉ | 1753/5920 [25:29<46:49, 1.48it/s, tok/s=14566.4]
Inference: 30%|██▉ | 1753/5920 [25:29<46:49, 1.48it/s, tok/s=14567.5]
Inference: 30%|██▉ | 1754/5920 [25:29<46:49, 1.48it/s, tok/s=14570.9]
Inference: 30%|██▉ | 1755/5920 [25:29<36:41, 1.89it/s, tok/s=14570.9]
Inference: 30%|██▉ | 1755/5920 [25:29<36:41, 1.89it/s, tok/s=14570.8]
Inference: 30%|██▉ | 1756/5920 [25:29<36:41, 1.89it/s, tok/s=14574.1]
Inference: 30%|██▉ | 1757/5920 [25:30<25:45, 2.69it/s, tok/s=14574.1]
Inference: 30%|██▉ | 1757/5920 [25:30<25:45, 2.69it/s, tok/s=14579.7]
Inference: 30%|██▉ | 1758/5920 [25:30<29:51, 2.32it/s, tok/s=14579.7]
Inference: 30%|██▉ | 1758/5920 [25:30<29:51, 2.32it/s, tok/s=14577.8]
Inference: 30%|██▉ | 1759/5920 [25:31<30:40, 2.26it/s, tok/s=14577.8]
Inference: 30%|██▉ | 1759/5920 [25:31<30:40, 2.26it/s, tok/s=14579.7]
Inference: 30%|██▉ | 1760/5920 [25:31<31:29, 2.20it/s, tok/s=14579.7]
Inference: 30%|██▉ | 1760/5920 [25:31<31:29, 2.20it/s, tok/s=14580.4]
Inference: 30%|██▉ | 1761/5920 [25:32<43:04, 1.61it/s, tok/s=14580.4]
Inference: 30%|██▉ | 1761/5920 [25:32<43:04, 1.61it/s, tok/s=14575.4]
Inference: 30%|██▉ | 1762/5920 [25:32<43:03, 1.61it/s, tok/s=14590.7]
Inference: 30%|██▉ | 1763/5920 [25:35<1:00:29, 1.15it/s, tok/s=14590.7]
Inference: 30%|██▉ | 1763/5920 [25:35<1:00:29, 1.15it/s, tok/s=14582.4]
Inference: 30%|██▉ | 1764/5920 [25:35<53:09, 1.30it/s, tok/s=14582.4]
Inference: 30%|██▉ | 1764/5920 [25:35<53:09, 1.30it/s, tok/s=14582.1]
Inference: 30%|██▉ | 1765/5920 [25:37<1:19:25, 1.15s/it, tok/s=14582.1]
Inference: 30%|██▉ | 1765/5920 [25:37<1:19:25, 1.15s/it, tok/s=14568.9]
Inference: 30%|██▉ | 1766/5920 [25:38<1:04:18, 1.08it/s, tok/s=14568.9]
Inference: 30%|██▉ | 1766/5920 [25:38<1:04:18, 1.08it/s, tok/s=14575.7]
Inference: 30%|██▉ | 1767/5920 [25:40<1:38:09, 1.42s/it, tok/s=14575.7]
Inference: 30%|██▉ | 1767/5920 [25:40<1:38:09, 1.42s/it, tok/s=14556.4]
Inference: 30%|██▉ | 1768/5920 [25:41<1:13:07, 1.06s/it, tok/s=14556.4]
Inference: 30%|██▉ | 1768/5920 [25:41<1:13:07, 1.06s/it, tok/s=14560.8]
Inference: 30%|██▉ | 1769/5920 [25:42<1:20:17, 1.16s/it, tok/s=14560.8]
Inference: 30%|██▉ | 1769/5920 [25:42<1:20:17, 1.16s/it, tok/s=14550.5]
Inference: 30%|██▉ | 1770/5920 [25:43<1:16:02, 1.10s/it, tok/s=14550.5]
Inference: 30%|██▉ | 1770/5920 [25:43<1:16:02, 1.10s/it, tok/s=14545.5]
Inference: 30%|██▉ | 1771/5920 [25:44<1:09:11, 1.00s/it, tok/s=14545.5]
Inference: 30%|██▉ | 1771/5920 [25:44<1:09:11, 1.00s/it, tok/s=14542.4]
Inference: 30%|██▉ | 1772/5920 [25:44<1:09:10, 1.00s/it, tok/s=14560.8]
Inference: 30%|██▉ | 1773/5920 [25:44<1:09:09, 1.00s/it, tok/s=14567.5]
Inference: 30%|██▉ | 1774/5920 [25:45<50:03, 1.38it/s, tok/s=14567.5]
Inference: 30%|██▉ | 1774/5920 [25:45<50:03, 1.38it/s, tok/s=14572.4]
Inference: 30%|██▉ | 1775/5920 [25:46<43:43, 1.58it/s, tok/s=14572.4]
Inference: 30%|██▉ | 1775/5920 [25:46<43:43, 1.58it/s, tok/s=14584.9]
Inference: 30%|███ | 1776/5920 [25:46<46:04, 1.50it/s, tok/s=14584.9]
Inference: 30%|███ | 1776/5920 [25:46<46:04, 1.50it/s, tok/s=14595.4]
Inference: 30%|███ | 1777/5920 [25:46<46:04, 1.50it/s, tok/s=14602.2]
Inference: 30%|███ | 1778/5920 [25:47<35:48, 1.93it/s, tok/s=14602.2]
Inference: 30%|███ | 1778/5920 [25:47<35:48, 1.93it/s, tok/s=14602.6]
Inference: 30%|███ | 1779/5920 [25:48<51:41, 1.34it/s, tok/s=14602.6]
Inference: 30%|███ | 1779/5920 [25:48<51:41, 1.34it/s, tok/s=14605.7]
Inference: 30%|███ | 1780/5920 [25:49<44:37, 1.55it/s, tok/s=14605.7]
Inference: 30%|███ | 1780/5920 [25:49<44:37, 1.55it/s, tok/s=14610.0]
Inference: 30%|███ | 1781/5920 [25:49<44:36, 1.55it/s, tok/s=14615.5]
Inference: 30%|███ | 1782/5920 [25:49<44:36, 1.55it/s, tok/s=14628.0]
Inference: 30%|███ | 1783/5920 [25:49<26:39, 2.59it/s, tok/s=14628.0]
Inference: 30%|███ | 1783/5920 [25:49<26:39, 2.59it/s, tok/s=14638.8]
Inference: 30%|███ | 1784/5920 [25:49<23:50, 2.89it/s, tok/s=14638.8]
Inference: 30%|███ | 1784/5920 [25:49<23:50, 2.89it/s, tok/s=14642.9]
Inference: 30%|███ | 1785/5920 [25:50<26:22, 2.61it/s, tok/s=14642.9]
Inference: 30%|███ | 1785/5920 [25:50<26:22, 2.61it/s, tok/s=14642.0]
Inference: 30%|███ | 1786/5920 [25:51<43:52, 1.57it/s, tok/s=14642.0]
Inference: 30%|███ | 1786/5920 [25:51<43:52, 1.57it/s, tok/s=14631.3]
Inference: 30%|███ | 1787/5920 [25:51<43:51, 1.57it/s, tok/s=14636.9]
Inference: 30%|███ | 1788/5920 [25:52<39:19, 1.75it/s, tok/s=14636.9]
Inference: 30%|███ | 1788/5920 [25:52<39:19, 1.75it/s, tok/s=14647.2]
Inference: 30%|███ | 1789/5920 [25:56<1:22:06, 1.19s/it, tok/s=14647.2]
Inference: 30%|███ | 1789/5920 [25:56<1:22:06, 1.19s/it, tok/s=14622.8]
Inference: 30%|███ | 1790/5920 [25:56<1:03:59, 1.08it/s, tok/s=14622.8]
Inference: 30%|███ | 1790/5920 [25:56<1:03:59, 1.08it/s, tok/s=14628.5]
Inference: 30%|███ | 1791/5920 [25:56<51:34, 1.33it/s, tok/s=14628.5]
Inference: 30%|███ | 1791/5920 [25:56<51:34, 1.33it/s, tok/s=14633.5]
Inference: 30%|███ | 1792/5920 [25:57<58:14, 1.18it/s, tok/s=14633.5]
Inference: 30%|███ | 1792/5920 [25:57<58:14, 1.18it/s, tok/s=14629.7]
Inference: 30%|███ | 1793/5920 [25:59<1:10:30, 1.03s/it, tok/s=14629.7]
Inference: 30%|███ | 1793/5920 [25:59<1:10:30, 1.03s/it, tok/s=14627.7]
Inference: 30%|███ | 1794/5920 [25:59<59:11, 1.16it/s, tok/s=14627.7]
Inference: 30%|███ | 1794/5920 [25:59<59:11, 1.16it/s, tok/s=14626.8]
Inference: 30%|███ | 1795/5920 [25:59<44:42, 1.54it/s, tok/s=14626.8]
Inference: 30%|███ | 1795/5920 [25:59<44:42, 1.54it/s, tok/s=14629.5]
Inference: 30%|███ | 1796/5920 [26:00<51:18, 1.34it/s, tok/s=14629.5]
Inference: 30%|███ | 1796/5920 [26:00<51:18, 1.34it/s, tok/s=14630.5]
Inference: 30%|███ | 1797/5920 [26:01<55:43, 1.23it/s, tok/s=14630.5]
Inference: 30%|███ | 1797/5920 [26:01<55:43, 1.23it/s, tok/s=14635.5]
Inference: 30%|███ | 1798/5920 [26:02<1:00:06, 1.14it/s, tok/s=14635.5]
Inference: 30%|███ | 1798/5920 [26:02<1:00:06, 1.14it/s, tok/s=14629.5]
Inference: 30%|███ | 1799/5920 [26:03<1:05:21, 1.05it/s, tok/s=14629.5]
Inference: 30%|███ | 1799/5920 [26:03<1:05:21, 1.05it/s, tok/s=14625.3]
Inference: 30%|███ | 1800/5920 [26:03<1:05:20, 1.05it/s, tok/s=14641.7]
Inference: 30%|███ | 1801/5920 [26:04<50:17, 1.36it/s, tok/s=14641.7]
Inference: 30%|███ | 1801/5920 [26:04<50:17, 1.36it/s, tok/s=14637.4]
Inference: 30%|███ | 1802/5920 [26:05<48:49, 1.41it/s, tok/s=14637.4]
Inference: 30%|███ | 1802/5920 [26:05<48:49, 1.41it/s, tok/s=14637.6]
Inference: 30%|███ | 1803/5920 [26:06<53:13, 1.29it/s, tok/s=14637.6]
Inference: 30%|███ | 1803/5920 [26:06<53:13, 1.29it/s, tok/s=14641.4]
Inference: 30%|███ | 1804/5920 [26:06<48:02, 1.43it/s, tok/s=14641.4]
Inference: 30%|███ | 1804/5920 [26:06<48:02, 1.43it/s, tok/s=14643.7]
Inference: 30%|███ | 1805/5920 [26:07<52:11, 1.31it/s, tok/s=14643.7]
Inference: 30%|███ | 1805/5920 [26:07<52:11, 1.31it/s, tok/s=14639.0]
Inference: 31%|███ | 1806/5920 [26:08<52:44, 1.30it/s, tok/s=14639.0]
Inference: 31%|███ | 1806/5920 [26:08<52:44, 1.30it/s, tok/s=14635.0]
Inference: 31%|███ | 1807/5920 [26:08<52:43, 1.30it/s, tok/s=14651.1]
Inference: 31%|███ | 1808/5920 [26:09<52:09, 1.31it/s, tok/s=14651.1]
Inference: 31%|███ | 1808/5920 [26:09<52:09, 1.31it/s, tok/s=14648.1]
Inference: 31%|███ | 1809/5920 [26:10<54:14, 1.26it/s, tok/s=14648.1]
Inference: 31%|███ | 1809/5920 [26:10<54:14, 1.26it/s, tok/s=14655.1]
Inference: 31%|███ | 1810/5920 [26:11<47:36, 1.44it/s, tok/s=14655.1]
Inference: 31%|███ | 1810/5920 [26:11<47:36, 1.44it/s, tok/s=14655.0]
Inference: 31%|███ | 1811/5920 [26:11<44:36, 1.54it/s, tok/s=14655.0]
Inference: 31%|███ | 1811/5920 [26:11<44:36, 1.54it/s, tok/s=14655.2]
Inference: 31%|███ | 1812/5920 [26:11<35:12, 1.94it/s, tok/s=14655.2]
Inference: 31%|███ | 1812/5920 [26:11<35:12, 1.94it/s, tok/s=14669.1]
Inference: 31%|███ | 1813/5920 [26:13<1:03:38, 1.08it/s, tok/s=14669.1]
Inference: 31%|███ | 1813/5920 [26:13<1:03:38, 1.08it/s, tok/s=14654.7]
Inference: 31%|███ | 1814/5920 [26:15<1:22:05, 1.20s/it, tok/s=14654.7]
Inference: 31%|███ | 1814/5920 [26:15<1:22:05, 1.20s/it, tok/s=14644.3]
Inference: 31%|███ | 1815/5920 [26:16<1:11:13, 1.04s/it, tok/s=14644.3]
Inference: 31%|███ | 1815/5920 [26:16<1:11:13, 1.04s/it, tok/s=14647.3]
Inference: 31%|███ | 1816/5920 [26:17<1:12:53, 1.07s/it, tok/s=14647.3]
Inference: 31%|███ | 1816/5920 [26:17<1:12:53, 1.07s/it, tok/s=14644.8]
Inference: 31%|███ | 1817/5920 [26:18<1:16:17, 1.12s/it, tok/s=14644.8]
Inference: 31%|███ | 1817/5920 [26:18<1:16:17, 1.12s/it, tok/s=14643.6]
Inference: 31%|███ | 1818/5920 [26:20<1:35:26, 1.40s/it, tok/s=14643.6]
Inference: 31%|███ | 1818/5920 [26:20<1:35:26, 1.40s/it, tok/s=14631.4]
Inference: 31%|███ | 1819/5920 [26:21<1:17:55, 1.14s/it, tok/s=14631.4]
Inference: 31%|███ | 1819/5920 [26:21<1:17:55, 1.14s/it, tok/s=14638.1]
Inference: 31%|███ | 1820/5920 [26:23<1:42:47, 1.50s/it, tok/s=14638.1]
Inference: 31%|███ | 1820/5920 [26:23<1:42:47, 1.50s/it, tok/s=14632.2]
Inference: 31%|███ | 1821/5920 [26:24<1:17:44, 1.14s/it, tok/s=14632.2]
Inference: 31%|███ | 1821/5920 [26:24<1:17:44, 1.14s/it, tok/s=14636.6]
Inference: 31%|███ | 1822/5920 [26:28<2:26:43, 2.15s/it, tok/s=14636.6]
Inference: 31%|███ | 1822/5920 [26:28<2:26:43, 2.15s/it, tok/s=14613.9]
Inference: 31%|███ | 1823/5920 [26:30<2:25:39, 2.13s/it, tok/s=14613.9]
Inference: 31%|███ | 1823/5920 [26:30<2:25:39, 2.13s/it, tok/s=14612.5]
Inference: 31%|███ | 1824/5920 [26:31<1:56:18, 1.70s/it, tok/s=14612.5]
Inference: 31%|███ | 1824/5920 [26:31<1:56:18, 1.70s/it, tok/s=14624.3]
Inference: 31%|███ | 1825/5920 [26:31<1:24:10, 1.23s/it, tok/s=14624.3]
Inference: 31%|███ | 1825/5920 [26:31<1:24:10, 1.23s/it, tok/s=14640.5]
Inference: 31%|███ | 1826/5920 [26:34<2:04:50, 1.83s/it, tok/s=14640.5]
Inference: 31%|███ | 1826/5920 [26:34<2:04:50, 1.83s/it, tok/s=14627.8]
Inference: 31%|███ | 1827/5920 [26:36<2:08:00, 1.88s/it, tok/s=14627.8]
Inference: 31%|███ | 1827/5920 [26:36<2:08:00, 1.88s/it, tok/s=14615.7]
Inference: 31%|███ | 1828/5920 [26:36<1:34:45, 1.39s/it, tok/s=14615.7]
Inference: 31%|███ | 1828/5920 [26:36<1:34:45, 1.39s/it, tok/s=14619.7]
Inference: 31%|███ | 1829/5920 [26:41<2:47:26, 2.46s/it, tok/s=14619.7]
Inference: 31%|███ | 1829/5920 [26:41<2:47:26, 2.46s/it, tok/s=14581.2]
Inference: 31%|███ | 1830/5920 [26:43<2:21:57, 2.08s/it, tok/s=14581.2]
Inference: 31%|███ | 1830/5920 [26:43<2:21:57, 2.08s/it, tok/s=14576.5]
Inference: 31%|███ | 1831/5920 [26:45<2:22:34, 2.09s/it, tok/s=14576.5]
Inference: 31%|███ | 1831/5920 [26:45<2:22:34, 2.09s/it, tok/s=14561.0]
Inference: 31%|███ | 1832/5920 [26:45<1:44:15, 1.53s/it, tok/s=14561.0]
Inference: 31%|███ | 1832/5920 [26:45<1:44:15, 1.53s/it, tok/s=14568.7]
Inference: 31%|███ | 1833/5920 [26:46<1:26:49, 1.27s/it, tok/s=14568.7]
Inference: 31%|███ | 1833/5920 [26:46<1:26:49, 1.27s/it, tok/s=14569.4]
Inference: 31%|███ | 1834/5920 [26:46<1:07:52, 1.00it/s, tok/s=14569.4]
Inference: 31%|███ | 1834/5920 [26:46<1:07:52, 1.00it/s, tok/s=14572.9]
Inference: 31%|███ | 1835/5920 [26:48<1:28:18, 1.30s/it, tok/s=14572.9]
Inference: 31%|███ | 1835/5920 [26:48<1:28:18, 1.30s/it, tok/s=14569.1]
Inference: 31%|███ | 1836/5920 [26:49<1:30:01, 1.32s/it, tok/s=14569.1]
Inference: 31%|███ | 1836/5920 [26:49<1:30:01, 1.32s/it, tok/s=14572.4]
Inference: 31%|███ | 1837/5920 [26:51<1:41:58, 1.50s/it, tok/s=14572.4]
Inference: 31%|███ | 1837/5920 [26:51<1:41:58, 1.50s/it, tok/s=14570.2]
Inference: 31%|███ | 1838/5920 [26:52<1:16:55, 1.13s/it, tok/s=14570.2]
Inference: 31%|███ | 1838/5920 [26:52<1:16:55, 1.13s/it, tok/s=14582.5]
Inference: 31%|███ | 1839/5920 [26:53<1:19:52, 1.17s/it, tok/s=14582.5]
Inference: 31%|███ | 1839/5920 [26:53<1:19:52, 1.17s/it, tok/s=14579.2]
Inference: 31%|███ | 1840/5920 [26:53<1:03:33, 1.07it/s, tok/s=14579.2]
Inference: 31%|███ | 1840/5920 [26:53<1:03:33, 1.07it/s, tok/s=14590.5]
Inference: 31%|███ | 1841/5920 [26:54<53:34, 1.27it/s, tok/s=14590.5]
Inference: 31%|███ | 1841/5920 [26:54<53:34, 1.27it/s, tok/s=14604.4]
Inference: 31%|███ | 1842/5920 [26:57<1:53:53, 1.68s/it, tok/s=14604.4]
Inference: 31%|███ | 1842/5920 [26:57<1:53:53, 1.68s/it, tok/s=14579.4]
Inference: 31%|███ | 1843/5920 [26:58<1:41:28, 1.49s/it, tok/s=14579.4]
Inference: 31%|███ | 1843/5920 [26:58<1:41:28, 1.49s/it, tok/s=14577.8]
Inference: 31%|███ | 1844/5920 [27:00<1:39:34, 1.47s/it, tok/s=14577.8]
Inference: 31%|███ | 1844/5920 [27:00<1:39:34, 1.47s/it, tok/s=14575.7]
Inference: 31%|███ | 1845/5920 [27:03<2:18:37, 2.04s/it, tok/s=14575.7]
Inference: 31%|███ | 1845/5920 [27:03<2:18:37, 2.04s/it, tok/s=14551.8]
Inference: 31%|███ | 1846/5920 [27:06<2:38:24, 2.33s/it, tok/s=14551.8]
Inference: 31%|███ | 1846/5920 [27:06<2:38:24, 2.33s/it, tok/s=14541.5]
Inference: 31%|███ | 1847/5920 [27:10<3:00:31, 2.66s/it, tok/s=14541.5]
Inference: 31%|███ | 1847/5920 [27:10<3:00:31, 2.66s/it, tok/s=14525.7]
Inference: 31%|███ | 1848/5920 [27:12<2:58:58, 2.64s/it, tok/s=14525.7]
Inference: 31%|███ | 1848/5920 [27:12<2:58:58, 2.64s/it, tok/s=14518.2]
Inference: 31%|███ | 1849/5920 [27:14<2:41:16, 2.38s/it, tok/s=14518.2]
Inference: 31%|███ | 1849/5920 [27:14<2:41:16, 2.38s/it, tok/s=14519.3]
Inference: 31%|███▏ | 1850/5920 [27:15<2:17:19, 2.02s/it, tok/s=14519.3]
Inference: 31%|███▏ | 1850/5920 [27:15<2:17:19, 2.02s/it, tok/s=14527.1]
Inference: 31%|███▏ | 1851/5920 [27:16<1:42:18, 1.51s/it, tok/s=14527.1]
Inference: 31%|███▏ | 1851/5920 [27:16<1:42:18, 1.51s/it, tok/s=14532.6]
Inference: 31%|███▏ | 1852/5920 [27:17<1:37:41, 1.44s/it, tok/s=14532.6]
Inference: 31%|███▏ | 1852/5920 [27:17<1:37:41, 1.44s/it, tok/s=14536.3]
Inference: 31%|███▏ | 1853/5920 [27:21<2:30:00, 2.21s/it, tok/s=14536.3]
Inference: 31%|███▏ | 1853/5920 [27:21<2:30:00, 2.21s/it, tok/s=14514.4]
Inference: 31%|███▏ | 1854/5920 [27:21<1:52:21, 1.66s/it, tok/s=14514.4]
Inference: 31%|███▏ | 1854/5920 [27:21<1:52:21, 1.66s/it, tok/s=14519.2]
Inference: 31%|███▏ | 1855/5920 [27:22<1:24:42, 1.25s/it, tok/s=14519.2]
Inference: 31%|███▏ | 1855/5920 [27:22<1:24:42, 1.25s/it, tok/s=14526.8]
Inference: 31%|███▏ | 1856/5920 [27:25<2:10:12, 1.92s/it, tok/s=14526.8]
Inference: 31%|███▏ | 1856/5920 [27:25<2:10:12, 1.92s/it, tok/s=14504.1]
Inference: 31%|███▏ | 1857/5920 [27:30<3:05:50, 2.74s/it, tok/s=14504.1]
Inference: 31%|███▏ | 1857/5920 [27:30<3:05:50, 2.74s/it, tok/s=14469.8]
Inference: 31%|███▏ | 1858/5920 [27:30<2:14:14, 1.98s/it, tok/s=14469.8]
Inference: 31%|███▏ | 1858/5920 [27:30<2:14:14, 1.98s/it, tok/s=14474.7]
Inference: 31%|███▏ | 1859/5920 [27:32<2:10:22, 1.93s/it, tok/s=14474.7]
Inference: 31%|███▏ | 1859/5920 [27:32<2:10:22, 1.93s/it, tok/s=14474.1]
Inference: 31%|███▏ | 1860/5920 [27:37<3:12:38, 2.85s/it, tok/s=14474.1]
Inference: 31%|███▏ | 1860/5920 [27:37<3:12:38, 2.85s/it, tok/s=14438.5]
Inference: 31%|███▏ | 1861/5920 [27:37<2:26:07, 2.16s/it, tok/s=14438.5]
Inference: 31%|███▏ | 1861/5920 [27:37<2:26:07, 2.16s/it, tok/s=14451.9]
Inference: 31%|███▏ | 1862/5920 [27:38<2:05:29, 1.86s/it, tok/s=14451.9]
Inference: 31%|███▏ | 1862/5920 [27:38<2:05:29, 1.86s/it, tok/s=14448.8]
Inference: 31%|███▏ | 1863/5920 [27:40<1:53:33, 1.68s/it, tok/s=14448.8]
Inference: 31%|███▏ | 1863/5920 [27:40<1:53:33, 1.68s/it, tok/s=14447.6]
Inference: 31%|███▏ | 1864/5920 [27:42<2:12:47, 1.96s/it, tok/s=14447.6]
Inference: 31%|███▏ | 1864/5920 [27:42<2:12:47, 1.96s/it, tok/s=14431.9]
Inference: 32%|███▏ | 1865/5920 [27:43<1:44:48, 1.55s/it, tok/s=14431.9]
Inference: 32%|███▏ | 1865/5920 [27:43<1:44:48, 1.55s/it, tok/s=14434.4]
Inference: 32%|███▏ | 1866/5920 [27:45<2:06:35, 1.87s/it, tok/s=14434.4]
Inference: 32%|███▏ | 1866/5920 [27:45<2:06:35, 1.87s/it, tok/s=14418.9]
Inference: 32%|███▏ | 1867/5920 [27:47<2:00:52, 1.79s/it, tok/s=14418.9]
Inference: 32%|███▏ | 1867/5920 [27:47<2:00:52, 1.79s/it, tok/s=14413.9]
Inference: 32%|███▏ | 1868/5920 [27:51<2:39:06, 2.36s/it, tok/s=14413.9]
Inference: 32%|███▏ | 1868/5920 [27:51<2:39:06, 2.36s/it, tok/s=14400.2]
Inference: 32%|███▏ | 1869/5920 [27:52<2:23:58, 2.13s/it, tok/s=14400.2]
Inference: 32%|███▏ | 1869/5920 [27:52<2:23:58, 2.13s/it, tok/s=14404.3]
Inference: 32%|███▏ | 1870/5920 [27:53<1:49:15, 1.62s/it, tok/s=14404.3]
Inference: 32%|███▏ | 1870/5920 [27:53<1:49:15, 1.62s/it, tok/s=14410.8]
Inference: 32%|███▏ | 1871/5920 [27:55<1:53:53, 1.69s/it, tok/s=14410.8]
Inference: 32%|███▏ | 1871/5920 [27:55<1:53:53, 1.69s/it, tok/s=14403.5]
Inference: 32%|███▏ | 1872/5920 [27:58<2:22:29, 2.11s/it, tok/s=14403.5]
Inference: 32%|███▏ | 1872/5920 [27:58<2:22:29, 2.11s/it, tok/s=14384.1]
Inference: 32%|███▏ | 1873/5920 [27:58<1:47:13, 1.59s/it, tok/s=14384.1]
Inference: 32%|███▏ | 1873/5920 [27:58<1:47:13, 1.59s/it, tok/s=14397.5]
Inference: 32%|███▏ | 1874/5920 [28:02<2:37:23, 2.33s/it, tok/s=14397.5]
Inference: 32%|███▏ | 1874/5920 [28:02<2:37:23, 2.33s/it, tok/s=14370.9]
Inference: 32%|███▏ | 1875/5920 [28:04<2:36:36, 2.32s/it, tok/s=14370.9]
Inference: 32%|███▏ | 1875/5920 [28:04<2:36:36, 2.32s/it, tok/s=14369.1]
Inference: 32%|███▏ | 1876/5920 [28:06<2:30:29, 2.23s/it, tok/s=14369.1]
Inference: 32%|███▏ | 1876/5920 [28:06<2:30:29, 2.23s/it, tok/s=14362.5]
Inference: 32%|███▏ | 1877/5920 [28:09<2:26:14, 2.17s/it, tok/s=14362.5]
Inference: 32%|███▏ | 1877/5920 [28:09<2:26:14, 2.17s/it, tok/s=14352.7]
Inference: 32%|███▏ | 1878/5920 [28:09<2:26:12, 2.17s/it, tok/s=14359.3]
Inference: 32%|███▏ | 1879/5920 [28:10<1:39:54, 1.48s/it, tok/s=14359.3]
Inference: 32%|███▏ | 1879/5920 [28:10<1:39:54, 1.48s/it, tok/s=14363.1]
Inference: 32%|███▏ | 1880/5920 [28:10<1:21:10, 1.21s/it, tok/s=14363.1]
Inference: 32%|███▏ | 1880/5920 [28:10<1:21:10, 1.21s/it, tok/s=14377.8]
Inference: 32%|███▏ | 1881/5920 [28:11<1:12:15, 1.07s/it, tok/s=14377.8]
Inference: 32%|███▏ | 1881/5920 [28:11<1:12:15, 1.07s/it, tok/s=14387.1]
Inference: 32%|███▏ | 1882/5920 [28:16<2:16:11, 2.02s/it, tok/s=14387.1]
Inference: 32%|███▏ | 1882/5920 [28:16<2:16:11, 2.02s/it, tok/s=14365.6]
Inference: 32%|███▏ | 1883/5920 [28:16<1:45:25, 1.57s/it, tok/s=14365.6]
Inference: 32%|███▏ | 1883/5920 [28:16<1:45:25, 1.57s/it, tok/s=14369.8]
Inference: 32%|███▏ | 1884/5920 [28:16<1:20:55, 1.20s/it, tok/s=14369.8]
Inference: 32%|███▏ | 1884/5920 [28:16<1:20:55, 1.20s/it, tok/s=14382.0]
Inference: 32%|███▏ | 1885/5920 [28:17<1:07:26, 1.00s/it, tok/s=14382.0]
Inference: 32%|███▏ | 1885/5920 [28:17<1:07:26, 1.00s/it, tok/s=14386.7]
Inference: 32%|███▏ | 1886/5920 [28:21<2:14:21, 2.00s/it, tok/s=14386.7]
Inference: 32%|███▏ | 1886/5920 [28:21<2:14:21, 2.00s/it, tok/s=14356.8]
Inference: 32%|███▏ | 1887/5920 [28:22<1:52:11, 1.67s/it, tok/s=14356.8]
Inference: 32%|███▏ | 1887/5920 [28:22<1:52:11, 1.67s/it, tok/s=14358.4]
Inference: 32%|███▏ | 1888/5920 [28:24<2:03:19, 1.84s/it, tok/s=14358.4]
Inference: 32%|███▏ | 1888/5920 [28:24<2:03:19, 1.84s/it, tok/s=14356.4]
Inference: 32%|███▏ | 1889/5920 [28:27<2:31:15, 2.25s/it, tok/s=14356.4]
Inference: 32%|███▏ | 1889/5920 [28:27<2:31:15, 2.25s/it, tok/s=14340.8]
Inference: 32%|███▏ | 1890/5920 [28:29<2:14:46, 2.01s/it, tok/s=14340.8]
Inference: 32%|███▏ | 1890/5920 [28:29<2:14:46, 2.01s/it, tok/s=14345.2]
Inference: 32%|███▏ | 1891/5920 [28:39<4:57:00, 4.42s/it, tok/s=14345.2]
Inference: 32%|███▏ | 1891/5920 [28:39<4:57:00, 4.42s/it, tok/s=14270.9]
Inference: 32%|███▏ | 1892/5920 [28:40<3:47:50, 3.39s/it, tok/s=14270.9]
Inference: 32%|███▏ | 1892/5920 [28:40<3:47:50, 3.39s/it, tok/s=14270.6]
Inference: 32%|███▏ | 1893/5920 [28:43<3:36:48, 3.23s/it, tok/s=14270.6]
Inference: 32%|███▏ | 1893/5920 [28:43<3:36:48, 3.23s/it, tok/s=14264.5]
Inference: 32%|███▏ | 1894/5920 [28:43<3:36:44, 3.23s/it, tok/s=14271.7]
Inference: 32%|███▏ | 1895/5920 [28:44<2:09:17, 1.93s/it, tok/s=14271.7]
Inference: 32%|███▏ | 1895/5920 [28:44<2:09:17, 1.93s/it, tok/s=14267.7]
Inference: 32%|███▏ | 1896/5920 [28:45<1:55:39, 1.72s/it, tok/s=14267.7]
Inference: 32%|███▏ | 1896/5920 [28:45<1:55:39, 1.72s/it, tok/s=14269.3]
Inference: 32%|███▏ | 1897/5920 [28:48<2:25:29, 2.17s/it, tok/s=14269.3]
Inference: 32%|███▏ | 1897/5920 [28:48<2:25:29, 2.17s/it, tok/s=14249.0]
Inference: 32%|███▏ | 1898/5920 [28:49<1:53:01, 1.69s/it, tok/s=14249.0]
Inference: 32%|███▏ | 1898/5920 [28:49<1:53:01, 1.69s/it, tok/s=14258.6]
Inference: 32%|███▏ | 1899/5920 [28:49<1:28:58, 1.33s/it, tok/s=14258.6]
Inference: 32%|███▏ | 1899/5920 [28:49<1:28:58, 1.33s/it, tok/s=14260.8]
Inference: 32%|███▏ | 1900/5920 [28:50<1:14:14, 1.11s/it, tok/s=14260.8]
Inference: 32%|███▏ | 1900/5920 [28:50<1:14:14, 1.11s/it, tok/s=14273.2]
Inference: 32%|███▏ | 1901/5920 [28:50<59:34, 1.12it/s, tok/s=14273.2]
Inference: 32%|███▏ | 1901/5920 [28:50<59:34, 1.12it/s, tok/s=14282.3]
Inference: 32%|███▏ | 1902/5920 [28:59<3:36:04, 3.23s/it, tok/s=14282.3]
Inference: 32%|███▏ | 1902/5920 [28:59<3:36:04, 3.23s/it, tok/s=14216.9]
Inference: 32%|███▏ | 1903/5920 [29:03<3:53:24, 3.49s/it, tok/s=14216.9]
Inference: 32%|███▏ | 1903/5920 [29:03<3:53:24, 3.49s/it, tok/s=14185.5]
Inference: 32%|███▏ | 1904/5920 [29:06<3:42:00, 3.32s/it, tok/s=14185.5]
Inference: 32%|███▏ | 1904/5920 [29:06<3:42:00, 3.32s/it, tok/s=14177.4]
Inference: 32%|███▏ | 1905/5920 [29:08<3:24:10, 3.05s/it, tok/s=14177.4]
Inference: 32%|███▏ | 1905/5920 [29:08<3:24:10, 3.05s/it, tok/s=14174.4]
Inference: 32%|███▏ | 1906/5920 [29:10<2:55:00, 2.62s/it, tok/s=14174.4]
Inference: 32%|███▏ | 1906/5920 [29:10<2:55:00, 2.62s/it, tok/s=14171.5]
Inference: 32%|███▏ | 1907/5920 [29:11<2:27:11, 2.20s/it, tok/s=14171.5]
Inference: 32%|███▏ | 1907/5920 [29:11<2:27:11, 2.20s/it, tok/s=14164.1]
Inference: 32%|███▏ | 1908/5920 [29:12<1:57:07, 1.75s/it, tok/s=14164.1]
Inference: 32%|███▏ | 1908/5920 [29:12<1:57:07, 1.75s/it, tok/s=14160.7]
Inference: 32%|███▏ | 1909/5920 [29:13<1:41:44, 1.52s/it, tok/s=14160.7]
Inference: 32%|███▏ | 1909/5920 [29:13<1:41:44, 1.52s/it, tok/s=14163.4]
Inference: 32%|███▏ | 1910/5920 [29:13<1:15:22, 1.13s/it, tok/s=14163.4]
Inference: 32%|███▏ | 1910/5920 [29:13<1:15:22, 1.13s/it, tok/s=14169.6]
Inference: 32%|███▏ | 1911/5920 [29:16<1:52:18, 1.68s/it, tok/s=14169.6]
Inference: 32%|███▏ | 1911/5920 [29:16<1:52:18, 1.68s/it, tok/s=14161.9]
Inference: 32%|███▏ | 1912/5920 [29:18<2:00:40, 1.81s/it, tok/s=14161.9]
Inference: 32%|███▏ | 1912/5920 [29:18<2:00:40, 1.81s/it, tok/s=14153.1]
Inference: 32%|███▏ | 1913/5920 [29:19<1:42:03, 1.53s/it, tok/s=14153.1]
Inference: 32%|███▏ | 1913/5920 [29:19<1:42:03, 1.53s/it, tok/s=14155.2]
Inference: 32%|███▏ | 1914/5920 [29:19<1:20:45, 1.21s/it, tok/s=14155.2]
Inference: 32%|███▏ | 1914/5920 [29:19<1:20:45, 1.21s/it, tok/s=14168.6]
Inference: 32%|███▏ | 1915/5920 [29:21<1:28:38, 1.33s/it, tok/s=14168.6]
Inference: 32%|███▏ | 1915/5920 [29:21<1:28:38, 1.33s/it, tok/s=14167.5]
Inference: 32%|███▏ | 1916/5920 [29:24<2:09:37, 1.94s/it, tok/s=14167.5]
Inference: 32%|███▏ | 1916/5920 [29:24<2:09:37, 1.94s/it, tok/s=14148.8]
Inference: 32%|███▏ | 1917/5920 [29:25<1:36:07, 1.44s/it, tok/s=14148.8]
Inference: 32%|███▏ | 1917/5920 [29:25<1:36:07, 1.44s/it, tok/s=14162.3]
Inference: 32%|███▏ | 1918/5920 [29:25<1:21:14, 1.22s/it, tok/s=14162.3]
Inference: 32%|███▏ | 1918/5920 [29:25<1:21:14, 1.22s/it, tok/s=14168.5]
Inference: 32%|███▏ | 1919/5920 [29:25<1:21:13, 1.22s/it, tok/s=14184.9]
Inference: 32%|███▏ | 1920/5920 [29:26<48:55, 1.36it/s, tok/s=14184.9]
Inference: 32%|███▏ | 1920/5920 [29:26<48:55, 1.36it/s, tok/s=14192.4]
Inference: 32%|███▏ | 1921/5920 [29:26<48:54, 1.36it/s, tok/s=14200.6]
Inference: 32%|███▏ | 1922/5920 [29:27<48:52, 1.36it/s, tok/s=14200.6]
Inference: 32%|███▏ | 1922/5920 [29:27<48:52, 1.36it/s, tok/s=14191.6]
Inference: 32%|███▏ | 1923/5920 [29:27<40:24, 1.65it/s, tok/s=14191.6]
Inference: 32%|███▏ | 1923/5920 [29:27<40:24, 1.65it/s, tok/s=14202.0]
Inference: 32%|███▎ | 1924/5920 [29:29<56:11, 1.19it/s, tok/s=14202.0]
Inference: 32%|███▎ | 1924/5920 [29:29<56:11, 1.19it/s, tok/s=14191.5]
Inference: 33%|███▎ | 1925/5920 [29:29<56:11, 1.19it/s, tok/s=14193.2]
Inference: 33%|███▎ | 1926/5920 [29:30<46:23, 1.44it/s, tok/s=14193.2]
Inference: 33%|███▎ | 1926/5920 [29:30<46:23, 1.44it/s, tok/s=14188.6]
Inference: 33%|███▎ | 1927/5920 [29:30<41:47, 1.59it/s, tok/s=14188.6]
Inference: 33%|███▎ | 1927/5920 [29:30<41:47, 1.59it/s, tok/s=14194.8]
Inference: 33%|███▎ | 1928/5920 [29:31<42:23, 1.57it/s, tok/s=14194.8]
Inference: 33%|███▎ | 1928/5920 [29:31<42:23, 1.57it/s, tok/s=14197.6]
Inference: 33%|███▎ | 1929/5920 [29:32<54:15, 1.23it/s, tok/s=14197.6]
Inference: 33%|███▎ | 1929/5920 [29:32<54:15, 1.23it/s, tok/s=14199.7]
Inference: 33%|███▎ | 1930/5920 [29:35<1:21:39, 1.23s/it, tok/s=14199.7]
Inference: 33%|███▎ | 1930/5920 [29:35<1:21:39, 1.23s/it, tok/s=14190.2]
Inference: 33%|███▎ | 1931/5920 [29:38<1:55:33, 1.74s/it, tok/s=14190.2]
Inference: 33%|███▎ | 1931/5920 [29:38<1:55:33, 1.74s/it, tok/s=14171.3]
Inference: 33%|███▎ | 1932/5920 [29:38<1:28:40, 1.33s/it, tok/s=14171.3]
Inference: 33%|███▎ | 1932/5920 [29:38<1:28:40, 1.33s/it, tok/s=14177.4]
Inference: 33%|███▎ | 1933/5920 [29:38<1:28:39, 1.33s/it, tok/s=14187.1]
Inference: 33%|███▎ | 1934/5920 [29:39<58:57, 1.13it/s, tok/s=14187.1]
Inference: 33%|███▎ | 1934/5920 [29:39<58:57, 1.13it/s, tok/s=14199.3]
Inference: 33%|███▎ | 1935/5920 [29:40<1:12:28, 1.09s/it, tok/s=14199.3]
Inference: 33%|███▎ | 1935/5920 [29:40<1:12:28, 1.09s/it, tok/s=14197.5]
Inference: 33%|███▎ | 1936/5920 [29:42<1:20:46, 1.22s/it, tok/s=14197.5]
Inference: 33%|███▎ | 1936/5920 [29:42<1:20:46, 1.22s/it, tok/s=14187.5]
Inference: 33%|███▎ | 1937/5920 [29:43<1:19:14, 1.19s/it, tok/s=14187.5]
Inference: 33%|███▎ | 1937/5920 [29:43<1:19:14, 1.19s/it, tok/s=14180.9]
Inference: 33%|███▎ | 1938/5920 [29:45<1:37:34, 1.47s/it, tok/s=14180.9]
Inference: 33%|███▎ | 1938/5920 [29:45<1:37:34, 1.47s/it, tok/s=14165.7]
Inference: 33%|███▎ | 1939/5920 [29:47<1:43:44, 1.56s/it, tok/s=14165.7]
Inference: 33%|███▎ | 1939/5920 [29:47<1:43:44, 1.56s/it, tok/s=14168.3]
Inference: 33%|███▎ | 1940/5920 [29:47<1:18:31, 1.18s/it, tok/s=14168.3]
Inference: 33%|███▎ | 1940/5920 [29:47<1:18:31, 1.18s/it, tok/s=14175.2]
Inference: 33%|███▎ | 1941/5920 [29:48<1:10:55, 1.07s/it, tok/s=14175.2]
Inference: 33%|███▎ | 1941/5920 [29:48<1:10:55, 1.07s/it, tok/s=14177.0]
Inference: 33%|███▎ | 1942/5920 [29:50<1:33:03, 1.40s/it, tok/s=14177.0]
Inference: 33%|███▎ | 1942/5920 [29:50<1:33:03, 1.40s/it, tok/s=14161.9]
Inference: 33%|███▎ | 1943/5920 [29:51<1:16:25, 1.15s/it, tok/s=14161.9]
Inference: 33%|███▎ | 1943/5920 [29:51<1:16:25, 1.15s/it, tok/s=14174.4]
Inference: 33%|███▎ | 1944/5920 [29:52<1:08:09, 1.03s/it, tok/s=14174.4]
Inference: 33%|███▎ | 1944/5920 [29:52<1:08:09, 1.03s/it, tok/s=14180.5]
Inference: 33%|███▎ | 1945/5920 [29:53<1:15:05, 1.13s/it, tok/s=14180.5]
Inference: 33%|███▎ | 1945/5920 [29:53<1:15:05, 1.13s/it, tok/s=14171.8]
Inference: 33%|███▎ | 1946/5920 [29:55<1:40:29, 1.52s/it, tok/s=14171.8]
Inference: 33%|███▎ | 1946/5920 [29:55<1:40:29, 1.52s/it, tok/s=14164.6]
Inference: 33%|███▎ | 1947/5920 [29:56<1:22:00, 1.24s/it, tok/s=14164.6]
Inference: 33%|███▎ | 1947/5920 [29:56<1:22:00, 1.24s/it, tok/s=14176.8]
Inference: 33%|███▎ | 1948/5920 [29:57<1:09:47, 1.05s/it, tok/s=14176.8]
Inference: 33%|███▎ | 1948/5920 [29:57<1:09:47, 1.05s/it, tok/s=14173.8]
Inference: 33%|███▎ | 1949/5920 [29:58<1:15:38, 1.14s/it, tok/s=14173.8]
Inference: 33%|███▎ | 1949/5920 [29:58<1:15:38, 1.14s/it, tok/s=14172.7]
Inference: 33%|███▎ | 1950/5920 [29:58<57:08, 1.16it/s, tok/s=14172.7]
Inference: 33%|███▎ | 1950/5920 [29:58<57:08, 1.16it/s, tok/s=14173.3]
Inference: 33%|███▎ | 1951/5920 [30:00<1:13:00, 1.10s/it, tok/s=14173.3]
Inference: 33%|███▎ | 1951/5920 [30:00<1:13:00, 1.10s/it, tok/s=14177.0]
Inference: 33%|███▎ | 1952/5920 [30:01<1:10:31, 1.07s/it, tok/s=14177.0]
Inference: 33%|███▎ | 1952/5920 [30:01<1:10:31, 1.07s/it, tok/s=14171.8]
Inference: 33%|███▎ | 1953/5920 [30:02<1:23:10, 1.26s/it, tok/s=14171.8]
Inference: 33%|███▎ | 1953/5920 [30:02<1:23:10, 1.26s/it, tok/s=14174.0]
Inference: 33%|███▎ | 1954/5920 [30:06<2:10:31, 1.97s/it, tok/s=14174.0]
Inference: 33%|███▎ | 1954/5920 [30:06<2:10:31, 1.97s/it, tok/s=14152.3]
Inference: 33%|███▎ | 1955/5920 [30:07<1:46:30, 1.61s/it, tok/s=14152.3]
Inference: 33%|███▎ | 1955/5920 [30:07<1:46:30, 1.61s/it, tok/s=14148.1]
Inference: 33%|███▎ | 1956/5920 [30:09<1:57:23, 1.78s/it, tok/s=14148.1]
Inference: 33%|███▎ | 1956/5920 [30:09<1:57:23, 1.78s/it, tok/s=14133.7]
Inference: 33%|███▎ | 1957/5920 [30:11<1:52:49, 1.71s/it, tok/s=14133.7]
Inference: 33%|███▎ | 1957/5920 [30:11<1:52:49, 1.71s/it, tok/s=14124.2]
Inference: 33%|███▎ | 1958/5920 [30:13<2:15:50, 2.06s/it, tok/s=14124.2]
Inference: 33%|███▎ | 1958/5920 [30:13<2:15:50, 2.06s/it, tok/s=14118.5]
Inference: 33%|███▎ | 1959/5920 [30:14<1:52:26, 1.70s/it, tok/s=14118.5]
Inference: 33%|███▎ | 1959/5920 [30:14<1:52:26, 1.70s/it, tok/s=14121.0]
Inference: 33%|███▎ | 1960/5920 [30:15<1:25:39, 1.30s/it, tok/s=14121.0]
Inference: 33%|███▎ | 1960/5920 [30:15<1:25:39, 1.30s/it, tok/s=14126.5]
Inference: 33%|███▎ | 1961/5920 [30:15<1:14:28, 1.13s/it, tok/s=14126.5]
Inference: 33%|███▎ | 1961/5920 [30:15<1:14:28, 1.13s/it, tok/s=14123.6]
Inference: 33%|███▎ | 1962/5920 [30:17<1:32:19, 1.40s/it, tok/s=14123.6]
Inference: 33%|███▎ | 1962/5920 [30:17<1:32:19, 1.40s/it, tok/s=14124.4]
Inference: 33%|███▎ | 1963/5920 [30:18<1:12:48, 1.10s/it, tok/s=14124.4]
Inference: 33%|███▎ | 1963/5920 [30:18<1:12:48, 1.10s/it, tok/s=14133.0]
Inference: 33%|███▎ | 1964/5920 [30:19<1:14:28, 1.13s/it, tok/s=14133.0]
Inference: 33%|███▎ | 1964/5920 [30:19<1:14:28, 1.13s/it, tok/s=14140.3]
Inference: 33%|███▎ | 1965/5920 [30:19<1:14:27, 1.13s/it, tok/s=14156.2]
Inference: 33%|███▎ | 1966/5920 [30:19<42:30, 1.55it/s, tok/s=14156.2]
Inference: 33%|███▎ | 1966/5920 [30:19<42:30, 1.55it/s, tok/s=14166.4]
Inference: 33%|███▎ | 1967/5920 [30:21<1:02:14, 1.06it/s, tok/s=14166.4]
Inference: 33%|███▎ | 1967/5920 [30:21<1:02:14, 1.06it/s, tok/s=14161.3]
Inference: 33%|███▎ | 1968/5920 [30:23<1:23:21, 1.27s/it, tok/s=14161.3]
Inference: 33%|███▎ | 1968/5920 [30:23<1:23:21, 1.27s/it, tok/s=14151.1]
Inference: 33%|███▎ | 1969/5920 [30:23<1:23:20, 1.27s/it, tok/s=14166.5]
Inference: 33%|███▎ | 1970/5920 [30:26<1:20:22, 1.22s/it, tok/s=14166.5]
Inference: 33%|███▎ | 1970/5920 [30:26<1:20:22, 1.22s/it, tok/s=14156.7]
Inference: 33%|███▎ | 1971/5920 [30:26<1:07:57, 1.03s/it, tok/s=14156.7]
Inference: 33%|███▎ | 1971/5920 [30:26<1:07:57, 1.03s/it, tok/s=14165.4]
Inference: 33%|███▎ | 1972/5920 [30:28<1:16:58, 1.17s/it, tok/s=14165.4]
Inference: 33%|███▎ | 1972/5920 [30:28<1:16:58, 1.17s/it, tok/s=14169.7]
Inference: 33%|███▎ | 1973/5920 [30:28<1:04:40, 1.02it/s, tok/s=14169.7]
Inference: 33%|███▎ | 1973/5920 [30:28<1:04:40, 1.02it/s, tok/s=14180.2]
Inference: 33%|███▎ | 1974/5920 [30:30<1:27:43, 1.33s/it, tok/s=14180.2]
Inference: 33%|███▎ | 1974/5920 [30:30<1:27:43, 1.33s/it, tok/s=14164.7]
Inference: 33%|███▎ | 1975/5920 [30:30<1:27:41, 1.33s/it, tok/s=14176.3]
Inference: 33%|███▎ | 1976/5920 [30:30<52:14, 1.26it/s, tok/s=14176.3]
Inference: 33%|███▎ | 1976/5920 [30:30<52:14, 1.26it/s, tok/s=14191.6]
Inference: 33%|███▎ | 1977/5920 [30:31<49:53, 1.32it/s, tok/s=14191.6]
Inference: 33%|███▎ | 1977/5920 [30:31<49:53, 1.32it/s, tok/s=14197.3]
Inference: 33%|███▎ | 1978/5920 [30:32<48:56, 1.34it/s, tok/s=14197.3]
Inference: 33%|███▎ | 1978/5920 [30:32<48:56, 1.34it/s, tok/s=14208.4]
Inference: 33%|███▎ | 1979/5920 [30:35<1:24:51, 1.29s/it, tok/s=14208.4]
Inference: 33%|███▎ | 1979/5920 [30:35<1:24:51, 1.29s/it, tok/s=14203.1]
Inference: 33%|███▎ | 1980/5920 [30:35<1:07:15, 1.02s/it, tok/s=14203.1]
Inference: 33%|███▎ | 1980/5920 [30:35<1:07:15, 1.02s/it, tok/s=14217.1]
Inference: 33%|███▎ | 1981/5920 [30:35<55:46, 1.18it/s, tok/s=14217.1]
Inference: 33%|███▎ | 1981/5920 [30:35<55:46, 1.18it/s, tok/s=14216.7]
Inference: 33%|███▎ | 1982/5920 [30:36<56:14, 1.17it/s, tok/s=14216.7]
Inference: 33%|███▎ | 1982/5920 [30:36<56:14, 1.17it/s, tok/s=14226.4]
Inference: 33%|███▎ | 1983/5920 [30:37<55:56, 1.17it/s, tok/s=14226.4]
Inference: 33%|███▎ | 1983/5920 [30:37<55:56, 1.17it/s, tok/s=14228.1]
Inference: 34%|███▎ | 1984/5920 [30:39<1:12:02, 1.10s/it, tok/s=14228.1]
Inference: 34%|███▎ | 1984/5920 [30:39<1:12:02, 1.10s/it, tok/s=14223.9]
Inference: 34%|███▎ | 1985/5920 [30:39<55:54, 1.17it/s, tok/s=14223.9]
Inference: 34%|███▎ | 1985/5920 [30:39<55:54, 1.17it/s, tok/s=14233.3]
Inference: 34%|███▎ | 1986/5920 [30:41<1:20:19, 1.23s/it, tok/s=14233.3]
Inference: 34%|███▎ | 1986/5920 [30:41<1:20:19, 1.23s/it, tok/s=14228.9]
Inference: 34%|███▎ | 1987/5920 [30:41<1:01:31, 1.07it/s, tok/s=14228.9]
Inference: 34%|███▎ | 1987/5920 [30:41<1:01:31, 1.07it/s, tok/s=14243.3]
Inference: 34%|███▎ | 1988/5920 [30:43<1:05:27, 1.00it/s, tok/s=14243.3]
Inference: 34%|███▎ | 1988/5920 [30:43<1:05:27, 1.00it/s, tok/s=14242.0]
Inference: 34%|███▎ | 1989/5920 [30:43<53:47, 1.22it/s, tok/s=14242.0]
Inference: 34%|███▎ | 1989/5920 [30:43<53:47, 1.22it/s, tok/s=14252.9]
Inference: 34%|███▎ | 1990/5920 [30:43<40:40, 1.61it/s, tok/s=14252.9]
Inference: 34%|███▎ | 1990/5920 [30:43<40:40, 1.61it/s, tok/s=14268.1]
Inference: 34%|███▎ | 1991/5920 [30:43<40:39, 1.61it/s, tok/s=14270.2]
Inference: 34%|███▎ | 1992/5920 [30:44<38:21, 1.71it/s, tok/s=14270.2]
Inference: 34%|███▎ | 1992/5920 [30:44<38:21, 1.71it/s, tok/s=14271.2]
Inference: 34%|███▎ | 1993/5920 [30:47<1:09:18, 1.06s/it, tok/s=14271.2]
Inference: 34%|███▎ | 1993/5920 [30:47<1:09:18, 1.06s/it, tok/s=14254.5]
Inference: 34%|███▎ | 1994/5920 [30:53<2:35:59, 2.38s/it, tok/s=14254.5]
Inference: 34%|███▎ | 1994/5920 [30:53<2:35:59, 2.38s/it, tok/s=14210.2]
Inference: 34%|███▎ | 1995/5920 [30:54<2:17:13, 2.10s/it, tok/s=14210.2]
Inference: 34%|███▎ | 1995/5920 [30:54<2:17:13, 2.10s/it, tok/s=14214.4]
Inference: 34%|███▎ | 1996/5920 [30:57<2:29:46, 2.29s/it, tok/s=14214.4]
Inference: 34%|███▎ | 1996/5920 [30:57<2:29:46, 2.29s/it, tok/s=14208.3]
Inference: 34%|███▎ | 1997/5920 [30:57<1:50:44, 1.69s/it, tok/s=14208.3]
Inference: 34%|███▎ | 1997/5920 [30:57<1:50:44, 1.69s/it, tok/s=14222.7]
Inference: 34%|███▍ | 1998/5920 [30:58<1:40:33, 1.54s/it, tok/s=14222.7]
Inference: 34%|███▍ | 1998/5920 [30:58<1:40:33, 1.54s/it, tok/s=14216.0]
Inference: 34%|███▍ | 1999/5920 [30:58<1:40:32, 1.54s/it, tok/s=14219.0]
Inference: 34%|███▍ | 2000/5920 [30:59<1:02:36, 1.04it/s, tok/s=14219.0]
Inference: 34%|███▍ | 2000/5920 [30:59<1:02:36, 1.04it/s, tok/s=14231.4]
Inference: 34%|███▍ | 2001/5920 [31:01<1:24:38, 1.30s/it, tok/s=14231.4]
Inference: 34%|███▍ | 2001/5920 [31:01<1:24:38, 1.30s/it, tok/s=14229.7]
Inference: 34%|███▍ | 2002/5920 [31:02<1:23:20, 1.28s/it, tok/s=14229.7]
Inference: 34%|███▍ | 2002/5920 [31:02<1:23:20, 1.28s/it, tok/s=14230.9]
Inference: 34%|███▍ | 2003/5920 [31:03<1:16:59, 1.18s/it, tok/s=14230.9]
Inference: 34%|███▍ | 2003/5920 [31:03<1:16:59, 1.18s/it, tok/s=14240.1]
Inference: 34%|███▍ | 2004/5920 [31:04<1:09:15, 1.06s/it, tok/s=14240.1]
Inference: 34%|███▍ | 2004/5920 [31:04<1:09:15, 1.06s/it, tok/s=14237.1]
Inference: 34%|███▍ | 2005/5920 [31:04<57:10, 1.14it/s, tok/s=14237.1]
Inference: 34%|███▍ | 2005/5920 [31:04<57:10, 1.14it/s, tok/s=14250.1]
Inference: 34%|███▍ | 2006/5920 [31:07<1:30:37, 1.39s/it, tok/s=14250.1]
Inference: 34%|███▍ | 2006/5920 [31:07<1:30:37, 1.39s/it, tok/s=14242.7]
Inference: 34%|███▍ | 2007/5920 [31:08<1:18:27, 1.20s/it, tok/s=14242.7]
Inference: 34%|███▍ | 2007/5920 [31:08<1:18:27, 1.20s/it, tok/s=14239.5]
Inference: 34%|███▍ | 2008/5920 [31:09<1:20:36, 1.24s/it, tok/s=14239.5]
Inference: 34%|███▍ | 2008/5920 [31:09<1:20:36, 1.24s/it, tok/s=14245.6]
Inference: 34%|███▍ | 2009/5920 [31:11<1:30:17, 1.39s/it, tok/s=14245.6]
Inference: 34%|███▍ | 2009/5920 [31:11<1:30:17, 1.39s/it, tok/s=14247.4]
Inference: 34%|███▍ | 2010/5920 [31:15<2:20:26, 2.16s/it, tok/s=14247.4]
Inference: 34%|███▍ | 2010/5920 [31:15<2:20:26, 2.16s/it, tok/s=14226.7]
Inference: 34%|███▍ | 2011/5920 [31:15<1:41:01, 1.55s/it, tok/s=14226.7]
Inference: 34%|███▍ | 2011/5920 [31:15<1:41:01, 1.55s/it, tok/s=14241.8]
Inference: 34%|███▍ | 2012/5920 [31:16<1:35:45, 1.47s/it, tok/s=14241.8]
Inference: 34%|███▍ | 2012/5920 [31:16<1:35:45, 1.47s/it, tok/s=14233.9]
Inference: 34%|███▍ | 2013/5920 [31:16<1:35:44, 1.47s/it, tok/s=14236.5]
Inference: 34%|███▍ | 2014/5920 [31:17<1:07:17, 1.03s/it, tok/s=14236.5]
Inference: 34%|███▍ | 2014/5920 [31:17<1:07:17, 1.03s/it, tok/s=14237.4]
Inference: 34%|███▍ | 2015/5920 [31:18<55:48, 1.17it/s, tok/s=14237.4]
Inference: 34%|███▍ | 2015/5920 [31:18<55:48, 1.17it/s, tok/s=14243.7]
Inference: 34%|███▍ | 2016/5920 [31:19<1:03:45, 1.02it/s, tok/s=14243.7]
Inference: 34%|███▍ | 2016/5920 [31:19<1:03:45, 1.02it/s, tok/s=14249.7]
Inference: 34%|███▍ | 2017/5920 [31:22<1:38:46, 1.52s/it, tok/s=14249.7]
Inference: 34%|███▍ | 2017/5920 [31:22<1:38:46, 1.52s/it, tok/s=14240.7]
Inference: 34%|███▍ | 2018/5920 [31:23<1:31:47, 1.41s/it, tok/s=14240.7]
Inference: 34%|███▍ | 2018/5920 [31:23<1:31:47, 1.41s/it, tok/s=14246.0]
Inference: 34%|███▍ | 2019/5920 [31:24<1:22:17, 1.27s/it, tok/s=14246.0]
Inference: 34%|███▍ | 2019/5920 [31:24<1:22:17, 1.27s/it, tok/s=14255.2]
Inference: 34%|███▍ | 2020/5920 [31:24<1:07:10, 1.03s/it, tok/s=14255.2]
Inference: 34%|███▍ | 2020/5920 [31:24<1:07:10, 1.03s/it, tok/s=14259.9]
Inference: 34%|███▍ | 2021/5920 [31:25<51:26, 1.26it/s, tok/s=14259.9]
Inference: 34%|███▍ | 2021/5920 [31:25<51:26, 1.26it/s, tok/s=14260.8]
Inference: 34%|███▍ | 2022/5920 [31:25<51:25, 1.26it/s, tok/s=14262.5]
Inference: 34%|███▍ | 2023/5920 [31:25<39:37, 1.64it/s, tok/s=14262.5]
Inference: 34%|███▍ | 2023/5920 [31:25<39:37, 1.64it/s, tok/s=14267.1]
Inference: 34%|███▍ | 2024/5920 [31:27<56:50, 1.14it/s, tok/s=14267.1]
Inference: 34%|███▍ | 2024/5920 [31:27<56:50, 1.14it/s, tok/s=14266.5]
Inference: 34%|███▍ | 2025/5920 [31:29<1:10:13, 1.08s/it, tok/s=14266.5]
Inference: 34%|███▍ | 2025/5920 [31:29<1:10:13, 1.08s/it, tok/s=14269.9]
Inference: 34%|███▍ | 2026/5920 [31:30<1:04:11, 1.01it/s, tok/s=14269.9]
Inference: 34%|███▍ | 2026/5920 [31:30<1:04:11, 1.01it/s, tok/s=14277.7]
Inference: 34%|███▍ | 2027/5920 [31:30<59:16, 1.09it/s, tok/s=14277.7]
Inference: 34%|███▍ | 2027/5920 [31:30<59:16, 1.09it/s, tok/s=14284.8]
Inference: 34%|███▍ | 2028/5920 [31:31<53:27, 1.21it/s, tok/s=14284.8]
Inference: 34%|███▍ | 2028/5920 [31:31<53:27, 1.21it/s, tok/s=14296.3]
Inference: 34%|███▍ | 2029/5920 [31:31<40:13, 1.61it/s, tok/s=14296.3]
Inference: 34%|███▍ | 2029/5920 [31:31<40:13, 1.61it/s, tok/s=14303.2]
Inference: 34%|███▍ | 2030/5920 [31:32<54:03, 1.20it/s, tok/s=14303.2]
Inference: 34%|███▍ | 2030/5920 [31:32<54:03, 1.20it/s, tok/s=14305.0]
Inference: 34%|███▍ | 2031/5920 [31:33<56:24, 1.15it/s, tok/s=14305.0]
Inference: 34%|███▍ | 2031/5920 [31:33<56:24, 1.15it/s, tok/s=14304.9]
Inference: 34%|███▍ | 2032/5920 [31:34<1:01:55, 1.05it/s, tok/s=14304.9]
Inference: 34%|███▍ | 2032/5920 [31:34<1:01:55, 1.05it/s, tok/s=14309.9]
Inference: 34%|███▍ | 2033/5920 [31:35<47:27, 1.36it/s, tok/s=14309.9]
Inference: 34%|███▍ | 2033/5920 [31:35<47:27, 1.36it/s, tok/s=14320.9]
Inference: 34%|███▍ | 2034/5920 [31:37<1:13:05, 1.13s/it, tok/s=14320.9]
Inference: 34%|███▍ | 2034/5920 [31:37<1:13:05, 1.13s/it, tok/s=14321.3]
Inference: 34%|███▍ | 2035/5920 [31:37<1:03:01, 1.03it/s, tok/s=14321.3]
Inference: 34%|███▍ | 2035/5920 [31:37<1:03:01, 1.03it/s, tok/s=14320.2]
Inference: 34%|███▍ | 2036/5920 [31:38<48:17, 1.34it/s, tok/s=14320.2]
Inference: 34%|███▍ | 2036/5920 [31:38<48:17, 1.34it/s, tok/s=14333.6]
Inference: 34%|███▍ | 2037/5920 [31:40<1:25:42, 1.32s/it, tok/s=14333.6]
Inference: 34%|███▍ | 2037/5920 [31:40<1:25:42, 1.32s/it, tok/s=14322.5]
Inference: 34%|███▍ | 2038/5920 [31:41<1:20:51, 1.25s/it, tok/s=14322.5]
Inference: 34%|███▍ | 2038/5920 [31:41<1:20:51, 1.25s/it, tok/s=14326.6]
Inference: 34%|███▍ | 2039/5920 [31:41<1:20:50, 1.25s/it, tok/s=14342.3]
Inference: 34%|███▍ | 2040/5920 [31:46<1:58:45, 1.84s/it, tok/s=14342.3]
Inference: 34%|███▍ | 2040/5920 [31:46<1:58:45, 1.84s/it, tok/s=14320.3]
Inference: 34%|███▍ | 2041/5920 [31:52<2:52:17, 2.67s/it, tok/s=14320.3]
Inference: 34%|███▍ | 2041/5920 [31:52<2:52:17, 2.67s/it, tok/s=14297.3]
Inference: 34%|███▍ | 2042/5920 [31:54<2:52:34, 2.67s/it, tok/s=14297.3]
Inference: 34%|███▍ | 2042/5920 [31:54<2:52:34, 2.67s/it, tok/s=14293.1]
Inference: 35%|███▍ | 2043/5920 [31:56<2:39:48, 2.47s/it, tok/s=14293.1]
Inference: 35%|███▍ | 2043/5920 [31:56<2:39:48, 2.47s/it, tok/s=14281.2]
Inference: 35%|███▍ | 2044/5920 [32:00<3:10:24, 2.95s/it, tok/s=14281.2]
Inference: 35%|███▍ | 2044/5920 [32:00<3:10:24, 2.95s/it, tok/s=14263.8]
Inference: 35%|███▍ | 2045/5920 [32:01<2:37:53, 2.44s/it, tok/s=14263.8]
Inference: 35%|███▍ | 2045/5920 [32:01<2:37:53, 2.44s/it, tok/s=14270.7]
Inference: 35%|███▍ | 2046/5920 [32:03<2:25:57, 2.26s/it, tok/s=14270.7]
Inference: 35%|███▍ | 2046/5920 [32:03<2:25:57, 2.26s/it, tok/s=14265.8]
Inference: 35%|███▍ | 2047/5920 [32:06<2:28:13, 2.30s/it, tok/s=14265.8]
Inference: 35%|███▍ | 2047/5920 [32:06<2:28:13, 2.30s/it, tok/s=14260.5]
Inference: 35%|███▍ | 2048/5920 [32:06<1:54:01, 1.77s/it, tok/s=14260.5]
Inference: 35%|███▍ | 2048/5920 [32:06<1:54:01, 1.77s/it, tok/s=14259.9]
Inference: 35%|███▍ | 2049/5920 [32:06<1:25:40, 1.33s/it, tok/s=14259.9]
Inference: 35%|███▍ | 2049/5920 [32:06<1:25:40, 1.33s/it, tok/s=14260.1]
Inference: 35%|███▍ | 2050/5920 [32:09<1:42:45, 1.59s/it, tok/s=14260.1]
Inference: 35%|███▍ | 2050/5920 [32:09<1:42:45, 1.59s/it, tok/s=14246.5]
Inference: 35%|███▍ | 2051/5920 [32:10<1:33:33, 1.45s/it, tok/s=14246.5]
Inference: 35%|███▍ | 2051/5920 [32:10<1:33:33, 1.45s/it, tok/s=14241.7]
Inference: 35%|███▍ | 2052/5920 [32:16<2:55:56, 2.73s/it, tok/s=14241.7]
Inference: 35%|███▍ | 2052/5920 [32:16<2:55:56, 2.73s/it, tok/s=14215.1]
Inference: 35%|███▍ | 2053/5920 [32:17<2:25:25, 2.26s/it, tok/s=14215.1]
Inference: 35%|███▍ | 2053/5920 [32:17<2:25:25, 2.26s/it, tok/s=14209.2]
Inference: 35%|███▍ | 2054/5920 [32:17<1:46:04, 1.65s/it, tok/s=14209.2]
Inference: 35%|███▍ | 2054/5920 [32:17<1:46:04, 1.65s/it, tok/s=14209.8]
Inference: 35%|███▍ | 2055/5920 [32:18<1:26:24, 1.34s/it, tok/s=14209.8]
Inference: 35%|███▍ | 2055/5920 [32:18<1:26:24, 1.34s/it, tok/s=14213.3]
Inference: 35%|███▍ | 2056/5920 [32:18<1:17:35, 1.20s/it, tok/s=14213.3]
Inference: 35%|███▍ | 2056/5920 [32:18<1:17:35, 1.20s/it, tok/s=14209.5]
Inference: 35%|███▍ | 2057/5920 [32:19<1:00:17, 1.07it/s, tok/s=14209.5]
Inference: 35%|███▍ | 2057/5920 [32:19<1:00:17, 1.07it/s, tok/s=14222.8]
Inference: 35%|███▍ | 2058/5920 [32:20<1:08:37, 1.07s/it, tok/s=14222.8]
Inference: 35%|███▍ | 2058/5920 [32:20<1:08:37, 1.07s/it, tok/s=14215.4]
Inference: 35%|███▍ | 2059/5920 [32:21<57:39, 1.12it/s, tok/s=14215.4]
Inference: 35%|███▍ | 2059/5920 [32:21<57:39, 1.12it/s, tok/s=14227.0]
Inference: 35%|███▍ | 2060/5920 [32:21<42:27, 1.52it/s, tok/s=14227.0]
Inference: 35%|███▍ | 2060/5920 [32:21<42:27, 1.52it/s, tok/s=14240.7]
Inference: 35%|███▍ | 2061/5920 [32:21<39:44, 1.62it/s, tok/s=14240.7]
Inference: 35%|███▍ | 2061/5920 [32:21<39:44, 1.62it/s, tok/s=14252.4]
Inference: 35%|███▍ | 2062/5920 [32:21<39:43, 1.62it/s, tok/s=14259.1]
Inference: 35%|███▍ | 2063/5920 [32:24<1:01:41, 1.04it/s, tok/s=14259.1]
Inference: 35%|███▍ | 2063/5920 [32:24<1:01:41, 1.04it/s, tok/s=14255.2]
Inference: 35%|███▍ | 2064/5920 [32:25<1:07:05, 1.04s/it, tok/s=14255.2]
Inference: 35%|███▍ | 2064/5920 [32:25<1:07:05, 1.04s/it, tok/s=14248.8]
Inference: 35%|███▍ | 2065/5920 [32:26<56:14, 1.14it/s, tok/s=14248.8]
Inference: 35%|███▍ | 2065/5920 [32:26<56:14, 1.14it/s, tok/s=14248.0]
Inference: 35%|███▍ | 2066/5920 [32:26<49:26, 1.30it/s, tok/s=14248.0]
Inference: 35%|███▍ | 2066/5920 [32:26<49:26, 1.30it/s, tok/s=14253.8]
Inference: 35%|███▍ | 2067/5920 [32:29<1:27:33, 1.36s/it, tok/s=14253.8]
Inference: 35%|███▍ | 2067/5920 [32:29<1:27:33, 1.36s/it, tok/s=14246.3]
Inference: 35%|███▍ | 2068/5920 [32:31<1:42:49, 1.60s/it, tok/s=14246.3]
Inference: 35%|███▍ | 2068/5920 [32:31<1:42:49, 1.60s/it, tok/s=14240.3]
Inference: 35%|███▍ | 2069/5920 [32:38<3:10:48, 2.97s/it, tok/s=14240.3]
Inference: 35%|███▍ | 2069/5920 [32:38<3:10:48, 2.97s/it, tok/s=14201.5]
Inference: 35%|███▍ | 2070/5920 [32:38<3:10:45, 2.97s/it, tok/s=14203.5]
Inference: 35%|███▍ | 2071/5920 [32:40<2:18:12, 2.15s/it, tok/s=14203.5]
Inference: 35%|███▍ | 2071/5920 [32:40<2:18:12, 2.15s/it, tok/s=14195.7]
Inference: 35%|███▌ | 2072/5920 [32:40<1:47:53, 1.68s/it, tok/s=14195.7]
Inference: 35%|███▌ | 2072/5920 [32:40<1:47:53, 1.68s/it, tok/s=14201.8]
Inference: 35%|███▌ | 2073/5920 [32:43<2:13:50, 2.09s/it, tok/s=14201.8]
Inference: 35%|███▌ | 2073/5920 [32:43<2:13:50, 2.09s/it, tok/s=14187.3]
Inference: 35%|███▌ | 2074/5920 [32:44<1:55:06, 1.80s/it, tok/s=14187.3]
Inference: 35%|███▌ | 2074/5920 [32:44<1:55:06, 1.80s/it, tok/s=14195.3]
Inference: 35%|███▌ | 2075/5920 [32:48<2:28:41, 2.32s/it, tok/s=14195.3]
Inference: 35%|███▌ | 2075/5920 [32:48<2:28:41, 2.32s/it, tok/s=14184.2]
Inference: 35%|███▌ | 2076/5920 [32:49<1:58:58, 1.86s/it, tok/s=14184.2]
Inference: 35%|███▌ | 2076/5920 [32:49<1:58:58, 1.86s/it, tok/s=14194.5]
Inference: 35%|███▌ | 2077/5920 [32:53<2:41:35, 2.52s/it, tok/s=14194.5]
Inference: 35%|███▌ | 2077/5920 [32:53<2:41:35, 2.52s/it, tok/s=14177.4]
Inference: 35%|███▌ | 2078/5920 [32:53<1:59:12, 1.86s/it, tok/s=14177.4]
Inference: 35%|███▌ | 2078/5920 [32:53<1:59:12, 1.86s/it, tok/s=14177.7]
Inference: 35%|███▌ | 2079/5920 [32:54<1:36:50, 1.51s/it, tok/s=14177.7]
Inference: 35%|███▌ | 2079/5920 [32:54<1:36:50, 1.51s/it, tok/s=14187.6]
Inference: 35%|███▌ | 2080/5920 [32:55<1:35:42, 1.50s/it, tok/s=14187.6]
Inference: 35%|███▌ | 2080/5920 [32:55<1:35:42, 1.50s/it, tok/s=14180.2]
Inference: 35%|███▌ | 2081/5920 [32:56<1:23:17, 1.30s/it, tok/s=14180.2]
Inference: 35%|███▌ | 2081/5920 [32:56<1:23:17, 1.30s/it, tok/s=14189.5]
Inference: 35%|███▌ | 2082/5920 [32:57<1:12:24, 1.13s/it, tok/s=14189.5]
Inference: 35%|███▌ | 2082/5920 [32:57<1:12:24, 1.13s/it, tok/s=14197.4]
Inference: 35%|███▌ | 2083/5920 [33:04<3:01:59, 2.85s/it, tok/s=14197.4]
Inference: 35%|███▌ | 2083/5920 [33:04<3:01:59, 2.85s/it, tok/s=14156.9]
Inference: 35%|███▌ | 2084/5920 [33:04<2:09:38, 2.03s/it, tok/s=14156.9]
Inference: 35%|███▌ | 2084/5920 [33:04<2:09:38, 2.03s/it, tok/s=14163.0]
Inference: 35%|███▌ | 2085/5920 [33:04<1:42:49, 1.61s/it, tok/s=14163.0]
Inference: 35%|███▌ | 2085/5920 [33:04<1:42:49, 1.61s/it, tok/s=14167.2]
Inference: 35%|███▌ | 2086/5920 [33:05<1:22:50, 1.30s/it, tok/s=14167.2]
Inference: 35%|███▌ | 2086/5920 [33:05<1:22:50, 1.30s/it, tok/s=14174.1]
Inference: 35%|███▌ | 2087/5920 [33:06<1:07:44, 1.06s/it, tok/s=14174.1]
Inference: 35%|███▌ | 2087/5920 [33:06<1:07:44, 1.06s/it, tok/s=14185.7]
Inference: 35%|███▌ | 2088/5920 [33:08<1:37:32, 1.53s/it, tok/s=14185.7]
Inference: 35%|███▌ | 2088/5920 [33:08<1:37:32, 1.53s/it, tok/s=14181.5]
Inference: 35%|███▌ | 2089/5920 [33:09<1:20:45, 1.26s/it, tok/s=14181.5]
Inference: 35%|███▌ | 2089/5920 [33:09<1:20:45, 1.26s/it, tok/s=14187.9]
Inference: 35%|███▌ | 2090/5920 [33:09<1:06:36, 1.04s/it, tok/s=14187.9]
Inference: 35%|███▌ | 2090/5920 [33:09<1:06:36, 1.04s/it, tok/s=14199.2]
Inference: 35%|███▌ | 2091/5920 [33:10<54:01, 1.18it/s, tok/s=14199.2]
Inference: 35%|███▌ | 2091/5920 [33:10<54:01, 1.18it/s, tok/s=14207.7]
Inference: 35%|███▌ | 2092/5920 [33:11<58:58, 1.08it/s, tok/s=14207.7]
Inference: 35%|███▌ | 2092/5920 [33:11<58:58, 1.08it/s, tok/s=14206.8]
Inference: 35%|███▌ | 2093/5920 [33:12<1:00:35, 1.05it/s, tok/s=14206.8]
Inference: 35%|███▌ | 2093/5920 [33:12<1:00:35, 1.05it/s, tok/s=14208.2]
Inference: 35%|███▌ | 2094/5920 [33:13<1:05:09, 1.02s/it, tok/s=14208.2]
Inference: 35%|███▌ | 2094/5920 [33:13<1:05:09, 1.02s/it, tok/s=14214.1]
Inference: 35%|███▌ | 2095/5920 [33:15<1:16:01, 1.19s/it, tok/s=14214.1]
Inference: 35%|███▌ | 2095/5920 [33:15<1:16:01, 1.19s/it, tok/s=14212.0]
Inference: 35%|███▌ | 2096/5920 [33:16<1:24:20, 1.32s/it, tok/s=14212.0]
Inference: 35%|███▌ | 2096/5920 [33:16<1:24:20, 1.32s/it, tok/s=14204.8]
Inference: 35%|███▌ | 2097/5920 [33:17<1:10:24, 1.11s/it, tok/s=14204.8]
Inference: 35%|███▌ | 2097/5920 [33:17<1:10:24, 1.11s/it, tok/s=14203.5]
Inference: 35%|███▌ | 2098/5920 [33:19<1:28:02, 1.38s/it, tok/s=14203.5]
Inference: 35%|███▌ | 2098/5920 [33:19<1:28:02, 1.38s/it, tok/s=14192.3]
Inference: 35%|███▌ | 2099/5920 [33:21<1:37:37, 1.53s/it, tok/s=14192.3]
Inference: 35%|███▌ | 2099/5920 [33:21<1:37:37, 1.53s/it, tok/s=14193.9]
Inference: 35%|███▌ | 2100/5920 [33:22<1:26:00, 1.35s/it, tok/s=14193.9]
Inference: 35%|███▌ | 2100/5920 [33:22<1:26:00, 1.35s/it, tok/s=14194.1]
Inference: 35%|███▌ | 2101/5920 [33:24<1:50:52, 1.74s/it, tok/s=14194.1]
Inference: 35%|███▌ | 2101/5920 [33:24<1:50:52, 1.74s/it, tok/s=14179.4]
Inference: 36%|███▌ | 2102/5920 [33:27<2:00:39, 1.90s/it, tok/s=14179.4]
Inference: 36%|███▌ | 2102/5920 [33:27<2:00:39, 1.90s/it, tok/s=14166.4]
Inference: 36%|███▌ | 2103/5920 [33:31<2:54:13, 2.74s/it, tok/s=14166.4]
Inference: 36%|███▌ | 2103/5920 [33:31<2:54:13, 2.74s/it, tok/s=14148.2]
Inference: 36%|███▌ | 2104/5920 [33:31<2:54:10, 2.74s/it, tok/s=14155.7]
Inference: 36%|███▌ | 2105/5920 [33:34<2:19:07, 2.19s/it, tok/s=14155.7]
Inference: 36%|███▌ | 2105/5920 [33:34<2:19:07, 2.19s/it, tok/s=14142.9]
Inference: 36%|███▌ | 2106/5920 [33:38<2:38:20, 2.49s/it, tok/s=14142.9]
Inference: 36%|███▌ | 2106/5920 [33:38<2:38:20, 2.49s/it, tok/s=14120.9]
Inference: 36%|███▌ | 2107/5920 [33:39<2:08:59, 2.03s/it, tok/s=14120.9]
Inference: 36%|███▌ | 2107/5920 [33:39<2:08:59, 2.03s/it, tok/s=14123.5]
Inference: 36%|███▌ | 2108/5920 [33:40<2:02:10, 1.92s/it, tok/s=14123.5]
Inference: 36%|███▌ | 2108/5920 [33:40<2:02:10, 1.92s/it, tok/s=14123.0]
Inference: 36%|███▌ | 2109/5920 [33:40<1:32:23, 1.45s/it, tok/s=14123.0]
Inference: 36%|███▌ | 2109/5920 [33:40<1:32:23, 1.45s/it, tok/s=14130.5]
Inference: 36%|███▌ | 2110/5920 [33:41<1:09:24, 1.09s/it, tok/s=14130.5]
Inference: 36%|███▌ | 2110/5920 [33:41<1:09:24, 1.09s/it, tok/s=14144.0]
Inference: 36%|███▌ | 2111/5920 [33:41<58:16, 1.09it/s, tok/s=14144.0]
Inference: 36%|███▌ | 2111/5920 [33:41<58:16, 1.09it/s, tok/s=14144.4]
Inference: 36%|███▌ | 2112/5920 [33:42<56:58, 1.11it/s, tok/s=14144.4]
Inference: 36%|███▌ | 2112/5920 [33:42<56:58, 1.11it/s, tok/s=14141.0]
Inference: 36%|███▌ | 2113/5920 [33:46<1:56:14, 1.83s/it, tok/s=14141.0]
Inference: 36%|███▌ | 2113/5920 [33:46<1:56:14, 1.83s/it, tok/s=14116.3]
Inference: 36%|███▌ | 2114/5920 [33:48<2:05:01, 1.97s/it, tok/s=14116.3]
Inference: 36%|███▌ | 2114/5920 [33:48<2:05:01, 1.97s/it, tok/s=14103.4]
Inference: 36%|███▌ | 2115/5920 [33:50<2:02:50, 1.94s/it, tok/s=14103.4]
Inference: 36%|███▌ | 2115/5920 [33:50<2:02:50, 1.94s/it, tok/s=14093.3]
Inference: 36%|███▌ | 2116/5920 [33:53<2:16:19, 2.15s/it, tok/s=14093.3]
Inference: 36%|███▌ | 2116/5920 [33:53<2:16:19, 2.15s/it, tok/s=14084.3]
Inference: 36%|███▌ | 2117/5920 [33:53<1:40:17, 1.58s/it, tok/s=14084.3]
Inference: 36%|███▌ | 2117/5920 [33:53<1:40:17, 1.58s/it, tok/s=14097.3]
Inference: 36%|███▌ | 2118/5920 [33:56<2:10:51, 2.07s/it, tok/s=14097.3]
Inference: 36%|███▌ | 2118/5920 [33:56<2:10:51, 2.07s/it, tok/s=14090.0]
Inference: 36%|███▌ | 2119/5920 [33:59<2:19:07, 2.20s/it, tok/s=14090.0]
Inference: 36%|███▌ | 2119/5920 [33:59<2:19:07, 2.20s/it, tok/s=14085.5]
Inference: 36%|███▌ | 2120/5920 [34:00<2:00:56, 1.91s/it, tok/s=14085.5]
Inference: 36%|███▌ | 2120/5920 [34:00<2:00:56, 1.91s/it, tok/s=14090.3]
Inference: 36%|███▌ | 2121/5920 [34:02<2:07:39, 2.02s/it, tok/s=14090.3]
Inference: 36%|███▌ | 2121/5920 [34:02<2:07:39, 2.02s/it, tok/s=14083.2]
Inference: 36%|███▌ | 2122/5920 [34:05<2:13:09, 2.10s/it, tok/s=14083.2]
Inference: 36%|███▌ | 2122/5920 [34:05<2:13:09, 2.10s/it, tok/s=14074.9]
Inference: 36%|███▌ | 2123/5920 [34:05<1:40:43, 1.59s/it, tok/s=14074.9]
Inference: 36%|███▌ | 2123/5920 [34:05<1:40:43, 1.59s/it, tok/s=14074.6]
Inference: 36%|███▌ | 2124/5920 [34:05<1:14:18, 1.17s/it, tok/s=14074.6]
Inference: 36%|███▌ | 2124/5920 [34:05<1:14:18, 1.17s/it, tok/s=14085.3]
Inference: 36%|███▌ | 2125/5920 [34:07<1:21:43, 1.29s/it, tok/s=14085.3]
Inference: 36%|███▌ | 2125/5920 [34:07<1:21:43, 1.29s/it, tok/s=14089.4]
Inference: 36%|███▌ | 2126/5920 [34:07<1:21:42, 1.29s/it, tok/s=14091.6]
Inference: 36%|███▌ | 2127/5920 [34:07<54:45, 1.15it/s, tok/s=14091.6]
Inference: 36%|███▌ | 2127/5920 [34:07<54:45, 1.15it/s, tok/s=14100.7]
Inference: 36%|███▌ | 2128/5920 [34:11<1:37:47, 1.55s/it, tok/s=14100.7]
Inference: 36%|███▌ | 2128/5920 [34:11<1:37:47, 1.55s/it, tok/s=14088.6]
Inference: 36%|███▌ | 2129/5920 [34:12<1:29:08, 1.41s/it, tok/s=14088.6]
Inference: 36%|███▌ | 2129/5920 [34:12<1:29:08, 1.41s/it, tok/s=14085.0]
Inference: 36%|███▌ | 2130/5920 [34:15<1:46:26, 1.69s/it, tok/s=14085.0]
Inference: 36%|███▌ | 2130/5920 [34:15<1:46:26, 1.69s/it, tok/s=14077.5]
Inference: 36%|███▌ | 2131/5920 [34:16<1:49:44, 1.74s/it, tok/s=14077.5]
Inference: 36%|███▌ | 2131/5920 [34:16<1:49:44, 1.74s/it, tok/s=14079.5]
Inference: 36%|███▌ | 2132/5920 [34:17<1:26:49, 1.38s/it, tok/s=14079.5]
Inference: 36%|███▌ | 2132/5920 [34:17<1:26:49, 1.38s/it, tok/s=14080.1]
Inference: 36%|███▌ | 2133/5920 [34:19<1:49:25, 1.73s/it, tok/s=14080.1]
Inference: 36%|███▌ | 2133/5920 [34:19<1:49:25, 1.73s/it, tok/s=14071.0]
Inference: 36%|███▌ | 2134/5920 [34:24<2:32:51, 2.42s/it, tok/s=14071.0]
Inference: 36%|███▌ | 2134/5920 [34:24<2:32:51, 2.42s/it, tok/s=14051.7]
Inference: 36%|███▌ | 2135/5920 [34:24<2:32:48, 2.42s/it, tok/s=14063.7]
Inference: 36%|███▌ | 2136/5920 [34:24<1:35:41, 1.52s/it, tok/s=14063.7]
Inference: 36%|███▌ | 2136/5920 [34:24<1:35:41, 1.52s/it, tok/s=14062.0]
Inference: 36%|███▌ | 2137/5920 [34:29<2:19:54, 2.22s/it, tok/s=14062.0]
Inference: 36%|███▌ | 2137/5920 [34:29<2:19:54, 2.22s/it, tok/s=14046.8]
Inference: 36%|███▌ | 2138/5920 [34:31<2:24:47, 2.30s/it, tok/s=14046.8]
Inference: 36%|███▌ | 2138/5920 [34:31<2:24:47, 2.30s/it, tok/s=14040.0]
Inference: 36%|███▌ | 2139/5920 [34:32<1:51:10, 1.76s/it, tok/s=14040.0]
Inference: 36%|███▌ | 2139/5920 [34:32<1:51:10, 1.76s/it, tok/s=14045.8]
Inference: 36%|███▌ | 2140/5920 [34:32<1:28:10, 1.40s/it, tok/s=14045.8]
Inference: 36%|███▌ | 2140/5920 [34:32<1:28:10, 1.40s/it, tok/s=14056.6]
Inference: 36%|███▌ | 2141/5920 [34:32<1:28:08, 1.40s/it, tok/s=14070.9]
Inference: 36%|███▌ | 2142/5920 [34:32<51:40, 1.22it/s, tok/s=14070.9]
Inference: 36%|███▌ | 2142/5920 [34:32<51:40, 1.22it/s, tok/s=14074.2]
Inference: 36%|███▌ | 2143/5920 [34:32<40:45, 1.54it/s, tok/s=14074.2]
Inference: 36%|███▌ | 2143/5920 [34:32<40:45, 1.54it/s, tok/s=14081.8]
Inference: 36%|███▌ | 2144/5920 [34:35<1:05:08, 1.04s/it, tok/s=14081.8]
Inference: 36%|███▌ | 2144/5920 [34:35<1:05:08, 1.04s/it, tok/s=14080.2]
Inference: 36%|███▌ | 2145/5920 [34:35<53:01, 1.19it/s, tok/s=14080.2]
Inference: 36%|███▌ | 2145/5920 [34:35<53:01, 1.19it/s, tok/s=14086.8]
Inference: 36%|███▋ | 2146/5920 [34:42<2:39:46, 2.54s/it, tok/s=14086.8]
Inference: 36%|███▋ | 2146/5920 [34:42<2:39:46, 2.54s/it, tok/s=14048.1]
Inference: 36%|███▋ | 2147/5920 [34:43<2:08:22, 2.04s/it, tok/s=14048.1]
Inference: 36%|███▋ | 2147/5920 [34:43<2:08:22, 2.04s/it, tok/s=14053.5]
Inference: 36%|███▋ | 2148/5920 [34:44<2:01:37, 1.93s/it, tok/s=14053.5]
Inference: 36%|███▋ | 2148/5920 [34:44<2:01:37, 1.93s/it, tok/s=14054.8]
Inference: 36%|███▋ | 2149/5920 [34:53<4:01:22, 3.84s/it, tok/s=14054.8]
Inference: 36%|███▋ | 2149/5920 [34:53<4:01:22, 3.84s/it, tok/s=14005.9]
Inference: 36%|███▋ | 2150/5920 [34:57<4:12:01, 4.01s/it, tok/s=14005.9]
Inference: 36%|███▋ | 2150/5920 [34:57<4:12:01, 4.01s/it, tok/s=13989.5]
Inference: 36%|███▋ | 2151/5920 [34:58<3:16:42, 3.13s/it, tok/s=13989.5]
Inference: 36%|███▋ | 2151/5920 [34:58<3:16:42, 3.13s/it, tok/s=13996.9]
Inference: 36%|███▋ | 2152/5920 [34:59<2:39:34, 2.54s/it, tok/s=13996.9]
Inference: 36%|███▋ | 2152/5920 [34:59<2:39:34, 2.54s/it, tok/s=13997.5]
Inference: 36%|███▋ | 2153/5920 [35:01<2:16:33, 2.18s/it, tok/s=13997.5]
Inference: 36%|███▋ | 2153/5920 [35:01<2:16:33, 2.18s/it, tok/s=14003.2]
Inference: 36%|███▋ | 2154/5920 [35:02<1:58:01, 1.88s/it, tok/s=14003.2]
Inference: 36%|███▋ | 2154/5920 [35:02<1:58:01, 1.88s/it, tok/s=13999.3]
Inference: 36%|███▋ | 2155/5920 [35:06<2:41:23, 2.57s/it, tok/s=13999.3]
Inference: 36%|███▋ | 2155/5920 [35:06<2:41:23, 2.57s/it, tok/s=13979.8]
Inference: 36%|███▋ | 2156/5920 [35:07<2:18:57, 2.22s/it, tok/s=13979.8]
Inference: 36%|███▋ | 2156/5920 [35:07<2:18:57, 2.22s/it, tok/s=13973.3]
Inference: 36%|███▋ | 2157/5920 [35:08<1:55:58, 1.85s/it, tok/s=13973.3]
Inference: 36%|███▋ | 2157/5920 [35:08<1:55:58, 1.85s/it, tok/s=13975.1]
Inference: 36%|███▋ | 2158/5920 [35:09<1:36:10, 1.53s/it, tok/s=13975.1]
Inference: 36%|███▋ | 2158/5920 [35:09<1:36:10, 1.53s/it, tok/s=13984.2]
Inference: 36%|███▋ | 2159/5920 [35:11<1:41:33, 1.62s/it, tok/s=13984.2]
Inference: 36%|███▋ | 2159/5920 [35:11<1:41:33, 1.62s/it, tok/s=13986.6]
Inference: 36%|███▋ | 2160/5920 [35:12<1:28:23, 1.41s/it, tok/s=13986.6]
Inference: 36%|███▋ | 2160/5920 [35:12<1:28:23, 1.41s/it, tok/s=13982.9]
Inference: 37%|███▋ | 2161/5920 [35:15<2:02:28, 1.95s/it, tok/s=13982.9]
Inference: 37%|███▋ | 2161/5920 [35:15<2:02:28, 1.95s/it, tok/s=13975.1]
Inference: 37%|███▋ | 2162/5920 [35:17<2:01:41, 1.94s/it, tok/s=13975.1]
Inference: 37%|███▋ | 2162/5920 [35:17<2:01:41, 1.94s/it, tok/s=13976.6]
Inference: 37%|███▋ | 2163/5920 [35:18<1:35:50, 1.53s/it, tok/s=13976.6]
Inference: 37%|███▋ | 2163/5920 [35:18<1:35:50, 1.53s/it, tok/s=13987.2]
Inference: 37%|███▋ | 2164/5920 [35:18<1:16:08, 1.22s/it, tok/s=13987.2]
Inference: 37%|███▋ | 2164/5920 [35:18<1:16:08, 1.22s/it, tok/s=13994.9]
Inference: 37%|███▋ | 2165/5920 [35:22<2:09:52, 2.08s/it, tok/s=13994.9]
Inference: 37%|███▋ | 2165/5920 [35:22<2:09:52, 2.08s/it, tok/s=13978.6]
Inference: 37%|███▋ | 2166/5920 [35:25<2:20:21, 2.24s/it, tok/s=13978.6]
Inference: 37%|███▋ | 2166/5920 [35:25<2:20:21, 2.24s/it, tok/s=13965.0]
Inference: 37%|███▋ | 2167/5920 [35:25<2:20:19, 2.24s/it, tok/s=13978.7]
Inference: 37%|███▋ | 2168/5920 [35:26<1:30:52, 1.45s/it, tok/s=13978.7]
Inference: 37%|███▋ | 2168/5920 [35:26<1:30:52, 1.45s/it, tok/s=13986.6]
Inference: 37%|███▋ | 2169/5920 [35:27<1:23:29, 1.34s/it, tok/s=13986.6]
Inference: 37%|███▋ | 2169/5920 [35:27<1:23:29, 1.34s/it, tok/s=13983.2]
Inference: 37%|███▋ | 2170/5920 [35:28<1:21:22, 1.30s/it, tok/s=13983.2]
Inference: 37%|███▋ | 2170/5920 [35:28<1:21:22, 1.30s/it, tok/s=13989.0]
Inference: 37%|███▋ | 2171/5920 [35:30<1:32:19, 1.48s/it, tok/s=13989.0]
Inference: 37%|███▋ | 2171/5920 [35:30<1:32:19, 1.48s/it, tok/s=13990.3]
Inference: 37%|███▋ | 2172/5920 [35:30<1:11:55, 1.15s/it, tok/s=13990.3]
Inference: 37%|███▋ | 2172/5920 [35:30<1:11:55, 1.15s/it, tok/s=13999.4]
Inference: 37%|███▋ | 2173/5920 [35:31<1:00:30, 1.03it/s, tok/s=13999.4]
Inference: 37%|███▋ | 2173/5920 [35:31<1:00:30, 1.03it/s, tok/s=14007.0]
Inference: 37%|███▋ | 2174/5920 [35:34<1:38:09, 1.57s/it, tok/s=14007.0]
Inference: 37%|███▋ | 2174/5920 [35:34<1:38:09, 1.57s/it, tok/s=13991.2]
Inference: 37%|███▋ | 2175/5920 [35:38<2:22:26, 2.28s/it, tok/s=13991.2]
Inference: 37%|███▋ | 2175/5920 [35:38<2:22:26, 2.28s/it, tok/s=13974.1]
Inference: 37%|███▋ | 2176/5920 [35:39<1:51:39, 1.79s/it, tok/s=13974.1]
Inference: 37%|███▋ | 2176/5920 [35:39<1:51:39, 1.79s/it, tok/s=13978.3]
Inference: 37%|███▋ | 2177/5920 [35:42<2:12:56, 2.13s/it, tok/s=13978.3]
Inference: 37%|███▋ | 2177/5920 [35:42<2:12:56, 2.13s/it, tok/s=13973.3]
Inference: 37%|███▋ | 2178/5920 [35:45<2:39:31, 2.56s/it, tok/s=13973.3]
Inference: 37%|███▋ | 2178/5920 [35:45<2:39:31, 2.56s/it, tok/s=13964.2]
Inference: 37%|███▋ | 2179/5920 [35:46<2:12:32, 2.13s/it, tok/s=13964.2]
Inference: 37%|███▋ | 2179/5920 [35:46<2:12:32, 2.13s/it, tok/s=13967.4]
Inference: 37%|███▋ | 2180/5920 [35:46<1:35:54, 1.54s/it, tok/s=13967.4]
Inference: 37%|███▋ | 2180/5920 [35:46<1:35:54, 1.54s/it, tok/s=13975.5]
Inference: 37%|███▋ | 2181/5920 [35:47<1:22:30, 1.32s/it, tok/s=13975.5]
Inference: 37%|███▋ | 2181/5920 [35:47<1:22:30, 1.32s/it, tok/s=13982.0]
Inference: 37%|███▋ | 2182/5920 [35:51<1:59:48, 1.92s/it, tok/s=13982.0]
Inference: 37%|███▋ | 2182/5920 [35:51<1:59:48, 1.92s/it, tok/s=13974.5]
Inference: 37%|███▋ | 2183/5920 [35:51<1:34:58, 1.52s/it, tok/s=13974.5]
Inference: 37%|███▋ | 2183/5920 [35:51<1:34:58, 1.52s/it, tok/s=13978.3]
Inference: 37%|███▋ | 2184/5920 [35:51<1:10:39, 1.13s/it, tok/s=13978.3]
Inference: 37%|███▋ | 2184/5920 [35:51<1:10:39, 1.13s/it, tok/s=13986.7]
Inference: 37%|███▋ | 2185/5920 [35:51<1:10:38, 1.13s/it, tok/s=14000.3]
Inference: 37%|███▋ | 2186/5920 [35:52<50:56, 1.22it/s, tok/s=14000.3]
Inference: 37%|███▋ | 2186/5920 [35:52<50:56, 1.22it/s, tok/s=14006.5]
Inference: 37%|███▋ | 2187/5920 [35:54<59:00, 1.05it/s, tok/s=14006.5]
Inference: 37%|███▋ | 2187/5920 [35:54<59:00, 1.05it/s, tok/s=14011.9]
Inference: 37%|███▋ | 2188/5920 [35:56<1:25:18, 1.37s/it, tok/s=14011.9]
Inference: 37%|███▋ | 2188/5920 [35:56<1:25:18, 1.37s/it, tok/s=14005.7]
Inference: 37%|███▋ | 2189/5920 [35:56<1:25:17, 1.37s/it, tok/s=14016.4]
Inference: 37%|███▋ | 2190/5920 [35:57<53:49, 1.15it/s, tok/s=14016.4]
Inference: 37%|███▋ | 2190/5920 [35:57<53:49, 1.15it/s, tok/s=14017.3]
Inference: 37%|███▋ | 2191/5920 [35:57<52:18, 1.19it/s, tok/s=14017.3]
Inference: 37%|███▋ | 2191/5920 [35:57<52:18, 1.19it/s, tok/s=14026.4]
Inference: 37%|███▋ | 2192/5920 [36:02<1:46:57, 1.72s/it, tok/s=14026.4]
Inference: 37%|███▋ | 2192/5920 [36:02<1:46:57, 1.72s/it, tok/s=14009.9]
Inference: 37%|███▋ | 2193/5920 [36:04<2:03:40, 1.99s/it, tok/s=14009.9]
Inference: 37%|███▋ | 2193/5920 [36:04<2:03:40, 1.99s/it, tok/s=13995.3]
Inference: 37%|███▋ | 2194/5920 [36:08<2:27:27, 2.37s/it, tok/s=13995.3]
Inference: 37%|███▋ | 2194/5920 [36:08<2:27:27, 2.37s/it, tok/s=13984.8]
Inference: 37%|███▋ | 2195/5920 [36:08<1:50:06, 1.77s/it, tok/s=13984.8]
Inference: 37%|███▋ | 2195/5920 [36:08<1:50:06, 1.77s/it, tok/s=13997.3]
Inference: 37%|███▋ | 2196/5920 [36:09<1:37:59, 1.58s/it, tok/s=13997.3]
Inference: 37%|███▋ | 2196/5920 [36:09<1:37:59, 1.58s/it, tok/s=13994.5]
Inference: 37%|███▋ | 2197/5920 [36:10<1:28:21, 1.42s/it, tok/s=13994.5]
Inference: 37%|███▋ | 2197/5920 [36:10<1:28:21, 1.42s/it, tok/s=13992.4]
Inference: 37%|███▋ | 2198/5920 [36:10<1:07:37, 1.09s/it, tok/s=13992.4]
Inference: 37%|███▋ | 2198/5920 [36:10<1:07:37, 1.09s/it, tok/s=13994.9]
Inference: 37%|███▋ | 2199/5920 [36:11<50:51, 1.22it/s, tok/s=13994.9]
Inference: 37%|███▋ | 2199/5920 [36:11<50:51, 1.22it/s, tok/s=14007.6]
Inference: 37%|███▋ | 2200/5920 [36:11<50:25, 1.23it/s, tok/s=14007.6]
Inference: 37%|███▋ | 2200/5920 [36:11<50:25, 1.23it/s, tok/s=14014.4]
Inference: 37%|███▋ | 2201/5920 [36:12<54:34, 1.14it/s, tok/s=14014.4]
Inference: 37%|███▋ | 2201/5920 [36:12<54:34, 1.14it/s, tok/s=14013.1]
Inference: 37%|███▋ | 2202/5920 [36:18<2:14:12, 2.17s/it, tok/s=14013.1]
Inference: 37%|███▋ | 2202/5920 [36:18<2:14:12, 2.17s/it, tok/s=13993.7]
Inference: 37%|███▋ | 2203/5920 [36:18<1:44:33, 1.69s/it, tok/s=13993.7]
Inference: 37%|███▋ | 2203/5920 [36:18<1:44:33, 1.69s/it, tok/s=13995.1]
Inference: 37%|███▋ | 2204/5920 [36:18<1:44:31, 1.69s/it, tok/s=13997.8]
Inference: 37%|███▋ | 2205/5920 [36:19<1:03:19, 1.02s/it, tok/s=13997.8]
Inference: 37%|███▋ | 2205/5920 [36:19<1:03:19, 1.02s/it, tok/s=13998.9]
Inference: 37%|███▋ | 2206/5920 [36:20<1:00:43, 1.02it/s, tok/s=13998.9]
Inference: 37%|███▋ | 2206/5920 [36:20<1:00:43, 1.02it/s, tok/s=14007.3]
Inference: 37%|███▋ | 2207/5920 [36:21<1:10:43, 1.14s/it, tok/s=14007.3]
Inference: 37%|███▋ | 2207/5920 [36:21<1:10:43, 1.14s/it, tok/s=14010.8]
Inference: 37%|███▋ | 2208/5920 [36:25<1:56:55, 1.89s/it, tok/s=14010.8]
Inference: 37%|███▋ | 2208/5920 [36:25<1:56:55, 1.89s/it, tok/s=13994.4]
Inference: 37%|███▋ | 2209/5920 [36:26<1:41:03, 1.63s/it, tok/s=13994.4]
Inference: 37%|███▋ | 2209/5920 [36:26<1:41:03, 1.63s/it, tok/s=14001.9]
Inference: 37%|███▋ | 2210/5920 [36:28<1:44:55, 1.70s/it, tok/s=14001.9]
Inference: 37%|███▋ | 2210/5920 [36:28<1:44:55, 1.70s/it, tok/s=13998.7]
Inference: 37%|███▋ | 2211/5920 [36:33<2:45:52, 2.68s/it, tok/s=13998.7]
Inference: 37%|███▋ | 2211/5920 [36:33<2:45:52, 2.68s/it, tok/s=13969.7]
Inference: 37%|███▋ | 2212/5920 [36:33<2:02:37, 1.98s/it, tok/s=13969.7]
Inference: 37%|███▋ | 2212/5920 [36:33<2:02:37, 1.98s/it, tok/s=13981.7]
Inference: 37%|███▋ | 2213/5920 [36:34<1:46:11, 1.72s/it, tok/s=13981.7]
Inference: 37%|███▋ | 2213/5920 [36:34<1:46:11, 1.72s/it, tok/s=13979.3]
Inference: 37%|███▋ | 2214/5920 [36:38<2:18:24, 2.24s/it, tok/s=13979.3]
Inference: 37%|███▋ | 2214/5920 [36:38<2:18:24, 2.24s/it, tok/s=13965.7]
Inference: 37%|███▋ | 2215/5920 [36:41<2:29:12, 2.42s/it, tok/s=13965.7]
Inference: 37%|███▋ | 2215/5920 [36:41<2:29:12, 2.42s/it, tok/s=13961.5]
Inference: 37%|███▋ | 2216/5920 [36:44<2:51:23, 2.78s/it, tok/s=13961.5]
Inference: 37%|███▋ | 2216/5920 [36:44<2:51:23, 2.78s/it, tok/s=13947.5]
Inference: 37%|███▋ | 2217/5920 [36:45<2:10:51, 2.12s/it, tok/s=13947.5]
Inference: 37%|███▋ | 2217/5920 [36:45<2:10:51, 2.12s/it, tok/s=13957.4]
Inference: 37%|███▋ | 2218/5920 [36:46<1:53:20, 1.84s/it, tok/s=13957.4]
Inference: 37%|███▋ | 2218/5920 [36:46<1:53:20, 1.84s/it, tok/s=13954.0]
Inference: 37%|███▋ | 2219/5920 [36:46<1:21:45, 1.33s/it, tok/s=13954.0]
Inference: 37%|███▋ | 2219/5920 [36:46<1:21:45, 1.33s/it, tok/s=13967.0]
Inference: 38%|███▊ | 2220/5920 [36:48<1:38:55, 1.60s/it, tok/s=13967.0]
Inference: 38%|███▊ | 2220/5920 [36:48<1:38:55, 1.60s/it, tok/s=13956.7]
Inference: 38%|███▊ | 2221/5920 [36:48<1:38:53, 1.60s/it, tok/s=13969.5]
Inference: 38%|███▊ | 2222/5920 [36:50<1:11:20, 1.16s/it, tok/s=13969.5]
Inference: 38%|███▊ | 2222/5920 [36:50<1:11:20, 1.16s/it, tok/s=13975.5]
Inference: 38%|███▊ | 2223/5920 [36:51<1:09:53, 1.13s/it, tok/s=13975.5]
Inference: 38%|███▊ | 2223/5920 [36:51<1:09:53, 1.13s/it, tok/s=13981.4]
Inference: 38%|███▊ | 2224/5920 [36:52<1:08:26, 1.11s/it, tok/s=13981.4]
Inference: 38%|███▊ | 2224/5920 [36:52<1:08:26, 1.11s/it, tok/s=13985.9]
Inference: 38%|███▊ | 2225/5920 [36:53<1:14:13, 1.21s/it, tok/s=13985.9]
Inference: 38%|███▊ | 2225/5920 [36:53<1:14:13, 1.21s/it, tok/s=13986.6]
Inference: 38%|███▊ | 2226/5920 [36:53<56:02, 1.10it/s, tok/s=13986.6]
Inference: 38%|███▊ | 2226/5920 [36:53<56:02, 1.10it/s, tok/s=13994.4]
Inference: 38%|███▊ | 2227/5920 [36:53<56:01, 1.10it/s, tok/s=13999.1]
Inference: 38%|███▊ | 2228/5920 [36:54<34:27, 1.79it/s, tok/s=13999.1]
Inference: 38%|███▊ | 2228/5920 [36:54<34:27, 1.79it/s, tok/s=14006.9]
Inference: 38%|███▊ | 2229/5920 [36:54<28:30, 2.16it/s, tok/s=14006.9]
Inference: 38%|███▊ | 2229/5920 [36:54<28:30, 2.16it/s, tok/s=14019.6]
Inference: 38%|███▊ | 2230/5920 [36:57<1:05:02, 1.06s/it, tok/s=14019.6]
Inference: 38%|███▊ | 2230/5920 [36:57<1:05:02, 1.06s/it, tok/s=14006.9]
Inference: 38%|███▊ | 2231/5920 [36:59<1:21:18, 1.32s/it, tok/s=14006.9]
Inference: 38%|███▊ | 2231/5920 [36:59<1:21:18, 1.32s/it, tok/s=14007.5]
Inference: 38%|███▊ | 2232/5920 [37:00<1:29:28, 1.46s/it, tok/s=14007.5]
Inference: 38%|███▊ | 2232/5920 [37:00<1:29:28, 1.46s/it, tok/s=14005.2]
Inference: 38%|███▊ | 2233/5920 [37:05<2:23:17, 2.33s/it, tok/s=14005.2]
Inference: 38%|███▊ | 2233/5920 [37:05<2:23:17, 2.33s/it, tok/s=13980.4]
Inference: 38%|███▊ | 2234/5920 [37:07<2:24:35, 2.35s/it, tok/s=13980.4]
Inference: 38%|███▊ | 2234/5920 [37:07<2:24:35, 2.35s/it, tok/s=13973.2]
Inference: 38%|███▊ | 2235/5920 [37:08<1:53:32, 1.85s/it, tok/s=13973.2]
Inference: 38%|███▊ | 2235/5920 [37:08<1:53:32, 1.85s/it, tok/s=13983.0]
Inference: 38%|███▊ | 2236/5920 [37:09<1:32:08, 1.50s/it, tok/s=13983.0]
Inference: 38%|███▊ | 2236/5920 [37:09<1:32:08, 1.50s/it, tok/s=13988.3]
Inference: 38%|███▊ | 2237/5920 [37:12<2:09:01, 2.10s/it, tok/s=13988.3]
Inference: 38%|███▊ | 2237/5920 [37:12<2:09:01, 2.10s/it, tok/s=13971.7]
Inference: 38%|███▊ | 2238/5920 [37:16<2:31:40, 2.47s/it, tok/s=13971.7]
Inference: 38%|███▊ | 2238/5920 [37:16<2:31:40, 2.47s/it, tok/s=13956.0]
Inference: 38%|███▊ | 2239/5920 [37:17<2:16:39, 2.23s/it, tok/s=13956.0]
Inference: 38%|███▊ | 2239/5920 [37:17<2:16:39, 2.23s/it, tok/s=13950.0]
Inference: 38%|███▊ | 2240/5920 [37:20<2:32:45, 2.49s/it, tok/s=13950.0]
Inference: 38%|███▊ | 2240/5920 [37:20<2:32:45, 2.49s/it, tok/s=13936.0]
Inference: 38%|███▊ | 2241/5920 [37:22<2:20:34, 2.29s/it, tok/s=13936.0]
Inference: 38%|███▊ | 2241/5920 [37:22<2:20:34, 2.29s/it, tok/s=13934.1]
Inference: 38%|███▊ | 2242/5920 [37:23<2:02:10, 1.99s/it, tok/s=13934.1]
Inference: 38%|███▊ | 2242/5920 [37:23<2:02:10, 1.99s/it, tok/s=13931.1]
Inference: 38%|███▊ | 2243/5920 [37:31<3:48:22, 3.73s/it, tok/s=13931.1]
Inference: 38%|███▊ | 2243/5920 [37:31<3:48:22, 3.73s/it, tok/s=13886.7]
Inference: 38%|███▊ | 2244/5920 [37:31<3:48:18, 3.73s/it, tok/s=13890.3]
Inference: 38%|███▊ | 2245/5920 [37:33<2:20:56, 2.30s/it, tok/s=13890.3]
Inference: 38%|███▊ | 2245/5920 [37:33<2:20:56, 2.30s/it, tok/s=13886.8]
Inference: 38%|███▊ | 2246/5920 [37:35<2:20:46, 2.30s/it, tok/s=13886.8]
Inference: 38%|███▊ | 2246/5920 [37:35<2:20:46, 2.30s/it, tok/s=13881.6]
Inference: 38%|███▊ | 2247/5920 [37:36<2:01:21, 1.98s/it, tok/s=13881.6]
Inference: 38%|███▊ | 2247/5920 [37:36<2:01:21, 1.98s/it, tok/s=13885.4]
Inference: 38%|███▊ | 2248/5920 [37:36<1:33:52, 1.53s/it, tok/s=13885.4]
Inference: 38%|███▊ | 2248/5920 [37:36<1:33:52, 1.53s/it, tok/s=13895.0]
Inference: 38%|███▊ | 2249/5920 [37:38<1:46:10, 1.74s/it, tok/s=13895.0]
Inference: 38%|███▊ | 2249/5920 [37:38<1:46:10, 1.74s/it, tok/s=13891.2]
Inference: 38%|███▊ | 2250/5920 [37:42<2:19:37, 2.28s/it, tok/s=13891.2]
Inference: 38%|███▊ | 2250/5920 [37:42<2:19:37, 2.28s/it, tok/s=13882.1]
Inference: 38%|███▊ | 2251/5920 [37:44<2:20:10, 2.29s/it, tok/s=13882.1]
Inference: 38%|███▊ | 2251/5920 [37:44<2:20:10, 2.29s/it, tok/s=13872.3]
Inference: 38%|███▊ | 2252/5920 [37:45<1:47:13, 1.75s/it, tok/s=13872.3]
Inference: 38%|███▊ | 2252/5920 [37:45<1:47:13, 1.75s/it, tok/s=13880.1]
Inference: 38%|███▊ | 2253/5920 [37:48<2:08:57, 2.11s/it, tok/s=13880.1]
Inference: 38%|███▊ | 2253/5920 [37:48<2:08:57, 2.11s/it, tok/s=13870.9]
Inference: 38%|███▊ | 2254/5920 [37:48<1:41:01, 1.65s/it, tok/s=13870.9]
Inference: 38%|███▊ | 2254/5920 [37:48<1:41:01, 1.65s/it, tok/s=13876.2]
Inference: 38%|███▊ | 2255/5920 [37:51<2:05:53, 2.06s/it, tok/s=13876.2]
Inference: 38%|███▊ | 2255/5920 [37:51<2:05:53, 2.06s/it, tok/s=13866.4]
Inference: 38%|███▊ | 2256/5920 [37:57<3:01:11, 2.97s/it, tok/s=13866.4]
Inference: 38%|███▊ | 2256/5920 [37:57<3:01:11, 2.97s/it, tok/s=13844.1]
Inference: 38%|███▊ | 2257/5920 [37:57<2:10:30, 2.14s/it, tok/s=13844.1]
Inference: 38%|███▊ | 2257/5920 [37:57<2:10:30, 2.14s/it, tok/s=13852.1]
Inference: 38%|███▊ | 2258/5920 [37:58<1:49:17, 1.79s/it, tok/s=13852.1]
Inference: 38%|███▊ | 2258/5920 [37:58<1:49:17, 1.79s/it, tok/s=13851.7]
Inference: 38%|███▊ | 2259/5920 [38:03<2:54:18, 2.86s/it, tok/s=13851.7]
Inference: 38%|███▊ | 2259/5920 [38:03<2:54:18, 2.86s/it, tok/s=13830.1]
Inference: 38%|███▊ | 2260/5920 [38:04<2:18:03, 2.26s/it, tok/s=13830.1]
Inference: 38%|███▊ | 2260/5920 [38:04<2:18:03, 2.26s/it, tok/s=13837.2]
Inference: 38%|███▊ | 2261/5920 [38:09<3:13:09, 3.17s/it, tok/s=13837.2]
Inference: 38%|███▊ | 2261/5920 [38:09<3:13:09, 3.17s/it, tok/s=13813.2]
Inference: 38%|███▊ | 2262/5920 [38:10<2:27:31, 2.42s/it, tok/s=13813.2]
Inference: 38%|███▊ | 2262/5920 [38:10<2:27:31, 2.42s/it, tok/s=13818.4]
Inference: 38%|███▊ | 2263/5920 [38:10<2:27:29, 2.42s/it, tok/s=13826.5]
Inference: 38%|███▊ | 2264/5920 [38:11<1:36:24, 1.58s/it, tok/s=13826.5]
Inference: 38%|███▊ | 2264/5920 [38:11<1:36:24, 1.58s/it, tok/s=13829.9]
Inference: 38%|███▊ | 2265/5920 [38:11<1:36:22, 1.58s/it, tok/s=13842.3]
Inference: 38%|███▊ | 2266/5920 [38:14<1:29:08, 1.46s/it, tok/s=13842.3]
Inference: 38%|███▊ | 2266/5920 [38:14<1:29:08, 1.46s/it, tok/s=13831.6]
Inference: 38%|███▊ | 2267/5920 [38:16<1:36:45, 1.59s/it, tok/s=13831.6]
Inference: 38%|███▊ | 2267/5920 [38:16<1:36:45, 1.59s/it, tok/s=13828.7]
Inference: 38%|███▊ | 2268/5920 [38:16<1:24:24, 1.39s/it, tok/s=13828.7]
Inference: 38%|███▊ | 2268/5920 [38:16<1:24:24, 1.39s/it, tok/s=13828.5]
Inference: 38%|███▊ | 2269/5920 [38:18<1:20:22, 1.32s/it, tok/s=13828.5]
Inference: 38%|███▊ | 2269/5920 [38:18<1:20:22, 1.32s/it, tok/s=13826.5]
Inference: 38%|███▊ | 2270/5920 [38:18<1:06:22, 1.09s/it, tok/s=13826.5]
Inference: 38%|███▊ | 2270/5920 [38:18<1:06:22, 1.09s/it, tok/s=13836.0]
Inference: 38%|███▊ | 2271/5920 [38:20<1:26:38, 1.42s/it, tok/s=13836.0]
Inference: 38%|███▊ | 2271/5920 [38:20<1:26:38, 1.42s/it, tok/s=13834.5]
Inference: 38%|███▊ | 2272/5920 [38:21<1:15:25, 1.24s/it, tok/s=13834.5]
Inference: 38%|███▊ | 2272/5920 [38:21<1:15:25, 1.24s/it, tok/s=13834.4]
Inference: 38%|███▊ | 2273/5920 [38:23<1:26:45, 1.43s/it, tok/s=13834.4]
Inference: 38%|███▊ | 2273/5920 [38:23<1:26:45, 1.43s/it, tok/s=13831.4]
Inference: 38%|███▊ | 2274/5920 [38:25<1:34:00, 1.55s/it, tok/s=13831.4]
Inference: 38%|███▊ | 2274/5920 [38:25<1:34:00, 1.55s/it, tok/s=13824.5]
Inference: 38%|███▊ | 2275/5920 [38:25<1:11:47, 1.18s/it, tok/s=13824.5]
Inference: 38%|███▊ | 2275/5920 [38:25<1:11:47, 1.18s/it, tok/s=13826.9]
Inference: 38%|███▊ | 2276/5920 [38:25<1:11:46, 1.18s/it, tok/s=13831.3]
Inference: 38%|███▊ | 2277/5920 [38:25<1:11:44, 1.18s/it, tok/s=13839.4]
Inference: 38%|███▊ | 2278/5920 [38:32<1:48:14, 1.78s/it, tok/s=13839.4]
Inference: 38%|███▊ | 2278/5920 [38:32<1:48:14, 1.78s/it, tok/s=13810.1]
Inference: 38%|███▊ | 2279/5920 [38:32<1:31:05, 1.50s/it, tok/s=13810.1]
Inference: 38%|███▊ | 2279/5920 [38:32<1:31:05, 1.50s/it, tok/s=13814.9]
Inference: 39%|███▊ | 2280/5920 [38:33<1:23:46, 1.38s/it, tok/s=13814.9]
Inference: 39%|███▊ | 2280/5920 [38:33<1:23:46, 1.38s/it, tok/s=13821.7]
Inference: 39%|███▊ | 2281/5920 [38:35<1:21:29, 1.34s/it, tok/s=13821.7]
Inference: 39%|███▊ | 2281/5920 [38:35<1:21:29, 1.34s/it, tok/s=13820.9]
Inference: 39%|███▊ | 2282/5920 [38:37<1:40:15, 1.65s/it, tok/s=13820.9]
Inference: 39%|███▊ | 2282/5920 [38:37<1:40:15, 1.65s/it, tok/s=13815.8]
Inference: 39%|███▊ | 2283/5920 [38:38<1:22:19, 1.36s/it, tok/s=13815.8]
Inference: 39%|███▊ | 2283/5920 [38:38<1:22:19, 1.36s/it, tok/s=13823.6]
Inference: 39%|███▊ | 2284/5920 [38:39<1:28:05, 1.45s/it, tok/s=13823.6]
Inference: 39%|███▊ | 2284/5920 [38:39<1:28:05, 1.45s/it, tok/s=13822.7]
Inference: 39%|███▊ | 2285/5920 [38:40<1:08:39, 1.13s/it, tok/s=13822.7]
Inference: 39%|███▊ | 2285/5920 [38:40<1:08:39, 1.13s/it, tok/s=13831.2]
Inference: 39%|███▊ | 2286/5920 [38:40<1:08:38, 1.13s/it, tok/s=13840.6]
Inference: 39%|███▊ | 2287/5920 [38:43<1:18:26, 1.30s/it, tok/s=13840.6]
Inference: 39%|███▊ | 2287/5920 [38:43<1:18:26, 1.30s/it, tok/s=13834.4]
Inference: 39%|███▊ | 2288/5920 [38:43<1:07:55, 1.12s/it, tok/s=13834.4]
Inference: 39%|███▊ | 2288/5920 [38:43<1:07:55, 1.12s/it, tok/s=13835.4]
Inference: 39%|███▊ | 2289/5920 [38:43<52:10, 1.16it/s, tok/s=13835.4]
Inference: 39%|███▊ | 2289/5920 [38:43<52:10, 1.16it/s, tok/s=13842.4]
Inference: 39%|███▊ | 2290/5920 [38:44<41:00, 1.48it/s, tok/s=13842.4]
Inference: 39%|███▊ | 2290/5920 [38:44<41:00, 1.48it/s, tok/s=13844.8]
Inference: 39%|███▊ | 2291/5920 [38:44<40:20, 1.50it/s, tok/s=13844.8]
Inference: 39%|███▊ | 2291/5920 [38:44<40:20, 1.50it/s, tok/s=13844.7]
Inference: 39%|███▊ | 2292/5920 [38:45<44:58, 1.34it/s, tok/s=13844.7]
Inference: 39%|███▊ | 2292/5920 [38:45<44:58, 1.34it/s, tok/s=13848.6]
Inference: 39%|███▊ | 2293/5920 [38:47<1:07:21, 1.11s/it, tok/s=13848.6]
Inference: 39%|███▊ | 2293/5920 [38:47<1:07:21, 1.11s/it, tok/s=13846.0]
Inference: 39%|███▉ | 2294/5920 [38:49<1:18:45, 1.30s/it, tok/s=13846.0]
Inference: 39%|███▉ | 2294/5920 [38:49<1:18:45, 1.30s/it, tok/s=13841.4]
Inference: 39%|███▉ | 2295/5920 [38:55<2:43:34, 2.71s/it, tok/s=13841.4]
Inference: 39%|███▉ | 2295/5920 [38:55<2:43:34, 2.71s/it, tok/s=13811.0]
Inference: 39%|███▉ | 2296/5920 [38:56<2:21:14, 2.34s/it, tok/s=13811.0]
Inference: 39%|███▉ | 2296/5920 [38:56<2:21:14, 2.34s/it, tok/s=13814.2]
Inference: 39%|███▉ | 2297/5920 [38:57<1:56:53, 1.94s/it, tok/s=13814.2]
Inference: 39%|███▉ | 2297/5920 [38:57<1:56:53, 1.94s/it, tok/s=13821.4]
Inference: 39%|███▉ | 2298/5920 [38:57<1:56:51, 1.94s/it, tok/s=13834.4]
Inference: 39%|███▉ | 2299/5920 [38:58<1:56:49, 1.94s/it, tok/s=13844.8]
Inference: 39%|███▉ | 2300/5920 [39:00<1:15:35, 1.25s/it, tok/s=13844.8]
Inference: 39%|███▉ | 2300/5920 [39:00<1:15:35, 1.25s/it, tok/s=13841.6]
Inference: 39%|███▉ | 2301/5920 [39:01<1:16:36, 1.27s/it, tok/s=13841.6]
Inference: 39%|███▉ | 2301/5920 [39:01<1:16:36, 1.27s/it, tok/s=13838.5]
Inference: 39%|███▉ | 2302/5920 [39:05<1:54:47, 1.90s/it, tok/s=13838.5]
Inference: 39%|███▉ | 2302/5920 [39:05<1:54:47, 1.90s/it, tok/s=13824.3]
Inference: 39%|███▉ | 2303/5920 [39:05<1:28:51, 1.47s/it, tok/s=13824.3]
Inference: 39%|███▉ | 2303/5920 [39:05<1:28:51, 1.47s/it, tok/s=13836.1]
Inference: 39%|███▉ | 2304/5920 [39:05<1:28:50, 1.47s/it, tok/s=13849.1]
Inference: 39%|███▉ | 2305/5920 [39:07<1:19:04, 1.31s/it, tok/s=13849.1]
Inference: 39%|███▉ | 2305/5920 [39:07<1:19:04, 1.31s/it, tok/s=13841.3]
Inference: 39%|███▉ | 2306/5920 [39:08<1:06:53, 1.11s/it, tok/s=13841.3]
Inference: 39%|███▉ | 2306/5920 [39:08<1:06:53, 1.11s/it, tok/s=13850.1]
Inference: 39%|███▉ | 2307/5920 [39:13<2:02:44, 2.04s/it, tok/s=13850.1]
Inference: 39%|███▉ | 2307/5920 [39:13<2:02:44, 2.04s/it, tok/s=13831.4]
Inference: 39%|███▉ | 2308/5920 [39:13<1:36:05, 1.60s/it, tok/s=13831.4]
Inference: 39%|███▉ | 2308/5920 [39:13<1:36:05, 1.60s/it, tok/s=13839.6]
Inference: 39%|███▉ | 2309/5920 [39:14<1:22:33, 1.37s/it, tok/s=13839.6]
Inference: 39%|███▉ | 2309/5920 [39:14<1:22:33, 1.37s/it, tok/s=13845.8]
Inference: 39%|███▉ | 2310/5920 [39:15<1:15:16, 1.25s/it, tok/s=13845.8]
Inference: 39%|███▉ | 2310/5920 [39:15<1:15:16, 1.25s/it, tok/s=13845.6]
Inference: 39%|███▉ | 2311/5920 [39:15<55:43, 1.08it/s, tok/s=13845.6]
Inference: 39%|███▉ | 2311/5920 [39:15<55:43, 1.08it/s, tok/s=13849.2]
Inference: 39%|███▉ | 2312/5920 [39:20<2:09:37, 2.16s/it, tok/s=13849.2]
Inference: 39%|███▉ | 2312/5920 [39:20<2:09:37, 2.16s/it, tok/s=13829.5]
Inference: 39%|███▉ | 2313/5920 [39:20<1:35:55, 1.60s/it, tok/s=13829.5]
Inference: 39%|███▉ | 2313/5920 [39:20<1:35:55, 1.60s/it, tok/s=13839.4]
Inference: 39%|███▉ | 2314/5920 [39:21<1:21:31, 1.36s/it, tok/s=13839.4]
Inference: 39%|███▉ | 2314/5920 [39:21<1:21:31, 1.36s/it, tok/s=13840.0]
Inference: 39%|███▉ | 2315/5920 [39:22<1:15:06, 1.25s/it, tok/s=13840.0]
Inference: 39%|███▉ | 2315/5920 [39:22<1:15:06, 1.25s/it, tok/s=13843.5]
Inference: 39%|███▉ | 2316/5920 [39:24<1:37:17, 1.62s/it, tok/s=13843.5]
Inference: 39%|███▉ | 2316/5920 [39:24<1:37:17, 1.62s/it, tok/s=13840.8]
Inference: 39%|███▉ | 2317/5920 [39:32<3:19:42, 3.33s/it, tok/s=13840.8]
Inference: 39%|███▉ | 2317/5920 [39:32<3:19:42, 3.33s/it, tok/s=13801.7]
Inference: 39%|███▉ | 2318/5920 [39:35<3:14:07, 3.23s/it, tok/s=13801.7]
Inference: 39%|███▉ | 2318/5920 [39:35<3:14:07, 3.23s/it, tok/s=13787.3]
Inference: 39%|███▉ | 2319/5920 [39:37<3:00:43, 3.01s/it, tok/s=13787.3]
Inference: 39%|███▉ | 2319/5920 [39:37<3:00:43, 3.01s/it, tok/s=13779.1]
Inference: 39%|███▉ | 2320/5920 [39:39<2:30:57, 2.52s/it, tok/s=13779.1]
Inference: 39%|███▉ | 2320/5920 [39:39<2:30:57, 2.52s/it, tok/s=13781.5]
Inference: 39%|███▉ | 2321/5920 [39:40<2:09:46, 2.16s/it, tok/s=13781.5]
Inference: 39%|███▉ | 2321/5920 [39:40<2:09:46, 2.16s/it, tok/s=13784.6]
Inference: 39%|███▉ | 2322/5920 [39:40<1:39:39, 1.66s/it, tok/s=13784.6]
Inference: 39%|███▉ | 2322/5920 [39:40<1:39:39, 1.66s/it, tok/s=13787.5]
Inference: 39%|███▉ | 2323/5920 [39:43<1:52:02, 1.87s/it, tok/s=13787.5]
Inference: 39%|███▉ | 2323/5920 [39:43<1:52:02, 1.87s/it, tok/s=13779.9]
Inference: 39%|███▉ | 2324/5920 [39:43<1:23:21, 1.39s/it, tok/s=13779.9]
Inference: 39%|███▉ | 2324/5920 [39:43<1:23:21, 1.39s/it, tok/s=13782.1]
Inference: 39%|███▉ | 2325/5920 [39:44<1:15:20, 1.26s/it, tok/s=13782.1]
Inference: 39%|███▉ | 2325/5920 [39:44<1:15:20, 1.26s/it, tok/s=13786.2]
Inference: 39%|███▉ | 2326/5920 [39:45<1:03:34, 1.06s/it, tok/s=13786.2]
Inference: 39%|███▉ | 2326/5920 [39:45<1:03:34, 1.06s/it, tok/s=13787.0]
Inference: 39%|███▉ | 2327/5920 [39:47<1:23:55, 1.40s/it, tok/s=13787.0]
Inference: 39%|███▉ | 2327/5920 [39:47<1:23:55, 1.40s/it, tok/s=13785.4]
Inference: 39%|███▉ | 2328/5920 [39:47<1:10:20, 1.17s/it, tok/s=13785.4]
Inference: 39%|███▉ | 2328/5920 [39:47<1:10:20, 1.17s/it, tok/s=13785.5]
Inference: 39%|███▉ | 2329/5920 [39:48<1:02:56, 1.05s/it, tok/s=13785.5]
Inference: 39%|███▉ | 2329/5920 [39:48<1:02:56, 1.05s/it, tok/s=13785.9]
Inference: 39%|███▉ | 2330/5920 [39:49<51:33, 1.16it/s, tok/s=13785.9]
Inference: 39%|███▉ | 2330/5920 [39:49<51:33, 1.16it/s, tok/s=13796.1]
Inference: 39%|███▉ | 2331/5920 [39:50<1:06:23, 1.11s/it, tok/s=13796.1]
Inference: 39%|███▉ | 2331/5920 [39:50<1:06:23, 1.11s/it, tok/s=13798.9]
Inference: 39%|███▉ | 2332/5920 [39:50<1:06:22, 1.11s/it, tok/s=13803.1]
Inference: 39%|███▉ | 2333/5920 [39:51<42:21, 1.41it/s, tok/s=13803.1]
Inference: 39%|███▉ | 2333/5920 [39:51<42:21, 1.41it/s, tok/s=13809.7]
Inference: 39%|███▉ | 2334/5920 [39:51<38:21, 1.56it/s, tok/s=13809.7]
Inference: 39%|███▉ | 2334/5920 [39:51<38:21, 1.56it/s, tok/s=13817.0]
Inference: 39%|███▉ | 2335/5920 [39:54<1:10:37, 1.18s/it, tok/s=13817.0]
Inference: 39%|███▉ | 2335/5920 [39:54<1:10:37, 1.18s/it, tok/s=13813.1]
Inference: 39%|███▉ | 2336/5920 [39:55<1:15:15, 1.26s/it, tok/s=13813.1]
Inference: 39%|███▉ | 2336/5920 [39:55<1:15:15, 1.26s/it, tok/s=13810.0]
Inference: 39%|███▉ | 2337/5920 [39:58<1:29:34, 1.50s/it, tok/s=13810.0]
Inference: 39%|███▉ | 2337/5920 [39:58<1:29:34, 1.50s/it, tok/s=13803.3]
Inference: 39%|███▉ | 2338/5920 [39:58<1:19:08, 1.33s/it, tok/s=13803.3]
Inference: 39%|███▉ | 2338/5920 [39:58<1:19:08, 1.33s/it, tok/s=13803.9]
Inference: 40%|███▉ | 2339/5920 [40:01<1:34:14, 1.58s/it, tok/s=13803.9]
Inference: 40%|███▉ | 2339/5920 [40:01<1:34:14, 1.58s/it, tok/s=13796.4]
Inference: 40%|███▉ | 2340/5920 [40:03<1:45:26, 1.77s/it, tok/s=13796.4]
Inference: 40%|███▉ | 2340/5920 [40:03<1:45:26, 1.77s/it, tok/s=13795.6]
Inference: 40%|███▉ | 2341/5920 [40:05<1:44:05, 1.74s/it, tok/s=13795.6]
Inference: 40%|███▉ | 2341/5920 [40:05<1:44:05, 1.74s/it, tok/s=13791.3]
Inference: 40%|███▉ | 2342/5920 [40:05<1:25:34, 1.43s/it, tok/s=13791.3]
Inference: 40%|███▉ | 2342/5920 [40:05<1:25:34, 1.43s/it, tok/s=13797.5]
Inference: 40%|███▉ | 2343/5920 [40:05<1:25:32, 1.43s/it, tok/s=13807.1]
Inference: 40%|███▉ | 2344/5920 [40:06<1:02:50, 1.05s/it, tok/s=13807.1]
Inference: 40%|███▉ | 2344/5920 [40:06<1:02:50, 1.05s/it, tok/s=13810.2]
Inference: 40%|███▉ | 2345/5920 [40:07<51:20, 1.16it/s, tok/s=13810.2]
Inference: 40%|███▉ | 2345/5920 [40:07<51:20, 1.16it/s, tok/s=13809.0]
Inference: 40%|███▉ | 2346/5920 [40:07<45:26, 1.31it/s, tok/s=13809.0]
Inference: 40%|███▉ | 2346/5920 [40:07<45:26, 1.31it/s, tok/s=13813.9]
Inference: 40%|███▉ | 2347/5920 [40:09<58:41, 1.01it/s, tok/s=13813.9]
Inference: 40%|███▉ | 2347/5920 [40:09<58:41, 1.01it/s, tok/s=13816.0]
Inference: 40%|███▉ | 2348/5920 [40:10<1:04:43, 1.09s/it, tok/s=13816.0]
Inference: 40%|███▉ | 2348/5920 [40:10<1:04:43, 1.09s/it, tok/s=13814.3]
Inference: 40%|███▉ | 2349/5920 [40:12<1:16:46, 1.29s/it, tok/s=13814.3]
Inference: 40%|███▉ | 2349/5920 [40:12<1:16:46, 1.29s/it, tok/s=13809.4]
Inference: 40%|███▉ | 2350/5920 [40:13<1:09:34, 1.17s/it, tok/s=13809.4]
Inference: 40%|███▉ | 2350/5920 [40:13<1:09:34, 1.17s/it, tok/s=13811.3]
Inference: 40%|███▉ | 2351/5920 [40:15<1:21:35, 1.37s/it, tok/s=13811.3]
Inference: 40%|███▉ | 2351/5920 [40:15<1:21:35, 1.37s/it, tok/s=13804.9]
Inference: 40%|███▉ | 2352/5920 [40:15<1:21:33, 1.37s/it, tok/s=13815.6]
Inference: 40%|███▉ | 2353/5920 [40:15<47:25, 1.25it/s, tok/s=13815.6]
Inference: 40%|███▉ | 2353/5920 [40:15<47:25, 1.25it/s, tok/s=13820.2]
Inference: 40%|███▉ | 2354/5920 [40:15<47:24, 1.25it/s, tok/s=13827.2]
Inference: 40%|███▉ | 2355/5920 [40:16<42:29, 1.40it/s, tok/s=13827.2]
Inference: 40%|███▉ | 2355/5920 [40:16<42:29, 1.40it/s, tok/s=13830.8]
Inference: 40%|███▉ | 2356/5920 [40:17<48:16, 1.23it/s, tok/s=13830.8]
Inference: 40%|███▉ | 2356/5920 [40:17<48:16, 1.23it/s, tok/s=13835.1]
Inference: 40%|███▉ | 2357/5920 [40:18<45:46, 1.30it/s, tok/s=13835.1]
Inference: 40%|███▉ | 2357/5920 [40:18<45:46, 1.30it/s, tok/s=13843.9]
Inference: 40%|███▉ | 2358/5920 [40:19<49:33, 1.20it/s, tok/s=13843.9]
Inference: 40%|███▉ | 2358/5920 [40:19<49:33, 1.20it/s, tok/s=13844.4]
Inference: 40%|███▉ | 2359/5920 [40:20<53:02, 1.12it/s, tok/s=13844.4]
Inference: 40%|███▉ | 2359/5920 [40:20<53:02, 1.12it/s, tok/s=13848.9]
Inference: 40%|███▉ | 2360/5920 [40:21<1:02:45, 1.06s/it, tok/s=13848.9]
Inference: 40%|███▉ | 2360/5920 [40:21<1:02:45, 1.06s/it, tok/s=13850.0]
Inference: 40%|███▉ | 2361/5920 [40:23<1:09:44, 1.18s/it, tok/s=13850.0]
Inference: 40%|███▉ | 2361/5920 [40:23<1:09:44, 1.18s/it, tok/s=13842.1]
Inference: 40%|███▉ | 2362/5920 [40:24<1:04:24, 1.09s/it, tok/s=13842.1]
Inference: 40%|███▉ | 2362/5920 [40:24<1:04:24, 1.09s/it, tok/s=13837.7]
Inference: 40%|███▉ | 2363/5920 [40:24<1:04:23, 1.09s/it, tok/s=13843.4]
Inference: 40%|███▉ | 2364/5920 [40:25<51:10, 1.16it/s, tok/s=13843.4]
Inference: 40%|███▉ | 2364/5920 [40:25<51:10, 1.16it/s, tok/s=13845.2]
Inference: 40%|███▉ | 2365/5920 [40:26<50:14, 1.18it/s, tok/s=13845.2]
Inference: 40%|███▉ | 2365/5920 [40:26<50:14, 1.18it/s, tok/s=13848.0]
Inference: 40%|███▉ | 2366/5920 [40:27<56:09, 1.05it/s, tok/s=13848.0]
Inference: 40%|███▉ | 2366/5920 [40:27<56:09, 1.05it/s, tok/s=13841.4]
Inference: 40%|███▉ | 2367/5920 [40:27<43:20, 1.37it/s, tok/s=13841.4]
Inference: 40%|███▉ | 2367/5920 [40:27<43:20, 1.37it/s, tok/s=13846.2]
Inference: 40%|████ | 2368/5920 [40:28<44:33, 1.33it/s, tok/s=13846.2]
Inference: 40%|████ | 2368/5920 [40:28<44:33, 1.33it/s, tok/s=13852.0]
Inference: 40%|████ | 2369/5920 [40:30<1:02:59, 1.06s/it, tok/s=13852.0]
Inference: 40%|████ | 2369/5920 [40:30<1:02:59, 1.06s/it, tok/s=13852.2]
Inference: 40%|████ | 2370/5920 [40:30<1:02:58, 1.06s/it, tok/s=13858.4]
Inference: 40%|████ | 2371/5920 [40:30<42:45, 1.38it/s, tok/s=13858.4]
Inference: 40%|████ | 2371/5920 [40:30<42:45, 1.38it/s, tok/s=13865.5]
Inference: 40%|████ | 2372/5920 [40:32<49:59, 1.18it/s, tok/s=13865.5]
Inference: 40%|████ | 2372/5920 [40:32<49:59, 1.18it/s, tok/s=13863.1]
Inference: 40%|████ | 2373/5920 [40:34<1:05:36, 1.11s/it, tok/s=13863.1]
Inference: 40%|████ | 2373/5920 [40:34<1:05:36, 1.11s/it, tok/s=13861.7]
Inference: 40%|████ | 2374/5920 [40:34<56:35, 1.04it/s, tok/s=13861.7]
Inference: 40%|████ | 2374/5920 [40:34<56:35, 1.04it/s, tok/s=13865.0]
Inference: 40%|████ | 2375/5920 [40:35<50:57, 1.16it/s, tok/s=13865.0]
Inference: 40%|████ | 2375/5920 [40:35<50:57, 1.16it/s, tok/s=13873.9]
Inference: 40%|████ | 2376/5920 [40:35<44:29, 1.33it/s, tok/s=13873.9]
Inference: 40%|████ | 2376/5920 [40:35<44:29, 1.33it/s, tok/s=13879.2]
Inference: 40%|████ | 2377/5920 [40:36<49:46, 1.19it/s, tok/s=13879.2]
Inference: 40%|████ | 2377/5920 [40:36<49:46, 1.19it/s, tok/s=13883.0]
Inference: 40%|████ | 2378/5920 [40:37<54:30, 1.08it/s, tok/s=13883.0]
Inference: 40%|████ | 2378/5920 [40:37<54:30, 1.08it/s, tok/s=13883.8]
Inference: 40%|████ | 2379/5920 [40:38<43:50, 1.35it/s, tok/s=13883.8]
Inference: 40%|████ | 2379/5920 [40:38<43:50, 1.35it/s, tok/s=13886.1]
Inference: 40%|████ | 2380/5920 [40:39<46:11, 1.28it/s, tok/s=13886.1]
Inference: 40%|████ | 2380/5920 [40:39<46:11, 1.28it/s, tok/s=13884.9]
Inference: 40%|████ | 2381/5920 [40:39<36:34, 1.61it/s, tok/s=13884.9]
Inference: 40%|████ | 2381/5920 [40:39<36:34, 1.61it/s, tok/s=13894.7]
Inference: 40%|████ | 2382/5920 [40:40<51:20, 1.15it/s, tok/s=13894.7]
Inference: 40%|████ | 2382/5920 [40:40<51:20, 1.15it/s, tok/s=13896.5]
Inference: 40%|████ | 2383/5920 [40:42<1:15:31, 1.28s/it, tok/s=13896.5]
Inference: 40%|████ | 2383/5920 [40:42<1:15:31, 1.28s/it, tok/s=13884.3]
Inference: 40%|████ | 2384/5920 [40:43<1:01:06, 1.04s/it, tok/s=13884.3]
Inference: 40%|████ | 2384/5920 [40:43<1:01:06, 1.04s/it, tok/s=13891.9]
Inference: 40%|████ | 2385/5920 [40:43<48:45, 1.21it/s, tok/s=13891.9]
Inference: 40%|████ | 2385/5920 [40:43<48:45, 1.21it/s, tok/s=13890.6]
Inference: 40%|████ | 2386/5920 [40:43<48:44, 1.21it/s, tok/s=13901.3]
Inference: 40%|████ | 2387/5920 [40:44<31:10, 1.89it/s, tok/s=13901.3]
Inference: 40%|████ | 2387/5920 [40:44<31:10, 1.89it/s, tok/s=13904.3]
Inference: 40%|████ | 2388/5920 [40:44<28:13, 2.09it/s, tok/s=13904.3]
Inference: 40%|████ | 2388/5920 [40:44<28:13, 2.09it/s, tok/s=13908.1]
Inference: 40%|████ | 2389/5920 [40:46<47:15, 1.25it/s, tok/s=13908.1]
Inference: 40%|████ | 2389/5920 [40:46<47:15, 1.25it/s, tok/s=13903.7]
Inference: 40%|████ | 2390/5920 [40:46<45:05, 1.30it/s, tok/s=13903.7]
Inference: 40%|████ | 2390/5920 [40:46<45:05, 1.30it/s, tok/s=13911.8]
Inference: 40%|████ | 2391/5920 [40:47<48:20, 1.22it/s, tok/s=13911.8]
Inference: 40%|████ | 2391/5920 [40:47<48:20, 1.22it/s, tok/s=13907.0]
Inference: 40%|████ | 2392/5920 [40:47<36:57, 1.59it/s, tok/s=13907.0]
Inference: 40%|████ | 2392/5920 [40:47<36:57, 1.59it/s, tok/s=13918.5]
Inference: 40%|████ | 2393/5920 [40:51<1:26:43, 1.48s/it, tok/s=13918.5]
Inference: 40%|████ | 2393/5920 [40:51<1:26:43, 1.48s/it, tok/s=13909.2]
Inference: 40%|████ | 2394/5920 [40:52<1:19:42, 1.36s/it, tok/s=13909.2]
Inference: 40%|████ | 2394/5920 [40:52<1:19:42, 1.36s/it, tok/s=13910.1]
Inference: 40%|████ | 2395/5920 [40:52<1:02:07, 1.06s/it, tok/s=13910.1]
Inference: 40%|████ | 2395/5920 [40:52<1:02:07, 1.06s/it, tok/s=13908.6]
Inference: 40%|████ | 2396/5920 [40:54<1:13:15, 1.25s/it, tok/s=13908.6]
Inference: 40%|████ | 2396/5920 [40:54<1:13:15, 1.25s/it, tok/s=13905.0]
Inference: 40%|████ | 2397/5920 [40:54<1:13:13, 1.25s/it, tok/s=13910.6]
Inference: 41%|████ | 2398/5920 [40:55<44:39, 1.31it/s, tok/s=13910.6]
Inference: 41%|████ | 2398/5920 [40:55<44:39, 1.31it/s, tok/s=13909.2]
Inference: 41%|████ | 2399/5920 [40:56<50:45, 1.16it/s, tok/s=13909.2]
Inference: 41%|████ | 2399/5920 [40:56<50:45, 1.16it/s, tok/s=13914.8]
Inference: 41%|████ | 2400/5920 [40:56<40:03, 1.46it/s, tok/s=13914.8]
Inference: 41%|████ | 2400/5920 [40:56<40:03, 1.46it/s, tok/s=13918.3]
Inference: 41%|████ | 2401/5920 [40:58<1:00:51, 1.04s/it, tok/s=13918.3]
Inference: 41%|████ | 2401/5920 [40:58<1:00:51, 1.04s/it, tok/s=13916.9]
Inference: 41%|████ | 2402/5920 [40:58<47:02, 1.25it/s, tok/s=13916.9]
Inference: 41%|████ | 2402/5920 [40:58<47:02, 1.25it/s, tok/s=13919.3]
Inference: 41%|████ | 2403/5920 [41:00<1:09:57, 1.19s/it, tok/s=13919.3]
Inference: 41%|████ | 2403/5920 [41:00<1:09:57, 1.19s/it, tok/s=13911.9]
Inference: 41%|████ | 2404/5920 [41:00<54:00, 1.08it/s, tok/s=13911.9]
Inference: 41%|████ | 2404/5920 [41:00<54:00, 1.08it/s, tok/s=13911.0]
Inference: 41%|████ | 2405/5920 [41:01<53:59, 1.08it/s, tok/s=13921.9]
Inference: 41%|████ | 2406/5920 [41:01<53:58, 1.08it/s, tok/s=13927.3]
Inference: 41%|████ | 2407/5920 [41:01<33:16, 1.76it/s, tok/s=13927.3]
Inference: 41%|████ | 2407/5920 [41:01<33:16, 1.76it/s, tok/s=13935.0]
Inference: 41%|████ | 2408/5920 [41:02<28:58, 2.02it/s, tok/s=13935.0]
Inference: 41%|████ | 2408/5920 [41:02<28:58, 2.02it/s, tok/s=13940.5]
Inference: 41%|████ | 2409/5920 [41:04<53:06, 1.10it/s, tok/s=13940.5]
Inference: 41%|████ | 2409/5920 [41:04<53:06, 1.10it/s, tok/s=13939.7]
Inference: 41%|████ | 2410/5920 [41:04<44:03, 1.33it/s, tok/s=13939.7]
Inference: 41%|████ | 2410/5920 [41:04<44:03, 1.33it/s, tok/s=13943.8]
Inference: 41%|████ | 2411/5920 [41:06<59:47, 1.02s/it, tok/s=13943.8]
Inference: 41%|████ | 2411/5920 [41:06<59:47, 1.02s/it, tok/s=13944.5]
Inference: 41%|████ | 2412/5920 [41:06<59:46, 1.02s/it, tok/s=13945.1]
Inference: 41%|████ | 2413/5920 [41:06<41:13, 1.42it/s, tok/s=13945.1]
Inference: 41%|████ | 2413/5920 [41:06<41:13, 1.42it/s, tok/s=13947.1]
Inference: 41%|████ | 2414/5920 [41:07<40:52, 1.43it/s, tok/s=13947.1]
Inference: 41%|████ | 2414/5920 [41:07<40:52, 1.43it/s, tok/s=13955.5]
Inference: 41%|████ | 2415/5920 [41:07<32:18, 1.81it/s, tok/s=13955.5]
Inference: 41%|████ | 2415/5920 [41:07<32:18, 1.81it/s, tok/s=13966.0]
Inference: 41%|████ | 2416/5920 [41:07<26:20, 2.22it/s, tok/s=13966.0]
Inference: 41%|████ | 2416/5920 [41:07<26:20, 2.22it/s, tok/s=13965.7]
Inference: 41%|████ | 2417/5920 [41:08<29:11, 2.00it/s, tok/s=13965.7]
Inference: 41%|████ | 2417/5920 [41:08<29:11, 2.00it/s, tok/s=13973.2]
Inference: 41%|████ | 2418/5920 [41:10<52:51, 1.10it/s, tok/s=13973.2]
Inference: 41%|████ | 2418/5920 [41:10<52:51, 1.10it/s, tok/s=13972.9]
Inference: 41%|████ | 2419/5920 [41:11<53:28, 1.09it/s, tok/s=13972.9]
Inference: 41%|████ | 2419/5920 [41:11<53:28, 1.09it/s, tok/s=13979.0]
Inference: 41%|████ | 2420/5920 [41:13<1:08:46, 1.18s/it, tok/s=13979.0]
Inference: 41%|████ | 2420/5920 [41:13<1:08:46, 1.18s/it, tok/s=13974.4]
Inference: 41%|████ | 2421/5920 [41:14<1:09:51, 1.20s/it, tok/s=13974.4]
Inference: 41%|████ | 2421/5920 [41:14<1:09:51, 1.20s/it, tok/s=13970.3]
Inference: 41%|████ | 2422/5920 [41:15<1:07:00, 1.15s/it, tok/s=13970.3]
Inference: 41%|████ | 2422/5920 [41:15<1:07:00, 1.15s/it, tok/s=13975.6]
Inference: 41%|████ | 2423/5920 [41:17<1:22:30, 1.42s/it, tok/s=13975.6]
Inference: 41%|████ | 2423/5920 [41:17<1:22:30, 1.42s/it, tok/s=13976.2]
Inference: 41%|████ | 2424/5920 [41:18<1:11:08, 1.22s/it, tok/s=13976.2]
Inference: 41%|████ | 2424/5920 [41:18<1:11:08, 1.22s/it, tok/s=13972.5]
Inference: 41%|████ | 2425/5920 [41:19<1:09:49, 1.20s/it, tok/s=13972.5]
Inference: 41%|████ | 2425/5920 [41:19<1:09:49, 1.20s/it, tok/s=13977.9]
Inference: 41%|████ | 2426/5920 [41:20<1:01:48, 1.06s/it, tok/s=13977.9]
Inference: 41%|████ | 2426/5920 [41:20<1:01:48, 1.06s/it, tok/s=13974.8]
Inference: 41%|████ | 2427/5920 [41:20<1:01:47, 1.06s/it, tok/s=13975.2]
Inference: 41%|████ | 2428/5920 [41:20<35:45, 1.63it/s, tok/s=13975.2]
Inference: 41%|████ | 2428/5920 [41:20<35:45, 1.63it/s, tok/s=13979.3]
Inference: 41%|████ | 2429/5920 [41:20<29:11, 1.99it/s, tok/s=13979.3]
Inference: 41%|████ | 2429/5920 [41:20<29:11, 1.99it/s, tok/s=13983.5]
Inference: 41%|████ | 2430/5920 [41:21<30:08, 1.93it/s, tok/s=13983.5]
Inference: 41%|████ | 2430/5920 [41:21<30:08, 1.93it/s, tok/s=13988.9]
Inference: 41%|████ | 2431/5920 [41:22<35:58, 1.62it/s, tok/s=13988.9]
Inference: 41%|████ | 2431/5920 [41:22<35:58, 1.62it/s, tok/s=13994.5]
Inference: 41%|████ | 2432/5920 [41:23<46:31, 1.25it/s, tok/s=13994.5]
Inference: 41%|████ | 2432/5920 [41:23<46:31, 1.25it/s, tok/s=13999.5]
Inference: 41%|████ | 2433/5920 [41:23<40:41, 1.43it/s, tok/s=13999.5]
Inference: 41%|████ | 2433/5920 [41:23<40:41, 1.43it/s, tok/s=13997.4]
Inference: 41%|████ | 2434/5920 [41:24<46:45, 1.24it/s, tok/s=13997.4]
Inference: 41%|████ | 2434/5920 [41:24<46:45, 1.24it/s, tok/s=13992.5]
Inference: 41%|████ | 2435/5920 [41:25<41:34, 1.40it/s, tok/s=13992.5]
Inference: 41%|████ | 2435/5920 [41:25<41:34, 1.40it/s, tok/s=13999.1]
Inference: 41%|████ | 2436/5920 [41:26<57:42, 1.01it/s, tok/s=13999.1]
Inference: 41%|████ | 2436/5920 [41:26<57:42, 1.01it/s, tok/s=13990.9]
Inference: 41%|████ | 2437/5920 [41:27<45:25, 1.28it/s, tok/s=13990.9]
Inference: 41%|████ | 2437/5920 [41:27<45:25, 1.28it/s, tok/s=13999.7]
Inference: 41%|████ | 2438/5920 [41:27<39:51, 1.46it/s, tok/s=13999.7]
Inference: 41%|████ | 2438/5920 [41:27<39:51, 1.46it/s, tok/s=14002.1]
Inference: 41%|████ | 2439/5920 [41:27<39:50, 1.46it/s, tok/s=14008.0]
Inference: 41%|████ | 2440/5920 [41:28<26:06, 2.22it/s, tok/s=14008.0]
Inference: 41%|████ | 2440/5920 [41:28<26:06, 2.22it/s, tok/s=14010.4]
Inference: 41%|████ | 2441/5920 [41:28<23:57, 2.42it/s, tok/s=14010.4]
Inference: 41%|████ | 2441/5920 [41:28<23:57, 2.42it/s, tok/s=14016.4]
Inference: 41%|████▏ | 2442/5920 [41:28<26:21, 2.20it/s, tok/s=14016.4]
Inference: 41%|████▏ | 2442/5920 [41:28<26:21, 2.20it/s, tok/s=14023.7]
Inference: 41%|████▏ | 2443/5920 [41:30<36:19, 1.60it/s, tok/s=14023.7]
Inference: 41%|████▏ | 2443/5920 [41:30<36:19, 1.60it/s, tok/s=14028.9]
Inference: 41%|████▏ | 2444/5920 [41:30<28:50, 2.01it/s, tok/s=14028.9]
Inference: 41%|████▏ | 2444/5920 [41:30<28:50, 2.01it/s, tok/s=14039.6]
Inference: 41%|████▏ | 2445/5920 [41:34<1:30:56, 1.57s/it, tok/s=14039.6]
Inference: 41%|████▏ | 2445/5920 [41:34<1:30:56, 1.57s/it, tok/s=14027.8]
Inference: 41%|████▏ | 2446/5920 [41:35<1:18:03, 1.35s/it, tok/s=14027.8]
Inference: 41%|████▏ | 2446/5920 [41:35<1:18:03, 1.35s/it, tok/s=14035.3]
Inference: 41%|████▏ | 2447/5920 [41:35<1:04:50, 1.12s/it, tok/s=14035.3]
Inference: 41%|████▏ | 2447/5920 [41:35<1:04:50, 1.12s/it, tok/s=14044.1]
Inference: 41%|████▏ | 2448/5920 [41:36<53:32, 1.08it/s, tok/s=14044.1]
Inference: 41%|████▏ | 2448/5920 [41:36<53:32, 1.08it/s, tok/s=14053.6]
Inference: 41%|████▏ | 2449/5920 [41:37<57:45, 1.00it/s, tok/s=14053.6]
Inference: 41%|████▏ | 2449/5920 [41:37<57:45, 1.00it/s, tok/s=14047.5]
Inference: 41%|████▏ | 2450/5920 [41:39<1:11:22, 1.23s/it, tok/s=14047.5]
Inference: 41%|████▏ | 2450/5920 [41:39<1:11:22, 1.23s/it, tok/s=14049.5]
Inference: 41%|████▏ | 2451/5920 [41:39<57:34, 1.00it/s, tok/s=14049.5]
Inference: 41%|████▏ | 2451/5920 [41:39<57:34, 1.00it/s, tok/s=14057.2]
Inference: 41%|████▏ | 2452/5920 [41:41<1:08:17, 1.18s/it, tok/s=14057.2]
Inference: 41%|████▏ | 2452/5920 [41:41<1:08:17, 1.18s/it, tok/s=14055.6]
Inference: 41%|████▏ | 2453/5920 [41:41<1:08:16, 1.18s/it, tok/s=14067.4]
Inference: 41%|████▏ | 2454/5920 [41:42<52:14, 1.11it/s, tok/s=14067.4]
Inference: 41%|████▏ | 2454/5920 [41:42<52:14, 1.11it/s, tok/s=14066.2]
Inference: 41%|████▏ | 2455/5920 [41:45<1:21:23, 1.41s/it, tok/s=14066.2]
Inference: 41%|████▏ | 2455/5920 [41:45<1:21:23, 1.41s/it, tok/s=14050.5]
Inference: 41%|████▏ | 2456/5920 [41:45<1:03:00, 1.09s/it, tok/s=14050.5]
Inference: 41%|████▏ | 2456/5920 [41:45<1:03:00, 1.09s/it, tok/s=14049.9]
Inference: 42%|████▏ | 2457/5920 [41:46<53:36, 1.08it/s, tok/s=14049.9]
Inference: 42%|████▏ | 2457/5920 [41:46<53:36, 1.08it/s, tok/s=14057.7]
Inference: 42%|████▏ | 2458/5920 [41:46<41:08, 1.40it/s, tok/s=14057.7]
Inference: 42%|████▏ | 2458/5920 [41:46<41:08, 1.40it/s, tok/s=14068.6]
Inference: 42%|████▏ | 2459/5920 [41:48<1:03:32, 1.10s/it, tok/s=14068.6]
Inference: 42%|████▏ | 2459/5920 [41:48<1:03:32, 1.10s/it, tok/s=14058.1]
Inference: 42%|████▏ | 2460/5920 [41:49<1:11:59, 1.25s/it, tok/s=14058.1]
Inference: 42%|████▏ | 2460/5920 [41:49<1:11:59, 1.25s/it, tok/s=14050.0]
Inference: 42%|████▏ | 2461/5920 [41:50<1:01:14, 1.06s/it, tok/s=14050.0]
Inference: 42%|████▏ | 2461/5920 [41:50<1:01:14, 1.06s/it, tok/s=14057.7]
Inference: 42%|████▏ | 2462/5920 [41:51<54:34, 1.06it/s, tok/s=14057.7]
Inference: 42%|████▏ | 2462/5920 [41:51<54:34, 1.06it/s, tok/s=14059.5]
Inference: 42%|████▏ | 2463/5920 [41:51<42:31, 1.36it/s, tok/s=14059.5]
Inference: 42%|████▏ | 2463/5920 [41:51<42:31, 1.36it/s, tok/s=14064.9]
Inference: 42%|████▏ | 2464/5920 [41:52<44:46, 1.29it/s, tok/s=14064.9]
Inference: 42%|████▏ | 2464/5920 [41:52<44:46, 1.29it/s, tok/s=14072.1]
Inference: 42%|████▏ | 2465/5920 [41:52<37:59, 1.52it/s, tok/s=14072.1]
Inference: 42%|████▏ | 2465/5920 [41:52<37:59, 1.52it/s, tok/s=14081.9]
Inference: 42%|████▏ | 2466/5920 [41:52<28:34, 2.01it/s, tok/s=14081.9]
Inference: 42%|████▏ | 2466/5920 [41:52<28:34, 2.01it/s, tok/s=14081.9]
Inference: 42%|████▏ | 2467/5920 [41:53<28:39, 2.01it/s, tok/s=14081.9]
Inference: 42%|████▏ | 2467/5920 [41:53<28:39, 2.01it/s, tok/s=14083.9]
Inference: 42%|████▏ | 2468/5920 [41:54<34:21, 1.67it/s, tok/s=14083.9]
Inference: 42%|████▏ | 2468/5920 [41:54<34:21, 1.67it/s, tok/s=14086.8]
Inference: 42%|████▏ | 2469/5920 [41:54<28:49, 2.00it/s, tok/s=14086.8]
Inference: 42%|████▏ | 2469/5920 [41:54<28:49, 2.00it/s, tok/s=14086.1]
Inference: 42%|████▏ | 2470/5920 [41:54<28:48, 2.00it/s, tok/s=14086.7]
Inference: 42%|████▏ | 2471/5920 [41:55<31:49, 1.81it/s, tok/s=14086.7]
Inference: 42%|████▏ | 2471/5920 [41:55<31:49, 1.81it/s, tok/s=14081.0]
Inference: 42%|████▏ | 2472/5920 [41:56<29:15, 1.96it/s, tok/s=14081.0]
Inference: 42%|████▏ | 2472/5920 [41:56<29:15, 1.96it/s, tok/s=14079.7]
Inference: 42%|████▏ | 2473/5920 [41:56<29:15, 1.96it/s, tok/s=14080.2]
Inference: 42%|████▏ | 2474/5920 [41:56<20:34, 2.79it/s, tok/s=14080.2]
Inference: 42%|████▏ | 2474/5920 [41:56<20:34, 2.79it/s, tok/s=14089.1]
Inference: 42%|████▏ | 2475/5920 [41:57<26:31, 2.16it/s, tok/s=14089.1]
Inference: 42%|████▏ | 2475/5920 [41:57<26:31, 2.16it/s, tok/s=14094.2]
Inference: 42%|████▏ | 2476/5920 [41:57<28:32, 2.01it/s, tok/s=14094.2]
Inference: 42%|████▏ | 2476/5920 [41:57<28:32, 2.01it/s, tok/s=14092.0]
Inference: 42%|████▏ | 2477/5920 [41:57<28:31, 2.01it/s, tok/s=14096.9]
Inference: 42%|████▏ | 2478/5920 [41:57<28:31, 2.01it/s, tok/s=14096.9]
Inference: 42%|████▏ | 2479/5920 [41:58<20:04, 2.86it/s, tok/s=14096.9]
Inference: 42%|████▏ | 2479/5920 [41:58<20:04, 2.86it/s, tok/s=14094.3]
Inference: 42%|████▏ | 2480/5920 [41:58<18:03, 3.17it/s, tok/s=14094.3]
Inference: 42%|████▏ | 2480/5920 [41:58<18:03, 3.17it/s, tok/s=14105.2]
Inference: 42%|████▏ | 2481/5920 [41:59<30:37, 1.87it/s, tok/s=14105.2]
Inference: 42%|████▏ | 2481/5920 [41:59<30:37, 1.87it/s, tok/s=14098.6]
Inference: 42%|████▏ | 2482/5920 [42:00<26:53, 2.13it/s, tok/s=14098.6]
Inference: 42%|████▏ | 2482/5920 [42:00<26:53, 2.13it/s, tok/s=14097.9]
Inference: 42%|████▏ | 2483/5920 [42:00<25:16, 2.27it/s, tok/s=14097.9]
Inference: 42%|████▏ | 2483/5920 [42:00<25:16, 2.27it/s, tok/s=14107.0]
Inference: 42%|████▏ | 2484/5920 [42:01<34:29, 1.66it/s, tok/s=14107.0]
Inference: 42%|████▏ | 2484/5920 [42:01<34:29, 1.66it/s, tok/s=14111.8]
Inference: 42%|████▏ | 2485/5920 [42:01<34:28, 1.66it/s, tok/s=14123.3]
Inference: 42%|████▏ | 2486/5920 [42:02<25:48, 2.22it/s, tok/s=14123.3]
Inference: 42%|████▏ | 2486/5920 [42:02<25:48, 2.22it/s, tok/s=14127.9]
Inference: 42%|████▏ | 2487/5920 [42:03<36:04, 1.59it/s, tok/s=14127.9]
Inference: 42%|████▏ | 2487/5920 [42:03<36:04, 1.59it/s, tok/s=14122.1]
Inference: 42%|████▏ | 2488/5920 [42:03<29:10, 1.96it/s, tok/s=14122.1]
Inference: 42%|████▏ | 2488/5920 [42:03<29:10, 1.96it/s, tok/s=14132.5]
Inference: 42%|████▏ | 2489/5920 [42:04<31:44, 1.80it/s, tok/s=14132.5]
Inference: 42%|████▏ | 2489/5920 [42:04<31:44, 1.80it/s, tok/s=14140.6]
Inference: 42%|████▏ | 2490/5920 [42:04<27:22, 2.09it/s, tok/s=14140.6]
Inference: 42%|████▏ | 2490/5920 [42:04<27:22, 2.09it/s, tok/s=14139.8]
Inference: 42%|████▏ | 2491/5920 [42:05<32:29, 1.76it/s, tok/s=14139.8]
Inference: 42%|████▏ | 2491/5920 [42:05<32:29, 1.76it/s, tok/s=14147.3]
Inference: 42%|████▏ | 2492/5920 [42:06<38:54, 1.47it/s, tok/s=14147.3]
Inference: 42%|████▏ | 2492/5920 [42:06<38:54, 1.47it/s, tok/s=14142.9]
Inference: 42%|████▏ | 2493/5920 [42:06<38:54, 1.47it/s, tok/s=14154.1]
Inference: 42%|████▏ | 2494/5920 [42:06<31:05, 1.84it/s, tok/s=14154.1]
Inference: 42%|████▏ | 2494/5920 [42:06<31:05, 1.84it/s, tok/s=14151.0]
Inference: 42%|████▏ | 2495/5920 [42:07<30:22, 1.88it/s, tok/s=14151.0]
Inference: 42%|████▏ | 2495/5920 [42:07<30:22, 1.88it/s, tok/s=14160.2]
Inference: 42%|████▏ | 2496/5920 [42:09<51:19, 1.11it/s, tok/s=14160.2]
Inference: 42%|████▏ | 2496/5920 [42:09<51:19, 1.11it/s, tok/s=14161.2]
Inference: 42%|████▏ | 2497/5920 [42:09<41:48, 1.36it/s, tok/s=14161.2]
Inference: 42%|████▏ | 2497/5920 [42:09<41:48, 1.36it/s, tok/s=14160.2]
Inference: 42%|████▏ | 2498/5920 [42:10<38:20, 1.49it/s, tok/s=14160.2]
Inference: 42%|████▏ | 2498/5920 [42:10<38:20, 1.49it/s, tok/s=14163.6]
Inference: 42%|████▏ | 2499/5920 [42:10<39:42, 1.44it/s, tok/s=14163.6]
Inference: 42%|████▏ | 2499/5920 [42:10<39:42, 1.44it/s, tok/s=14160.5]
Inference: 42%|████▏ | 2500/5920 [42:10<39:41, 1.44it/s, tok/s=14170.8]
Inference: 42%|████▏ | 2501/5920 [42:11<33:56, 1.68it/s, tok/s=14170.8]
Inference: 42%|████▏ | 2501/5920 [42:11<33:56, 1.68it/s, tok/s=14171.5]
Inference: 42%|████▏ | 2502/5920 [42:11<33:55, 1.68it/s, tok/s=14172.1]
Inference: 42%|████▏ | 2503/5920 [42:12<23:51, 2.39it/s, tok/s=14172.1]
Inference: 42%|████▏ | 2503/5920 [42:12<23:51, 2.39it/s, tok/s=14175.0]
Inference: 42%|████▏ | 2504/5920 [42:13<31:33, 1.80it/s, tok/s=14175.0]
Inference: 42%|████▏ | 2504/5920 [42:13<31:33, 1.80it/s, tok/s=14181.0]
Inference: 42%|████▏ | 2505/5920 [42:13<29:33, 1.93it/s, tok/s=14181.0]
Inference: 42%|████▏ | 2505/5920 [42:13<29:33, 1.93it/s, tok/s=14179.3]
Inference: 42%|████▏ | 2506/5920 [42:14<40:41, 1.40it/s, tok/s=14179.3]
Inference: 42%|████▏ | 2506/5920 [42:14<40:41, 1.40it/s, tok/s=14172.7]
Inference: 42%|████▏ | 2507/5920 [42:16<1:01:24, 1.08s/it, tok/s=14172.7]
Inference: 42%|████▏ | 2507/5920 [42:16<1:01:24, 1.08s/it, tok/s=14162.4]
Inference: 42%|████▏ | 2508/5920 [42:17<46:55, 1.21it/s, tok/s=14162.4]
Inference: 42%|████▏ | 2508/5920 [42:17<46:55, 1.21it/s, tok/s=14162.5]
Inference: 42%|████▏ | 2509/5920 [42:19<1:09:07, 1.22s/it, tok/s=14162.5]
Inference: 42%|████▏ | 2509/5920 [42:19<1:09:07, 1.22s/it, tok/s=14161.6]
Inference: 42%|████▏ | 2510/5920 [42:19<55:58, 1.02it/s, tok/s=14161.6]
Inference: 42%|████▏ | 2510/5920 [42:19<55:58, 1.02it/s, tok/s=14171.2]
Inference: 42%|████▏ | 2511/5920 [42:20<58:30, 1.03s/it, tok/s=14171.2]
Inference: 42%|████▏ | 2511/5920 [42:20<58:30, 1.03s/it, tok/s=14174.3]
Inference: 42%|████▏ | 2512/5920 [42:21<54:10, 1.05it/s, tok/s=14174.3]
Inference: 42%|████▏ | 2512/5920 [42:21<54:10, 1.05it/s, tok/s=14180.5]
Inference: 42%|████▏ | 2513/5920 [42:22<49:21, 1.15it/s, tok/s=14180.5]
Inference: 42%|████▏ | 2513/5920 [42:22<49:21, 1.15it/s, tok/s=14177.5]
Inference: 42%|████▏ | 2514/5920 [42:24<1:08:01, 1.20s/it, tok/s=14177.5]
Inference: 42%|████▏ | 2514/5920 [42:24<1:08:01, 1.20s/it, tok/s=14171.9]
Inference: 42%|████▏ | 2515/5920 [42:24<1:08:00, 1.20s/it, tok/s=14183.6]
Inference: 42%|████▎ | 2516/5920 [42:25<59:13, 1.04s/it, tok/s=14183.6]
Inference: 42%|████▎ | 2516/5920 [42:25<59:13, 1.04s/it, tok/s=14182.5]
Inference: 43%|████▎ | 2517/5920 [42:26<59:12, 1.04s/it, tok/s=14182.4]
Inference: 43%|████▎ | 2518/5920 [42:26<42:13, 1.34it/s, tok/s=14182.4]
Inference: 43%|████▎ | 2518/5920 [42:26<42:13, 1.34it/s, tok/s=14180.5]
Inference: 43%|████▎ | 2519/5920 [42:26<35:33, 1.59it/s, tok/s=14180.5]
Inference: 43%|████▎ | 2519/5920 [42:26<35:33, 1.59it/s, tok/s=14180.2]
Inference: 43%|████▎ | 2520/5920 [42:27<38:00, 1.49it/s, tok/s=14180.2]
Inference: 43%|████▎ | 2520/5920 [42:27<38:00, 1.49it/s, tok/s=14187.5]
Inference: 43%|████▎ | 2521/5920 [42:27<37:59, 1.49it/s, tok/s=14188.8]
Inference: 43%|████▎ | 2522/5920 [42:28<37:38, 1.50it/s, tok/s=14188.8]
Inference: 43%|████▎ | 2522/5920 [42:28<37:38, 1.50it/s, tok/s=14182.7]
Inference: 43%|████▎ | 2523/5920 [42:29<31:56, 1.77it/s, tok/s=14182.7]
Inference: 43%|████▎ | 2523/5920 [42:29<31:56, 1.77it/s, tok/s=14190.3]
Inference: 43%|████▎ | 2524/5920 [42:29<31:55, 1.77it/s, tok/s=14202.2]
Inference: 43%|████▎ | 2525/5920 [42:29<23:35, 2.40it/s, tok/s=14202.2]
Inference: 43%|████▎ | 2525/5920 [42:29<23:35, 2.40it/s, tok/s=14201.8]
Inference: 43%|████▎ | 2526/5920 [42:30<30:07, 1.88it/s, tok/s=14201.8]
Inference: 43%|████▎ | 2526/5920 [42:30<30:07, 1.88it/s, tok/s=14197.1]
Inference: 43%|████▎ | 2527/5920 [42:30<26:17, 2.15it/s, tok/s=14197.1]
Inference: 43%|████▎ | 2527/5920 [42:30<26:17, 2.15it/s, tok/s=14197.5]
Inference: 43%|████▎ | 2528/5920 [42:30<21:28, 2.63it/s, tok/s=14197.5]
Inference: 43%|████▎ | 2528/5920 [42:30<21:28, 2.63it/s, tok/s=14197.4]
Inference: 43%|████▎ | 2529/5920 [42:31<20:10, 2.80it/s, tok/s=14197.4]
Inference: 43%|████▎ | 2529/5920 [42:31<20:10, 2.80it/s, tok/s=14207.7]
Inference: 43%|████▎ | 2530/5920 [42:31<20:25, 2.77it/s, tok/s=14207.7]
Inference: 43%|████▎ | 2530/5920 [42:31<20:25, 2.77it/s, tok/s=14206.7]
Inference: 43%|████▎ | 2531/5920 [42:31<17:55, 3.15it/s, tok/s=14206.7]
Inference: 43%|████▎ | 2531/5920 [42:31<17:55, 3.15it/s, tok/s=14207.1]
Inference: 43%|████▎ | 2532/5920 [42:31<15:08, 3.73it/s, tok/s=14207.1]
Inference: 43%|████▎ | 2532/5920 [42:31<15:08, 3.73it/s, tok/s=14218.1]
Inference: 43%|████▎ | 2533/5920 [42:31<12:41, 4.45it/s, tok/s=14218.1]
Inference: 43%|████▎ | 2533/5920 [42:31<12:41, 4.45it/s, tok/s=14218.0]
Inference: 43%|████▎ | 2534/5920 [42:32<15:57, 3.53it/s, tok/s=14218.0]
Inference: 43%|████▎ | 2534/5920 [42:32<15:57, 3.53it/s, tok/s=14227.5]
Inference: 43%|████▎ | 2535/5920 [42:33<25:49, 2.18it/s, tok/s=14227.5]
Inference: 43%|████▎ | 2535/5920 [42:33<25:49, 2.18it/s, tok/s=14233.9]
Inference: 43%|████▎ | 2536/5920 [42:33<21:08, 2.67it/s, tok/s=14233.9]
Inference: 43%|████▎ | 2536/5920 [42:33<21:08, 2.67it/s, tok/s=14234.4]
Inference: 43%|████▎ | 2537/5920 [42:33<20:01, 2.82it/s, tok/s=14234.4]
Inference: 43%|████▎ | 2537/5920 [42:33<20:01, 2.82it/s, tok/s=14240.6]
Inference: 43%|████▎ | 2538/5920 [42:34<21:42, 2.60it/s, tok/s=14240.6]
Inference: 43%|████▎ | 2538/5920 [42:34<21:42, 2.60it/s, tok/s=14249.9]
Inference: 43%|████▎ | 2539/5920 [42:34<17:30, 3.22it/s, tok/s=14249.9]
Inference: 43%|████▎ | 2539/5920 [42:34<17:30, 3.22it/s, tok/s=14249.8]
Inference: 43%|████▎ | 2540/5920 [42:34<21:33, 2.61it/s, tok/s=14249.8]
Inference: 43%|████▎ | 2540/5920 [42:34<21:33, 2.61it/s, tok/s=14247.9]
Inference: 43%|████▎ | 2541/5920 [42:35<19:31, 2.88it/s, tok/s=14247.9]
Inference: 43%|████▎ | 2541/5920 [42:35<19:31, 2.88it/s, tok/s=14246.9]
Inference: 43%|████▎ | 2542/5920 [42:35<16:51, 3.34it/s, tok/s=14246.9]
Inference: 43%|████▎ | 2542/5920 [42:35<16:51, 3.34it/s, tok/s=14250.4]
Inference: 43%|████▎ | 2543/5920 [42:35<16:51, 3.34it/s, tok/s=14251.3]
Inference: 43%|████▎ | 2544/5920 [42:35<15:30, 3.63it/s, tok/s=14251.3]
Inference: 43%|████▎ | 2544/5920 [42:35<15:30, 3.63it/s, tok/s=14260.6]
Inference: 43%|████▎ | 2545/5920 [42:36<22:34, 2.49it/s, tok/s=14260.6]
Inference: 43%|████▎ | 2545/5920 [42:36<22:34, 2.49it/s, tok/s=14260.8]
Inference: 43%|████▎ | 2546/5920 [42:37<24:53, 2.26it/s, tok/s=14260.8]
Inference: 43%|████▎ | 2546/5920 [42:37<24:53, 2.26it/s, tok/s=14258.8]
Inference: 43%|████▎ | 2547/5920 [42:37<27:51, 2.02it/s, tok/s=14258.8]
Inference: 43%|████▎ | 2547/5920 [42:37<27:51, 2.02it/s, tok/s=14255.8]
Inference: 43%|████▎ | 2548/5920 [42:37<22:21, 2.51it/s, tok/s=14255.8]
Inference: 43%|████▎ | 2548/5920 [42:37<22:21, 2.51it/s, tok/s=14256.2]
Inference: 43%|████▎ | 2549/5920 [42:37<22:20, 2.51it/s, tok/s=14267.8]
Inference: 43%|████▎ | 2550/5920 [42:39<33:59, 1.65it/s, tok/s=14267.8]
Inference: 43%|████▎ | 2550/5920 [42:39<33:59, 1.65it/s, tok/s=14263.2]
Inference: 43%|████▎ | 2551/5920 [42:40<31:26, 1.79it/s, tok/s=14263.2]
Inference: 43%|████▎ | 2551/5920 [42:40<31:26, 1.79it/s, tok/s=14261.3]
Inference: 43%|████▎ | 2552/5920 [42:40<30:38, 1.83it/s, tok/s=14261.3]
Inference: 43%|████▎ | 2552/5920 [42:40<30:38, 1.83it/s, tok/s=14269.5]
Inference: 43%|████▎ | 2553/5920 [42:41<29:22, 1.91it/s, tok/s=14269.5]
Inference: 43%|████▎ | 2553/5920 [42:41<29:22, 1.91it/s, tok/s=14268.1]
Inference: 43%|████▎ | 2554/5920 [42:41<23:44, 2.36it/s, tok/s=14268.1]
Inference: 43%|████▎ | 2554/5920 [42:41<23:44, 2.36it/s, tok/s=14267.7]
Inference: 43%|████▎ | 2555/5920 [42:41<23:43, 2.36it/s, tok/s=14268.2]
Inference: 43%|████▎ | 2556/5920 [42:41<16:13, 3.46it/s, tok/s=14268.2]
Inference: 43%|████▎ | 2556/5920 [42:41<16:13, 3.46it/s, tok/s=14267.5]
Inference: 43%|████▎ | 2557/5920 [42:42<21:29, 2.61it/s, tok/s=14267.5]
Inference: 43%|████▎ | 2557/5920 [42:42<21:29, 2.61it/s, tok/s=14274.7]
Inference: 43%|████▎ | 2558/5920 [42:42<21:29, 2.61it/s, tok/s=14275.3]
Inference: 43%|████▎ | 2559/5920 [42:42<14:00, 4.00it/s, tok/s=14275.3]
Inference: 43%|████▎ | 2559/5920 [42:42<14:00, 4.00it/s, tok/s=14275.7]
Inference: 43%|████▎ | 2560/5920 [42:42<15:32, 3.60it/s, tok/s=14275.7]
Inference: 43%|████▎ | 2560/5920 [42:42<15:32, 3.60it/s, tok/s=14277.8]
Inference: 43%|████▎ | 2561/5920 [42:42<16:58, 3.30it/s, tok/s=14277.8]
Inference: 43%|████▎ | 2561/5920 [42:42<16:58, 3.30it/s, tok/s=14276.3]
Inference: 43%|████▎ | 2562/5920 [42:43<16:51, 3.32it/s, tok/s=14276.3]
Inference: 43%|████▎ | 2562/5920 [42:43<16:51, 3.32it/s, tok/s=14282.4]
Inference: 43%|████▎ | 2563/5920 [42:44<31:56, 1.75it/s, tok/s=14282.4]
Inference: 43%|████▎ | 2563/5920 [42:44<31:56, 1.75it/s, tok/s=14275.6]
Inference: 43%|████▎ | 2564/5920 [42:44<26:14, 2.13it/s, tok/s=14275.6]
Inference: 43%|████▎ | 2564/5920 [42:44<26:14, 2.13it/s, tok/s=14275.0]
Inference: 43%|████▎ | 2565/5920 [42:45<22:40, 2.47it/s, tok/s=14275.0]
Inference: 43%|████▎ | 2565/5920 [42:45<22:40, 2.47it/s, tok/s=14275.8]
Inference: 43%|████▎ | 2566/5920 [42:45<19:00, 2.94it/s, tok/s=14275.8]
Inference: 43%|████▎ | 2566/5920 [42:45<19:00, 2.94it/s, tok/s=14275.7]
Inference: 43%|████▎ | 2567/5920 [42:45<21:42, 2.57it/s, tok/s=14275.7]
Inference: 43%|████▎ | 2567/5920 [42:45<21:42, 2.57it/s, tok/s=14277.0]
Inference: 43%|████▎ | 2568/5920 [42:47<42:02, 1.33it/s, tok/s=14277.0]
Inference: 43%|████▎ | 2568/5920 [42:47<42:02, 1.33it/s, tok/s=14279.7]
Inference: 43%|████▎ | 2569/5920 [42:47<34:35, 1.61it/s, tok/s=14279.7]
Inference: 43%|████▎ | 2569/5920 [42:47<34:35, 1.61it/s, tok/s=14279.2]
Inference: 43%|████▎ | 2570/5920 [42:47<26:03, 2.14it/s, tok/s=14279.2]
Inference: 43%|████▎ | 2570/5920 [42:47<26:03, 2.14it/s, tok/s=14279.2]
Inference: 43%|████▎ | 2571/5920 [42:48<23:09, 2.41it/s, tok/s=14279.2]
Inference: 43%|████▎ | 2571/5920 [42:48<23:09, 2.41it/s, tok/s=14285.4]
Inference: 43%|████▎ | 2572/5920 [42:48<23:08, 2.41it/s, tok/s=14285.3]
Inference: 43%|████▎ | 2573/5920 [42:48<15:25, 3.62it/s, tok/s=14285.3]
Inference: 43%|████▎ | 2573/5920 [42:48<15:25, 3.62it/s, tok/s=14285.2]
Inference: 43%|████▎ | 2574/5920 [42:48<20:48, 2.68it/s, tok/s=14285.2]
Inference: 43%|████▎ | 2574/5920 [42:48<20:48, 2.68it/s, tok/s=14281.9]
Inference: 43%|████▎ | 2575/5920 [42:49<21:23, 2.61it/s, tok/s=14281.9]
Inference: 43%|████▎ | 2575/5920 [42:49<21:23, 2.61it/s, tok/s=14280.1]
Inference: 44%|████▎ | 2576/5920 [42:49<17:14, 3.23it/s, tok/s=14280.1]
Inference: 44%|████▎ | 2576/5920 [42:49<17:14, 3.23it/s, tok/s=14281.4]
Inference: 44%|████▎ | 2577/5920 [42:49<18:00, 3.10it/s, tok/s=14281.4]
Inference: 44%|████▎ | 2577/5920 [42:49<18:00, 3.10it/s, tok/s=14291.2]
Inference: 44%|████▎ | 2578/5920 [42:50<17:26, 3.19it/s, tok/s=14291.2]
Inference: 44%|████▎ | 2578/5920 [42:50<17:26, 3.19it/s, tok/s=14290.4]
Inference: 44%|████▎ | 2579/5920 [42:50<17:26, 3.19it/s, tok/s=14301.1]
Inference: 44%|████▎ | 2580/5920 [42:50<18:15, 3.05it/s, tok/s=14301.1]
Inference: 44%|████▎ | 2580/5920 [42:50<18:15, 3.05it/s, tok/s=14304.3]
Inference: 44%|████▎ | 2581/5920 [42:51<21:14, 2.62it/s, tok/s=14304.3]
Inference: 44%|████▎ | 2581/5920 [42:51<21:14, 2.62it/s, tok/s=14303.2]
Inference: 44%|████▎ | 2582/5920 [42:51<21:14, 2.62it/s, tok/s=14303.5]
Inference: 44%|████▎ | 2583/5920 [42:51<16:03, 3.46it/s, tok/s=14303.5]
Inference: 44%|████▎ | 2583/5920 [42:51<16:03, 3.46it/s, tok/s=14309.6]
Inference: 44%|████▎ | 2584/5920 [42:51<16:02, 3.46it/s, tok/s=14309.9]
Inference: 44%|████▎ | 2585/5920 [42:52<14:11, 3.92it/s, tok/s=14309.9]
Inference: 44%|████▎ | 2585/5920 [42:52<14:11, 3.92it/s, tok/s=14308.5]
Inference: 44%|████▎ | 2586/5920 [42:52<14:11, 3.92it/s, tok/s=14308.9]
Inference: 44%|████▎ | 2587/5920 [42:52<10:19, 5.38it/s, tok/s=14308.9]
Inference: 44%|████▎ | 2587/5920 [42:52<10:19, 5.38it/s, tok/s=14310.7]
Inference: 44%|████▎ | 2588/5920 [42:52<15:29, 3.59it/s, tok/s=14310.7]
Inference: 44%|████▎ | 2588/5920 [42:52<15:29, 3.59it/s, tok/s=14308.3]
Inference: 44%|████▎ | 2589/5920 [42:53<25:53, 2.14it/s, tok/s=14308.3]
Inference: 44%|████▎ | 2589/5920 [42:53<25:53, 2.14it/s, tok/s=14302.6]
Inference: 44%|████▍ | 2590/5920 [42:54<24:59, 2.22it/s, tok/s=14302.6]
Inference: 44%|████▍ | 2590/5920 [42:54<24:59, 2.22it/s, tok/s=14301.1]
Inference: 44%|████▍ | 2591/5920 [42:54<20:11, 2.75it/s, tok/s=14301.1]
Inference: 44%|████▍ | 2591/5920 [42:54<20:11, 2.75it/s, tok/s=14312.2]
Inference: 44%|████▍ | 2592/5920 [42:55<27:32, 2.01it/s, tok/s=14312.2]
Inference: 44%|████▍ | 2592/5920 [42:55<27:32, 2.01it/s, tok/s=14312.0]
Inference: 44%|████▍ | 2593/5920 [42:56<31:33, 1.76it/s, tok/s=14312.0]
Inference: 44%|████▍ | 2593/5920 [42:56<31:33, 1.76it/s, tok/s=14309.6]
Inference: 44%|████▍ | 2594/5920 [42:56<31:32, 1.76it/s, tok/s=14314.0]
Inference: 44%|████▍ | 2595/5920 [42:56<24:53, 2.23it/s, tok/s=14314.0]
Inference: 44%|████▍ | 2595/5920 [42:56<24:53, 2.23it/s, tok/s=14313.4]
Inference: 44%|████▍ | 2596/5920 [42:57<25:30, 2.17it/s, tok/s=14313.4]
Inference: 44%|████▍ | 2596/5920 [42:57<25:30, 2.17it/s, tok/s=14322.4]
Inference: 44%|████▍ | 2597/5920 [42:57<21:14, 2.61it/s, tok/s=14322.4]
Inference: 44%|████▍ | 2597/5920 [42:57<21:14, 2.61it/s, tok/s=14322.0]
Inference: 44%|████▍ | 2598/5920 [42:57<21:14, 2.61it/s, tok/s=14322.4]
Inference: 44%|████▍ | 2599/5920 [42:58<24:05, 2.30it/s, tok/s=14322.4]
Inference: 44%|████▍ | 2599/5920 [42:58<24:05, 2.30it/s, tok/s=14318.8]
Inference: 44%|████▍ | 2600/5920 [42:58<25:47, 2.15it/s, tok/s=14318.8]
Inference: 44%|████▍ | 2600/5920 [42:58<25:47, 2.15it/s, tok/s=14316.1]
Inference: 44%|████▍ | 2601/5920 [42:59<31:45, 1.74it/s, tok/s=14316.1]
Inference: 44%|████▍ | 2601/5920 [42:59<31:45, 1.74it/s, tok/s=14311.5]
Inference: 44%|████▍ | 2602/5920 [43:00<28:00, 1.97it/s, tok/s=14311.5]
Inference: 44%|████▍ | 2602/5920 [43:00<28:00, 1.97it/s, tok/s=14310.3]
Inference: 44%|████▍ | 2603/5920 [43:00<23:35, 2.34it/s, tok/s=14310.3]
Inference: 44%|████▍ | 2603/5920 [43:00<23:35, 2.34it/s, tok/s=14309.6]
Inference: 44%|████▍ | 2604/5920 [43:01<38:42, 1.43it/s, tok/s=14309.6]
Inference: 44%|████▍ | 2604/5920 [43:01<38:42, 1.43it/s, tok/s=14302.3]
Inference: 44%|████▍ | 2605/5920 [43:02<38:41, 1.43it/s, tok/s=14302.3]
Inference: 44%|████▍ | 2605/5920 [43:02<38:41, 1.43it/s, tok/s=14300.2]
Inference: 44%|████▍ | 2606/5920 [43:02<33:20, 1.66it/s, tok/s=14300.2]
Inference: 44%|████▍ | 2606/5920 [43:02<33:20, 1.66it/s, tok/s=14307.5]
Inference: 44%|████▍ | 2607/5920 [43:03<31:58, 1.73it/s, tok/s=14307.5]
Inference: 44%|████▍ | 2607/5920 [43:03<31:58, 1.73it/s, tok/s=14309.8]
Inference: 44%|████▍ | 2608/5920 [43:03<26:17, 2.10it/s, tok/s=14309.8]
Inference: 44%|████▍ | 2608/5920 [43:03<26:17, 2.10it/s, tok/s=14320.2]
Inference: 44%|████▍ | 2609/5920 [43:05<43:49, 1.26it/s, tok/s=14320.2]
Inference: 44%|████▍ | 2609/5920 [43:05<43:49, 1.26it/s, tok/s=14322.5]
Inference: 44%|████▍ | 2610/5920 [43:05<44:07, 1.25it/s, tok/s=14322.5]
Inference: 44%|████▍ | 2610/5920 [43:05<44:07, 1.25it/s, tok/s=14318.7]
Inference: 44%|████▍ | 2611/5920 [43:06<33:03, 1.67it/s, tok/s=14318.7]
Inference: 44%|████▍ | 2611/5920 [43:06<33:03, 1.67it/s, tok/s=14318.8]
Inference: 44%|████▍ | 2612/5920 [43:07<47:16, 1.17it/s, tok/s=14318.8]
Inference: 44%|████▍ | 2612/5920 [43:07<47:16, 1.17it/s, tok/s=14311.4]
Inference: 44%|████▍ | 2613/5920 [43:08<44:42, 1.23it/s, tok/s=14311.4]
Inference: 44%|████▍ | 2613/5920 [43:08<44:42, 1.23it/s, tok/s=14308.0]
Inference: 44%|████▍ | 2614/5920 [43:08<44:42, 1.23it/s, tok/s=14315.7]
Inference: 44%|████▍ | 2615/5920 [43:08<28:30, 1.93it/s, tok/s=14315.7]
Inference: 44%|████▍ | 2615/5920 [43:08<28:30, 1.93it/s, tok/s=14319.6]
Inference: 44%|████▍ | 2616/5920 [43:08<23:06, 2.38it/s, tok/s=14319.6]
Inference: 44%|████▍ | 2616/5920 [43:08<23:06, 2.38it/s, tok/s=14320.2]
Inference: 44%|████▍ | 2617/5920 [43:08<20:52, 2.64it/s, tok/s=14320.2]
Inference: 44%|████▍ | 2617/5920 [43:08<20:52, 2.64it/s, tok/s=14319.1]
Inference: 44%|████▍ | 2618/5920 [43:09<20:31, 2.68it/s, tok/s=14319.1]
Inference: 44%|████▍ | 2618/5920 [43:09<20:31, 2.68it/s, tok/s=14318.1]
Inference: 44%|████▍ | 2619/5920 [43:09<16:51, 3.26it/s, tok/s=14318.1]
Inference: 44%|████▍ | 2619/5920 [43:09<16:51, 3.26it/s, tok/s=14328.5]
Inference: 44%|████▍ | 2620/5920 [43:09<16:51, 3.26it/s, tok/s=14328.8]
Inference: 44%|████▍ | 2621/5920 [43:09<13:30, 4.07it/s, tok/s=14328.8]
Inference: 44%|████▍ | 2621/5920 [43:09<13:30, 4.07it/s, tok/s=14328.9]
Inference: 44%|████▍ | 2622/5920 [43:09<13:03, 4.21it/s, tok/s=14328.9]
Inference: 44%|████▍ | 2622/5920 [43:09<13:03, 4.21it/s, tok/s=14329.1]
Inference: 44%|████▍ | 2623/5920 [43:10<16:57, 3.24it/s, tok/s=14329.1]
Inference: 44%|████▍ | 2623/5920 [43:10<16:57, 3.24it/s, tok/s=14326.6]
Inference: 44%|████▍ | 2624/5920 [43:11<25:44, 2.13it/s, tok/s=14326.6]
Inference: 44%|████▍ | 2624/5920 [43:11<25:44, 2.13it/s, tok/s=14322.2]
Inference: 44%|████▍ | 2625/5920 [43:11<21:34, 2.54it/s, tok/s=14322.2]
Inference: 44%|████▍ | 2625/5920 [43:11<21:34, 2.54it/s, tok/s=14322.2]
Inference: 44%|████▍ | 2626/5920 [43:11<20:34, 2.67it/s, tok/s=14322.2]
Inference: 44%|████▍ | 2626/5920 [43:11<20:34, 2.67it/s, tok/s=14321.4]
Inference: 44%|████▍ | 2627/5920 [43:12<20:34, 2.67it/s, tok/s=14321.7]
Inference: 44%|████▍ | 2628/5920 [43:12<19:25, 2.82it/s, tok/s=14321.7]
Inference: 44%|████▍ | 2628/5920 [43:12<19:25, 2.82it/s, tok/s=14319.0]
Inference: 44%|████▍ | 2629/5920 [43:12<16:25, 3.34it/s, tok/s=14319.0]
Inference: 44%|████▍ | 2629/5920 [43:12<16:25, 3.34it/s, tok/s=14318.8]
Inference: 44%|████▍ | 2630/5920 [43:14<35:41, 1.54it/s, tok/s=14318.8]
Inference: 44%|████▍ | 2630/5920 [43:14<35:41, 1.54it/s, tok/s=14310.3]
Inference: 44%|████▍ | 2631/5920 [43:15<37:26, 1.46it/s, tok/s=14310.3]
Inference: 44%|████▍ | 2631/5920 [43:15<37:26, 1.46it/s, tok/s=14306.7]
Inference: 44%|████▍ | 2632/5920 [43:15<33:33, 1.63it/s, tok/s=14306.7]
Inference: 44%|████▍ | 2632/5920 [43:15<33:33, 1.63it/s, tok/s=14305.0]
Inference: 44%|████▍ | 2633/5920 [43:15<26:48, 2.04it/s, tok/s=14305.0]
Inference: 44%|████▍ | 2633/5920 [43:15<26:48, 2.04it/s, tok/s=14304.6]
Inference: 44%|████▍ | 2634/5920 [43:16<24:28, 2.24it/s, tok/s=14304.6]
Inference: 44%|████▍ | 2634/5920 [43:16<24:28, 2.24it/s, tok/s=14313.7]
Inference: 45%|████▍ | 2635/5920 [43:16<21:11, 2.58it/s, tok/s=14313.7]
Inference: 45%|████▍ | 2635/5920 [43:16<21:11, 2.58it/s, tok/s=14314.6]
Inference: 45%|████▍ | 2636/5920 [43:16<21:10, 2.58it/s, tok/s=14315.1]
Inference: 45%|████▍ | 2637/5920 [43:16<21:10, 2.58it/s, tok/s=14326.4]
Inference: 45%|████▍ | 2638/5920 [43:16<11:11, 4.89it/s, tok/s=14326.4]
Inference: 45%|████▍ | 2638/5920 [43:16<11:11, 4.89it/s, tok/s=14337.2]
Inference: 45%|████▍ | 2639/5920 [43:16<11:52, 4.60it/s, tok/s=14337.2]
Inference: 45%|████▍ | 2639/5920 [43:16<11:52, 4.60it/s, tok/s=14336.2]
Inference: 45%|████▍ | 2640/5920 [43:16<11:33, 4.73it/s, tok/s=14336.2]
Inference: 45%|████▍ | 2640/5920 [43:16<11:33, 4.73it/s, tok/s=14335.9]
Inference: 45%|████▍ | 2641/5920 [43:19<40:09, 1.36it/s, tok/s=14335.9]
Inference: 45%|████▍ | 2641/5920 [43:19<40:09, 1.36it/s, tok/s=14328.5]
Inference: 45%|████▍ | 2642/5920 [43:19<36:13, 1.51it/s, tok/s=14328.5]
Inference: 45%|████▍ | 2642/5920 [43:19<36:13, 1.51it/s, tok/s=14334.0]
Inference: 45%|████▍ | 2643/5920 [43:20<31:44, 1.72it/s, tok/s=14334.0]
Inference: 45%|████▍ | 2643/5920 [43:20<31:44, 1.72it/s, tok/s=14332.5]
Inference: 45%|████▍ | 2644/5920 [43:20<28:06, 1.94it/s, tok/s=14332.5]
Inference: 45%|████▍ | 2644/5920 [43:20<28:06, 1.94it/s, tok/s=14331.2]
Inference: 45%|████▍ | 2645/5920 [43:21<33:12, 1.64it/s, tok/s=14331.2]
Inference: 45%|████▍ | 2645/5920 [43:21<33:12, 1.64it/s, tok/s=14338.2]
Inference: 45%|████▍ | 2646/5920 [43:21<33:11, 1.64it/s, tok/s=14343.2]
Inference: 45%|████▍ | 2647/5920 [43:21<22:32, 2.42it/s, tok/s=14343.2]
Inference: 45%|████▍ | 2647/5920 [43:21<22:32, 2.42it/s, tok/s=14341.8]
Inference: 45%|████▍ | 2648/5920 [43:21<22:32, 2.42it/s, tok/s=14342.2]
Inference: 45%|████▍ | 2649/5920 [43:21<15:17, 3.57it/s, tok/s=14342.2]
Inference: 45%|████▍ | 2649/5920 [43:21<15:17, 3.57it/s, tok/s=14342.2]
Inference: 45%|████▍ | 2650/5920 [43:21<13:56, 3.91it/s, tok/s=14342.2]
Inference: 45%|████▍ | 2650/5920 [43:21<13:56, 3.91it/s, tok/s=14341.7]
Inference: 45%|████▍ | 2651/5920 [43:22<19:18, 2.82it/s, tok/s=14341.7]
Inference: 45%|████▍ | 2651/5920 [43:22<19:18, 2.82it/s, tok/s=14346.8]
Inference: 45%|████▍ | 2652/5920 [43:22<17:13, 3.16it/s, tok/s=14346.8]
Inference: 45%|████▍ | 2652/5920 [43:22<17:13, 3.16it/s, tok/s=14354.4]
Inference: 45%|████▍ | 2653/5920 [43:22<14:31, 3.75it/s, tok/s=14354.4]
Inference: 45%|████▍ | 2653/5920 [43:22<14:31, 3.75it/s, tok/s=14355.2]
Inference: 45%|████▍ | 2654/5920 [43:23<12:45, 4.27it/s, tok/s=14355.2]
Inference: 45%|████▍ | 2654/5920 [43:23<12:45, 4.27it/s, tok/s=14355.0]
Inference: 45%|████▍ | 2655/5920 [43:23<11:25, 4.76it/s, tok/s=14355.0]
Inference: 45%|████▍ | 2655/5920 [43:23<11:25, 4.76it/s, tok/s=14354.7]
Inference: 45%|████▍ | 2656/5920 [43:23<10:32, 5.16it/s, tok/s=14354.7]
Inference: 45%|████▍ | 2656/5920 [43:23<10:32, 5.16it/s, tok/s=14354.6]
Inference: 45%|████▍ | 2657/5920 [43:23<12:56, 4.20it/s, tok/s=14354.6]
Inference: 45%|████▍ | 2657/5920 [43:23<12:56, 4.20it/s, tok/s=14364.3]
Inference: 45%|████▍ | 2658/5920 [43:24<16:46, 3.24it/s, tok/s=14364.3]
Inference: 45%|████▍ | 2658/5920 [43:24<16:46, 3.24it/s, tok/s=14364.3]
Inference: 45%|████▍ | 2659/5920 [43:24<16:46, 3.24it/s, tok/s=14366.0]
Inference: 45%|████▍ | 2660/5920 [43:24<11:36, 4.68it/s, tok/s=14366.0]
Inference: 45%|████▍ | 2660/5920 [43:24<11:36, 4.68it/s, tok/s=14376.0]
Inference: 45%|████▍ | 2661/5920 [43:25<16:26, 3.30it/s, tok/s=14376.0]
Inference: 45%|████▍ | 2661/5920 [43:25<16:26, 3.30it/s, tok/s=14380.0]
Inference: 45%|████▍ | 2662/5920 [43:25<23:49, 2.28it/s, tok/s=14380.0]
Inference: 45%|████▍ | 2662/5920 [43:25<23:49, 2.28it/s, tok/s=14383.5]
Inference: 45%|████▍ | 2663/5920 [43:26<22:16, 2.44it/s, tok/s=14383.5]
Inference: 45%|████▍ | 2663/5920 [43:26<22:16, 2.44it/s, tok/s=14383.0]
Inference: 45%|████▌ | 2664/5920 [43:26<18:15, 2.97it/s, tok/s=14383.0]
Inference: 45%|████▌ | 2664/5920 [43:26<18:15, 2.97it/s, tok/s=14393.7]
Inference: 45%|████▌ | 2665/5920 [43:27<24:25, 2.22it/s, tok/s=14393.7]
Inference: 45%|████▌ | 2665/5920 [43:27<24:25, 2.22it/s, tok/s=14390.5]
Inference: 45%|████▌ | 2666/5920 [43:27<21:06, 2.57it/s, tok/s=14390.5]
Inference: 45%|████▌ | 2666/5920 [43:27<21:06, 2.57it/s, tok/s=14389.8]
Inference: 45%|████▌ | 2667/5920 [43:27<21:42, 2.50it/s, tok/s=14389.8]
Inference: 45%|████▌ | 2667/5920 [43:27<21:42, 2.50it/s, tok/s=14399.0]
Inference: 45%|████▌ | 2668/5920 [43:28<33:05, 1.64it/s, tok/s=14399.0]
Inference: 45%|████▌ | 2668/5920 [43:28<33:05, 1.64it/s, tok/s=14393.4]
Inference: 45%|████▌ | 2669/5920 [43:29<30:18, 1.79it/s, tok/s=14393.4]
Inference: 45%|████▌ | 2669/5920 [43:29<30:18, 1.79it/s, tok/s=14402.5]
Inference: 45%|████▌ | 2670/5920 [43:29<27:19, 1.98it/s, tok/s=14402.5]
Inference: 45%|████▌ | 2670/5920 [43:29<27:19, 1.98it/s, tok/s=14401.4]
Inference: 45%|████▌ | 2671/5920 [43:29<27:18, 1.98it/s, tok/s=14402.3]
Inference: 45%|████▌ | 2672/5920 [43:30<21:19, 2.54it/s, tok/s=14402.3]
Inference: 45%|████▌ | 2672/5920 [43:30<21:19, 2.54it/s, tok/s=14400.2]
Inference: 45%|████▌ | 2673/5920 [43:31<28:58, 1.87it/s, tok/s=14400.2]
Inference: 45%|████▌ | 2673/5920 [43:31<28:58, 1.87it/s, tok/s=14406.4]
Inference: 45%|████▌ | 2674/5920 [43:31<28:05, 1.93it/s, tok/s=14406.4]
Inference: 45%|████▌ | 2674/5920 [43:31<28:05, 1.93it/s, tok/s=14415.4]
Inference: 45%|████▌ | 2675/5920 [43:31<23:55, 2.26it/s, tok/s=14415.4]
Inference: 45%|████▌ | 2675/5920 [43:31<23:55, 2.26it/s, tok/s=14415.0]
Inference: 45%|████▌ | 2676/5920 [43:32<24:14, 2.23it/s, tok/s=14415.0]
Inference: 45%|████▌ | 2676/5920 [43:32<24:14, 2.23it/s, tok/s=14413.9]
Inference: 45%|████▌ | 2677/5920 [43:32<19:27, 2.78it/s, tok/s=14413.9]
Inference: 45%|████▌ | 2677/5920 [43:32<19:27, 2.78it/s, tok/s=14415.5]
Inference: 45%|████▌ | 2678/5920 [43:32<16:55, 3.19it/s, tok/s=14415.5]
Inference: 45%|████▌ | 2678/5920 [43:32<16:55, 3.19it/s, tok/s=14415.0]
Inference: 45%|████▌ | 2679/5920 [43:32<15:34, 3.47it/s, tok/s=14415.0]
Inference: 45%|████▌ | 2679/5920 [43:32<15:34, 3.47it/s, tok/s=14422.2]
Inference: 45%|████▌ | 2680/5920 [43:33<26:23, 2.05it/s, tok/s=14422.2]
Inference: 45%|████▌ | 2680/5920 [43:33<26:23, 2.05it/s, tok/s=14422.8]
Inference: 45%|████▌ | 2681/5920 [43:34<21:44, 2.48it/s, tok/s=14422.8]
Inference: 45%|████▌ | 2681/5920 [43:34<21:44, 2.48it/s, tok/s=14422.7]
Inference: 45%|████▌ | 2682/5920 [43:34<21:43, 2.48it/s, tok/s=14423.5]
Inference: 45%|████▌ | 2683/5920 [43:34<21:43, 2.48it/s, tok/s=14425.6]
Inference: 45%|████▌ | 2684/5920 [43:34<16:01, 3.36it/s, tok/s=14425.6]
Inference: 45%|████▌ | 2684/5920 [43:34<16:01, 3.36it/s, tok/s=14429.7]
Inference: 45%|████▌ | 2685/5920 [43:35<18:12, 2.96it/s, tok/s=14429.7]
Inference: 45%|████▌ | 2685/5920 [43:35<18:12, 2.96it/s, tok/s=14433.4]
Inference: 45%|████▌ | 2686/5920 [43:35<19:02, 2.83it/s, tok/s=14433.4]
Inference: 45%|████▌ | 2686/5920 [43:35<19:02, 2.83it/s, tok/s=14431.9]
Inference: 45%|████▌ | 2687/5920 [43:36<29:47, 1.81it/s, tok/s=14431.9]
Inference: 45%|████▌ | 2687/5920 [43:36<29:47, 1.81it/s, tok/s=14436.6]
Inference: 45%|████▌ | 2688/5920 [43:36<24:43, 2.18it/s, tok/s=14436.6]
Inference: 45%|████▌ | 2688/5920 [43:36<24:43, 2.18it/s, tok/s=14447.1]
Inference: 45%|████▌ | 2689/5920 [43:37<23:52, 2.26it/s, tok/s=14447.1]
Inference: 45%|████▌ | 2689/5920 [43:37<23:52, 2.26it/s, tok/s=14445.2]
Inference: 45%|████▌ | 2690/5920 [43:37<19:57, 2.70it/s, tok/s=14445.2]
Inference: 45%|████▌ | 2690/5920 [43:37<19:57, 2.70it/s, tok/s=14453.5]
Inference: 45%|████▌ | 2691/5920 [43:37<16:42, 3.22it/s, tok/s=14453.5]
Inference: 45%|████▌ | 2691/5920 [43:37<16:42, 3.22it/s, tok/s=14463.8]
Inference: 45%|████▌ | 2692/5920 [43:37<14:11, 3.79it/s, tok/s=14463.8]
Inference: 45%|████▌ | 2692/5920 [43:37<14:11, 3.79it/s, tok/s=14463.8]
Inference: 45%|████▌ | 2693/5920 [43:38<26:04, 2.06it/s, tok/s=14463.8]
Inference: 45%|████▌ | 2693/5920 [43:38<26:04, 2.06it/s, tok/s=14458.6]
Inference: 46%|████▌ | 2694/5920 [43:39<26:18, 2.04it/s, tok/s=14458.6]
Inference: 46%|████▌ | 2694/5920 [43:39<26:18, 2.04it/s, tok/s=14457.1]
Inference: 46%|████▌ | 2695/5920 [43:40<34:57, 1.54it/s, tok/s=14457.1]
Inference: 46%|████▌ | 2695/5920 [43:40<34:57, 1.54it/s, tok/s=14452.3]
Inference: 46%|████▌ | 2696/5920 [43:40<34:56, 1.54it/s, tok/s=14463.7]
Inference: 46%|████▌ | 2697/5920 [43:40<20:58, 2.56it/s, tok/s=14463.7]
Inference: 46%|████▌ | 2697/5920 [43:40<20:58, 2.56it/s, tok/s=14474.3]
Inference: 46%|████▌ | 2698/5920 [43:40<20:58, 2.56it/s, tok/s=14475.1]
Inference: 46%|████▌ | 2699/5920 [43:40<13:56, 3.85it/s, tok/s=14475.1]
Inference: 46%|████▌ | 2699/5920 [43:40<13:56, 3.85it/s, tok/s=14475.0]
Inference: 46%|████▌ | 2700/5920 [43:41<23:08, 2.32it/s, tok/s=14475.0]
Inference: 46%|████▌ | 2700/5920 [43:41<23:08, 2.32it/s, tok/s=14469.8]
Inference: 46%|████▌ | 2701/5920 [43:43<38:42, 1.39it/s, tok/s=14469.8]
Inference: 46%|████▌ | 2701/5920 [43:43<38:42, 1.39it/s, tok/s=14461.3]
Inference: 46%|████▌ | 2702/5920 [43:43<33:05, 1.62it/s, tok/s=14461.3]
Inference: 46%|████▌ | 2702/5920 [43:43<33:05, 1.62it/s, tok/s=14460.4]
Inference: 46%|████▌ | 2703/5920 [43:43<33:04, 1.62it/s, tok/s=14460.4]
Inference: 46%|████▌ | 2704/5920 [43:44<24:50, 2.16it/s, tok/s=14460.4]
Inference: 46%|████▌ | 2704/5920 [43:44<24:50, 2.16it/s, tok/s=14458.6]
Inference: 46%|████▌ | 2705/5920 [43:44<26:51, 2.00it/s, tok/s=14458.6]
Inference: 46%|████▌ | 2705/5920 [43:44<26:51, 2.00it/s, tok/s=14466.6]
Inference: 46%|████▌ | 2706/5920 [43:44<26:50, 2.00it/s, tok/s=14467.0]
Inference: 46%|████▌ | 2707/5920 [43:45<19:53, 2.69it/s, tok/s=14467.0]
Inference: 46%|████▌ | 2707/5920 [43:45<19:53, 2.69it/s, tok/s=14466.6]
Inference: 46%|████▌ | 2708/5920 [43:45<19:52, 2.69it/s, tok/s=14467.2]
Inference: 46%|████▌ | 2709/5920 [43:46<22:16, 2.40it/s, tok/s=14467.2]
Inference: 46%|████▌ | 2709/5920 [43:46<22:16, 2.40it/s, tok/s=14473.6]
Inference: 46%|████▌ | 2710/5920 [43:46<20:29, 2.61it/s, tok/s=14473.6]
Inference: 46%|████▌ | 2710/5920 [43:46<20:29, 2.61it/s, tok/s=14473.0]
Inference: 46%|████▌ | 2711/5920 [43:46<20:28, 2.61it/s, tok/s=14473.5]
Inference: 46%|████▌ | 2712/5920 [43:46<20:28, 2.61it/s, tok/s=14474.3]
Inference: 46%|████▌ | 2713/5920 [43:46<13:37, 3.92it/s, tok/s=14474.3]
Inference: 46%|████▌ | 2713/5920 [43:46<13:37, 3.92it/s, tok/s=14483.8]
Inference: 46%|████▌ | 2714/5920 [43:47<14:55, 3.58it/s, tok/s=14483.8]
Inference: 46%|████▌ | 2714/5920 [43:47<14:55, 3.58it/s, tok/s=14482.6]
Inference: 46%|████▌ | 2715/5920 [43:47<18:07, 2.95it/s, tok/s=14482.6]
Inference: 46%|████▌ | 2715/5920 [43:47<18:07, 2.95it/s, tok/s=14480.2]
Inference: 46%|████▌ | 2716/5920 [43:48<18:12, 2.93it/s, tok/s=14480.2]
Inference: 46%|████▌ | 2716/5920 [43:48<18:12, 2.93it/s, tok/s=14478.6]
Inference: 46%|████▌ | 2717/5920 [43:48<17:30, 3.05it/s, tok/s=14478.6]
Inference: 46%|████▌ | 2717/5920 [43:48<17:30, 3.05it/s, tok/s=14478.4]
Inference: 46%|████▌ | 2718/5920 [43:48<17:29, 3.05it/s, tok/s=14489.1]
Inference: 46%|████▌ | 2719/5920 [43:48<14:44, 3.62it/s, tok/s=14489.1]
Inference: 46%|████▌ | 2719/5920 [43:48<14:44, 3.62it/s, tok/s=14487.8]
Inference: 46%|████▌ | 2720/5920 [43:49<14:54, 3.58it/s, tok/s=14487.8]
Inference: 46%|████▌ | 2720/5920 [43:49<14:54, 3.58it/s, tok/s=14486.6]
Inference: 46%|████▌ | 2721/5920 [43:49<16:11, 3.29it/s, tok/s=14486.6]
Inference: 46%|████▌ | 2721/5920 [43:49<16:11, 3.29it/s, tok/s=14485.2]
Inference: 46%|████▌ | 2722/5920 [43:50<22:36, 2.36it/s, tok/s=14485.2]
Inference: 46%|████▌ | 2722/5920 [43:50<22:36, 2.36it/s, tok/s=14488.8]
Inference: 46%|████▌ | 2723/5920 [43:50<18:12, 2.93it/s, tok/s=14488.8]
Inference: 46%|████▌ | 2723/5920 [43:50<18:12, 2.93it/s, tok/s=14488.7]
Inference: 46%|████▌ | 2724/5920 [43:50<17:48, 2.99it/s, tok/s=14488.7]
Inference: 46%|████▌ | 2724/5920 [43:50<17:48, 2.99it/s, tok/s=14487.9]
Inference: 46%|████▌ | 2725/5920 [43:51<21:42, 2.45it/s, tok/s=14487.9]
Inference: 46%|████▌ | 2725/5920 [43:51<21:42, 2.45it/s, tok/s=14485.1]
Inference: 46%|████▌ | 2726/5920 [43:51<22:56, 2.32it/s, tok/s=14485.1]
Inference: 46%|████▌ | 2726/5920 [43:51<22:56, 2.32it/s, tok/s=14483.2]
Inference: 46%|████▌ | 2727/5920 [43:52<21:59, 2.42it/s, tok/s=14483.2]
Inference: 46%|████▌ | 2727/5920 [43:52<21:59, 2.42it/s, tok/s=14487.2]
Inference: 46%|████▌ | 2728/5920 [43:53<31:08, 1.71it/s, tok/s=14487.2]
Inference: 46%|████▌ | 2728/5920 [43:53<31:08, 1.71it/s, tok/s=14491.8]
Inference: 46%|████▌ | 2729/5920 [43:53<25:38, 2.07it/s, tok/s=14491.8]
Inference: 46%|████▌ | 2729/5920 [43:53<25:38, 2.07it/s, tok/s=14495.5]
Inference: 46%|████▌ | 2730/5920 [43:53<23:19, 2.28it/s, tok/s=14495.5]
Inference: 46%|████▌ | 2730/5920 [43:53<23:19, 2.28it/s, tok/s=14505.1]
Inference: 46%|████▌ | 2731/5920 [43:53<21:23, 2.49it/s, tok/s=14505.1]
Inference: 46%|████▌ | 2731/5920 [43:53<21:23, 2.49it/s, tok/s=14503.8]
Inference: 46%|████▌ | 2732/5920 [43:54<18:40, 2.85it/s, tok/s=14503.8]
Inference: 46%|████▌ | 2732/5920 [43:54<18:40, 2.85it/s, tok/s=14503.1]
Inference: 46%|████▌ | 2733/5920 [43:54<15:45, 3.37it/s, tok/s=14503.1]
Inference: 46%|████▌ | 2733/5920 [43:54<15:45, 3.37it/s, tok/s=14503.5]
Inference: 46%|████▌ | 2734/5920 [43:54<15:45, 3.37it/s, tok/s=14509.6]
Inference: 46%|████▌ | 2735/5920 [43:54<12:01, 4.41it/s, tok/s=14509.6]
Inference: 46%|████▌ | 2735/5920 [43:54<12:01, 4.41it/s, tok/s=14508.9]
Inference: 46%|████▌ | 2736/5920 [43:54<12:01, 4.41it/s, tok/s=14508.9]
Inference: 46%|████▌ | 2736/5920 [43:54<12:01, 4.41it/s, tok/s=14512.8]
Inference: 46%|████▌ | 2737/5920 [43:55<18:29, 2.87it/s, tok/s=14512.8]
Inference: 46%|████▌ | 2737/5920 [43:55<18:29, 2.87it/s, tok/s=14509.7]
Inference: 46%|████▋ | 2738/5920 [43:56<29:04, 1.82it/s, tok/s=14509.7]
Inference: 46%|████▋ | 2738/5920 [43:56<29:04, 1.82it/s, tok/s=14504.1]
Inference: 46%|████▋ | 2739/5920 [43:57<30:59, 1.71it/s, tok/s=14504.1]
Inference: 46%|████▋ | 2739/5920 [43:57<30:59, 1.71it/s, tok/s=14505.9]
Inference: 46%|████▋ | 2740/5920 [43:57<23:55, 2.22it/s, tok/s=14505.9]
Inference: 46%|████▋ | 2740/5920 [43:57<23:55, 2.22it/s, tok/s=14516.8]
Inference: 46%|████▋ | 2741/5920 [43:57<25:12, 2.10it/s, tok/s=14516.8]
Inference: 46%|████▋ | 2741/5920 [43:57<25:12, 2.10it/s, tok/s=14514.3]
Inference: 46%|████▋ | 2742/5920 [43:58<21:29, 2.47it/s, tok/s=14514.3]
Inference: 46%|████▋ | 2742/5920 [43:58<21:29, 2.47it/s, tok/s=14513.7]
Inference: 46%|████▋ | 2743/5920 [43:58<21:28, 2.47it/s, tok/s=14524.4]
Inference: 46%|████▋ | 2744/5920 [43:58<21:28, 2.47it/s, tok/s=14524.5]
Inference: 46%|████▋ | 2745/5920 [43:58<10:52, 4.87it/s, tok/s=14524.5]
Inference: 46%|████▋ | 2745/5920 [43:58<10:52, 4.87it/s, tok/s=14525.8]
Inference: 46%|████▋ | 2746/5920 [43:58<10:18, 5.13it/s, tok/s=14525.8]
Inference: 46%|████▋ | 2746/5920 [43:58<10:18, 5.13it/s, tok/s=14526.6]
Inference: 46%|████▋ | 2747/5920 [43:58<13:51, 3.82it/s, tok/s=14526.6]
Inference: 46%|████▋ | 2747/5920 [43:58<13:51, 3.82it/s, tok/s=14524.3]
Inference: 46%|████▋ | 2748/5920 [43:59<13:01, 4.06it/s, tok/s=14524.3]
Inference: 46%|████▋ | 2748/5920 [43:59<13:01, 4.06it/s, tok/s=14524.0]
Inference: 46%|████▋ | 2749/5920 [43:59<13:00, 4.06it/s, tok/s=14532.8]
Inference: 46%|████▋ | 2750/5920 [44:00<19:42, 2.68it/s, tok/s=14532.8]
Inference: 46%|████▋ | 2750/5920 [44:00<19:42, 2.68it/s, tok/s=14527.4]
Inference: 46%|████▋ | 2751/5920 [44:00<19:42, 2.68it/s, tok/s=14527.8]
Inference: 46%|████▋ | 2752/5920 [44:00<19:41, 2.68it/s, tok/s=14536.9]
Inference: 47%|████▋ | 2753/5920 [44:01<16:23, 3.22it/s, tok/s=14536.9]
Inference: 47%|████▋ | 2753/5920 [44:01<16:23, 3.22it/s, tok/s=14534.1]
Inference: 47%|████▋ | 2754/5920 [44:01<14:47, 3.57it/s, tok/s=14534.1]
Inference: 47%|████▋ | 2754/5920 [44:01<14:47, 3.57it/s, tok/s=14534.7]
Inference: 47%|████▋ | 2755/5920 [44:01<18:15, 2.89it/s, tok/s=14534.7]
Inference: 47%|████▋ | 2755/5920 [44:01<18:15, 2.89it/s, tok/s=14532.6]
Inference: 47%|████▋ | 2756/5920 [44:01<16:21, 3.22it/s, tok/s=14532.6]
Inference: 47%|████▋ | 2756/5920 [44:01<16:21, 3.22it/s, tok/s=14531.9]
Inference: 47%|████▋ | 2757/5920 [44:02<15:02, 3.50it/s, tok/s=14531.9]
Inference: 47%|████▋ | 2757/5920 [44:02<15:02, 3.50it/s, tok/s=14531.2]
Inference: 47%|████▋ | 2758/5920 [44:02<13:47, 3.82it/s, tok/s=14531.2]
Inference: 47%|████▋ | 2758/5920 [44:02<13:47, 3.82it/s, tok/s=14530.8]
Inference: 47%|████▋ | 2759/5920 [44:03<19:26, 2.71it/s, tok/s=14530.8]
Inference: 47%|████▋ | 2759/5920 [44:03<19:26, 2.71it/s, tok/s=14538.6]
Inference: 47%|████▋ | 2760/5920 [44:03<15:45, 3.34it/s, tok/s=14538.6]
Inference: 47%|████▋ | 2760/5920 [44:03<15:45, 3.34it/s, tok/s=14538.7]
Inference: 47%|████▋ | 2761/5920 [44:03<15:45, 3.34it/s, tok/s=14539.2]
Inference: 47%|████▋ | 2762/5920 [44:03<12:12, 4.31it/s, tok/s=14539.2]
Inference: 47%|████▋ | 2762/5920 [44:03<12:12, 4.31it/s, tok/s=14538.5]
Inference: 47%|████▋ | 2763/5920 [44:03<11:55, 4.41it/s, tok/s=14538.5]
Inference: 47%|████▋ | 2763/5920 [44:03<11:55, 4.41it/s, tok/s=14538.1]
Inference: 47%|████▋ | 2764/5920 [44:03<11:55, 4.41it/s, tok/s=14544.1]
Inference: 47%|████▋ | 2765/5920 [44:04<12:08, 4.33it/s, tok/s=14544.1]
Inference: 47%|████▋ | 2765/5920 [44:04<12:08, 4.33it/s, tok/s=14547.2]
Inference: 47%|████▋ | 2766/5920 [44:04<12:39, 4.15it/s, tok/s=14547.2]
Inference: 47%|████▋ | 2766/5920 [44:04<12:39, 4.15it/s, tok/s=14546.3]
Inference: 47%|████▋ | 2767/5920 [44:04<12:15, 4.28it/s, tok/s=14546.3]
Inference: 47%|████▋ | 2767/5920 [44:04<12:15, 4.28it/s, tok/s=14556.6]
Inference: 47%|████▋ | 2768/5920 [44:04<12:15, 4.28it/s, tok/s=14557.5]
Inference: 47%|████▋ | 2769/5920 [44:04<12:15, 4.28it/s, tok/s=14557.9]
Inference: 47%|████▋ | 2770/5920 [44:05<16:11, 3.24it/s, tok/s=14557.9]
Inference: 47%|████▋ | 2770/5920 [44:05<16:11, 3.24it/s, tok/s=14552.6]
Inference: 47%|████▋ | 2771/5920 [44:06<16:45, 3.13it/s, tok/s=14552.6]
Inference: 47%|████▋ | 2771/5920 [44:06<16:45, 3.13it/s, tok/s=14556.2]
Inference: 47%|████▋ | 2772/5920 [44:06<15:37, 3.36it/s, tok/s=14556.2]
Inference: 47%|████▋ | 2772/5920 [44:06<15:37, 3.36it/s, tok/s=14555.5]
Inference: 47%|████▋ | 2773/5920 [44:06<15:29, 3.38it/s, tok/s=14555.5]
Inference: 47%|████▋ | 2773/5920 [44:06<15:29, 3.38it/s, tok/s=14554.4]
Inference: 47%|████▋ | 2774/5920 [44:07<25:29, 2.06it/s, tok/s=14554.4]
Inference: 47%|████▋ | 2774/5920 [44:07<25:29, 2.06it/s, tok/s=14555.1]
Inference: 47%|████▋ | 2775/5920 [44:07<21:29, 2.44it/s, tok/s=14555.1]
Inference: 47%|████▋ | 2775/5920 [44:07<21:29, 2.44it/s, tok/s=14555.5]
Inference: 47%|████▋ | 2776/5920 [44:08<20:47, 2.52it/s, tok/s=14555.5]
Inference: 47%|████▋ | 2776/5920 [44:08<20:47, 2.52it/s, tok/s=14554.0]
Inference: 47%|████▋ | 2777/5920 [44:08<21:55, 2.39it/s, tok/s=14554.0]
Inference: 47%|████▋ | 2777/5920 [44:08<21:55, 2.39it/s, tok/s=14552.5]
Inference: 47%|████▋ | 2778/5920 [44:09<21:58, 2.38it/s, tok/s=14552.5]
Inference: 47%|████▋ | 2778/5920 [44:09<21:58, 2.38it/s, tok/s=14550.8]
Inference: 47%|████▋ | 2779/5920 [44:09<24:43, 2.12it/s, tok/s=14550.8]
Inference: 47%|████▋ | 2779/5920 [44:09<24:43, 2.12it/s, tok/s=14549.1]
Inference: 47%|████▋ | 2780/5920 [44:10<30:50, 1.70it/s, tok/s=14549.1]
Inference: 47%|████▋ | 2780/5920 [44:10<30:50, 1.70it/s, tok/s=14546.2]
Inference: 47%|████▋ | 2781/5920 [44:10<25:57, 2.01it/s, tok/s=14546.2]
Inference: 47%|████▋ | 2781/5920 [44:10<25:57, 2.01it/s, tok/s=14545.5]
Inference: 47%|████▋ | 2782/5920 [44:11<24:46, 2.11it/s, tok/s=14545.5]
Inference: 47%|████▋ | 2782/5920 [44:11<24:46, 2.11it/s, tok/s=14550.4]
Inference: 47%|████▋ | 2783/5920 [44:12<31:15, 1.67it/s, tok/s=14550.4]
Inference: 47%|████▋ | 2783/5920 [44:12<31:15, 1.67it/s, tok/s=14546.4]
Inference: 47%|████▋ | 2784/5920 [44:12<26:54, 1.94it/s, tok/s=14546.4]
Inference: 47%|████▋ | 2784/5920 [44:12<26:54, 1.94it/s, tok/s=14552.4]
Inference: 47%|████▋ | 2785/5920 [44:12<21:22, 2.44it/s, tok/s=14552.4]
Inference: 47%|████▋ | 2785/5920 [44:12<21:22, 2.44it/s, tok/s=14558.2]
Inference: 47%|████▋ | 2786/5920 [44:12<17:14, 3.03it/s, tok/s=14558.2]
Inference: 47%|████▋ | 2786/5920 [44:12<17:14, 3.03it/s, tok/s=14558.1]
Inference: 47%|████▋ | 2787/5920 [44:12<17:14, 3.03it/s, tok/s=14559.0]
Inference: 47%|████▋ | 2788/5920 [44:12<11:24, 4.58it/s, tok/s=14559.0]
Inference: 47%|████▋ | 2788/5920 [44:12<11:24, 4.58it/s, tok/s=14561.2]
Inference: 47%|████▋ | 2789/5920 [44:13<10:27, 4.99it/s, tok/s=14561.2]
Inference: 47%|████▋ | 2789/5920 [44:13<10:27, 4.99it/s, tok/s=14560.9]
Inference: 47%|████▋ | 2790/5920 [44:13<15:40, 3.33it/s, tok/s=14560.9]
Inference: 47%|████▋ | 2790/5920 [44:13<15:40, 3.33it/s, tok/s=14558.3]
Inference: 47%|████▋ | 2791/5920 [44:14<19:40, 2.65it/s, tok/s=14558.3]
Inference: 47%|████▋ | 2791/5920 [44:14<19:40, 2.65it/s, tok/s=14556.1]
Inference: 47%|████▋ | 2792/5920 [44:14<18:28, 2.82it/s, tok/s=14556.1]
Inference: 47%|████▋ | 2792/5920 [44:14<18:28, 2.82it/s, tok/s=14556.3]
Inference: 47%|████▋ | 2793/5920 [44:15<19:48, 2.63it/s, tok/s=14556.3]
Inference: 47%|████▋ | 2793/5920 [44:15<19:48, 2.63it/s, tok/s=14555.4]
Inference: 47%|████▋ | 2794/5920 [44:15<19:48, 2.63it/s, tok/s=14556.3]
Inference: 47%|████▋ | 2795/5920 [44:15<16:45, 3.11it/s, tok/s=14556.3]
Inference: 47%|████▋ | 2795/5920 [44:15<16:45, 3.11it/s, tok/s=14565.1]
Inference: 47%|████▋ | 2796/5920 [44:15<16:48, 3.10it/s, tok/s=14565.1]
Inference: 47%|████▋ | 2796/5920 [44:15<16:48, 3.10it/s, tok/s=14564.4]
Inference: 47%|████▋ | 2797/5920 [44:15<16:47, 3.10it/s, tok/s=14565.7]
Inference: 47%|████▋ | 2798/5920 [44:16<12:08, 4.29it/s, tok/s=14565.7]
Inference: 47%|████▋ | 2798/5920 [44:16<12:08, 4.29it/s, tok/s=14569.4]
Inference: 47%|████▋ | 2799/5920 [44:16<14:35, 3.57it/s, tok/s=14569.4]
Inference: 47%|████▋ | 2799/5920 [44:16<14:35, 3.57it/s, tok/s=14567.8]
Inference: 47%|████▋ | 2800/5920 [44:16<15:08, 3.43it/s, tok/s=14567.8]
Inference: 47%|████▋ | 2800/5920 [44:16<15:08, 3.43it/s, tok/s=14566.8]
Inference: 47%|████▋ | 2801/5920 [44:17<16:47, 3.10it/s, tok/s=14566.8]
Inference: 47%|████▋ | 2801/5920 [44:17<16:47, 3.10it/s, tok/s=14565.5]
Inference: 47%|████▋ | 2802/5920 [44:18<24:44, 2.10it/s, tok/s=14565.5]
Inference: 47%|████▋ | 2802/5920 [44:18<24:44, 2.10it/s, tok/s=14561.0]
Inference: 47%|████▋ | 2803/5920 [44:18<24:44, 2.10it/s, tok/s=14562.3]
Inference: 47%|████▋ | 2804/5920 [44:18<24:43, 2.10it/s, tok/s=14565.9]
Inference: 47%|████▋ | 2805/5920 [44:18<13:19, 3.89it/s, tok/s=14565.9]
Inference: 47%|████▋ | 2805/5920 [44:18<13:19, 3.89it/s, tok/s=14566.0]
Inference: 47%|████▋ | 2806/5920 [44:18<13:19, 3.89it/s, tok/s=14567.2]
Inference: 47%|████▋ | 2807/5920 [44:19<15:50, 3.28it/s, tok/s=14567.2]
Inference: 47%|████▋ | 2807/5920 [44:19<15:50, 3.28it/s, tok/s=14568.0]
Inference: 47%|████▋ | 2808/5920 [44:20<23:59, 2.16it/s, tok/s=14568.0]
Inference: 47%|████▋ | 2808/5920 [44:20<23:59, 2.16it/s, tok/s=14563.0]
Inference: 47%|████▋ | 2809/5920 [44:20<24:34, 2.11it/s, tok/s=14563.0]
Inference: 47%|████▋ | 2809/5920 [44:20<24:34, 2.11it/s, tok/s=14560.5]
Inference: 47%|████▋ | 2810/5920 [44:20<20:39, 2.51it/s, tok/s=14560.5]
Inference: 47%|████▋ | 2810/5920 [44:20<20:39, 2.51it/s, tok/s=14560.7]
Inference: 47%|████▋ | 2811/5920 [44:21<23:28, 2.21it/s, tok/s=14560.7]
Inference: 47%|████▋ | 2811/5920 [44:21<23:28, 2.21it/s, tok/s=14557.8]
Inference: 48%|████▊ | 2812/5920 [44:22<27:36, 1.88it/s, tok/s=14557.8]
Inference: 48%|████▊ | 2812/5920 [44:22<27:36, 1.88it/s, tok/s=14554.1]
Inference: 48%|████▊ | 2813/5920 [44:22<27:35, 1.88it/s, tok/s=14564.3]
Inference: 48%|████▊ | 2814/5920 [44:22<27:35, 1.88it/s, tok/s=14565.1]
Inference: 48%|████▊ | 2815/5920 [44:23<23:56, 2.16it/s, tok/s=14565.1]
Inference: 48%|████▊ | 2815/5920 [44:23<23:56, 2.16it/s, tok/s=14559.2]
Inference: 48%|████▊ | 2816/5920 [44:24<32:37, 1.59it/s, tok/s=14559.2]
Inference: 48%|████▊ | 2816/5920 [44:24<32:37, 1.59it/s, tok/s=14553.1]
Inference: 48%|████▊ | 2817/5920 [44:24<26:47, 1.93it/s, tok/s=14553.1]
Inference: 48%|████▊ | 2817/5920 [44:24<26:47, 1.93it/s, tok/s=14562.3]
Inference: 48%|████▊ | 2818/5920 [44:24<26:46, 1.93it/s, tok/s=14562.5]
Inference: 48%|████▊ | 2819/5920 [44:25<23:56, 2.16it/s, tok/s=14562.5]
Inference: 48%|████▊ | 2819/5920 [44:25<23:56, 2.16it/s, tok/s=14559.8]
Inference: 48%|████▊ | 2820/5920 [44:26<26:04, 1.98it/s, tok/s=14559.8]
Inference: 48%|████▊ | 2820/5920 [44:26<26:04, 1.98it/s, tok/s=14558.5]
Inference: 48%|████▊ | 2821/5920 [44:26<25:56, 1.99it/s, tok/s=14558.5]
Inference: 48%|████▊ | 2821/5920 [44:26<25:56, 1.99it/s, tok/s=14557.4]
Inference: 48%|████▊ | 2822/5920 [44:28<42:57, 1.20it/s, tok/s=14557.4]
Inference: 48%|████▊ | 2822/5920 [44:28<42:57, 1.20it/s, tok/s=14556.6]
Inference: 48%|████▊ | 2823/5920 [44:28<34:57, 1.48it/s, tok/s=14556.6]
Inference: 48%|████▊ | 2823/5920 [44:28<34:57, 1.48it/s, tok/s=14564.4]
Inference: 48%|████▊ | 2824/5920 [44:29<35:01, 1.47it/s, tok/s=14564.4]
Inference: 48%|████▊ | 2824/5920 [44:29<35:01, 1.47it/s, tok/s=14561.8]
Inference: 48%|████▊ | 2825/5920 [44:29<27:41, 1.86it/s, tok/s=14561.8]
Inference: 48%|████▊ | 2825/5920 [44:29<27:41, 1.86it/s, tok/s=14561.3]
Inference: 48%|████▊ | 2826/5920 [44:30<36:38, 1.41it/s, tok/s=14561.3]
Inference: 48%|████▊ | 2826/5920 [44:30<36:38, 1.41it/s, tok/s=14555.9]
Inference: 48%|████▊ | 2827/5920 [44:30<27:59, 1.84it/s, tok/s=14555.9]
Inference: 48%|████▊ | 2827/5920 [44:30<27:59, 1.84it/s, tok/s=14555.6]
Inference: 48%|████▊ | 2828/5920 [44:31<27:58, 1.84it/s, tok/s=14560.8]
Inference: 48%|████▊ | 2829/5920 [44:31<17:58, 2.87it/s, tok/s=14560.8]
Inference: 48%|████▊ | 2829/5920 [44:31<17:58, 2.87it/s, tok/s=14561.4]
Inference: 48%|████▊ | 2830/5920 [44:31<17:57, 2.87it/s, tok/s=14566.8]
Inference: 48%|████▊ | 2831/5920 [44:31<14:33, 3.53it/s, tok/s=14566.8]
Inference: 48%|████▊ | 2831/5920 [44:31<14:33, 3.53it/s, tok/s=14566.3]
Inference: 48%|████▊ | 2832/5920 [44:31<15:15, 3.37it/s, tok/s=14566.3]
Inference: 48%|████▊ | 2832/5920 [44:31<15:15, 3.37it/s, tok/s=14564.9]
Inference: 48%|████▊ | 2833/5920 [44:32<15:15, 3.37it/s, tok/s=14564.7]
Inference: 48%|████▊ | 2834/5920 [44:33<27:38, 1.86it/s, tok/s=14564.7]
Inference: 48%|████▊ | 2834/5920 [44:33<27:38, 1.86it/s, tok/s=14555.9]
Inference: 48%|████▊ | 2835/5920 [44:33<23:20, 2.20it/s, tok/s=14555.9]
Inference: 48%|████▊ | 2835/5920 [44:33<23:20, 2.20it/s, tok/s=14556.0]
Inference: 48%|████▊ | 2836/5920 [44:34<22:15, 2.31it/s, tok/s=14556.0]
Inference: 48%|████▊ | 2836/5920 [44:34<22:15, 2.31it/s, tok/s=14555.0]
Inference: 48%|████▊ | 2837/5920 [44:34<24:45, 2.07it/s, tok/s=14555.0]
Inference: 48%|████▊ | 2837/5920 [44:34<24:45, 2.07it/s, tok/s=14552.9]
Inference: 48%|████▊ | 2838/5920 [44:36<33:28, 1.53it/s, tok/s=14552.9]
Inference: 48%|████▊ | 2838/5920 [44:36<33:28, 1.53it/s, tok/s=14547.7]
Inference: 48%|████▊ | 2839/5920 [44:36<35:47, 1.43it/s, tok/s=14547.7]
Inference: 48%|████▊ | 2839/5920 [44:36<35:47, 1.43it/s, tok/s=14544.3]
Inference: 48%|████▊ | 2840/5920 [44:37<31:18, 1.64it/s, tok/s=14544.3]
Inference: 48%|████▊ | 2840/5920 [44:37<31:18, 1.64it/s, tok/s=14542.6]
Inference: 48%|████▊ | 2841/5920 [44:37<31:16, 1.64it/s, tok/s=14542.6]
Inference: 48%|████▊ | 2841/5920 [44:37<31:16, 1.64it/s, tok/s=14540.0]
Inference: 48%|████▊ | 2842/5920 [44:37<31:15, 1.64it/s, tok/s=14540.6]
Inference: 48%|████▊ | 2843/5920 [44:38<23:10, 2.21it/s, tok/s=14540.6]
Inference: 48%|████▊ | 2843/5920 [44:38<23:10, 2.21it/s, tok/s=14538.7]
Inference: 48%|████▊ | 2844/5920 [44:38<23:10, 2.21it/s, tok/s=14540.6]
Inference: 48%|████▊ | 2845/5920 [44:39<21:10, 2.42it/s, tok/s=14540.6]
Inference: 48%|████▊ | 2845/5920 [44:39<21:10, 2.42it/s, tok/s=14538.2]
Inference: 48%|████▊ | 2846/5920 [44:39<21:06, 2.43it/s, tok/s=14538.2]
Inference: 48%|████▊ | 2846/5920 [44:39<21:06, 2.43it/s, tok/s=14536.7]
Inference: 48%|████▊ | 2847/5920 [44:40<26:59, 1.90it/s, tok/s=14536.7]
Inference: 48%|████▊ | 2847/5920 [44:40<26:59, 1.90it/s, tok/s=14532.4]
Inference: 48%|████▊ | 2848/5920 [44:40<22:11, 2.31it/s, tok/s=14532.4]
Inference: 48%|████▊ | 2848/5920 [44:40<22:11, 2.31it/s, tok/s=14532.3]
Inference: 48%|████▊ | 2849/5920 [44:40<22:11, 2.31it/s, tok/s=14538.9]
Inference: 48%|████▊ | 2850/5920 [44:40<15:55, 3.21it/s, tok/s=14538.9]
Inference: 48%|████▊ | 2850/5920 [44:40<15:55, 3.21it/s, tok/s=14539.3]
Inference: 48%|████▊ | 2851/5920 [44:41<16:45, 3.05it/s, tok/s=14539.3]
Inference: 48%|████▊ | 2851/5920 [44:41<16:45, 3.05it/s, tok/s=14538.2]
Inference: 48%|████▊ | 2852/5920 [44:41<18:16, 2.80it/s, tok/s=14538.2]
Inference: 48%|████▊ | 2852/5920 [44:41<18:16, 2.80it/s, tok/s=14536.8]
Inference: 48%|████▊ | 2853/5920 [44:41<16:40, 3.06it/s, tok/s=14536.8]
Inference: 48%|████▊ | 2853/5920 [44:41<16:40, 3.06it/s, tok/s=14536.5]
Inference: 48%|████▊ | 2854/5920 [44:42<20:36, 2.48it/s, tok/s=14536.5]
Inference: 48%|████▊ | 2854/5920 [44:42<20:36, 2.48it/s, tok/s=14534.0]
Inference: 48%|████▊ | 2855/5920 [44:42<16:24, 3.11it/s, tok/s=14534.0]
Inference: 48%|████▊ | 2855/5920 [44:42<16:24, 3.11it/s, tok/s=14544.6]
Inference: 48%|████▊ | 2856/5920 [44:42<13:10, 3.88it/s, tok/s=14544.6]
Inference: 48%|████▊ | 2856/5920 [44:42<13:10, 3.88it/s, tok/s=14553.5]
Inference: 48%|████▊ | 2857/5920 [44:44<29:54, 1.71it/s, tok/s=14553.5]
Inference: 48%|████▊ | 2857/5920 [44:44<29:54, 1.71it/s, tok/s=14546.4]
Inference: 48%|████▊ | 2858/5920 [44:44<29:54, 1.71it/s, tok/s=14546.8]
Inference: 48%|████▊ | 2859/5920 [44:44<17:33, 2.90it/s, tok/s=14546.8]
Inference: 48%|████▊ | 2859/5920 [44:44<17:33, 2.90it/s, tok/s=14551.9]
Inference: 48%|████▊ | 2860/5920 [44:44<20:59, 2.43it/s, tok/s=14551.9]
Inference: 48%|████▊ | 2860/5920 [44:44<20:59, 2.43it/s, tok/s=14549.0]
Inference: 48%|████▊ | 2861/5920 [44:45<28:35, 1.78it/s, tok/s=14549.0]
Inference: 48%|████▊ | 2861/5920 [44:45<28:35, 1.78it/s, tok/s=14544.4]
Inference: 48%|████▊ | 2862/5920 [44:45<22:15, 2.29it/s, tok/s=14544.4]
Inference: 48%|████▊ | 2862/5920 [44:45<22:15, 2.29it/s, tok/s=14544.6]
Inference: 48%|████▊ | 2863/5920 [44:46<18:08, 2.81it/s, tok/s=14544.6]
Inference: 48%|████▊ | 2863/5920 [44:46<18:08, 2.81it/s, tok/s=14555.1]
Inference: 48%|████▊ | 2864/5920 [44:46<20:18, 2.51it/s, tok/s=14555.1]
Inference: 48%|████▊ | 2864/5920 [44:46<20:18, 2.51it/s, tok/s=14553.6]
Inference: 48%|████▊ | 2865/5920 [44:46<20:18, 2.51it/s, tok/s=14554.0]
Inference: 48%|████▊ | 2866/5920 [44:47<17:10, 2.96it/s, tok/s=14554.0]
Inference: 48%|████▊ | 2866/5920 [44:47<17:10, 2.96it/s, tok/s=14552.0]
Inference: 48%|████▊ | 2867/5920 [44:47<14:54, 3.41it/s, tok/s=14552.0]
Inference: 48%|████▊ | 2867/5920 [44:47<14:54, 3.41it/s, tok/s=14560.4]
Inference: 48%|████▊ | 2868/5920 [44:47<15:35, 3.26it/s, tok/s=14560.4]
Inference: 48%|████▊ | 2868/5920 [44:47<15:35, 3.26it/s, tok/s=14559.6]
Inference: 48%|████▊ | 2869/5920 [44:47<15:35, 3.26it/s, tok/s=14560.5]
Inference: 48%|████▊ | 2870/5920 [44:47<15:34, 3.26it/s, tok/s=14560.9]
Inference: 48%|████▊ | 2871/5920 [44:47<09:07, 5.57it/s, tok/s=14560.9]
Inference: 48%|████▊ | 2871/5920 [44:47<09:07, 5.57it/s, tok/s=14561.9]
Inference: 49%|████▊ | 2872/5920 [44:47<09:07, 5.57it/s, tok/s=14570.7]
Inference: 49%|████▊ | 2873/5920 [44:48<10:14, 4.96it/s, tok/s=14570.7]
Inference: 49%|████▊ | 2873/5920 [44:48<10:14, 4.96it/s, tok/s=14568.9]
Inference: 49%|████▊ | 2874/5920 [44:50<28:05, 1.81it/s, tok/s=14568.9]
Inference: 49%|████▊ | 2874/5920 [44:50<28:05, 1.81it/s, tok/s=14559.2]
Inference: 49%|████▊ | 2875/5920 [44:51<31:07, 1.63it/s, tok/s=14559.2]
Inference: 49%|████▊ | 2875/5920 [44:51<31:07, 1.63it/s, tok/s=14555.1]
Inference: 49%|████▊ | 2876/5920 [44:51<25:41, 1.97it/s, tok/s=14555.1]
Inference: 49%|████▊ | 2876/5920 [44:51<25:41, 1.97it/s, tok/s=14554.6]
Inference: 49%|████▊ | 2877/5920 [44:51<25:41, 1.97it/s, tok/s=14557.5]
Inference: 49%|████▊ | 2878/5920 [44:51<16:39, 3.04it/s, tok/s=14557.5]
Inference: 49%|████▊ | 2878/5920 [44:51<16:39, 3.04it/s, tok/s=14557.7]
Inference: 49%|████▊ | 2879/5920 [44:51<14:04, 3.60it/s, tok/s=14557.7]
Inference: 49%|████▊ | 2879/5920 [44:51<14:04, 3.60it/s, tok/s=14559.3]
Inference: 49%|████▊ | 2880/5920 [44:51<14:36, 3.47it/s, tok/s=14559.3]
Inference: 49%|████▊ | 2880/5920 [44:51<14:36, 3.47it/s, tok/s=14558.5]
Inference: 49%|████▊ | 2881/5920 [44:52<17:03, 2.97it/s, tok/s=14558.5]
Inference: 49%|████▊ | 2881/5920 [44:52<17:03, 2.97it/s, tok/s=14556.4]
Inference: 49%|████▊ | 2882/5920 [44:52<17:03, 2.97it/s, tok/s=14557.3]
Inference: 49%|████▊ | 2883/5920 [44:53<17:51, 2.83it/s, tok/s=14557.3]
Inference: 49%|████▊ | 2883/5920 [44:53<17:51, 2.83it/s, tok/s=14553.9]
Inference: 49%|████▊ | 2884/5920 [44:53<20:28, 2.47it/s, tok/s=14553.9]
Inference: 49%|████▊ | 2884/5920 [44:53<20:28, 2.47it/s, tok/s=14551.5]
Inference: 49%|████▊ | 2885/5920 [44:53<17:24, 2.91it/s, tok/s=14551.5]
Inference: 49%|████▊ | 2885/5920 [44:53<17:24, 2.91it/s, tok/s=14551.8]
Inference: 49%|████▉ | 2886/5920 [44:53<17:24, 2.91it/s, tok/s=14552.1]
Inference: 49%|████▉ | 2887/5920 [44:54<15:10, 3.33it/s, tok/s=14552.1]
Inference: 49%|████▉ | 2887/5920 [44:54<15:10, 3.33it/s, tok/s=14551.0]
Inference: 49%|████▉ | 2888/5920 [44:54<13:52, 3.64it/s, tok/s=14551.0]
Inference: 49%|████▉ | 2888/5920 [44:54<13:52, 3.64it/s, tok/s=14551.0]
Inference: 49%|████▉ | 2889/5920 [44:54<15:51, 3.19it/s, tok/s=14551.0]
Inference: 49%|████▉ | 2889/5920 [44:54<15:51, 3.19it/s, tok/s=14549.1]
Inference: 49%|████▉ | 2890/5920 [44:55<19:01, 2.66it/s, tok/s=14549.1]
Inference: 49%|████▉ | 2890/5920 [44:55<19:01, 2.66it/s, tok/s=14547.6]
Inference: 49%|████▉ | 2891/5920 [44:55<16:27, 3.07it/s, tok/s=14547.6]
Inference: 49%|████▉ | 2891/5920 [44:55<16:27, 3.07it/s, tok/s=14548.9]
Inference: 49%|████▉ | 2892/5920 [44:55<16:26, 3.07it/s, tok/s=14550.0]
Inference: 49%|████▉ | 2893/5920 [44:56<15:47, 3.20it/s, tok/s=14550.0]
Inference: 49%|████▉ | 2893/5920 [44:56<15:47, 3.20it/s, tok/s=14548.3]
Inference: 49%|████▉ | 2894/5920 [44:56<18:19, 2.75it/s, tok/s=14548.3]
Inference: 49%|████▉ | 2894/5920 [44:56<18:19, 2.75it/s, tok/s=14546.5]
Inference: 49%|████▉ | 2895/5920 [44:57<17:22, 2.90it/s, tok/s=14546.5]
Inference: 49%|████▉ | 2895/5920 [44:57<17:22, 2.90it/s, tok/s=14545.5]
Inference: 49%|████▉ | 2896/5920 [44:57<14:49, 3.40it/s, tok/s=14545.5]
Inference: 49%|████▉ | 2896/5920 [44:57<14:49, 3.40it/s, tok/s=14545.4]
Inference: 49%|████▉ | 2897/5920 [44:57<14:00, 3.60it/s, tok/s=14545.4]
Inference: 49%|████▉ | 2897/5920 [44:57<14:00, 3.60it/s, tok/s=14544.9]
Inference: 49%|████▉ | 2898/5920 [44:58<20:33, 2.45it/s, tok/s=14544.9]
Inference: 49%|████▉ | 2898/5920 [44:58<20:33, 2.45it/s, tok/s=14542.3]
Inference: 49%|████▉ | 2899/5920 [44:58<20:33, 2.45it/s, tok/s=14543.4]
Inference: 49%|████▉ | 2900/5920 [44:58<17:12, 2.93it/s, tok/s=14543.4]
Inference: 49%|████▉ | 2900/5920 [44:58<17:12, 2.93it/s, tok/s=14541.2]
Inference: 49%|████▉ | 2901/5920 [44:58<15:26, 3.26it/s, tok/s=14541.2]
Inference: 49%|████▉ | 2901/5920 [44:58<15:26, 3.26it/s, tok/s=14541.3]
Inference: 49%|████▉ | 2902/5920 [44:58<15:25, 3.26it/s, tok/s=14541.8]
Inference: 49%|████▉ | 2903/5920 [44:58<15:25, 3.26it/s, tok/s=14542.8]
Inference: 49%|████▉ | 2904/5920 [44:59<09:43, 5.17it/s, tok/s=14542.8]
Inference: 49%|████▉ | 2904/5920 [44:59<09:43, 5.17it/s, tok/s=14543.0]
Inference: 49%|████▉ | 2905/5920 [44:59<09:43, 5.17it/s, tok/s=14543.5]
Inference: 49%|████▉ | 2906/5920 [45:00<15:29, 3.24it/s, tok/s=14543.5]
Inference: 49%|████▉ | 2906/5920 [45:00<15:29, 3.24it/s, tok/s=14545.9]
Inference: 49%|████▉ | 2907/5920 [45:00<13:34, 3.70it/s, tok/s=14545.9]
Inference: 49%|████▉ | 2907/5920 [45:00<13:34, 3.70it/s, tok/s=14546.0]
Inference: 49%|████▉ | 2908/5920 [45:00<12:42, 3.95it/s, tok/s=14546.0]
Inference: 49%|████▉ | 2908/5920 [45:00<12:42, 3.95it/s, tok/s=14554.4]
Inference: 49%|████▉ | 2909/5920 [45:00<14:29, 3.46it/s, tok/s=14554.4]
Inference: 49%|████▉ | 2909/5920 [45:00<14:29, 3.46it/s, tok/s=14559.8]
Inference: 49%|████▉ | 2910/5920 [45:02<33:36, 1.49it/s, tok/s=14559.8]
Inference: 49%|████▉ | 2910/5920 [45:02<33:36, 1.49it/s, tok/s=14559.8]
Inference: 49%|████▉ | 2911/5920 [45:02<26:21, 1.90it/s, tok/s=14559.8]
Inference: 49%|████▉ | 2911/5920 [45:02<26:21, 1.90it/s, tok/s=14560.9]
Inference: 49%|████▉ | 2912/5920 [45:03<21:27, 2.34it/s, tok/s=14560.9]
Inference: 49%|████▉ | 2912/5920 [45:03<21:27, 2.34it/s, tok/s=14561.9]
Inference: 49%|████▉ | 2913/5920 [45:03<17:18, 2.90it/s, tok/s=14561.9]
Inference: 49%|████▉ | 2913/5920 [45:03<17:18, 2.90it/s, tok/s=14562.4]
Inference: 49%|████▉ | 2914/5920 [45:03<17:31, 2.86it/s, tok/s=14562.4]
Inference: 49%|████▉ | 2914/5920 [45:03<17:31, 2.86it/s, tok/s=14561.2]
Inference: 49%|████▉ | 2915/5920 [45:03<15:44, 3.18it/s, tok/s=14561.2]
Inference: 49%|████▉ | 2915/5920 [45:03<15:44, 3.18it/s, tok/s=14560.5]
Inference: 49%|████▉ | 2916/5920 [45:03<12:58, 3.86it/s, tok/s=14560.5]
Inference: 49%|████▉ | 2916/5920 [45:03<12:58, 3.86it/s, tok/s=14560.5]
Inference: 49%|████▉ | 2917/5920 [45:03<12:58, 3.86it/s, tok/s=14561.8]
Inference: 49%|████▉ | 2918/5920 [45:04<13:09, 3.80it/s, tok/s=14561.8]
Inference: 49%|████▉ | 2918/5920 [45:04<13:09, 3.80it/s, tok/s=14559.8]
Inference: 49%|████▉ | 2919/5920 [45:04<11:16, 4.44it/s, tok/s=14559.8]
Inference: 49%|████▉ | 2919/5920 [45:04<11:16, 4.44it/s, tok/s=14565.2]
Inference: 49%|████▉ | 2920/5920 [45:04<11:16, 4.44it/s, tok/s=14576.0]
Inference: 49%|████▉ | 2921/5920 [45:06<23:57, 2.09it/s, tok/s=14576.0]
Inference: 49%|████▉ | 2921/5920 [45:06<23:57, 2.09it/s, tok/s=14567.9]
Inference: 49%|████▉ | 2922/5920 [45:06<21:51, 2.29it/s, tok/s=14567.9]
Inference: 49%|████▉ | 2922/5920 [45:06<21:51, 2.29it/s, tok/s=14577.5]
Inference: 49%|████▉ | 2923/5920 [45:06<18:11, 2.75it/s, tok/s=14577.5]
Inference: 49%|████▉ | 2923/5920 [45:06<18:11, 2.75it/s, tok/s=14577.5]
Inference: 49%|████▉ | 2924/5920 [45:06<18:11, 2.75it/s, tok/s=14579.0]
Inference: 49%|████▉ | 2925/5920 [45:07<18:30, 2.70it/s, tok/s=14579.0]
Inference: 49%|████▉ | 2925/5920 [45:07<18:30, 2.70it/s, tok/s=14584.4]
Inference: 49%|████▉ | 2926/5920 [45:07<15:30, 3.22it/s, tok/s=14584.4]
Inference: 49%|████▉ | 2926/5920 [45:07<15:30, 3.22it/s, tok/s=14585.7]
Inference: 49%|████▉ | 2927/5920 [45:07<15:14, 3.27it/s, tok/s=14585.7]
Inference: 49%|████▉ | 2927/5920 [45:07<15:14, 3.27it/s, tok/s=14584.9]
Inference: 49%|████▉ | 2928/5920 [45:07<12:37, 3.95it/s, tok/s=14584.9]
Inference: 49%|████▉ | 2928/5920 [45:07<12:37, 3.95it/s, tok/s=14584.9]
Inference: 49%|████▉ | 2929/5920 [45:08<19:45, 2.52it/s, tok/s=14584.9]
Inference: 49%|████▉ | 2929/5920 [45:08<19:45, 2.52it/s, tok/s=14581.3]
Inference: 49%|████▉ | 2930/5920 [45:09<21:18, 2.34it/s, tok/s=14581.3]
Inference: 49%|████▉ | 2930/5920 [45:09<21:18, 2.34it/s, tok/s=14579.3]
Inference: 50%|████▉ | 2931/5920 [45:09<22:49, 2.18it/s, tok/s=14579.3]
Inference: 50%|████▉ | 2931/5920 [45:09<22:49, 2.18it/s, tok/s=14577.2]
Inference: 50%|████▉ | 2932/5920 [45:09<22:48, 2.18it/s, tok/s=14577.7]
Inference: 50%|████▉ | 2933/5920 [45:10<15:16, 3.26it/s, tok/s=14577.7]
Inference: 50%|████▉ | 2933/5920 [45:10<15:16, 3.26it/s, tok/s=14577.8]
Inference: 50%|████▉ | 2934/5920 [45:10<19:19, 2.58it/s, tok/s=14577.8]
Inference: 50%|████▉ | 2934/5920 [45:10<19:19, 2.58it/s, tok/s=14576.0]
Inference: 50%|████▉ | 2935/5920 [45:10<15:42, 3.17it/s, tok/s=14576.0]
Inference: 50%|████▉ | 2935/5920 [45:10<15:42, 3.17it/s, tok/s=14584.3]
Inference: 50%|████▉ | 2936/5920 [45:11<18:03, 2.75it/s, tok/s=14584.3]
Inference: 50%|████▉ | 2936/5920 [45:11<18:03, 2.75it/s, tok/s=14582.6]
Inference: 50%|████▉ | 2937/5920 [45:11<17:58, 2.76it/s, tok/s=14582.6]
Inference: 50%|████▉ | 2937/5920 [45:11<17:58, 2.76it/s, tok/s=14581.7]
Inference: 50%|████▉ | 2938/5920 [45:12<19:11, 2.59it/s, tok/s=14581.7]
Inference: 50%|████▉ | 2938/5920 [45:12<19:11, 2.59it/s, tok/s=14585.3]
Inference: 50%|████▉ | 2939/5920 [45:12<19:11, 2.59it/s, tok/s=14587.0]
Inference: 50%|████▉ | 2940/5920 [45:12<12:37, 3.93it/s, tok/s=14587.0]
Inference: 50%|████▉ | 2940/5920 [45:12<12:37, 3.93it/s, tok/s=14588.1]
Inference: 50%|████▉ | 2941/5920 [45:12<12:37, 3.93it/s, tok/s=14597.3]
Inference: 50%|████▉ | 2942/5920 [45:13<18:01, 2.75it/s, tok/s=14597.3]
Inference: 50%|████▉ | 2942/5920 [45:13<18:01, 2.75it/s, tok/s=14593.3]
Inference: 50%|████▉ | 2943/5920 [45:13<15:17, 3.25it/s, tok/s=14593.3]
Inference: 50%|████▉ | 2943/5920 [45:13<15:17, 3.25it/s, tok/s=14593.5]
Inference: 50%|████▉ | 2944/5920 [45:13<14:36, 3.39it/s, tok/s=14593.5]
Inference: 50%|████▉ | 2944/5920 [45:13<14:36, 3.39it/s, tok/s=14594.1]
Inference: 50%|████▉ | 2945/5920 [45:14<16:04, 3.08it/s, tok/s=14594.1]
Inference: 50%|████▉ | 2945/5920 [45:14<16:04, 3.08it/s, tok/s=14593.5]
Inference: 50%|████▉ | 2946/5920 [45:14<18:21, 2.70it/s, tok/s=14593.5]
Inference: 50%|████▉ | 2946/5920 [45:14<18:21, 2.70it/s, tok/s=14597.3]
Inference: 50%|████▉ | 2947/5920 [45:14<16:18, 3.04it/s, tok/s=14597.3]
Inference: 50%|████▉ | 2947/5920 [45:14<16:18, 3.04it/s, tok/s=14597.1]
Inference: 50%|████▉ | 2948/5920 [45:15<23:54, 2.07it/s, tok/s=14597.1]
Inference: 50%|████▉ | 2948/5920 [45:15<23:54, 2.07it/s, tok/s=14599.0]
Inference: 50%|████▉ | 2949/5920 [45:15<23:54, 2.07it/s, tok/s=14600.2]
Inference: 50%|████▉ | 2950/5920 [45:15<23:53, 2.07it/s, tok/s=14600.8]
Inference: 50%|████▉ | 2951/5920 [45:16<17:07, 2.89it/s, tok/s=14600.8]
Inference: 50%|████▉ | 2951/5920 [45:16<17:07, 2.89it/s, tok/s=14598.0]
Inference: 50%|████▉ | 2952/5920 [45:16<17:07, 2.89it/s, tok/s=14603.8]
Inference: 50%|████▉ | 2953/5920 [45:16<12:26, 3.98it/s, tok/s=14603.8]
Inference: 50%|████▉ | 2953/5920 [45:16<12:26, 3.98it/s, tok/s=14612.9]
Inference: 50%|████▉ | 2954/5920 [45:16<14:37, 3.38it/s, tok/s=14612.9]
Inference: 50%|████▉ | 2954/5920 [45:16<14:37, 3.38it/s, tok/s=14612.2]
Inference: 50%|████▉ | 2955/5920 [45:17<20:33, 2.40it/s, tok/s=14612.2]
Inference: 50%|████▉ | 2955/5920 [45:17<20:33, 2.40it/s, tok/s=14615.6]
Inference: 50%|████▉ | 2956/5920 [45:18<18:33, 2.66it/s, tok/s=14615.6]
Inference: 50%|████▉ | 2956/5920 [45:18<18:33, 2.66it/s, tok/s=14615.7]
Inference: 50%|████▉ | 2957/5920 [45:19<30:38, 1.61it/s, tok/s=14615.7]
Inference: 50%|████▉ | 2957/5920 [45:19<30:38, 1.61it/s, tok/s=14615.2]
Inference: 50%|████▉ | 2958/5920 [45:20<32:23, 1.52it/s, tok/s=14615.2]
Inference: 50%|████▉ | 2958/5920 [45:20<32:23, 1.52it/s, tok/s=14617.8]
Inference: 50%|████▉ | 2959/5920 [45:20<25:44, 1.92it/s, tok/s=14617.8]
Inference: 50%|████▉ | 2959/5920 [45:20<25:44, 1.92it/s, tok/s=14617.8]
Inference: 50%|█████ | 2960/5920 [45:20<20:07, 2.45it/s, tok/s=14617.8]
Inference: 50%|█████ | 2960/5920 [45:20<20:07, 2.45it/s, tok/s=14617.7]
Inference: 50%|█████ | 2961/5920 [45:20<20:47, 2.37it/s, tok/s=14617.7]
Inference: 50%|█████ | 2961/5920 [45:20<20:47, 2.37it/s, tok/s=14620.5]
Inference: 50%|█████ | 2962/5920 [45:21<20:31, 2.40it/s, tok/s=14620.5]
Inference: 50%|█████ | 2962/5920 [45:21<20:31, 2.40it/s, tok/s=14619.3]
Inference: 50%|█████ | 2963/5920 [45:22<29:39, 1.66it/s, tok/s=14619.3]
Inference: 50%|█████ | 2963/5920 [45:22<29:39, 1.66it/s, tok/s=14623.5]
Inference: 50%|█████ | 2964/5920 [45:22<26:31, 1.86it/s, tok/s=14623.5]
Inference: 50%|█████ | 2964/5920 [45:22<26:31, 1.86it/s, tok/s=14622.6]
Inference: 50%|█████ | 2965/5920 [45:22<21:29, 2.29it/s, tok/s=14622.6]
Inference: 50%|█████ | 2965/5920 [45:22<21:29, 2.29it/s, tok/s=14627.1]
Inference: 50%|█████ | 2966/5920 [45:23<23:52, 2.06it/s, tok/s=14627.1]
Inference: 50%|█████ | 2966/5920 [45:23<23:52, 2.06it/s, tok/s=14625.2]
Inference: 50%|█████ | 2967/5920 [45:23<18:48, 2.62it/s, tok/s=14625.2]
Inference: 50%|█████ | 2967/5920 [45:23<18:48, 2.62it/s, tok/s=14633.6]
Inference: 50%|█████ | 2968/5920 [45:24<18:12, 2.70it/s, tok/s=14633.6]
Inference: 50%|█████ | 2968/5920 [45:24<18:12, 2.70it/s, tok/s=14633.6]
Inference: 50%|█████ | 2969/5920 [45:24<18:47, 2.62it/s, tok/s=14633.6]
Inference: 50%|█████ | 2969/5920 [45:24<18:47, 2.62it/s, tok/s=14632.0]
Inference: 50%|█████ | 2970/5920 [45:24<18:46, 2.62it/s, tok/s=14633.0]
Inference: 50%|█████ | 2971/5920 [45:25<20:21, 2.41it/s, tok/s=14633.0]
Inference: 50%|█████ | 2971/5920 [45:25<20:21, 2.41it/s, tok/s=14629.9]
Inference: 50%|█████ | 2972/5920 [45:25<20:03, 2.45it/s, tok/s=14629.9]
Inference: 50%|█████ | 2972/5920 [45:25<20:03, 2.45it/s, tok/s=14629.3]
Inference: 50%|█████ | 2973/5920 [45:25<17:52, 2.75it/s, tok/s=14629.3]
Inference: 50%|█████ | 2973/5920 [45:25<17:52, 2.75it/s, tok/s=14629.4]
Inference: 50%|█████ | 2974/5920 [45:26<17:45, 2.77it/s, tok/s=14629.4]
Inference: 50%|█████ | 2974/5920 [45:26<17:45, 2.77it/s, tok/s=14633.5]
Inference: 50%|█████ | 2975/5920 [45:27<23:37, 2.08it/s, tok/s=14633.5]
Inference: 50%|█████ | 2975/5920 [45:27<23:37, 2.08it/s, tok/s=14630.1]
Inference: 50%|█████ | 2976/5920 [45:28<38:37, 1.27it/s, tok/s=14630.1]
Inference: 50%|█████ | 2976/5920 [45:28<38:37, 1.27it/s, tok/s=14622.3]
Inference: 50%|█████ | 2977/5920 [45:29<37:40, 1.30it/s, tok/s=14622.3]
Inference: 50%|█████ | 2977/5920 [45:29<37:40, 1.30it/s, tok/s=14618.9]
Inference: 50%|█████ | 2978/5920 [45:29<37:39, 1.30it/s, tok/s=14619.4]
Inference: 50%|█████ | 2979/5920 [45:29<21:57, 2.23it/s, tok/s=14619.4]
Inference: 50%|█████ | 2979/5920 [45:29<21:57, 2.23it/s, tok/s=14619.7]
Inference: 50%|█████ | 2980/5920 [45:30<24:12, 2.02it/s, tok/s=14619.7]
Inference: 50%|█████ | 2980/5920 [45:30<24:12, 2.02it/s, tok/s=14616.7]
Inference: 50%|█████ | 2981/5920 [45:30<19:51, 2.47it/s, tok/s=14616.7]
Inference: 50%|█████ | 2981/5920 [45:30<19:51, 2.47it/s, tok/s=14617.1]
Inference: 50%|█████ | 2982/5920 [45:31<36:22, 1.35it/s, tok/s=14617.1]
Inference: 50%|█████ | 2982/5920 [45:31<36:22, 1.35it/s, tok/s=14608.7]
Inference: 50%|█████ | 2983/5920 [45:31<36:22, 1.35it/s, tok/s=14609.4]
Inference: 50%|█████ | 2984/5920 [45:31<36:21, 1.35it/s, tok/s=14610.0]
Inference: 50%|█████ | 2985/5920 [45:32<19:27, 2.51it/s, tok/s=14610.0]
Inference: 50%|█████ | 2985/5920 [45:32<19:27, 2.51it/s, tok/s=14609.3]
Inference: 50%|█████ | 2986/5920 [45:32<19:27, 2.51it/s, tok/s=14609.5]
Inference: 50%|█████ | 2987/5920 [45:33<20:45, 2.36it/s, tok/s=14609.5]
Inference: 50%|█████ | 2987/5920 [45:33<20:45, 2.36it/s, tok/s=14605.8]
Inference: 50%|█████ | 2988/5920 [45:33<20:00, 2.44it/s, tok/s=14605.8]
Inference: 50%|█████ | 2988/5920 [45:33<20:00, 2.44it/s, tok/s=14604.4]
Inference: 50%|█████ | 2989/5920 [45:33<17:09, 2.85it/s, tok/s=14604.4]
Inference: 50%|█████ | 2989/5920 [45:33<17:09, 2.85it/s, tok/s=14612.9]
Inference: 51%|█████ | 2990/5920 [45:33<17:09, 2.85it/s, tok/s=14618.7]
Inference: 51%|█████ | 2991/5920 [45:34<17:33, 2.78it/s, tok/s=14618.7]
Inference: 51%|█████ | 2991/5920 [45:34<17:33, 2.78it/s, tok/s=14616.0]
Inference: 51%|█████ | 2992/5920 [45:34<18:26, 2.65it/s, tok/s=14616.0]
Inference: 51%|█████ | 2992/5920 [45:34<18:26, 2.65it/s, tok/s=14614.4]
Inference: 51%|█████ | 2993/5920 [45:35<19:12, 2.54it/s, tok/s=14614.4]
Inference: 51%|█████ | 2993/5920 [45:35<19:12, 2.54it/s, tok/s=14612.9]
Inference: 51%|█████ | 2994/5920 [45:35<19:11, 2.54it/s, tok/s=14613.4]
Inference: 51%|█████ | 2995/5920 [45:35<14:59, 3.25it/s, tok/s=14613.4]
Inference: 51%|█████ | 2995/5920 [45:35<14:59, 3.25it/s, tok/s=14621.9]
Inference: 51%|█████ | 2996/5920 [45:35<12:50, 3.80it/s, tok/s=14621.9]
Inference: 51%|█████ | 2996/5920 [45:35<12:50, 3.80it/s, tok/s=14622.2]
Inference: 51%|█████ | 2997/5920 [45:36<15:12, 3.20it/s, tok/s=14622.2]
Inference: 51%|█████ | 2997/5920 [45:36<15:12, 3.20it/s, tok/s=14622.0]
Inference: 51%|█████ | 2998/5920 [45:37<24:17, 2.00it/s, tok/s=14622.0]
Inference: 51%|█████ | 2998/5920 [45:37<24:17, 2.00it/s, tok/s=14623.5]
Inference: 51%|█████ | 2999/5920 [45:38<30:22, 1.60it/s, tok/s=14623.5]
Inference: 51%|█████ | 2999/5920 [45:38<30:22, 1.60it/s, tok/s=14618.8]
Inference: 51%|█████ | 3000/5920 [45:39<32:43, 1.49it/s, tok/s=14618.8]
Inference: 51%|█████ | 3000/5920 [45:39<32:43, 1.49it/s, tok/s=14620.7]
Inference: 51%|█████ | 3001/5920 [45:39<36:18, 1.34it/s, tok/s=14620.7]
Inference: 51%|█████ | 3001/5920 [45:39<36:18, 1.34it/s, tok/s=14616.5]
Inference: 51%|█████ | 3002/5920 [45:40<30:49, 1.58it/s, tok/s=14616.5]
Inference: 51%|█████ | 3002/5920 [45:40<30:49, 1.58it/s, tok/s=14615.3]
Inference: 51%|█████ | 3003/5920 [45:40<24:57, 1.95it/s, tok/s=14615.3]
Inference: 51%|█████ | 3003/5920 [45:40<24:57, 1.95it/s, tok/s=14615.4]
Inference: 51%|█████ | 3004/5920 [45:40<20:40, 2.35it/s, tok/s=14615.4]
Inference: 51%|█████ | 3004/5920 [45:40<20:40, 2.35it/s, tok/s=14619.9]
Inference: 51%|█████ | 3005/5920 [45:41<17:44, 2.74it/s, tok/s=14619.9]
Inference: 51%|█████ | 3005/5920 [45:41<17:44, 2.74it/s, tok/s=14619.9]
Inference: 51%|█████ | 3006/5920 [45:41<19:52, 2.44it/s, tok/s=14619.9]
Inference: 51%|█████ | 3006/5920 [45:41<19:52, 2.44it/s, tok/s=14618.3]
Inference: 51%|█████ | 3007/5920 [45:41<18:58, 2.56it/s, tok/s=14618.3]
Inference: 51%|█████ | 3007/5920 [45:41<18:58, 2.56it/s, tok/s=14617.5]
Inference: 51%|█████ | 3008/5920 [45:41<18:57, 2.56it/s, tok/s=14623.8]
Inference: 51%|█████ | 3009/5920 [45:41<11:21, 4.27it/s, tok/s=14623.8]
Inference: 51%|█████ | 3009/5920 [45:41<11:21, 4.27it/s, tok/s=14625.5]
Inference: 51%|█████ | 3010/5920 [45:42<15:03, 3.22it/s, tok/s=14625.5]
Inference: 51%|█████ | 3010/5920 [45:42<15:03, 3.22it/s, tok/s=14629.7]
Inference: 51%|█████ | 3011/5920 [45:43<22:32, 2.15it/s, tok/s=14629.7]
Inference: 51%|█████ | 3011/5920 [45:43<22:32, 2.15it/s, tok/s=14628.3]
Inference: 51%|█████ | 3012/5920 [45:43<18:37, 2.60it/s, tok/s=14628.3]
Inference: 51%|█████ | 3012/5920 [45:43<18:37, 2.60it/s, tok/s=14629.0]
Inference: 51%|█████ | 3013/5920 [45:43<18:36, 2.60it/s, tok/s=14632.1]
Inference: 51%|█████ | 3014/5920 [45:43<14:33, 3.33it/s, tok/s=14632.1]
Inference: 51%|█████ | 3014/5920 [45:43<14:33, 3.33it/s, tok/s=14631.4]
Inference: 51%|█████ | 3015/5920 [45:44<14:10, 3.42it/s, tok/s=14631.4]
Inference: 51%|█████ | 3015/5920 [45:44<14:10, 3.42it/s, tok/s=14630.6]
Inference: 51%|█████ | 3016/5920 [45:44<14:07, 3.43it/s, tok/s=14630.6]
Inference: 51%|█████ | 3016/5920 [45:44<14:07, 3.43it/s, tok/s=14632.5]
Inference: 51%|█████ | 3017/5920 [45:45<19:55, 2.43it/s, tok/s=14632.5]
Inference: 51%|█████ | 3017/5920 [45:45<19:55, 2.43it/s, tok/s=14629.0]
Inference: 51%|█████ | 3018/5920 [45:45<16:08, 3.00it/s, tok/s=14629.0]
Inference: 51%|█████ | 3018/5920 [45:45<16:08, 3.00it/s, tok/s=14629.2]
Inference: 51%|█████ | 3019/5920 [45:45<14:21, 3.37it/s, tok/s=14629.2]
Inference: 51%|█████ | 3019/5920 [45:45<14:21, 3.37it/s, tok/s=14629.6]
Inference: 51%|█████ | 3020/5920 [45:46<19:18, 2.50it/s, tok/s=14629.6]
Inference: 51%|█████ | 3020/5920 [45:46<19:18, 2.50it/s, tok/s=14627.0]
Inference: 51%|█████ | 3021/5920 [45:47<35:11, 1.37it/s, tok/s=14627.0]
Inference: 51%|█████ | 3021/5920 [45:47<35:11, 1.37it/s, tok/s=14619.9]
Inference: 51%|█████ | 3022/5920 [45:47<35:10, 1.37it/s, tok/s=14623.4]
Inference: 51%|█████ | 3023/5920 [45:48<27:18, 1.77it/s, tok/s=14623.4]
Inference: 51%|█████ | 3023/5920 [45:48<27:18, 1.77it/s, tok/s=14622.8]
Inference: 51%|█████ | 3024/5920 [45:48<21:49, 2.21it/s, tok/s=14622.8]
Inference: 51%|█████ | 3024/5920 [45:48<21:49, 2.21it/s, tok/s=14623.0]
Inference: 51%|█████ | 3025/5920 [45:49<23:04, 2.09it/s, tok/s=14623.0]
Inference: 51%|█████ | 3025/5920 [45:49<23:04, 2.09it/s, tok/s=14628.6]
Inference: 51%|█████ | 3026/5920 [45:49<23:03, 2.09it/s, tok/s=14629.0]
Inference: 51%|█████ | 3027/5920 [45:49<14:54, 3.23it/s, tok/s=14629.0]
Inference: 51%|█████ | 3027/5920 [45:49<14:54, 3.23it/s, tok/s=14629.4]
Inference: 51%|█████ | 3028/5920 [45:49<14:52, 3.24it/s, tok/s=14629.4]
Inference: 51%|█████ | 3028/5920 [45:49<14:52, 3.24it/s, tok/s=14628.1]
Inference: 51%|█████ | 3029/5920 [45:49<12:36, 3.82it/s, tok/s=14628.1]
Inference: 51%|█████ | 3029/5920 [45:49<12:36, 3.82it/s, tok/s=14632.4]
Inference: 51%|█████ | 3030/5920 [45:50<17:20, 2.78it/s, tok/s=14632.4]
Inference: 51%|█████ | 3030/5920 [45:50<17:20, 2.78it/s, tok/s=14629.4]
Inference: 51%|█████ | 3031/5920 [45:50<16:51, 2.86it/s, tok/s=14629.4]
Inference: 51%|█████ | 3031/5920 [45:50<16:51, 2.86it/s, tok/s=14628.6]
Inference: 51%|█████ | 3032/5920 [45:50<14:22, 3.35it/s, tok/s=14628.6]
Inference: 51%|█████ | 3032/5920 [45:50<14:22, 3.35it/s, tok/s=14628.3]
Inference: 51%|█████ | 3033/5920 [45:51<22:56, 2.10it/s, tok/s=14628.3]
Inference: 51%|█████ | 3033/5920 [45:51<22:56, 2.10it/s, tok/s=14624.0]
Inference: 51%|█████▏ | 3034/5920 [45:52<29:12, 1.65it/s, tok/s=14624.0]
Inference: 51%|█████▏ | 3034/5920 [45:52<29:12, 1.65it/s, tok/s=14621.9]
Inference: 51%|█████▏ | 3035/5920 [45:52<22:09, 2.17it/s, tok/s=14621.9]
Inference: 51%|█████▏ | 3035/5920 [45:52<22:09, 2.17it/s, tok/s=14622.8]
Inference: 51%|█████▏ | 3036/5920 [45:53<20:39, 2.33it/s, tok/s=14622.8]
Inference: 51%|█████▏ | 3036/5920 [45:53<20:39, 2.33it/s, tok/s=14627.8]
Inference: 51%|█████▏ | 3037/5920 [45:53<19:27, 2.47it/s, tok/s=14627.8]
Inference: 51%|█████▏ | 3037/5920 [45:53<19:27, 2.47it/s, tok/s=14634.1]
Inference: 51%|█████▏ | 3038/5920 [45:53<17:50, 2.69it/s, tok/s=14634.1]
Inference: 51%|█████▏ | 3038/5920 [45:53<17:50, 2.69it/s, tok/s=14633.0]
Inference: 51%|█████▏ | 3039/5920 [45:53<17:49, 2.69it/s, tok/s=14633.2]
Inference: 51%|█████▏ | 3040/5920 [45:54<13:53, 3.46it/s, tok/s=14633.2]
Inference: 51%|█████▏ | 3040/5920 [45:54<13:53, 3.46it/s, tok/s=14632.0]
Inference: 51%|█████▏ | 3041/5920 [45:54<12:13, 3.93it/s, tok/s=14632.0]
Inference: 51%|█████▏ | 3041/5920 [45:54<12:13, 3.93it/s, tok/s=14641.0]
Inference: 51%|█████▏ | 3042/5920 [45:55<26:55, 1.78it/s, tok/s=14641.0]
Inference: 51%|█████▏ | 3042/5920 [45:55<26:55, 1.78it/s, tok/s=14636.5]
Inference: 51%|█████▏ | 3043/5920 [45:57<40:05, 1.20it/s, tok/s=14636.5]
Inference: 51%|█████▏ | 3043/5920 [45:57<40:05, 1.20it/s, tok/s=14628.6]
Inference: 51%|█████▏ | 3044/5920 [45:57<32:06, 1.49it/s, tok/s=14628.6]
Inference: 51%|█████▏ | 3044/5920 [45:57<32:06, 1.49it/s, tok/s=14628.2]
Inference: 51%|█████▏ | 3045/5920 [45:58<28:42, 1.67it/s, tok/s=14628.2]
Inference: 51%|█████▏ | 3045/5920 [45:58<28:42, 1.67it/s, tok/s=14626.7]
Inference: 51%|█████▏ | 3046/5920 [45:58<23:15, 2.06it/s, tok/s=14626.7]
Inference: 51%|█████▏ | 3046/5920 [45:58<23:15, 2.06it/s, tok/s=14627.9]
Inference: 51%|█████▏ | 3047/5920 [45:58<18:15, 2.62it/s, tok/s=14627.9]
Inference: 51%|█████▏ | 3047/5920 [45:58<18:15, 2.62it/s, tok/s=14632.0]
Inference: 51%|█████▏ | 3048/5920 [45:58<17:32, 2.73it/s, tok/s=14632.0]
Inference: 51%|█████▏ | 3048/5920 [45:58<17:32, 2.73it/s, tok/s=14631.2]
Inference: 52%|█████▏ | 3049/5920 [45:59<20:49, 2.30it/s, tok/s=14631.2]
Inference: 52%|█████▏ | 3049/5920 [45:59<20:49, 2.30it/s, tok/s=14628.9]
Inference: 52%|█████▏ | 3050/5920 [45:59<21:36, 2.21it/s, tok/s=14628.9]
Inference: 52%|█████▏ | 3050/5920 [45:59<21:36, 2.21it/s, tok/s=14627.2]
Inference: 52%|█████▏ | 3051/5920 [46:00<22:24, 2.13it/s, tok/s=14627.2]
Inference: 52%|█████▏ | 3051/5920 [46:00<22:24, 2.13it/s, tok/s=14624.9]
Inference: 52%|█████▏ | 3052/5920 [46:01<26:40, 1.79it/s, tok/s=14624.9]
Inference: 52%|█████▏ | 3052/5920 [46:01<26:40, 1.79it/s, tok/s=14622.8]
Inference: 52%|█████▏ | 3053/5920 [46:02<34:03, 1.40it/s, tok/s=14622.8]
Inference: 52%|█████▏ | 3053/5920 [46:02<34:03, 1.40it/s, tok/s=14623.4]
Inference: 52%|█████▏ | 3054/5920 [46:02<34:02, 1.40it/s, tok/s=14624.4]
Inference: 52%|█████▏ | 3055/5920 [46:02<20:09, 2.37it/s, tok/s=14624.4]
Inference: 52%|█████▏ | 3055/5920 [46:02<20:09, 2.37it/s, tok/s=14624.1]
Inference: 52%|█████▏ | 3056/5920 [46:03<24:54, 1.92it/s, tok/s=14624.1]
Inference: 52%|█████▏ | 3056/5920 [46:03<24:54, 1.92it/s, tok/s=14623.8]
Inference: 52%|█████▏ | 3057/5920 [46:03<19:41, 2.42it/s, tok/s=14623.8]
Inference: 52%|█████▏ | 3057/5920 [46:03<19:41, 2.42it/s, tok/s=14623.9]
Inference: 52%|█████▏ | 3058/5920 [46:03<22:20, 2.14it/s, tok/s=14623.9]
Inference: 52%|█████▏ | 3058/5920 [46:03<22:20, 2.14it/s, tok/s=14624.5]
Inference: 52%|█████▏ | 3059/5920 [46:04<27:02, 1.76it/s, tok/s=14624.5]
Inference: 52%|█████▏ | 3059/5920 [46:04<27:02, 1.76it/s, tok/s=14621.1]
Inference: 52%|█████▏ | 3060/5920 [46:04<23:27, 2.03it/s, tok/s=14621.1]
Inference: 52%|█████▏ | 3060/5920 [46:04<23:27, 2.03it/s, tok/s=14620.6]
Inference: 52%|█████▏ | 3061/5920 [46:05<28:45, 1.66it/s, tok/s=14620.6]
Inference: 52%|█████▏ | 3061/5920 [46:05<28:45, 1.66it/s, tok/s=14621.6]
Inference: 52%|█████▏ | 3062/5920 [46:06<25:47, 1.85it/s, tok/s=14621.6]
Inference: 52%|█████▏ | 3062/5920 [46:06<25:47, 1.85it/s, tok/s=14625.2]
Inference: 52%|█████▏ | 3063/5920 [46:06<21:03, 2.26it/s, tok/s=14625.2]
Inference: 52%|█████▏ | 3063/5920 [46:06<21:03, 2.26it/s, tok/s=14625.7]
Inference: 52%|█████▏ | 3064/5920 [46:07<25:05, 1.90it/s, tok/s=14625.7]
Inference: 52%|█████▏ | 3064/5920 [46:07<25:05, 1.90it/s, tok/s=14622.3]
Inference: 52%|█████▏ | 3065/5920 [46:07<20:27, 2.33it/s, tok/s=14622.3]
Inference: 52%|█████▏ | 3065/5920 [46:07<20:27, 2.33it/s, tok/s=14621.7]
Inference: 52%|█████▏ | 3066/5920 [46:07<19:46, 2.41it/s, tok/s=14621.7]
Inference: 52%|█████▏ | 3066/5920 [46:07<19:46, 2.41it/s, tok/s=14620.8]
Inference: 52%|█████▏ | 3067/5920 [46:08<19:54, 2.39it/s, tok/s=14620.8]
Inference: 52%|█████▏ | 3067/5920 [46:08<19:54, 2.39it/s, tok/s=14619.5]
Inference: 52%|█████▏ | 3068/5920 [46:08<20:34, 2.31it/s, tok/s=14619.5]
Inference: 52%|█████▏ | 3068/5920 [46:08<20:34, 2.31it/s, tok/s=14617.7]
Inference: 52%|█████▏ | 3069/5920 [46:09<31:59, 1.49it/s, tok/s=14617.7]
Inference: 52%|█████▏ | 3069/5920 [46:09<31:59, 1.49it/s, tok/s=14617.7]
Inference: 52%|█████▏ | 3070/5920 [46:09<31:58, 1.49it/s, tok/s=14619.8]
Inference: 52%|█████▏ | 3071/5920 [46:10<22:17, 2.13it/s, tok/s=14619.8]
Inference: 52%|█████▏ | 3071/5920 [46:10<22:17, 2.13it/s, tok/s=14618.2]
Inference: 52%|█████▏ | 3072/5920 [46:10<18:07, 2.62it/s, tok/s=14618.2]
Inference: 52%|█████▏ | 3072/5920 [46:10<18:07, 2.62it/s, tok/s=14628.4]
Inference: 52%|█████▏ | 3073/5920 [46:10<15:07, 3.14it/s, tok/s=14628.4]
Inference: 52%|█████▏ | 3073/5920 [46:10<15:07, 3.14it/s, tok/s=14628.5]
Inference: 52%|█████▏ | 3074/5920 [46:11<18:24, 2.58it/s, tok/s=14628.5]
Inference: 52%|█████▏ | 3074/5920 [46:11<18:24, 2.58it/s, tok/s=14632.9]
Inference: 52%|█████▏ | 3075/5920 [46:11<21:57, 2.16it/s, tok/s=14632.9]
Inference: 52%|█████▏ | 3075/5920 [46:11<21:57, 2.16it/s, tok/s=14630.0]
Inference: 52%|█████▏ | 3076/5920 [46:12<20:32, 2.31it/s, tok/s=14630.0]
Inference: 52%|█████▏ | 3076/5920 [46:12<20:32, 2.31it/s, tok/s=14628.5]
Inference: 52%|█████▏ | 3077/5920 [46:12<20:32, 2.31it/s, tok/s=14638.1]
Inference: 52%|█████▏ | 3078/5920 [46:13<20:04, 2.36it/s, tok/s=14638.1]
Inference: 52%|█████▏ | 3078/5920 [46:13<20:04, 2.36it/s, tok/s=14634.6]
Inference: 52%|█████▏ | 3079/5920 [46:13<17:57, 2.64it/s, tok/s=14634.6]
Inference: 52%|█████▏ | 3079/5920 [46:13<17:57, 2.64it/s, tok/s=14634.0]
Inference: 52%|█████▏ | 3080/5920 [46:13<17:57, 2.64it/s, tok/s=14636.0]
Inference: 52%|█████▏ | 3081/5920 [46:13<14:44, 3.21it/s, tok/s=14636.0]
Inference: 52%|█████▏ | 3081/5920 [46:13<14:44, 3.21it/s, tok/s=14635.3]
Inference: 52%|█████▏ | 3082/5920 [46:13<14:44, 3.21it/s, tok/s=14635.9]
Inference: 52%|█████▏ | 3083/5920 [46:14<18:07, 2.61it/s, tok/s=14635.9]
Inference: 52%|█████▏ | 3083/5920 [46:14<18:07, 2.61it/s, tok/s=14631.4]
Inference: 52%|█████▏ | 3084/5920 [46:15<17:11, 2.75it/s, tok/s=14631.4]
Inference: 52%|█████▏ | 3084/5920 [46:15<17:11, 2.75it/s, tok/s=14630.4]
Inference: 52%|█████▏ | 3085/5920 [46:15<17:11, 2.75it/s, tok/s=14631.0]
Inference: 52%|█████▏ | 3086/5920 [46:15<17:50, 2.65it/s, tok/s=14631.0]
Inference: 52%|█████▏ | 3086/5920 [46:15<17:50, 2.65it/s, tok/s=14628.5]
Inference: 52%|█████▏ | 3087/5920 [46:15<15:14, 3.10it/s, tok/s=14628.5]
Inference: 52%|█████▏ | 3087/5920 [46:15<15:14, 3.10it/s, tok/s=14628.9]
Inference: 52%|█████▏ | 3088/5920 [46:16<14:41, 3.21it/s, tok/s=14628.9]
Inference: 52%|█████▏ | 3088/5920 [46:16<14:41, 3.21it/s, tok/s=14629.6]
Inference: 52%|█████▏ | 3089/5920 [46:16<15:42, 3.00it/s, tok/s=14629.6]
Inference: 52%|█████▏ | 3089/5920 [46:16<15:42, 3.00it/s, tok/s=14628.0]
Inference: 52%|█████▏ | 3090/5920 [46:17<21:33, 2.19it/s, tok/s=14628.0]
Inference: 52%|█████▏ | 3090/5920 [46:17<21:33, 2.19it/s, tok/s=14624.2]
Inference: 52%|█████▏ | 3091/5920 [46:17<17:06, 2.75it/s, tok/s=14624.2]
Inference: 52%|█████▏ | 3091/5920 [46:17<17:06, 2.75it/s, tok/s=14630.0]
Inference: 52%|█████▏ | 3092/5920 [46:18<24:43, 1.91it/s, tok/s=14630.0]
Inference: 52%|█████▏ | 3092/5920 [46:18<24:43, 1.91it/s, tok/s=14626.9]
Inference: 52%|█████▏ | 3093/5920 [46:19<31:45, 1.48it/s, tok/s=14626.9]
Inference: 52%|█████▏ | 3093/5920 [46:19<31:45, 1.48it/s, tok/s=14621.8]
Inference: 52%|█████▏ | 3094/5920 [46:19<25:37, 1.84it/s, tok/s=14621.8]
Inference: 52%|█████▏ | 3094/5920 [46:19<25:37, 1.84it/s, tok/s=14621.4]
Inference: 52%|█████▏ | 3095/5920 [46:20<30:58, 1.52it/s, tok/s=14621.4]
Inference: 52%|█████▏ | 3095/5920 [46:20<30:58, 1.52it/s, tok/s=14617.6]
Inference: 52%|█████▏ | 3096/5920 [46:20<23:54, 1.97it/s, tok/s=14617.6]
Inference: 52%|█████▏ | 3096/5920 [46:20<23:54, 1.97it/s, tok/s=14617.9]
Inference: 52%|█████▏ | 3097/5920 [46:21<24:19, 1.93it/s, tok/s=14617.9]
Inference: 52%|█████▏ | 3097/5920 [46:21<24:19, 1.93it/s, tok/s=14623.5]
Inference: 52%|█████▏ | 3098/5920 [46:21<24:18, 1.93it/s, tok/s=14623.9]
Inference: 52%|█████▏ | 3099/5920 [46:21<15:46, 2.98it/s, tok/s=14623.9]
Inference: 52%|█████▏ | 3099/5920 [46:21<15:46, 2.98it/s, tok/s=14628.8]
Inference: 52%|█████▏ | 3100/5920 [46:22<19:48, 2.37it/s, tok/s=14628.8]
Inference: 52%|█████▏ | 3100/5920 [46:22<19:48, 2.37it/s, tok/s=14626.0]
Inference: 52%|█████▏ | 3101/5920 [46:22<17:49, 2.64it/s, tok/s=14626.0]
Inference: 52%|█████▏ | 3101/5920 [46:22<17:49, 2.64it/s, tok/s=14627.4]
Inference: 52%|█████▏ | 3102/5920 [46:22<17:49, 2.64it/s, tok/s=14628.2]
Inference: 52%|█████▏ | 3103/5920 [46:23<16:12, 2.90it/s, tok/s=14628.2]
Inference: 52%|█████▏ | 3103/5920 [46:23<16:12, 2.90it/s, tok/s=14626.3]
Inference: 52%|█████▏ | 3104/5920 [46:24<26:06, 1.80it/s, tok/s=14626.3]
Inference: 52%|█████▏ | 3104/5920 [46:24<26:06, 1.80it/s, tok/s=14620.8]
Inference: 52%|█████▏ | 3105/5920 [46:25<30:04, 1.56it/s, tok/s=14620.8]
Inference: 52%|█████▏ | 3105/5920 [46:25<30:04, 1.56it/s, tok/s=14626.0]
Inference: 52%|█████▏ | 3106/5920 [46:25<25:35, 1.83it/s, tok/s=14626.0]
Inference: 52%|█████▏ | 3106/5920 [46:25<25:35, 1.83it/s, tok/s=14625.5]
Inference: 52%|█████▏ | 3107/5920 [46:25<21:54, 2.14it/s, tok/s=14625.5]
Inference: 52%|█████▏ | 3107/5920 [46:25<21:54, 2.14it/s, tok/s=14624.7]
Inference: 52%|█████▎ | 3108/5920 [46:26<21:24, 2.19it/s, tok/s=14624.7]
Inference: 52%|█████▎ | 3108/5920 [46:26<21:24, 2.19it/s, tok/s=14623.0]
Inference: 53%|█████▎ | 3109/5920 [46:26<21:29, 2.18it/s, tok/s=14623.0]
Inference: 53%|█████▎ | 3109/5920 [46:26<21:29, 2.18it/s, tok/s=14626.7]
Inference: 53%|█████▎ | 3110/5920 [46:26<18:41, 2.50it/s, tok/s=14626.7]
Inference: 53%|█████▎ | 3110/5920 [46:26<18:41, 2.50it/s, tok/s=14626.0]
Inference: 53%|█████▎ | 3111/5920 [46:28<28:41, 1.63it/s, tok/s=14626.0]
Inference: 53%|█████▎ | 3111/5920 [46:28<28:41, 1.63it/s, tok/s=14625.5]
Inference: 53%|█████▎ | 3112/5920 [46:28<28:41, 1.63it/s, tok/s=14625.7]
Inference: 53%|█████▎ | 3113/5920 [46:28<20:31, 2.28it/s, tok/s=14625.7]
Inference: 53%|█████▎ | 3113/5920 [46:28<20:31, 2.28it/s, tok/s=14624.3]
Inference: 53%|█████▎ | 3114/5920 [46:29<20:58, 2.23it/s, tok/s=14624.3]
Inference: 53%|█████▎ | 3114/5920 [46:29<20:58, 2.23it/s, tok/s=14622.2]
Inference: 53%|█████▎ | 3115/5920 [46:29<17:08, 2.73it/s, tok/s=14622.2]
Inference: 53%|█████▎ | 3115/5920 [46:29<17:08, 2.73it/s, tok/s=14622.5]
Inference: 53%|█████▎ | 3116/5920 [46:29<18:56, 2.47it/s, tok/s=14622.5]
Inference: 53%|█████▎ | 3116/5920 [46:29<18:56, 2.47it/s, tok/s=14630.3]
Inference: 53%|█████▎ | 3117/5920 [46:29<16:41, 2.80it/s, tok/s=14630.3]
Inference: 53%|█████▎ | 3117/5920 [46:29<16:41, 2.80it/s, tok/s=14635.1]
Inference: 53%|█████▎ | 3118/5920 [46:29<16:41, 2.80it/s, tok/s=14637.7]
Inference: 53%|█████▎ | 3119/5920 [46:30<15:25, 3.03it/s, tok/s=14637.7]
Inference: 53%|█████▎ | 3119/5920 [46:30<15:25, 3.03it/s, tok/s=14641.7]
Inference: 53%|█████▎ | 3120/5920 [46:31<20:45, 2.25it/s, tok/s=14641.7]
Inference: 53%|█████▎ | 3120/5920 [46:31<20:45, 2.25it/s, tok/s=14638.3]
Inference: 53%|█████▎ | 3121/5920 [46:31<19:36, 2.38it/s, tok/s=14638.3]
Inference: 53%|█████▎ | 3121/5920 [46:31<19:36, 2.38it/s, tok/s=14641.8]
Inference: 53%|█████▎ | 3122/5920 [46:31<19:36, 2.38it/s, tok/s=14642.1]
Inference: 53%|█████▎ | 3123/5920 [46:31<13:19, 3.50it/s, tok/s=14642.1]
Inference: 53%|█████▎ | 3123/5920 [46:31<13:19, 3.50it/s, tok/s=14641.9]
Inference: 53%|█████▎ | 3124/5920 [46:32<11:52, 3.93it/s, tok/s=14641.9]
Inference: 53%|█████▎ | 3124/5920 [46:32<11:52, 3.93it/s, tok/s=14647.1]
Inference: 53%|█████▎ | 3125/5920 [46:32<11:37, 4.00it/s, tok/s=14647.1]
Inference: 53%|█████▎ | 3125/5920 [46:32<11:37, 4.00it/s, tok/s=14655.8]
Inference: 53%|█████▎ | 3126/5920 [46:33<18:17, 2.55it/s, tok/s=14655.8]
Inference: 53%|█████▎ | 3126/5920 [46:33<18:17, 2.55it/s, tok/s=14652.9]
Inference: 53%|█████▎ | 3127/5920 [46:33<18:17, 2.55it/s, tok/s=14652.8]
Inference: 53%|█████▎ | 3128/5920 [46:33<14:33, 3.20it/s, tok/s=14652.8]
Inference: 53%|█████▎ | 3128/5920 [46:33<14:33, 3.20it/s, tok/s=14653.2]
Inference: 53%|█████▎ | 3129/5920 [46:33<14:30, 3.20it/s, tok/s=14653.2]
Inference: 53%|█████▎ | 3129/5920 [46:33<14:30, 3.20it/s, tok/s=14652.0]
Inference: 53%|█████▎ | 3130/5920 [46:34<15:58, 2.91it/s, tok/s=14652.0]
Inference: 53%|█████▎ | 3130/5920 [46:34<15:58, 2.91it/s, tok/s=14650.1]
Inference: 53%|█████▎ | 3131/5920 [46:34<15:57, 2.91it/s, tok/s=14651.2]
Inference: 53%|█████▎ | 3132/5920 [46:34<15:57, 2.91it/s, tok/s=14651.6]
Inference: 53%|█████▎ | 3133/5920 [46:34<08:37, 5.38it/s, tok/s=14651.6]
Inference: 53%|█████▎ | 3133/5920 [46:34<08:37, 5.38it/s, tok/s=14652.2]
Inference: 53%|█████▎ | 3134/5920 [46:34<08:37, 5.38it/s, tok/s=14661.3]
Inference: 53%|█████▎ | 3135/5920 [46:34<09:47, 4.74it/s, tok/s=14661.3]
Inference: 53%|█████▎ | 3135/5920 [46:34<09:47, 4.74it/s, tok/s=14661.4]
Inference: 53%|█████▎ | 3136/5920 [46:34<08:59, 5.16it/s, tok/s=14661.4]
Inference: 53%|█████▎ | 3136/5920 [46:34<08:59, 5.16it/s, tok/s=14661.6]
Inference: 53%|█████▎ | 3137/5920 [46:35<08:55, 5.20it/s, tok/s=14661.6]
Inference: 53%|█████▎ | 3137/5920 [46:35<08:55, 5.20it/s, tok/s=14665.1]
Inference: 53%|█████▎ | 3138/5920 [46:35<10:13, 4.53it/s, tok/s=14665.1]
Inference: 53%|█████▎ | 3138/5920 [46:35<10:13, 4.53it/s, tok/s=14663.8]
Inference: 53%|█████▎ | 3139/5920 [46:35<10:28, 4.42it/s, tok/s=14663.8]
Inference: 53%|█████▎ | 3139/5920 [46:35<10:28, 4.42it/s, tok/s=14665.0]
Inference: 53%|█████▎ | 3140/5920 [46:36<14:17, 3.24it/s, tok/s=14665.0]
Inference: 53%|█████▎ | 3140/5920 [46:36<14:17, 3.24it/s, tok/s=14662.6]
Inference: 53%|█████▎ | 3141/5920 [46:36<16:57, 2.73it/s, tok/s=14662.6]
Inference: 53%|█████▎ | 3141/5920 [46:36<16:57, 2.73it/s, tok/s=14661.0]
Inference: 53%|█████▎ | 3142/5920 [46:36<14:21, 3.22it/s, tok/s=14661.0]
Inference: 53%|█████▎ | 3142/5920 [46:36<14:21, 3.22it/s, tok/s=14661.1]
Inference: 53%|█████▎ | 3143/5920 [46:38<24:58, 1.85it/s, tok/s=14661.1]
Inference: 53%|█████▎ | 3143/5920 [46:38<24:58, 1.85it/s, tok/s=14661.9]
Inference: 53%|█████▎ | 3144/5920 [46:38<23:51, 1.94it/s, tok/s=14661.9]
Inference: 53%|█████▎ | 3144/5920 [46:38<23:51, 1.94it/s, tok/s=14660.3]
Inference: 53%|█████▎ | 3145/5920 [46:38<22:45, 2.03it/s, tok/s=14660.3]
Inference: 53%|█████▎ | 3145/5920 [46:38<22:45, 2.03it/s, tok/s=14662.8]
Inference: 53%|█████▎ | 3146/5920 [46:39<21:26, 2.16it/s, tok/s=14662.8]
Inference: 53%|█████▎ | 3146/5920 [46:39<21:26, 2.16it/s, tok/s=14661.6]
Inference: 53%|█████▎ | 3147/5920 [46:40<28:43, 1.61it/s, tok/s=14661.6]
Inference: 53%|█████▎ | 3147/5920 [46:40<28:43, 1.61it/s, tok/s=14657.0]
Inference: 53%|█████▎ | 3148/5920 [46:41<31:12, 1.48it/s, tok/s=14657.0]
Inference: 53%|█████▎ | 3148/5920 [46:41<31:12, 1.48it/s, tok/s=14662.6]
Inference: 53%|█████▎ | 3149/5920 [46:41<25:54, 1.78it/s, tok/s=14662.6]
Inference: 53%|█████▎ | 3149/5920 [46:41<25:54, 1.78it/s, tok/s=14661.8]
Inference: 53%|█████▎ | 3150/5920 [46:42<34:31, 1.34it/s, tok/s=14661.8]
Inference: 53%|█████▎ | 3150/5920 [46:42<34:31, 1.34it/s, tok/s=14657.9]
Inference: 53%|█████▎ | 3151/5920 [46:44<44:15, 1.04it/s, tok/s=14657.9]
Inference: 53%|█████▎ | 3151/5920 [46:44<44:15, 1.04it/s, tok/s=14656.8]
Inference: 53%|█████▎ | 3152/5920 [46:44<35:30, 1.30it/s, tok/s=14656.8]
Inference: 53%|█████▎ | 3152/5920 [46:44<35:30, 1.30it/s, tok/s=14655.5]
Inference: 53%|█████▎ | 3153/5920 [46:44<27:34, 1.67it/s, tok/s=14655.5]
Inference: 53%|█████▎ | 3153/5920 [46:44<27:34, 1.67it/s, tok/s=14655.6]
Inference: 53%|█████▎ | 3154/5920 [46:44<22:51, 2.02it/s, tok/s=14655.6]
Inference: 53%|█████▎ | 3154/5920 [46:44<22:51, 2.02it/s, tok/s=14657.6]
Inference: 53%|█████▎ | 3155/5920 [46:45<19:30, 2.36it/s, tok/s=14657.6]
Inference: 53%|█████▎ | 3155/5920 [46:45<19:30, 2.36it/s, tok/s=14657.3]
Inference: 53%|█████▎ | 3156/5920 [46:45<23:27, 1.96it/s, tok/s=14657.3]
Inference: 53%|█████▎ | 3156/5920 [46:45<23:27, 1.96it/s, tok/s=14655.4]
Inference: 53%|█████▎ | 3157/5920 [46:45<18:58, 2.43it/s, tok/s=14655.4]
Inference: 53%|█████▎ | 3157/5920 [46:45<18:58, 2.43it/s, tok/s=14655.0]
Inference: 53%|█████▎ | 3158/5920 [46:46<16:30, 2.79it/s, tok/s=14655.0]
Inference: 53%|█████▎ | 3158/5920 [46:46<16:30, 2.79it/s, tok/s=14655.5]
Inference: 53%|█████▎ | 3159/5920 [46:46<16:29, 2.79it/s, tok/s=14656.2]
Inference: 53%|█████▎ | 3160/5920 [46:46<16:29, 2.79it/s, tok/s=14656.5]
Inference: 53%|█████▎ | 3161/5920 [46:46<08:29, 5.42it/s, tok/s=14656.5]
Inference: 53%|█████▎ | 3161/5920 [46:46<08:29, 5.42it/s, tok/s=14656.4]
Inference: 53%|█████▎ | 3162/5920 [46:46<12:04, 3.81it/s, tok/s=14656.4]
Inference: 53%|█████▎ | 3162/5920 [46:46<12:04, 3.81it/s, tok/s=14654.9]
Inference: 53%|█████▎ | 3163/5920 [46:47<10:36, 4.33it/s, tok/s=14654.9]
Inference: 53%|█████▎ | 3163/5920 [46:47<10:36, 4.33it/s, tok/s=14659.6]
Inference: 53%|█████▎ | 3164/5920 [46:47<10:36, 4.33it/s, tok/s=14669.7]
Inference: 53%|█████▎ | 3165/5920 [46:47<12:49, 3.58it/s, tok/s=14669.7]
Inference: 53%|█████▎ | 3165/5920 [46:47<12:49, 3.58it/s, tok/s=14667.7]
Inference: 53%|█████▎ | 3166/5920 [46:47<12:49, 3.58it/s, tok/s=14668.9]
Inference: 53%|█████▎ | 3167/5920 [46:47<12:48, 3.58it/s, tok/s=14675.8]
Inference: 54%|█████▎ | 3168/5920 [46:48<09:30, 4.82it/s, tok/s=14675.8]
Inference: 54%|█████▎ | 3168/5920 [46:48<09:30, 4.82it/s, tok/s=14676.1]
Inference: 54%|█████▎ | 3169/5920 [46:48<14:53, 3.08it/s, tok/s=14676.1]
Inference: 54%|█████▎ | 3169/5920 [46:48<14:53, 3.08it/s, tok/s=14672.0]
Inference: 54%|█████▎ | 3170/5920 [46:51<31:52, 1.44it/s, tok/s=14672.0]
Inference: 54%|█████▎ | 3170/5920 [46:51<31:52, 1.44it/s, tok/s=14661.5]
Inference: 54%|█████▎ | 3171/5920 [46:51<31:51, 1.44it/s, tok/s=14664.4]
Inference: 54%|█████▎ | 3172/5920 [46:51<25:27, 1.80it/s, tok/s=14664.4]
Inference: 54%|█████▎ | 3172/5920 [46:51<25:27, 1.80it/s, tok/s=14662.6]
Inference: 54%|█████▎ | 3173/5920 [46:51<22:08, 2.07it/s, tok/s=14662.6]
Inference: 54%|█████▎ | 3173/5920 [46:51<22:08, 2.07it/s, tok/s=14663.4]
Inference: 54%|█████▎ | 3174/5920 [46:52<20:13, 2.26it/s, tok/s=14663.4]
Inference: 54%|█████▎ | 3174/5920 [46:52<20:13, 2.26it/s, tok/s=14664.3]
Inference: 54%|█████▎ | 3175/5920 [46:54<36:53, 1.24it/s, tok/s=14664.3]
Inference: 54%|█████▎ | 3175/5920 [46:54<36:53, 1.24it/s, tok/s=14655.6]
Inference: 54%|█████▎ | 3176/5920 [46:54<29:41, 1.54it/s, tok/s=14655.6]
Inference: 54%|█████▎ | 3176/5920 [46:54<29:41, 1.54it/s, tok/s=14663.1]
Inference: 54%|█████▎ | 3177/5920 [46:55<33:07, 1.38it/s, tok/s=14663.1]
Inference: 54%|█████▎ | 3177/5920 [46:55<33:07, 1.38it/s, tok/s=14659.9]
Inference: 54%|█████▎ | 3178/5920 [46:55<26:29, 1.73it/s, tok/s=14659.9]
Inference: 54%|█████▎ | 3178/5920 [46:55<26:29, 1.73it/s, tok/s=14659.3]
Inference: 54%|█████▎ | 3179/5920 [46:55<21:41, 2.11it/s, tok/s=14659.3]
Inference: 54%|█████▎ | 3179/5920 [46:55<21:41, 2.11it/s, tok/s=14660.8]
Inference: 54%|█████▎ | 3180/5920 [46:55<21:41, 2.11it/s, tok/s=14662.4]
Inference: 54%|█████▎ | 3181/5920 [46:56<17:09, 2.66it/s, tok/s=14662.4]
Inference: 54%|█████▎ | 3181/5920 [46:56<17:09, 2.66it/s, tok/s=14663.1]
Inference: 54%|█████▍ | 3182/5920 [46:56<17:09, 2.66it/s, tok/s=14664.2]
Inference: 54%|█████▍ | 3183/5920 [46:56<15:30, 2.94it/s, tok/s=14664.2]
Inference: 54%|█████▍ | 3183/5920 [46:56<15:30, 2.94it/s, tok/s=14667.2]
Inference: 54%|█████▍ | 3184/5920 [46:57<22:03, 2.07it/s, tok/s=14667.2]
Inference: 54%|█████▍ | 3184/5920 [46:57<22:03, 2.07it/s, tok/s=14672.3]
Inference: 54%|█████▍ | 3185/5920 [46:59<38:37, 1.18it/s, tok/s=14672.3]
Inference: 54%|█████▍ | 3185/5920 [46:59<38:37, 1.18it/s, tok/s=14672.1]
Inference: 54%|█████▍ | 3186/5920 [46:59<38:36, 1.18it/s, tok/s=14673.4]
Inference: 54%|█████▍ | 3187/5920 [47:00<27:25, 1.66it/s, tok/s=14673.4]
Inference: 54%|█████▍ | 3187/5920 [47:00<27:25, 1.66it/s, tok/s=14675.2]
Inference: 54%|█████▍ | 3188/5920 [47:01<29:13, 1.56it/s, tok/s=14675.2]
Inference: 54%|█████▍ | 3188/5920 [47:01<29:13, 1.56it/s, tok/s=14671.6]
Inference: 54%|█████▍ | 3189/5920 [47:01<29:13, 1.56it/s, tok/s=14672.0]
Inference: 54%|█████▍ | 3190/5920 [47:01<20:04, 2.27it/s, tok/s=14672.0]
Inference: 54%|█████▍ | 3190/5920 [47:01<20:04, 2.27it/s, tok/s=14678.6]
Inference: 54%|█████▍ | 3191/5920 [47:01<16:45, 2.72it/s, tok/s=14678.6]
Inference: 54%|█████▍ | 3191/5920 [47:01<16:45, 2.72it/s, tok/s=14678.5]
Inference: 54%|█████▍ | 3192/5920 [47:01<13:58, 3.25it/s, tok/s=14678.5]
Inference: 54%|█████▍ | 3192/5920 [47:01<13:58, 3.25it/s, tok/s=14688.6]
Inference: 54%|█████▍ | 3193/5920 [47:01<11:41, 3.88it/s, tok/s=14688.6]
Inference: 54%|█████▍ | 3193/5920 [47:01<11:41, 3.88it/s, tok/s=14688.5]
Inference: 54%|█████▍ | 3194/5920 [47:02<13:39, 3.33it/s, tok/s=14688.5]
Inference: 54%|█████▍ | 3194/5920 [47:02<13:39, 3.33it/s, tok/s=14686.7]
Inference: 54%|█████▍ | 3195/5920 [47:02<17:17, 2.63it/s, tok/s=14686.7]
Inference: 54%|█████▍ | 3195/5920 [47:02<17:17, 2.63it/s, tok/s=14684.1]
Inference: 54%|█████▍ | 3196/5920 [47:03<20:03, 2.26it/s, tok/s=14684.1]
Inference: 54%|█████▍ | 3196/5920 [47:03<20:03, 2.26it/s, tok/s=14691.7]
Inference: 54%|█████▍ | 3197/5920 [47:05<47:21, 1.04s/it, tok/s=14691.7]
Inference: 54%|█████▍ | 3197/5920 [47:05<47:21, 1.04s/it, tok/s=14680.8]
Inference: 54%|█████▍ | 3198/5920 [47:06<38:53, 1.17it/s, tok/s=14680.8]
Inference: 54%|█████▍ | 3198/5920 [47:06<38:53, 1.17it/s, tok/s=14679.1]
Inference: 54%|█████▍ | 3199/5920 [47:06<38:52, 1.17it/s, tok/s=14680.1]
Inference: 54%|█████▍ | 3200/5920 [47:06<27:10, 1.67it/s, tok/s=14680.1]
Inference: 54%|█████▍ | 3200/5920 [47:06<27:10, 1.67it/s, tok/s=14677.8]
Inference: 54%|█████▍ | 3201/5920 [47:07<23:48, 1.90it/s, tok/s=14677.8]
Inference: 54%|█████▍ | 3201/5920 [47:07<23:48, 1.90it/s, tok/s=14676.6]
Inference: 54%|█████▍ | 3202/5920 [47:08<30:24, 1.49it/s, tok/s=14676.6]
Inference: 54%|█████▍ | 3202/5920 [47:08<30:24, 1.49it/s, tok/s=14672.0]
Inference: 54%|█████▍ | 3203/5920 [47:08<30:23, 1.49it/s, tok/s=14672.5]
Inference: 54%|█████▍ | 3204/5920 [47:08<21:04, 2.15it/s, tok/s=14672.5]
Inference: 54%|█████▍ | 3204/5920 [47:08<21:04, 2.15it/s, tok/s=14673.8]
Inference: 54%|█████▍ | 3205/5920 [47:08<21:03, 2.15it/s, tok/s=14674.0]
Inference: 54%|█████▍ | 3206/5920 [47:09<19:48, 2.28it/s, tok/s=14674.0]
Inference: 54%|█████▍ | 3206/5920 [47:09<19:48, 2.28it/s, tok/s=14670.9]
Inference: 54%|█████▍ | 3207/5920 [47:09<19:15, 2.35it/s, tok/s=14670.9]
Inference: 54%|█████▍ | 3207/5920 [47:09<19:15, 2.35it/s, tok/s=14670.6]
Inference: 54%|█████▍ | 3208/5920 [47:09<19:14, 2.35it/s, tok/s=14670.8]
Inference: 54%|█████▍ | 3209/5920 [47:10<18:12, 2.48it/s, tok/s=14670.8]
Inference: 54%|█████▍ | 3209/5920 [47:10<18:12, 2.48it/s, tok/s=14669.4]
Inference: 54%|█████▍ | 3210/5920 [47:10<19:22, 2.33it/s, tok/s=14669.4]
Inference: 54%|█████▍ | 3210/5920 [47:10<19:22, 2.33it/s, tok/s=14672.6]
Inference: 54%|█████▍ | 3211/5920 [47:12<26:32, 1.70it/s, tok/s=14672.6]
Inference: 54%|█████▍ | 3211/5920 [47:12<26:32, 1.70it/s, tok/s=14667.8]
Inference: 54%|█████▍ | 3212/5920 [47:12<28:21, 1.59it/s, tok/s=14667.8]
Inference: 54%|█████▍ | 3212/5920 [47:12<28:21, 1.59it/s, tok/s=14672.0]
Inference: 54%|█████▍ | 3213/5920 [47:12<28:20, 1.59it/s, tok/s=14672.5]
Inference: 54%|█████▍ | 3214/5920 [47:13<18:07, 2.49it/s, tok/s=14672.5]
Inference: 54%|█████▍ | 3214/5920 [47:13<18:07, 2.49it/s, tok/s=14672.5]
Inference: 54%|█████▍ | 3215/5920 [47:14<27:29, 1.64it/s, tok/s=14672.5]
Inference: 54%|█████▍ | 3215/5920 [47:14<27:29, 1.64it/s, tok/s=14670.6]
Inference: 54%|█████▍ | 3216/5920 [47:14<27:28, 1.64it/s, tok/s=14671.0]
Inference: 54%|█████▍ | 3217/5920 [47:14<18:44, 2.40it/s, tok/s=14671.0]
Inference: 54%|█████▍ | 3217/5920 [47:14<18:44, 2.40it/s, tok/s=14671.1]
Inference: 54%|█████▍ | 3218/5920 [47:14<18:43, 2.40it/s, tok/s=14673.2]
Inference: 54%|█████▍ | 3219/5920 [47:15<17:38, 2.55it/s, tok/s=14673.2]
Inference: 54%|█████▍ | 3219/5920 [47:15<17:38, 2.55it/s, tok/s=14670.2]
Inference: 54%|█████▍ | 3220/5920 [47:15<17:37, 2.55it/s, tok/s=14670.4]
Inference: 54%|█████▍ | 3221/5920 [47:15<12:26, 3.62it/s, tok/s=14670.4]
Inference: 54%|█████▍ | 3221/5920 [47:15<12:26, 3.62it/s, tok/s=14670.7]
Inference: 54%|█████▍ | 3222/5920 [47:15<13:56, 3.23it/s, tok/s=14670.7]
Inference: 54%|█████▍ | 3222/5920 [47:15<13:56, 3.23it/s, tok/s=14669.7]
Inference: 54%|█████▍ | 3223/5920 [47:16<12:56, 3.47it/s, tok/s=14669.7]
Inference: 54%|█████▍ | 3223/5920 [47:16<12:56, 3.47it/s, tok/s=14669.4]
Inference: 54%|█████▍ | 3224/5920 [47:17<23:43, 1.89it/s, tok/s=14669.4]
Inference: 54%|█████▍ | 3224/5920 [47:17<23:43, 1.89it/s, tok/s=14664.0]
Inference: 54%|█████▍ | 3225/5920 [47:17<21:23, 2.10it/s, tok/s=14664.0]
Inference: 54%|█████▍ | 3225/5920 [47:17<21:23, 2.10it/s, tok/s=14664.1]
Inference: 54%|█████▍ | 3226/5920 [47:18<21:05, 2.13it/s, tok/s=14664.1]
Inference: 54%|█████▍ | 3226/5920 [47:18<21:05, 2.13it/s, tok/s=14672.4]
Inference: 55%|█████▍ | 3227/5920 [47:18<16:50, 2.66it/s, tok/s=14672.4]
Inference: 55%|█████▍ | 3227/5920 [47:18<16:50, 2.66it/s, tok/s=14682.5]
Inference: 55%|█████▍ | 3228/5920 [47:18<19:32, 2.30it/s, tok/s=14682.5]
Inference: 55%|█████▍ | 3228/5920 [47:18<19:32, 2.30it/s, tok/s=14688.7]
Inference: 55%|█████▍ | 3229/5920 [47:18<16:27, 2.73it/s, tok/s=14688.7]
Inference: 55%|█████▍ | 3229/5920 [47:18<16:27, 2.73it/s, tok/s=14690.5]
Inference: 55%|█████▍ | 3230/5920 [47:19<16:27, 2.73it/s, tok/s=14698.5]
Inference: 55%|█████▍ | 3231/5920 [47:19<16:26, 2.73it/s, tok/s=14707.0]
Inference: 55%|█████▍ | 3232/5920 [47:19<16:26, 2.73it/s, tok/s=14710.8]
Inference: 55%|█████▍ | 3233/5920 [47:19<16:26, 2.73it/s, tok/s=14711.9]
Inference: 55%|█████▍ | 3234/5920 [47:19<16:25, 2.73it/s, tok/s=14722.3]
Inference: 55%|█████▍ | 3235/5920 [47:19<08:56, 5.00it/s, tok/s=14722.3]
Inference: 55%|█████▍ | 3235/5920 [47:19<08:56, 5.00it/s, tok/s=14721.2]
Inference: 55%|█████▍ | 3236/5920 [47:20<11:55, 3.75it/s, tok/s=14721.2]
Inference: 55%|█████▍ | 3236/5920 [47:20<11:55, 3.75it/s, tok/s=14720.8]
Inference: 55%|█████▍ | 3237/5920 [47:20<11:55, 3.75it/s, tok/s=14721.3]
Inference: 55%|█████▍ | 3238/5920 [47:20<10:09, 4.40it/s, tok/s=14721.3]
Inference: 55%|█████▍ | 3238/5920 [47:20<10:09, 4.40it/s, tok/s=14725.1]
Inference: 55%|█████▍ | 3239/5920 [47:20<09:51, 4.53it/s, tok/s=14725.1]
Inference: 55%|█████▍ | 3239/5920 [47:20<09:51, 4.53it/s, tok/s=14725.7]
Inference: 55%|█████▍ | 3240/5920 [47:20<09:51, 4.53it/s, tok/s=14727.4]
Inference: 55%|█████▍ | 3241/5920 [47:22<14:44, 3.03it/s, tok/s=14727.4]
Inference: 55%|█████▍ | 3241/5920 [47:22<14:44, 3.03it/s, tok/s=14723.4]
Inference: 55%|█████▍ | 3242/5920 [47:22<12:56, 3.45it/s, tok/s=14723.4]
Inference: 55%|█████▍ | 3242/5920 [47:22<12:56, 3.45it/s, tok/s=14724.8]
Inference: 55%|█████▍ | 3243/5920 [47:22<12:56, 3.45it/s, tok/s=14728.0]
Inference: 55%|█████▍ | 3244/5920 [47:22<12:56, 3.45it/s, tok/s=14730.2]
Inference: 55%|█████▍ | 3245/5920 [47:22<09:46, 4.56it/s, tok/s=14730.2]
Inference: 55%|█████▍ | 3245/5920 [47:22<09:46, 4.56it/s, tok/s=14728.8]
Inference: 55%|█████▍ | 3246/5920 [47:22<09:45, 4.56it/s, tok/s=14729.5]
Inference: 55%|█████▍ | 3247/5920 [47:22<09:45, 4.56it/s, tok/s=14729.7]
Inference: 55%|█████▍ | 3248/5920 [47:22<07:07, 6.25it/s, tok/s=14729.7]
Inference: 55%|█████▍ | 3248/5920 [47:22<07:07, 6.25it/s, tok/s=14739.4]
Inference: 55%|█████▍ | 3249/5920 [47:22<07:07, 6.25it/s, tok/s=14740.4]
Inference: 55%|█████▍ | 3250/5920 [47:22<06:00, 7.40it/s, tok/s=14740.4]
Inference: 55%|█████▍ | 3250/5920 [47:22<06:00, 7.40it/s, tok/s=14747.2]
Inference: 55%|█████▍ | 3251/5920 [47:23<06:00, 7.40it/s, tok/s=14747.4]
Inference: 55%|█████▍ | 3252/5920 [47:23<06:13, 7.15it/s, tok/s=14747.4]
Inference: 55%|█████▍ | 3252/5920 [47:23<06:13, 7.15it/s, tok/s=14747.8]
Inference: 55%|█████▍ | 3253/5920 [47:23<07:41, 5.78it/s, tok/s=14747.8]
Inference: 55%|█████▍ | 3253/5920 [47:23<07:41, 5.78it/s, tok/s=14753.7]
Inference: 55%|█████▍ | 3254/5920 [47:24<14:01, 3.17it/s, tok/s=14753.7]
Inference: 55%|█████▍ | 3254/5920 [47:24<14:01, 3.17it/s, tok/s=14750.1]
Inference: 55%|█████▍ | 3255/5920 [47:24<13:08, 3.38it/s, tok/s=14750.1]
Inference: 55%|█████▍ | 3255/5920 [47:24<13:08, 3.38it/s, tok/s=14757.2]
Inference: 55%|█████▌ | 3256/5920 [47:24<13:07, 3.38it/s, tok/s=14767.9]
Inference: 55%|█████▌ | 3257/5920 [47:24<10:51, 4.09it/s, tok/s=14767.9]
Inference: 55%|█████▌ | 3257/5920 [47:24<10:51, 4.09it/s, tok/s=14768.3]
Inference: 55%|█████▌ | 3258/5920 [47:25<09:53, 4.49it/s, tok/s=14768.3]
Inference: 55%|█████▌ | 3258/5920 [47:25<09:53, 4.49it/s, tok/s=14777.7]
Inference: 55%|█████▌ | 3259/5920 [47:25<09:53, 4.49it/s, tok/s=14780.2]
Inference: 55%|█████▌ | 3260/5920 [47:25<07:39, 5.79it/s, tok/s=14780.2]
Inference: 55%|█████▌ | 3260/5920 [47:25<07:39, 5.79it/s, tok/s=14790.4]
Inference: 55%|█████▌ | 3261/5920 [47:25<07:39, 5.79it/s, tok/s=14791.0]
Inference: 55%|█████▌ | 3262/5920 [47:25<06:59, 6.34it/s, tok/s=14791.0]
Inference: 55%|█████▌ | 3262/5920 [47:25<06:59, 6.34it/s, tok/s=14790.4]
Inference: 55%|█████▌ | 3263/5920 [47:27<19:05, 2.32it/s, tok/s=14790.4]
Inference: 55%|█████▌ | 3263/5920 [47:27<19:05, 2.32it/s, tok/s=14783.0]
Inference: 55%|█████▌ | 3264/5920 [47:27<17:23, 2.54it/s, tok/s=14783.0]
Inference: 55%|█████▌ | 3264/5920 [47:27<17:23, 2.54it/s, tok/s=14782.5]
Inference: 55%|█████▌ | 3265/5920 [47:27<17:47, 2.49it/s, tok/s=14782.5]
Inference: 55%|█████▌ | 3265/5920 [47:27<17:47, 2.49it/s, tok/s=14780.9]
Inference: 55%|█████▌ | 3266/5920 [47:28<20:12, 2.19it/s, tok/s=14780.9]
Inference: 55%|█████▌ | 3266/5920 [47:28<20:12, 2.19it/s, tok/s=14782.7]
Inference: 55%|█████▌ | 3267/5920 [47:28<20:12, 2.19it/s, tok/s=14783.2]
Inference: 55%|█████▌ | 3268/5920 [47:28<20:11, 2.19it/s, tok/s=14784.0]
Inference: 55%|█████▌ | 3269/5920 [47:28<11:43, 3.77it/s, tok/s=14784.0]
Inference: 55%|█████▌ | 3269/5920 [47:28<11:43, 3.77it/s, tok/s=14783.3]
Inference: 55%|█████▌ | 3270/5920 [47:29<16:11, 2.73it/s, tok/s=14783.3]
Inference: 55%|█████▌ | 3270/5920 [47:29<16:11, 2.73it/s, tok/s=14786.9]
Inference: 55%|█████▌ | 3271/5920 [47:29<16:42, 2.64it/s, tok/s=14786.9]
Inference: 55%|█████▌ | 3271/5920 [47:29<16:42, 2.64it/s, tok/s=14785.6]
Inference: 55%|█████▌ | 3272/5920 [47:30<19:34, 2.25it/s, tok/s=14785.6]
Inference: 55%|█████▌ | 3272/5920 [47:30<19:34, 2.25it/s, tok/s=14782.6]
Inference: 55%|█████▌ | 3273/5920 [47:32<32:03, 1.38it/s, tok/s=14782.6]
Inference: 55%|█████▌ | 3273/5920 [47:32<32:03, 1.38it/s, tok/s=14776.0]
Inference: 55%|█████▌ | 3274/5920 [47:32<24:41, 1.79it/s, tok/s=14776.0]
Inference: 55%|█████▌ | 3274/5920 [47:32<24:41, 1.79it/s, tok/s=14776.3]
Inference: 55%|█████▌ | 3275/5920 [47:32<24:40, 1.79it/s, tok/s=14777.2]
Inference: 55%|█████▌ | 3276/5920 [47:32<16:05, 2.74it/s, tok/s=14777.2]
Inference: 55%|█████▌ | 3276/5920 [47:32<16:05, 2.74it/s, tok/s=14777.7]
Inference: 55%|█████▌ | 3277/5920 [47:32<16:52, 2.61it/s, tok/s=14777.7]
Inference: 55%|█████▌ | 3277/5920 [47:32<16:52, 2.61it/s, tok/s=14775.9]
Inference: 55%|█████▌ | 3278/5920 [47:32<16:51, 2.61it/s, tok/s=14776.7]
Inference: 55%|█████▌ | 3279/5920 [47:32<11:16, 3.90it/s, tok/s=14776.7]
Inference: 55%|█████▌ | 3279/5920 [47:32<11:16, 3.90it/s, tok/s=14778.3]
Inference: 55%|█████▌ | 3280/5920 [47:33<14:25, 3.05it/s, tok/s=14778.3]
Inference: 55%|█████▌ | 3280/5920 [47:33<14:25, 3.05it/s, tok/s=14776.1]
Inference: 55%|█████▌ | 3281/5920 [47:33<13:10, 3.34it/s, tok/s=14776.1]
Inference: 55%|█████▌ | 3281/5920 [47:33<13:10, 3.34it/s, tok/s=14775.7]
Inference: 55%|█████▌ | 3282/5920 [47:34<17:19, 2.54it/s, tok/s=14775.7]
Inference: 55%|█████▌ | 3282/5920 [47:34<17:19, 2.54it/s, tok/s=14772.8]
Inference: 55%|█████▌ | 3283/5920 [47:34<17:19, 2.54it/s, tok/s=14777.5]
Inference: 55%|█████▌ | 3284/5920 [47:34<14:32, 3.02it/s, tok/s=14777.5]
Inference: 55%|█████▌ | 3284/5920 [47:34<14:32, 3.02it/s, tok/s=14781.7]
Inference: 55%|█████▌ | 3285/5920 [47:35<14:05, 3.12it/s, tok/s=14781.7]
Inference: 55%|█████▌ | 3285/5920 [47:35<14:05, 3.12it/s, tok/s=14785.7]
Inference: 56%|█████▌ | 3286/5920 [47:35<18:44, 2.34it/s, tok/s=14785.7]
Inference: 56%|█████▌ | 3286/5920 [47:35<18:44, 2.34it/s, tok/s=14784.2]
Inference: 56%|█████▌ | 3287/5920 [47:36<15:56, 2.75it/s, tok/s=14784.2]
Inference: 56%|█████▌ | 3287/5920 [47:36<15:56, 2.75it/s, tok/s=14783.9]
Inference: 56%|█████▌ | 3288/5920 [47:36<16:48, 2.61it/s, tok/s=14783.9]
Inference: 56%|█████▌ | 3288/5920 [47:36<16:48, 2.61it/s, tok/s=14789.7]
Inference: 56%|█████▌ | 3289/5920 [47:37<18:17, 2.40it/s, tok/s=14789.7]
Inference: 56%|█████▌ | 3289/5920 [47:37<18:17, 2.40it/s, tok/s=14788.0]
Inference: 56%|█████▌ | 3290/5920 [47:37<16:02, 2.73it/s, tok/s=14788.0]
Inference: 56%|█████▌ | 3290/5920 [47:37<16:02, 2.73it/s, tok/s=14795.0]
Inference: 56%|█████▌ | 3291/5920 [47:37<16:01, 2.73it/s, tok/s=14795.7]
Inference: 56%|█████▌ | 3292/5920 [47:38<20:41, 2.12it/s, tok/s=14795.7]
Inference: 56%|█████▌ | 3292/5920 [47:38<20:41, 2.12it/s, tok/s=14790.9]
Inference: 56%|█████▌ | 3293/5920 [47:38<19:25, 2.25it/s, tok/s=14790.9]
Inference: 56%|█████▌ | 3293/5920 [47:38<19:25, 2.25it/s, tok/s=14790.5]
Inference: 56%|█████▌ | 3294/5920 [47:39<18:26, 2.37it/s, tok/s=14790.5]
Inference: 56%|█████▌ | 3294/5920 [47:39<18:26, 2.37it/s, tok/s=14789.9]
Inference: 56%|█████▌ | 3295/5920 [47:39<16:46, 2.61it/s, tok/s=14789.9]
Inference: 56%|█████▌ | 3295/5920 [47:39<16:46, 2.61it/s, tok/s=14789.1]
Inference: 56%|█████▌ | 3296/5920 [47:40<24:32, 1.78it/s, tok/s=14789.1]
Inference: 56%|█████▌ | 3296/5920 [47:40<24:32, 1.78it/s, tok/s=14785.2]
Inference: 56%|█████▌ | 3297/5920 [47:40<20:00, 2.18it/s, tok/s=14785.2]
Inference: 56%|█████▌ | 3297/5920 [47:40<20:00, 2.18it/s, tok/s=14784.7]
Inference: 56%|█████▌ | 3298/5920 [47:40<15:50, 2.76it/s, tok/s=14784.7]
Inference: 56%|█████▌ | 3298/5920 [47:40<15:50, 2.76it/s, tok/s=14784.8]
Inference: 56%|█████▌ | 3299/5920 [47:42<28:54, 1.51it/s, tok/s=14784.8]
Inference: 56%|█████▌ | 3299/5920 [47:42<28:54, 1.51it/s, tok/s=14780.1]
Inference: 56%|█████▌ | 3300/5920 [47:42<22:01, 1.98it/s, tok/s=14780.1]
Inference: 56%|█████▌ | 3300/5920 [47:42<22:01, 1.98it/s, tok/s=14781.1]
Inference: 56%|█████▌ | 3301/5920 [47:42<17:27, 2.50it/s, tok/s=14781.1]
Inference: 56%|█████▌ | 3301/5920 [47:42<17:27, 2.50it/s, tok/s=14781.0]
Inference: 56%|█████▌ | 3302/5920 [47:43<19:20, 2.26it/s, tok/s=14781.0]
Inference: 56%|█████▌ | 3302/5920 [47:43<19:20, 2.26it/s, tok/s=14785.8]
Inference: 56%|█████▌ | 3303/5920 [47:43<18:15, 2.39it/s, tok/s=14785.8]
Inference: 56%|█████▌ | 3303/5920 [47:43<18:15, 2.39it/s, tok/s=14784.7]
Inference: 56%|█████▌ | 3304/5920 [47:43<18:39, 2.34it/s, tok/s=14784.7]
Inference: 56%|█████▌ | 3304/5920 [47:43<18:39, 2.34it/s, tok/s=14783.1]
Inference: 56%|█████▌ | 3305/5920 [47:43<18:39, 2.34it/s, tok/s=14784.6]
Inference: 56%|█████▌ | 3306/5920 [47:43<18:38, 2.34it/s, tok/s=14789.5]
Inference: 56%|█████▌ | 3307/5920 [47:43<08:59, 4.84it/s, tok/s=14789.5]
Inference: 56%|█████▌ | 3307/5920 [47:43<08:59, 4.84it/s, tok/s=14790.9]
Inference: 56%|█████▌ | 3308/5920 [47:44<08:59, 4.84it/s, tok/s=14792.4]
Inference: 56%|█████▌ | 3309/5920 [47:44<09:26, 4.61it/s, tok/s=14792.4]
Inference: 56%|█████▌ | 3309/5920 [47:44<09:26, 4.61it/s, tok/s=14791.9]
Inference: 56%|█████▌ | 3310/5920 [47:44<09:44, 4.47it/s, tok/s=14791.9]
Inference: 56%|█████▌ | 3310/5920 [47:44<09:44, 4.47it/s, tok/s=14798.8]
Inference: 56%|█████▌ | 3311/5920 [47:45<12:10, 3.57it/s, tok/s=14798.8]
Inference: 56%|█████▌ | 3311/5920 [47:45<12:10, 3.57it/s, tok/s=14797.0]
Inference: 56%|█████▌ | 3312/5920 [47:45<10:46, 4.04it/s, tok/s=14797.0]
Inference: 56%|█████▌ | 3312/5920 [47:45<10:46, 4.04it/s, tok/s=14796.7]
Inference: 56%|█████▌ | 3313/5920 [47:45<13:17, 3.27it/s, tok/s=14796.7]
Inference: 56%|█████▌ | 3313/5920 [47:45<13:17, 3.27it/s, tok/s=14804.9]
Inference: 56%|█████▌ | 3314/5920 [47:45<13:16, 3.27it/s, tok/s=14805.8]
Inference: 56%|█████▌ | 3315/5920 [47:45<09:26, 4.60it/s, tok/s=14805.8]
Inference: 56%|█████▌ | 3315/5920 [47:45<09:26, 4.60it/s, tok/s=14806.0]
Inference: 56%|█████▌ | 3316/5920 [47:45<09:26, 4.60it/s, tok/s=14806.3]
Inference: 56%|█████▌ | 3317/5920 [47:46<07:17, 5.95it/s, tok/s=14806.3]
Inference: 56%|█████▌ | 3317/5920 [47:46<07:17, 5.95it/s, tok/s=14806.3]
Inference: 56%|█████▌ | 3318/5920 [47:46<10:36, 4.09it/s, tok/s=14806.3]
Inference: 56%|█████▌ | 3318/5920 [47:46<10:36, 4.09it/s, tok/s=14808.1]
Inference: 56%|█████▌ | 3319/5920 [47:46<10:04, 4.30it/s, tok/s=14808.1]
Inference: 56%|█████▌ | 3319/5920 [47:46<10:04, 4.30it/s, tok/s=14809.1]
Inference: 56%|█████▌ | 3320/5920 [47:46<08:47, 4.93it/s, tok/s=14809.1]
Inference: 56%|█████▌ | 3320/5920 [47:46<08:47, 4.93it/s, tok/s=14809.4]
Inference: 56%|█████▌ | 3321/5920 [47:47<08:47, 4.93it/s, tok/s=14816.2]
Inference: 56%|█████▌ | 3322/5920 [47:47<07:10, 6.04it/s, tok/s=14816.2]
Inference: 56%|█████▌ | 3322/5920 [47:47<07:10, 6.04it/s, tok/s=14816.1]
Inference: 56%|█████▌ | 3323/5920 [47:47<07:10, 6.04it/s, tok/s=14821.4]
Inference: 56%|█████▌ | 3324/5920 [47:47<07:09, 6.04it/s, tok/s=14822.4]
Inference: 56%|█████▌ | 3325/5920 [47:47<06:31, 6.62it/s, tok/s=14822.4]
Inference: 56%|█████▌ | 3325/5920 [47:47<06:31, 6.62it/s, tok/s=14821.2]
Inference: 56%|█████▌ | 3326/5920 [47:48<15:17, 2.83it/s, tok/s=14821.2]
Inference: 56%|█████▌ | 3326/5920 [47:48<15:17, 2.83it/s, tok/s=14816.6]
Inference: 56%|█████▌ | 3327/5920 [47:49<14:57, 2.89it/s, tok/s=14816.6]
Inference: 56%|█████▌ | 3327/5920 [47:49<14:57, 2.89it/s, tok/s=14815.4]
Inference: 56%|█████▌ | 3328/5920 [47:49<14:20, 3.01it/s, tok/s=14815.4]
Inference: 56%|█████▌ | 3328/5920 [47:49<14:20, 3.01it/s, tok/s=14817.6]
Inference: 56%|█████▌ | 3329/5920 [47:49<14:20, 3.01it/s, tok/s=14823.6]
Inference: 56%|█████▋ | 3330/5920 [47:49<11:07, 3.88it/s, tok/s=14823.6]
Inference: 56%|█████▋ | 3330/5920 [47:49<11:07, 3.88it/s, tok/s=14832.7]
Inference: 56%|█████▋ | 3331/5920 [47:51<20:53, 2.07it/s, tok/s=14832.7]
Inference: 56%|█████▋ | 3331/5920 [47:51<20:53, 2.07it/s, tok/s=14832.1]
Inference: 56%|█████▋ | 3332/5920 [47:51<24:49, 1.74it/s, tok/s=14832.1]
Inference: 56%|█████▋ | 3332/5920 [47:51<24:49, 1.74it/s, tok/s=14828.6]
Inference: 56%|█████▋ | 3333/5920 [47:52<27:59, 1.54it/s, tok/s=14828.6]
Inference: 56%|█████▋ | 3333/5920 [47:52<27:59, 1.54it/s, tok/s=14824.9]
Inference: 56%|█████▋ | 3334/5920 [47:53<32:20, 1.33it/s, tok/s=14824.9]
Inference: 56%|█████▋ | 3334/5920 [47:53<32:20, 1.33it/s, tok/s=14821.8]
Inference: 56%|█████▋ | 3335/5920 [47:53<32:19, 1.33it/s, tok/s=14822.0]
Inference: 56%|█████▋ | 3336/5920 [47:53<19:44, 2.18it/s, tok/s=14822.0]
Inference: 56%|█████▋ | 3336/5920 [47:53<19:44, 2.18it/s, tok/s=14823.4]
Inference: 56%|█████▋ | 3337/5920 [47:54<17:15, 2.49it/s, tok/s=14823.4]
Inference: 56%|█████▋ | 3337/5920 [47:54<17:15, 2.49it/s, tok/s=14828.3]
Inference: 56%|█████▋ | 3338/5920 [47:54<15:49, 2.72it/s, tok/s=14828.3]
Inference: 56%|█████▋ | 3338/5920 [47:54<15:49, 2.72it/s, tok/s=14828.1]
Inference: 56%|█████▋ | 3339/5920 [47:54<15:49, 2.72it/s, tok/s=14829.1]
Inference: 56%|█████▋ | 3340/5920 [47:54<15:49, 2.72it/s, tok/s=14829.5]
Inference: 56%|█████▋ | 3341/5920 [47:54<08:41, 4.95it/s, tok/s=14829.5]
Inference: 56%|█████▋ | 3341/5920 [47:54<08:41, 4.95it/s, tok/s=14829.5]
Inference: 56%|█████▋ | 3342/5920 [47:55<14:29, 2.96it/s, tok/s=14829.5]
Inference: 56%|█████▋ | 3342/5920 [47:55<14:29, 2.96it/s, tok/s=14825.9]
Inference: 56%|█████▋ | 3343/5920 [47:55<12:16, 3.50it/s, tok/s=14825.9]
Inference: 56%|█████▋ | 3343/5920 [47:55<12:16, 3.50it/s, tok/s=14834.9]
Inference: 56%|█████▋ | 3344/5920 [47:56<18:31, 2.32it/s, tok/s=14834.9]
Inference: 56%|█████▋ | 3344/5920 [47:56<18:31, 2.32it/s, tok/s=14831.8]
Inference: 57%|█████▋ | 3345/5920 [47:56<18:35, 2.31it/s, tok/s=14831.8]
Inference: 57%|█████▋ | 3345/5920 [47:56<18:35, 2.31it/s, tok/s=14837.8]
Inference: 57%|█████▋ | 3346/5920 [47:57<18:55, 2.27it/s, tok/s=14837.8]
Inference: 57%|█████▋ | 3346/5920 [47:57<18:55, 2.27it/s, tok/s=14844.3]
Inference: 57%|█████▋ | 3347/5920 [47:57<17:12, 2.49it/s, tok/s=14844.3]
Inference: 57%|█████▋ | 3347/5920 [47:57<17:12, 2.49it/s, tok/s=14845.1]
Inference: 57%|█████▋ | 3348/5920 [47:58<25:02, 1.71it/s, tok/s=14845.1]
Inference: 57%|█████▋ | 3348/5920 [47:58<25:02, 1.71it/s, tok/s=14840.2]
Inference: 57%|█████▋ | 3349/5920 [47:59<21:50, 1.96it/s, tok/s=14840.2]
Inference: 57%|█████▋ | 3349/5920 [47:59<21:50, 1.96it/s, tok/s=14839.2]
Inference: 57%|█████▋ | 3350/5920 [47:59<18:41, 2.29it/s, tok/s=14839.2]
Inference: 57%|█████▋ | 3350/5920 [47:59<18:41, 2.29it/s, tok/s=14838.6]
Inference: 57%|█████▋ | 3351/5920 [47:59<18:41, 2.29it/s, tok/s=14839.0]
Inference: 57%|█████▋ | 3352/5920 [47:59<15:11, 2.82it/s, tok/s=14839.0]
Inference: 57%|█████▋ | 3352/5920 [47:59<15:11, 2.82it/s, tok/s=14837.5]
Inference: 57%|█████▋ | 3353/5920 [48:00<16:14, 2.63it/s, tok/s=14837.5]
Inference: 57%|█████▋ | 3353/5920 [48:00<16:14, 2.63it/s, tok/s=14835.8]
Inference: 57%|█████▋ | 3354/5920 [48:00<17:31, 2.44it/s, tok/s=14835.8]
Inference: 57%|█████▋ | 3354/5920 [48:00<17:31, 2.44it/s, tok/s=14835.0]
Inference: 57%|█████▋ | 3355/5920 [48:00<17:31, 2.44it/s, tok/s=14837.8]
Inference: 57%|█████▋ | 3356/5920 [48:01<13:09, 3.25it/s, tok/s=14837.8]
Inference: 57%|█████▋ | 3356/5920 [48:01<13:09, 3.25it/s, tok/s=14837.0]
Inference: 57%|█████▋ | 3357/5920 [48:01<13:08, 3.25it/s, tok/s=14838.9]
Inference: 57%|█████▋ | 3358/5920 [48:01<09:40, 4.41it/s, tok/s=14838.9]
Inference: 57%|█████▋ | 3358/5920 [48:01<09:40, 4.41it/s, tok/s=14841.2]
Inference: 57%|█████▋ | 3359/5920 [48:01<09:40, 4.41it/s, tok/s=14846.5]
Inference: 57%|█████▋ | 3360/5920 [48:01<07:54, 5.40it/s, tok/s=14846.5]
Inference: 57%|█████▋ | 3360/5920 [48:01<07:54, 5.40it/s, tok/s=14856.1]
Inference: 57%|█████▋ | 3361/5920 [48:01<08:28, 5.04it/s, tok/s=14856.1]
Inference: 57%|█████▋ | 3361/5920 [48:01<08:28, 5.04it/s, tok/s=14855.4]
Inference: 57%|█████▋ | 3362/5920 [48:02<12:11, 3.50it/s, tok/s=14855.4]
Inference: 57%|█████▋ | 3362/5920 [48:02<12:11, 3.50it/s, tok/s=14856.7]
Inference: 57%|█████▋ | 3363/5920 [48:02<11:51, 3.60it/s, tok/s=14856.7]
Inference: 57%|█████▋ | 3363/5920 [48:02<11:51, 3.60it/s, tok/s=14865.0]
Inference: 57%|█████▋ | 3364/5920 [48:02<10:20, 4.12it/s, tok/s=14865.0]
Inference: 57%|█████▋ | 3364/5920 [48:02<10:20, 4.12it/s, tok/s=14865.4]
Inference: 57%|█████▋ | 3365/5920 [48:02<09:52, 4.31it/s, tok/s=14865.4]
Inference: 57%|█████▋ | 3365/5920 [48:02<09:52, 4.31it/s, tok/s=14865.2]
Inference: 57%|█████▋ | 3366/5920 [48:04<21:06, 2.02it/s, tok/s=14865.2]
Inference: 57%|█████▋ | 3366/5920 [48:04<21:06, 2.02it/s, tok/s=14866.6]
Inference: 57%|█████▋ | 3367/5920 [48:04<21:06, 2.02it/s, tok/s=14877.0]
Inference: 57%|█████▋ | 3368/5920 [48:04<13:06, 3.25it/s, tok/s=14877.0]
Inference: 57%|█████▋ | 3368/5920 [48:04<13:06, 3.25it/s, tok/s=14877.6]
Inference: 57%|█████▋ | 3369/5920 [48:05<22:03, 1.93it/s, tok/s=14877.6]
Inference: 57%|█████▋ | 3369/5920 [48:05<22:03, 1.93it/s, tok/s=14871.8]
Inference: 57%|█████▋ | 3370/5920 [48:05<17:48, 2.39it/s, tok/s=14871.8]
Inference: 57%|█████▋ | 3370/5920 [48:05<17:48, 2.39it/s, tok/s=14878.6]
Inference: 57%|█████▋ | 3371/5920 [48:06<20:54, 2.03it/s, tok/s=14878.6]
Inference: 57%|█████▋ | 3371/5920 [48:06<20:54, 2.03it/s, tok/s=14876.0]
Inference: 57%|█████▋ | 3372/5920 [48:06<18:20, 2.32it/s, tok/s=14876.0]
Inference: 57%|█████▋ | 3372/5920 [48:06<18:20, 2.32it/s, tok/s=14876.4]
Inference: 57%|█████▋ | 3373/5920 [48:06<18:19, 2.32it/s, tok/s=14876.6]
Inference: 57%|█████▋ | 3374/5920 [48:06<12:20, 3.44it/s, tok/s=14876.6]
Inference: 57%|█████▋ | 3374/5920 [48:06<12:20, 3.44it/s, tok/s=14882.6]
Inference: 57%|█████▋ | 3375/5920 [48:06<12:20, 3.44it/s, tok/s=14890.5]
Inference: 57%|█████▋ | 3376/5920 [48:07<11:20, 3.74it/s, tok/s=14890.5]
Inference: 57%|█████▋ | 3376/5920 [48:07<11:20, 3.74it/s, tok/s=14890.7]
Inference: 57%|█████▋ | 3377/5920 [48:07<10:46, 3.93it/s, tok/s=14890.7]
Inference: 57%|█████▋ | 3377/5920 [48:07<10:46, 3.93it/s, tok/s=14889.9]
Inference: 57%|█████▋ | 3378/5920 [48:07<10:46, 3.93it/s, tok/s=14897.7]
Inference: 57%|█████▋ | 3379/5920 [48:07<09:54, 4.27it/s, tok/s=14897.7]
Inference: 57%|█████▋ | 3379/5920 [48:07<09:54, 4.27it/s, tok/s=14896.2]
Inference: 57%|█████▋ | 3380/5920 [48:07<08:44, 4.84it/s, tok/s=14896.2]
Inference: 57%|█████▋ | 3380/5920 [48:07<08:44, 4.84it/s, tok/s=14903.5]
Inference: 57%|█████▋ | 3381/5920 [48:08<14:10, 2.99it/s, tok/s=14903.5]
Inference: 57%|█████▋ | 3381/5920 [48:08<14:10, 2.99it/s, tok/s=14905.6]
Inference: 57%|█████▋ | 3382/5920 [48:08<13:11, 3.21it/s, tok/s=14905.6]
Inference: 57%|█████▋ | 3382/5920 [48:08<13:11, 3.21it/s, tok/s=14905.8]
Inference: 57%|█████▋ | 3383/5920 [48:08<13:11, 3.21it/s, tok/s=14912.5]
Inference: 57%|█████▋ | 3384/5920 [48:09<11:29, 3.68it/s, tok/s=14912.5]
Inference: 57%|█████▋ | 3384/5920 [48:09<11:29, 3.68it/s, tok/s=14910.9]
Inference: 57%|█████▋ | 3385/5920 [48:09<12:23, 3.41it/s, tok/s=14910.9]
Inference: 57%|█████▋ | 3385/5920 [48:09<12:23, 3.41it/s, tok/s=14910.6]
Inference: 57%|█████▋ | 3386/5920 [48:10<14:56, 2.83it/s, tok/s=14910.6]
Inference: 57%|█████▋ | 3386/5920 [48:10<14:56, 2.83it/s, tok/s=14908.3]
Inference: 57%|█████▋ | 3387/5920 [48:10<14:56, 2.83it/s, tok/s=14908.8]
Inference: 57%|█████▋ | 3388/5920 [48:10<10:39, 3.96it/s, tok/s=14908.8]
Inference: 57%|█████▋ | 3388/5920 [48:10<10:39, 3.96it/s, tok/s=14910.3]
Inference: 57%|█████▋ | 3389/5920 [48:11<20:26, 2.06it/s, tok/s=14910.3]
Inference: 57%|█████▋ | 3389/5920 [48:11<20:26, 2.06it/s, tok/s=14904.6]
Inference: 57%|█████▋ | 3390/5920 [48:11<16:34, 2.55it/s, tok/s=14904.6]
Inference: 57%|█████▋ | 3390/5920 [48:11<16:34, 2.55it/s, tok/s=14904.7]
Inference: 57%|█████▋ | 3391/5920 [48:11<16:33, 2.55it/s, tok/s=14905.3]
Inference: 57%|█████▋ | 3392/5920 [48:12<12:48, 3.29it/s, tok/s=14905.3]
Inference: 57%|█████▋ | 3392/5920 [48:12<12:48, 3.29it/s, tok/s=14905.0]
Inference: 57%|█████▋ | 3393/5920 [48:12<12:48, 3.29it/s, tok/s=14912.1]
Inference: 57%|█████▋ | 3394/5920 [48:13<15:16, 2.76it/s, tok/s=14912.1]
Inference: 57%|█████▋ | 3394/5920 [48:13<15:16, 2.76it/s, tok/s=14908.4]
Inference: 57%|█████▋ | 3395/5920 [48:13<15:36, 2.70it/s, tok/s=14908.4]
Inference: 57%|█████▋ | 3395/5920 [48:13<15:36, 2.70it/s, tok/s=14907.3]
Inference: 57%|█████▋ | 3396/5920 [48:13<15:11, 2.77it/s, tok/s=14907.3]
Inference: 57%|█████▋ | 3396/5920 [48:13<15:11, 2.77it/s, tok/s=14906.3]
Inference: 57%|█████▋ | 3397/5920 [48:14<13:43, 3.06it/s, tok/s=14906.3]
Inference: 57%|█████▋ | 3397/5920 [48:14<13:43, 3.06it/s, tok/s=14908.2]
Inference: 57%|█████▋ | 3398/5920 [48:15<26:01, 1.62it/s, tok/s=14908.2]
Inference: 57%|█████▋ | 3398/5920 [48:15<26:01, 1.62it/s, tok/s=14903.8]
Inference: 57%|█████▋ | 3399/5920 [48:16<27:51, 1.51it/s, tok/s=14903.8]
Inference: 57%|█████▋ | 3399/5920 [48:16<27:51, 1.51it/s, tok/s=14901.0]
Inference: 57%|█████▋ | 3400/5920 [48:16<25:48, 1.63it/s, tok/s=14901.0]
Inference: 57%|█████▋ | 3400/5920 [48:16<25:48, 1.63it/s, tok/s=14898.9]
Inference: 57%|█████▋ | 3401/5920 [48:16<25:47, 1.63it/s, tok/s=14905.3]
Inference: 57%|█████▋ | 3402/5920 [48:16<25:47, 1.63it/s, tok/s=14911.3]
Inference: 57%|█████▋ | 3403/5920 [48:16<12:43, 3.29it/s, tok/s=14911.3]
Inference: 57%|█████▋ | 3403/5920 [48:16<12:43, 3.29it/s, tok/s=14911.4]
Inference: 57%|█████▊ | 3404/5920 [48:17<11:12, 3.74it/s, tok/s=14911.4]
Inference: 57%|█████▊ | 3404/5920 [48:17<11:12, 3.74it/s, tok/s=14911.5]
Inference: 58%|█████▊ | 3405/5920 [48:17<11:28, 3.65it/s, tok/s=14911.5]
Inference: 58%|█████▊ | 3405/5920 [48:17<11:28, 3.65it/s, tok/s=14912.6]
Inference: 58%|█████▊ | 3406/5920 [48:17<11:28, 3.65it/s, tok/s=14913.1]
Inference: 58%|█████▊ | 3407/5920 [48:17<11:42, 3.58it/s, tok/s=14913.1]
Inference: 58%|█████▊ | 3407/5920 [48:17<11:42, 3.58it/s, tok/s=14911.1]
Inference: 58%|█████▊ | 3408/5920 [48:18<15:11, 2.76it/s, tok/s=14911.1]
Inference: 58%|█████▊ | 3408/5920 [48:18<15:11, 2.76it/s, tok/s=14910.3]
Inference: 58%|█████▊ | 3409/5920 [48:19<15:30, 2.70it/s, tok/s=14910.3]
Inference: 58%|█████▊ | 3409/5920 [48:19<15:30, 2.70it/s, tok/s=14911.7]
Inference: 58%|█████▊ | 3410/5920 [48:20<24:59, 1.67it/s, tok/s=14911.7]
Inference: 58%|█████▊ | 3410/5920 [48:20<24:59, 1.67it/s, tok/s=14905.6]
Inference: 58%|█████▊ | 3411/5920 [48:20<21:40, 1.93it/s, tok/s=14905.6]
Inference: 58%|█████▊ | 3411/5920 [48:20<21:40, 1.93it/s, tok/s=14904.5]
Inference: 58%|█████▊ | 3412/5920 [48:20<21:40, 1.93it/s, tok/s=14904.7]
Inference: 58%|█████▊ | 3413/5920 [48:20<13:32, 3.09it/s, tok/s=14904.7]
Inference: 58%|█████▊ | 3413/5920 [48:20<13:32, 3.09it/s, tok/s=14905.1]
Inference: 58%|█████▊ | 3414/5920 [48:20<11:59, 3.49it/s, tok/s=14905.1]
Inference: 58%|█████▊ | 3414/5920 [48:20<11:59, 3.49it/s, tok/s=14905.0]
Inference: 58%|█████▊ | 3415/5920 [48:21<11:44, 3.55it/s, tok/s=14905.0]
Inference: 58%|█████▊ | 3415/5920 [48:21<11:44, 3.55it/s, tok/s=14906.8]
Inference: 58%|█████▊ | 3416/5920 [48:21<11:44, 3.55it/s, tok/s=14911.2]
Inference: 58%|█████▊ | 3417/5920 [48:21<13:18, 3.13it/s, tok/s=14911.2]
Inference: 58%|█████▊ | 3417/5920 [48:21<13:18, 3.13it/s, tok/s=14914.6]
Inference: 58%|█████▊ | 3418/5920 [48:22<12:26, 3.35it/s, tok/s=14914.6]
Inference: 58%|█████▊ | 3418/5920 [48:22<12:26, 3.35it/s, tok/s=14915.6]
Inference: 58%|█████▊ | 3419/5920 [48:22<12:26, 3.35it/s, tok/s=14916.3]
Inference: 58%|█████▊ | 3420/5920 [48:22<11:31, 3.61it/s, tok/s=14916.3]
Inference: 58%|█████▊ | 3420/5920 [48:22<11:31, 3.61it/s, tok/s=14914.8]
Inference: 58%|█████▊ | 3421/5920 [48:22<11:31, 3.61it/s, tok/s=14923.8]
Inference: 58%|█████▊ | 3422/5920 [48:23<10:46, 3.87it/s, tok/s=14923.8]
Inference: 58%|█████▊ | 3422/5920 [48:23<10:46, 3.87it/s, tok/s=14923.4]
Inference: 58%|█████▊ | 3423/5920 [48:23<15:03, 2.76it/s, tok/s=14923.4]
Inference: 58%|█████▊ | 3423/5920 [48:23<15:03, 2.76it/s, tok/s=14920.1]
Inference: 58%|█████▊ | 3424/5920 [48:24<19:17, 2.16it/s, tok/s=14920.1]
Inference: 58%|█████▊ | 3424/5920 [48:24<19:17, 2.16it/s, tok/s=14918.2]
Inference: 58%|█████▊ | 3425/5920 [48:25<20:58, 1.98it/s, tok/s=14918.2]
Inference: 58%|█████▊ | 3425/5920 [48:25<20:58, 1.98it/s, tok/s=14915.4]
Inference: 58%|█████▊ | 3426/5920 [48:25<20:58, 1.98it/s, tok/s=14920.3]
Inference: 58%|█████▊ | 3427/5920 [48:25<17:09, 2.42it/s, tok/s=14920.3]
Inference: 58%|█████▊ | 3427/5920 [48:25<17:09, 2.42it/s, tok/s=14920.6]
Inference: 58%|█████▊ | 3428/5920 [48:25<14:39, 2.83it/s, tok/s=14920.6]
Inference: 58%|█████▊ | 3428/5920 [48:25<14:39, 2.83it/s, tok/s=14920.2]
Inference: 58%|█████▊ | 3429/5920 [48:26<12:49, 3.24it/s, tok/s=14920.2]
Inference: 58%|█████▊ | 3429/5920 [48:26<12:49, 3.24it/s, tok/s=14921.6]
Inference: 58%|█████▊ | 3430/5920 [48:26<12:49, 3.24it/s, tok/s=14923.6]
Inference: 58%|█████▊ | 3431/5920 [48:26<09:15, 4.48it/s, tok/s=14923.6]
Inference: 58%|█████▊ | 3431/5920 [48:26<09:15, 4.48it/s, tok/s=14924.2]
Inference: 58%|█████▊ | 3432/5920 [48:26<09:07, 4.54it/s, tok/s=14924.2]
Inference: 58%|█████▊ | 3432/5920 [48:26<09:07, 4.54it/s, tok/s=14923.5]
Inference: 58%|█████▊ | 3433/5920 [48:27<11:36, 3.57it/s, tok/s=14923.5]
Inference: 58%|█████▊ | 3433/5920 [48:27<11:36, 3.57it/s, tok/s=14921.6]
Inference: 58%|█████▊ | 3434/5920 [48:27<09:43, 4.26it/s, tok/s=14921.6]
Inference: 58%|█████▊ | 3434/5920 [48:27<09:43, 4.26it/s, tok/s=14930.1]
Inference: 58%|█████▊ | 3435/5920 [48:27<08:20, 4.96it/s, tok/s=14930.1]
Inference: 58%|█████▊ | 3435/5920 [48:27<08:20, 4.96it/s, tok/s=14930.6]
Inference: 58%|█████▊ | 3436/5920 [48:27<08:20, 4.96it/s, tok/s=14935.8]
Inference: 58%|█████▊ | 3437/5920 [48:27<06:52, 6.02it/s, tok/s=14935.8]
Inference: 58%|█████▊ | 3437/5920 [48:27<06:52, 6.02it/s, tok/s=14944.1]
Inference: 58%|█████▊ | 3438/5920 [48:27<06:52, 6.02it/s, tok/s=14948.7]
Inference: 58%|█████▊ | 3439/5920 [48:27<05:07, 8.07it/s, tok/s=14948.7]
Inference: 58%|█████▊ | 3439/5920 [48:27<05:07, 8.07it/s, tok/s=14959.0]
Inference: 58%|█████▊ | 3440/5920 [48:27<05:07, 8.07it/s, tok/s=14961.3]
Inference: 58%|█████▊ | 3441/5920 [48:27<05:07, 8.07it/s, tok/s=14962.1]
Inference: 58%|█████▊ | 3442/5920 [48:27<04:43, 8.74it/s, tok/s=14962.1]
Inference: 58%|█████▊ | 3442/5920 [48:27<04:43, 8.74it/s, tok/s=14963.7]
Inference: 58%|█████▊ | 3443/5920 [48:28<04:43, 8.74it/s, tok/s=14965.3]
Inference: 58%|█████▊ | 3444/5920 [48:28<06:23, 6.45it/s, tok/s=14965.3]
Inference: 58%|█████▊ | 3444/5920 [48:28<06:23, 6.45it/s, tok/s=14964.6]
Inference: 58%|█████▊ | 3445/5920 [48:28<06:23, 6.45it/s, tok/s=14964.7]
Inference: 58%|█████▊ | 3446/5920 [48:28<05:07, 8.04it/s, tok/s=14964.7]
Inference: 58%|█████▊ | 3446/5920 [48:28<05:07, 8.04it/s, tok/s=14967.1]
Inference: 58%|█████▊ | 3447/5920 [48:28<05:07, 8.04it/s, tok/s=14969.0]
Inference: 58%|█████▊ | 3448/5920 [48:29<06:42, 6.14it/s, tok/s=14969.0]
Inference: 58%|█████▊ | 3448/5920 [48:29<06:42, 6.14it/s, tok/s=14967.2]
Inference: 58%|█████▊ | 3449/5920 [48:29<06:24, 6.43it/s, tok/s=14967.2]
Inference: 58%|█████▊ | 3449/5920 [48:29<06:24, 6.43it/s, tok/s=14968.3]
Inference: 58%|█████▊ | 3450/5920 [48:29<09:43, 4.24it/s, tok/s=14968.3]
Inference: 58%|█████▊ | 3450/5920 [48:29<09:43, 4.24it/s, tok/s=14969.5]
Inference: 58%|█████▊ | 3451/5920 [48:29<09:42, 4.24it/s, tok/s=14971.4]
Inference: 58%|█████▊ | 3452/5920 [48:30<08:49, 4.66it/s, tok/s=14971.4]
Inference: 58%|█████▊ | 3452/5920 [48:30<08:49, 4.66it/s, tok/s=14975.2]
Inference: 58%|█████▊ | 3453/5920 [48:30<10:04, 4.08it/s, tok/s=14975.2]
Inference: 58%|█████▊ | 3453/5920 [48:30<10:04, 4.08it/s, tok/s=14974.1]
Inference: 58%|█████▊ | 3454/5920 [48:30<08:44, 4.70it/s, tok/s=14974.1]
Inference: 58%|█████▊ | 3454/5920 [48:30<08:44, 4.70it/s, tok/s=14974.4]
Inference: 58%|█████▊ | 3455/5920 [48:30<08:44, 4.70it/s, tok/s=14975.8]
Inference: 58%|█████▊ | 3456/5920 [48:30<08:44, 4.70it/s, tok/s=14976.2]
Inference: 58%|█████▊ | 3457/5920 [48:31<08:25, 4.88it/s, tok/s=14976.2]
Inference: 58%|█████▊ | 3457/5920 [48:31<08:25, 4.88it/s, tok/s=14975.4]
Inference: 58%|█████▊ | 3458/5920 [48:31<08:47, 4.67it/s, tok/s=14975.4]
Inference: 58%|█████▊ | 3458/5920 [48:31<08:47, 4.67it/s, tok/s=14974.5]
Inference: 58%|█████▊ | 3459/5920 [48:31<08:07, 5.05it/s, tok/s=14974.5]
Inference: 58%|█████▊ | 3459/5920 [48:31<08:07, 5.05it/s, tok/s=14974.4]
Inference: 58%|█████▊ | 3460/5920 [48:31<07:10, 5.71it/s, tok/s=14974.4]
Inference: 58%|█████▊ | 3460/5920 [48:31<07:10, 5.71it/s, tok/s=14974.7]
Inference: 58%|█████▊ | 3461/5920 [48:31<08:20, 4.92it/s, tok/s=14974.7]
Inference: 58%|█████▊ | 3461/5920 [48:31<08:20, 4.92it/s, tok/s=14974.1]
Inference: 58%|█████▊ | 3462/5920 [48:32<08:07, 5.04it/s, tok/s=14974.1]
Inference: 58%|█████▊ | 3462/5920 [48:32<08:07, 5.04it/s, tok/s=14974.7]
Inference: 58%|█████▊ | 3463/5920 [48:32<08:07, 5.04it/s, tok/s=14975.1]
Inference: 59%|█████▊ | 3464/5920 [48:32<08:07, 5.04it/s, tok/s=14976.0]
Inference: 59%|█████▊ | 3465/5920 [48:32<06:05, 6.72it/s, tok/s=14976.0]
Inference: 59%|█████▊ | 3465/5920 [48:32<06:05, 6.72it/s, tok/s=14976.8]
Inference: 59%|█████▊ | 3466/5920 [48:32<06:05, 6.72it/s, tok/s=14985.1]
Inference: 59%|█████▊ | 3467/5920 [48:32<07:30, 5.45it/s, tok/s=14985.1]
Inference: 59%|█████▊ | 3467/5920 [48:32<07:30, 5.45it/s, tok/s=14983.1]
Inference: 59%|█████▊ | 3468/5920 [48:32<07:30, 5.45it/s, tok/s=14983.6]
Inference: 59%|█████▊ | 3469/5920 [48:33<07:09, 5.70it/s, tok/s=14983.6]
Inference: 59%|█████▊ | 3469/5920 [48:33<07:09, 5.70it/s, tok/s=14991.5]
Inference: 59%|█████▊ | 3470/5920 [48:33<08:09, 5.01it/s, tok/s=14991.5]
Inference: 59%|█████▊ | 3470/5920 [48:33<08:09, 5.01it/s, tok/s=14992.6]
Inference: 59%|█████▊ | 3471/5920 [48:33<08:09, 5.01it/s, tok/s=14992.6]
Inference: 59%|█████▊ | 3472/5920 [48:33<07:07, 5.73it/s, tok/s=14992.6]
Inference: 59%|█████▊ | 3472/5920 [48:33<07:07, 5.73it/s, tok/s=14992.3]
Inference: 59%|█████▊ | 3473/5920 [48:33<06:31, 6.26it/s, tok/s=14992.3]
Inference: 59%|█████▊ | 3473/5920 [48:33<06:31, 6.26it/s, tok/s=14992.4]
Inference: 59%|█████▊ | 3474/5920 [48:34<08:23, 4.86it/s, tok/s=14992.4]
Inference: 59%|█████▊ | 3474/5920 [48:34<08:23, 4.86it/s, tok/s=14991.3]
Inference: 59%|█████▊ | 3475/5920 [48:34<09:27, 4.31it/s, tok/s=14991.3]
Inference: 59%|█████▊ | 3475/5920 [48:34<09:27, 4.31it/s, tok/s=14992.2]
Inference: 59%|█████▊ | 3476/5920 [48:34<09:27, 4.31it/s, tok/s=14994.6]
Inference: 59%|█████▊ | 3477/5920 [48:34<06:32, 6.22it/s, tok/s=14994.6]
Inference: 59%|█████▊ | 3477/5920 [48:34<06:32, 6.22it/s, tok/s=14995.7]
Inference: 59%|█████▉ | 3478/5920 [48:34<07:22, 5.52it/s, tok/s=14995.7]
Inference: 59%|█████▉ | 3478/5920 [48:34<07:22, 5.52it/s, tok/s=14998.1]
Inference: 59%|█████▉ | 3479/5920 [48:35<08:48, 4.62it/s, tok/s=14998.1]
Inference: 59%|█████▉ | 3479/5920 [48:35<08:48, 4.62it/s, tok/s=14999.9]
Inference: 59%|█████▉ | 3480/5920 [48:35<10:21, 3.93it/s, tok/s=14999.9]
Inference: 59%|█████▉ | 3480/5920 [48:35<10:21, 3.93it/s, tok/s=15000.1]
Inference: 59%|█████▉ | 3481/5920 [48:35<11:03, 3.67it/s, tok/s=15000.1]
Inference: 59%|█████▉ | 3481/5920 [48:35<11:03, 3.67it/s, tok/s=14998.8]
Inference: 59%|█████▉ | 3482/5920 [48:36<14:19, 2.84it/s, tok/s=14998.8]
Inference: 59%|█████▉ | 3482/5920 [48:36<14:19, 2.84it/s, tok/s=14996.7]
Inference: 59%|█████▉ | 3483/5920 [48:36<14:18, 2.84it/s, tok/s=14997.0]
Inference: 59%|█████▉ | 3484/5920 [48:36<12:33, 3.23it/s, tok/s=14997.0]
Inference: 59%|█████▉ | 3484/5920 [48:36<12:33, 3.23it/s, tok/s=14997.1]
Inference: 59%|█████▉ | 3485/5920 [48:37<12:46, 3.18it/s, tok/s=14997.1]
Inference: 59%|█████▉ | 3485/5920 [48:37<12:46, 3.18it/s, tok/s=14996.5]
Inference: 59%|█████▉ | 3486/5920 [48:37<12:45, 3.18it/s, tok/s=14997.2]
Inference: 59%|█████▉ | 3487/5920 [48:37<11:04, 3.66it/s, tok/s=14997.2]
Inference: 59%|█████▉ | 3487/5920 [48:37<11:04, 3.66it/s, tok/s=14997.6]
Inference: 59%|█████▉ | 3488/5920 [48:37<09:48, 4.13it/s, tok/s=14997.6]
Inference: 59%|█████▉ | 3488/5920 [48:37<09:48, 4.13it/s, tok/s=15004.9]
Inference: 59%|█████▉ | 3489/5920 [48:37<09:48, 4.13it/s, tok/s=15005.5]
Inference: 59%|█████▉ | 3490/5920 [48:37<09:47, 4.13it/s, tok/s=15006.7]
Inference: 59%|█████▉ | 3491/5920 [48:37<09:47, 4.13it/s, tok/s=15007.1]
Inference: 59%|█████▉ | 3492/5920 [48:38<08:54, 4.54it/s, tok/s=15007.1]
Inference: 59%|█████▉ | 3492/5920 [48:38<08:54, 4.54it/s, tok/s=15007.9]
Inference: 59%|█████▉ | 3493/5920 [48:38<09:13, 4.38it/s, tok/s=15007.9]
Inference: 59%|█████▉ | 3493/5920 [48:38<09:13, 4.38it/s, tok/s=15009.9]
Inference: 59%|█████▉ | 3494/5920 [48:39<13:59, 2.89it/s, tok/s=15009.9]
Inference: 59%|█████▉ | 3494/5920 [48:39<13:59, 2.89it/s, tok/s=15007.4]
Inference: 59%|█████▉ | 3495/5920 [48:40<15:54, 2.54it/s, tok/s=15007.4]
Inference: 59%|█████▉ | 3495/5920 [48:40<15:54, 2.54it/s, tok/s=15005.5]
Inference: 59%|█████▉ | 3496/5920 [48:40<15:53, 2.54it/s, tok/s=15008.8]
Inference: 59%|█████▉ | 3497/5920 [48:40<11:02, 3.66it/s, tok/s=15008.8]
Inference: 59%|█████▉ | 3497/5920 [48:40<11:02, 3.66it/s, tok/s=15008.9]
Inference: 59%|█████▉ | 3498/5920 [48:40<10:24, 3.88it/s, tok/s=15008.9]
Inference: 59%|█████▉ | 3498/5920 [48:40<10:24, 3.88it/s, tok/s=15008.5]
Inference: 59%|█████▉ | 3499/5920 [48:41<16:38, 2.42it/s, tok/s=15008.5]
Inference: 59%|█████▉ | 3499/5920 [48:41<16:38, 2.42it/s, tok/s=15004.5]
Inference: 59%|█████▉ | 3500/5920 [48:42<17:50, 2.26it/s, tok/s=15004.5]
Inference: 59%|█████▉ | 3500/5920 [48:42<17:50, 2.26it/s, tok/s=15009.9]
Inference: 59%|█████▉ | 3501/5920 [48:42<17:50, 2.26it/s, tok/s=15010.4]
Inference: 59%|█████▉ | 3502/5920 [48:42<17:49, 2.26it/s, tok/s=15010.8]
Inference: 59%|█████▉ | 3503/5920 [48:42<10:17, 3.92it/s, tok/s=15010.8]
Inference: 59%|█████▉ | 3503/5920 [48:42<10:17, 3.92it/s, tok/s=15010.5]
Inference: 59%|█████▉ | 3504/5920 [48:42<10:14, 3.93it/s, tok/s=15010.5]
Inference: 59%|█████▉ | 3504/5920 [48:42<10:14, 3.93it/s, tok/s=15009.9]
Inference: 59%|█████▉ | 3505/5920 [48:43<15:08, 2.66it/s, tok/s=15009.9]
Inference: 59%|█████▉ | 3505/5920 [48:43<15:08, 2.66it/s, tok/s=15012.5]
Inference: 59%|█████▉ | 3506/5920 [48:43<13:59, 2.88it/s, tok/s=15012.5]
Inference: 59%|█████▉ | 3506/5920 [48:43<13:59, 2.88it/s, tok/s=15011.8]
Inference: 59%|█████▉ | 3507/5920 [48:43<13:59, 2.88it/s, tok/s=15011.9]
Inference: 59%|█████▉ | 3508/5920 [48:44<15:16, 2.63it/s, tok/s=15011.9]
Inference: 59%|█████▉ | 3508/5920 [48:44<15:16, 2.63it/s, tok/s=15008.5]
Inference: 59%|█████▉ | 3509/5920 [48:44<15:16, 2.63it/s, tok/s=15008.7]
Inference: 59%|█████▉ | 3510/5920 [48:44<12:03, 3.33it/s, tok/s=15008.7]
Inference: 59%|█████▉ | 3510/5920 [48:44<12:03, 3.33it/s, tok/s=15008.4]
Inference: 59%|█████▉ | 3511/5920 [48:45<12:03, 3.33it/s, tok/s=15008.6]
Inference: 59%|█████▉ | 3512/5920 [48:45<12:03, 3.33it/s, tok/s=15014.9]
Inference: 59%|█████▉ | 3513/5920 [48:45<08:09, 4.92it/s, tok/s=15014.9]
Inference: 59%|█████▉ | 3513/5920 [48:45<08:09, 4.92it/s, tok/s=15017.0]
Inference: 59%|█████▉ | 3514/5920 [48:45<08:09, 4.92it/s, tok/s=15017.7]
Inference: 59%|█████▉ | 3515/5920 [48:45<06:39, 6.02it/s, tok/s=15017.7]
Inference: 59%|█████▉ | 3515/5920 [48:45<06:39, 6.02it/s, tok/s=15017.6]
Inference: 59%|█████▉ | 3516/5920 [48:45<08:21, 4.79it/s, tok/s=15017.6]
Inference: 59%|█████▉ | 3516/5920 [48:45<08:21, 4.79it/s, tok/s=15022.7]
Inference: 59%|█████▉ | 3517/5920 [48:45<08:21, 4.79it/s, tok/s=15023.3]
Inference: 59%|█████▉ | 3518/5920 [48:46<09:24, 4.25it/s, tok/s=15023.3]
Inference: 59%|█████▉ | 3518/5920 [48:46<09:24, 4.25it/s, tok/s=15021.1]
Inference: 59%|█████▉ | 3519/5920 [48:46<09:24, 4.25it/s, tok/s=15022.6]
Inference: 59%|█████▉ | 3520/5920 [48:46<07:27, 5.36it/s, tok/s=15022.6]
Inference: 59%|█████▉ | 3520/5920 [48:46<07:27, 5.36it/s, tok/s=15022.6]
Inference: 59%|█████▉ | 3521/5920 [48:46<08:40, 4.61it/s, tok/s=15022.6]
Inference: 59%|█████▉ | 3521/5920 [48:46<08:40, 4.61it/s, tok/s=15021.4]
Inference: 59%|█████▉ | 3522/5920 [48:47<10:41, 3.74it/s, tok/s=15021.4]
Inference: 59%|█████▉ | 3522/5920 [48:47<10:41, 3.74it/s, tok/s=15019.6]
Inference: 60%|█████▉ | 3523/5920 [48:47<10:18, 3.88it/s, tok/s=15019.6]
Inference: 60%|█████▉ | 3523/5920 [48:47<10:18, 3.88it/s, tok/s=15021.0]
Inference: 60%|█████▉ | 3524/5920 [48:47<10:17, 3.88it/s, tok/s=15021.6]
Inference: 60%|█████▉ | 3525/5920 [48:47<09:37, 4.15it/s, tok/s=15021.6]
Inference: 60%|█████▉ | 3525/5920 [48:47<09:37, 4.15it/s, tok/s=15021.4]
Inference: 60%|█████▉ | 3526/5920 [48:48<08:25, 4.74it/s, tok/s=15021.4]
Inference: 60%|█████▉ | 3526/5920 [48:48<08:25, 4.74it/s, tok/s=15021.8]
Inference: 60%|█████▉ | 3527/5920 [48:48<07:32, 5.29it/s, tok/s=15021.8]
Inference: 60%|█████▉ | 3527/5920 [48:48<07:32, 5.29it/s, tok/s=15021.8]
Inference: 60%|█████▉ | 3528/5920 [48:48<12:48, 3.11it/s, tok/s=15021.8]
Inference: 60%|█████▉ | 3528/5920 [48:48<12:48, 3.11it/s, tok/s=15019.2]
Inference: 60%|█████▉ | 3529/5920 [48:49<14:58, 2.66it/s, tok/s=15019.2]
Inference: 60%|█████▉ | 3529/5920 [48:49<14:58, 2.66it/s, tok/s=15017.3]
Inference: 60%|█████▉ | 3530/5920 [48:49<14:57, 2.66it/s, tok/s=15017.4]
Inference: 60%|█████▉ | 3531/5920 [48:49<12:35, 3.16it/s, tok/s=15017.4]
Inference: 60%|█████▉ | 3531/5920 [48:49<12:35, 3.16it/s, tok/s=15016.1]
Inference: 60%|█████▉ | 3532/5920 [48:49<10:49, 3.68it/s, tok/s=15016.1]
Inference: 60%|█████▉ | 3532/5920 [48:49<10:49, 3.68it/s, tok/s=15019.0]
Inference: 60%|█████▉ | 3533/5920 [48:50<12:43, 3.13it/s, tok/s=15019.0]
Inference: 60%|█████▉ | 3533/5920 [48:50<12:43, 3.13it/s, tok/s=15019.2]
Inference: 60%|█████▉ | 3534/5920 [48:50<11:15, 3.53it/s, tok/s=15019.2]
Inference: 60%|█████▉ | 3534/5920 [48:50<11:15, 3.53it/s, tok/s=15019.2]
Inference: 60%|█████▉ | 3535/5920 [48:50<11:14, 3.53it/s, tok/s=15019.8]
Inference: 60%|█████▉ | 3536/5920 [48:51<12:47, 3.11it/s, tok/s=15019.8]
Inference: 60%|█████▉ | 3536/5920 [48:51<12:47, 3.11it/s, tok/s=15017.0]
Inference: 60%|█████▉ | 3537/5920 [48:51<12:46, 3.11it/s, tok/s=15018.7]
Inference: 60%|█████▉ | 3538/5920 [48:51<08:40, 4.58it/s, tok/s=15018.7]
Inference: 60%|█████▉ | 3538/5920 [48:51<08:40, 4.58it/s, tok/s=15018.9]
Inference: 60%|█████▉ | 3539/5920 [48:51<07:55, 5.01it/s, tok/s=15018.9]
Inference: 60%|█████▉ | 3539/5920 [48:51<07:55, 5.01it/s, tok/s=15021.2]
Inference: 60%|█████▉ | 3540/5920 [48:51<07:50, 5.06it/s, tok/s=15021.2]
Inference: 60%|█████▉ | 3540/5920 [48:51<07:50, 5.06it/s, tok/s=15021.6]
Inference: 60%|█████▉ | 3541/5920 [48:51<07:50, 5.06it/s, tok/s=15024.7]
Inference: 60%|█████▉ | 3542/5920 [48:52<07:06, 5.57it/s, tok/s=15024.7]
Inference: 60%|█████▉ | 3542/5920 [48:52<07:06, 5.57it/s, tok/s=15024.1]
Inference: 60%|█████▉ | 3543/5920 [48:52<07:06, 5.57it/s, tok/s=15024.7]
Inference: 60%|█████▉ | 3544/5920 [48:52<07:06, 5.57it/s, tok/s=15025.4]
Inference: 60%|█████▉ | 3545/5920 [48:52<07:12, 5.49it/s, tok/s=15025.4]
Inference: 60%|█████▉ | 3545/5920 [48:52<07:12, 5.49it/s, tok/s=15026.3]
Inference: 60%|█████▉ | 3546/5920 [48:52<06:48, 5.82it/s, tok/s=15026.3]
Inference: 60%|█████▉ | 3546/5920 [48:52<06:48, 5.82it/s, tok/s=15026.7]
Inference: 60%|█████▉ | 3547/5920 [48:53<08:16, 4.78it/s, tok/s=15026.7]
Inference: 60%|█████▉ | 3547/5920 [48:53<08:16, 4.78it/s, tok/s=15025.3]
Inference: 60%|█████▉ | 3548/5920 [48:53<08:04, 4.90it/s, tok/s=15025.3]
Inference: 60%|█████▉ | 3548/5920 [48:53<08:04, 4.90it/s, tok/s=15025.4]
Inference: 60%|█████▉ | 3549/5920 [48:53<08:03, 4.90it/s, tok/s=15025.7]
Inference: 60%|█████▉ | 3550/5920 [48:53<08:03, 4.90it/s, tok/s=15026.2]
Inference: 60%|█████▉ | 3551/5920 [48:53<05:07, 7.71it/s, tok/s=15026.2]
Inference: 60%|█████▉ | 3551/5920 [48:53<05:07, 7.71it/s, tok/s=15027.3]
Inference: 60%|██████ | 3552/5920 [48:53<07:15, 5.44it/s, tok/s=15027.3]
Inference: 60%|██████ | 3552/5920 [48:53<07:15, 5.44it/s, tok/s=15028.7]
Inference: 60%|██████ | 3553/5920 [48:54<14:13, 2.77it/s, tok/s=15028.7]
Inference: 60%|██████ | 3553/5920 [48:54<14:13, 2.77it/s, tok/s=15024.1]
Inference: 60%|██████ | 3554/5920 [48:55<14:13, 2.77it/s, tok/s=15024.4]
Inference: 60%|██████ | 3555/5920 [48:55<14:13, 2.77it/s, tok/s=15029.9]
Inference: 60%|██████ | 3556/5920 [48:55<14:12, 2.77it/s, tok/s=15033.0]
Inference: 60%|██████ | 3557/5920 [48:55<14:12, 2.77it/s, tok/s=15033.5]
Inference: 60%|██████ | 3558/5920 [48:55<06:15, 6.30it/s, tok/s=15033.5]
Inference: 60%|██████ | 3558/5920 [48:55<06:15, 6.30it/s, tok/s=15034.9]
Inference: 60%|██████ | 3559/5920 [48:55<06:14, 6.30it/s, tok/s=15035.1]
Inference: 60%|██████ | 3560/5920 [48:55<05:19, 7.38it/s, tok/s=15035.1]
Inference: 60%|██████ | 3560/5920 [48:55<05:19, 7.38it/s, tok/s=15036.2]
Inference: 60%|██████ | 3561/5920 [48:55<05:19, 7.38it/s, tok/s=15044.2]
Inference: 60%|██████ | 3562/5920 [48:55<05:19, 7.38it/s, tok/s=15044.8]
Inference: 60%|██████ | 3563/5920 [48:55<04:12, 9.32it/s, tok/s=15044.8]
Inference: 60%|██████ | 3563/5920 [48:55<04:12, 9.32it/s, tok/s=15049.1]
Inference: 60%|██████ | 3564/5920 [48:55<04:12, 9.32it/s, tok/s=15050.1]
Inference: 60%|██████ | 3565/5920 [48:55<03:55, 10.02it/s, tok/s=15050.1]
Inference: 60%|██████ | 3565/5920 [48:55<03:55, 10.02it/s, tok/s=15050.2]
Inference: 60%|██████ | 3566/5920 [48:55<03:54, 10.02it/s, tok/s=15050.2]
Inference: 60%|██████ | 3567/5920 [48:55<05:11, 7.54it/s, tok/s=15050.2]
Inference: 60%|██████ | 3567/5920 [48:55<05:11, 7.54it/s, tok/s=15050.2]
Inference: 60%|██████ | 3568/5920 [48:56<05:11, 7.54it/s, tok/s=15051.7]
Inference: 60%|██████ | 3569/5920 [48:56<06:50, 5.73it/s, tok/s=15051.7]
Inference: 60%|██████ | 3569/5920 [48:56<06:50, 5.73it/s, tok/s=15051.5]
Inference: 60%|██████ | 3570/5920 [48:56<08:40, 4.52it/s, tok/s=15051.5]
Inference: 60%|██████ | 3570/5920 [48:56<08:40, 4.52it/s, tok/s=15050.1]
Inference: 60%|██████ | 3571/5920 [48:56<08:40, 4.52it/s, tok/s=15053.3]
Inference: 60%|██████ | 3572/5920 [48:57<08:39, 4.52it/s, tok/s=15053.4]
Inference: 60%|██████ | 3573/5920 [48:57<05:42, 6.85it/s, tok/s=15053.4]
Inference: 60%|██████ | 3573/5920 [48:57<05:42, 6.85it/s, tok/s=15053.9]
Inference: 60%|██████ | 3574/5920 [48:57<05:42, 6.85it/s, tok/s=15055.2]
Inference: 60%|██████ | 3575/5920 [48:57<05:55, 6.59it/s, tok/s=15055.2]
Inference: 60%|██████ | 3575/5920 [48:57<05:55, 6.59it/s, tok/s=15055.6]
Inference: 60%|██████ | 3576/5920 [48:57<05:55, 6.59it/s, tok/s=15056.0]
Inference: 60%|██████ | 3577/5920 [48:57<05:18, 7.35it/s, tok/s=15056.0]
Inference: 60%|██████ | 3577/5920 [48:57<05:18, 7.35it/s, tok/s=15056.5]
Inference: 60%|██████ | 3578/5920 [48:57<05:18, 7.35it/s, tok/s=15055.6]
Inference: 60%|██████ | 3579/5920 [48:58<07:48, 5.00it/s, tok/s=15055.6]
Inference: 60%|██████ | 3579/5920 [48:58<07:48, 5.00it/s, tok/s=15054.0]
Inference: 60%|██████ | 3580/5920 [48:58<07:48, 5.00it/s, tok/s=15054.0]
Inference: 60%|██████ | 3581/5920 [48:58<06:59, 5.57it/s, tok/s=15054.0]
Inference: 60%|██████ | 3581/5920 [48:58<06:59, 5.57it/s, tok/s=15053.7]
Inference: 61%|██████ | 3582/5920 [48:58<06:59, 5.57it/s, tok/s=15056.8]
Inference: 61%|██████ | 3583/5920 [48:58<07:07, 5.46it/s, tok/s=15056.8]
Inference: 61%|██████ | 3583/5920 [48:58<07:07, 5.46it/s, tok/s=15058.8]
Inference: 61%|██████ | 3584/5920 [48:59<07:40, 5.07it/s, tok/s=15058.8]
Inference: 61%|██████ | 3584/5920 [48:59<07:40, 5.07it/s, tok/s=15059.1]
Inference: 61%|██████ | 3585/5920 [48:59<07:08, 5.45it/s, tok/s=15059.1]
Inference: 61%|██████ | 3585/5920 [48:59<07:08, 5.45it/s, tok/s=15062.1]
Inference: 61%|██████ | 3586/5920 [48:59<07:59, 4.87it/s, tok/s=15062.1]
Inference: 61%|██████ | 3586/5920 [48:59<07:59, 4.87it/s, tok/s=15063.4]
Inference: 61%|██████ | 3587/5920 [48:59<08:03, 4.83it/s, tok/s=15063.4]
Inference: 61%|██████ | 3587/5920 [48:59<08:03, 4.83it/s, tok/s=15064.0]
Inference: 61%|██████ | 3588/5920 [49:00<08:19, 4.67it/s, tok/s=15064.0]
Inference: 61%|██████ | 3588/5920 [49:00<08:19, 4.67it/s, tok/s=15063.4]
Inference: 61%|██████ | 3589/5920 [49:00<08:19, 4.67it/s, tok/s=15065.8]
Inference: 61%|██████ | 3590/5920 [49:00<06:26, 6.02it/s, tok/s=15065.8]
Inference: 61%|██████ | 3590/5920 [49:00<06:26, 6.02it/s, tok/s=15067.2]
Inference: 61%|██████ | 3591/5920 [49:00<06:26, 6.02it/s, tok/s=15067.7]
Inference: 61%|██████ | 3592/5920 [49:00<05:32, 7.01it/s, tok/s=15067.7]
Inference: 61%|██████ | 3592/5920 [49:00<05:32, 7.01it/s, tok/s=15070.6]
Inference: 61%|██████ | 3593/5920 [49:00<06:15, 6.19it/s, tok/s=15070.6]
Inference: 61%|██████ | 3593/5920 [49:00<06:15, 6.19it/s, tok/s=15070.0]
Inference: 61%|██████ | 3594/5920 [49:00<06:03, 6.40it/s, tok/s=15070.0]
Inference: 61%|██████ | 3594/5920 [49:00<06:03, 6.40it/s, tok/s=15072.9]
Inference: 61%|██████ | 3595/5920 [49:00<06:03, 6.40it/s, tok/s=15074.0]
Inference: 61%|██████ | 3596/5920 [49:01<05:55, 6.55it/s, tok/s=15074.0]
Inference: 61%|██████ | 3596/5920 [49:01<05:55, 6.55it/s, tok/s=15073.6]
Inference: 61%|██████ | 3597/5920 [49:01<06:25, 6.02it/s, tok/s=15073.6]
Inference: 61%|██████ | 3597/5920 [49:01<06:25, 6.02it/s, tok/s=15074.7]
Inference: 61%|██████ | 3598/5920 [49:02<11:30, 3.36it/s, tok/s=15074.7]
Inference: 61%|██████ | 3598/5920 [49:02<11:30, 3.36it/s, tok/s=15071.6]
Inference: 61%|██████ | 3599/5920 [49:02<17:06, 2.26it/s, tok/s=15071.6]
Inference: 61%|██████ | 3599/5920 [49:02<17:06, 2.26it/s, tok/s=15067.8]
Inference: 61%|██████ | 3600/5920 [49:03<13:50, 2.79it/s, tok/s=15067.8]
Inference: 61%|██████ | 3600/5920 [49:03<13:50, 2.79it/s, tok/s=15074.2]
Inference: 61%|██████ | 3601/5920 [49:03<13:05, 2.95it/s, tok/s=15074.2]
Inference: 61%|██████ | 3601/5920 [49:03<13:05, 2.95it/s, tok/s=15073.2]
Inference: 61%|██████ | 3602/5920 [49:03<13:05, 2.95it/s, tok/s=15073.7]
Inference: 61%|██████ | 3603/5920 [49:03<10:26, 3.70it/s, tok/s=15073.7]
Inference: 61%|██████ | 3603/5920 [49:03<10:26, 3.70it/s, tok/s=15072.8]
Inference: 61%|██████ | 3604/5920 [49:04<10:51, 3.55it/s, tok/s=15072.8]
Inference: 61%|██████ | 3604/5920 [49:04<10:51, 3.55it/s, tok/s=15073.0]
Inference: 61%|██████ | 3605/5920 [49:04<10:41, 3.61it/s, tok/s=15073.0]
Inference: 61%|██████ | 3605/5920 [49:04<10:41, 3.61it/s, tok/s=15072.4]
Inference: 61%|██████ | 3606/5920 [49:05<15:22, 2.51it/s, tok/s=15072.4]
Inference: 61%|██████ | 3606/5920 [49:05<15:22, 2.51it/s, tok/s=15072.7]
Inference: 61%|██████ | 3607/5920 [49:06<22:05, 1.74it/s, tok/s=15072.7]
Inference: 61%|██████ | 3607/5920 [49:06<22:05, 1.74it/s, tok/s=15068.4]
Inference: 61%|██████ | 3608/5920 [49:06<22:05, 1.74it/s, tok/s=15070.1]
Inference: 61%|██████ | 3609/5920 [49:07<21:23, 1.80it/s, tok/s=15070.1]
Inference: 61%|██████ | 3609/5920 [49:07<21:23, 1.80it/s, tok/s=15065.4]
Inference: 61%|██████ | 3610/5920 [49:07<21:18, 1.81it/s, tok/s=15065.4]
Inference: 61%|██████ | 3610/5920 [49:07<21:18, 1.81it/s, tok/s=15063.6]
Inference: 61%|██████ | 3611/5920 [49:07<21:18, 1.81it/s, tok/s=15064.4]
Inference: 61%|██████ | 3612/5920 [49:07<14:24, 2.67it/s, tok/s=15064.4]
Inference: 61%|██████ | 3612/5920 [49:07<14:24, 2.67it/s, tok/s=15064.1]
Inference: 61%|██████ | 3613/5920 [49:08<14:33, 2.64it/s, tok/s=15064.1]
Inference: 61%|██████ | 3613/5920 [49:08<14:33, 2.64it/s, tok/s=15065.4]
Inference: 61%|██████ | 3614/5920 [49:09<17:27, 2.20it/s, tok/s=15065.4]
Inference: 61%|██████ | 3614/5920 [49:09<17:27, 2.20it/s, tok/s=15072.1]
Inference: 61%|██████ | 3615/5920 [49:09<19:00, 2.02it/s, tok/s=15072.1]
Inference: 61%|██████ | 3615/5920 [49:09<19:00, 2.02it/s, tok/s=15070.9]
Inference: 61%|██████ | 3616/5920 [49:10<18:27, 2.08it/s, tok/s=15070.9]
Inference: 61%|██████ | 3616/5920 [49:10<18:27, 2.08it/s, tok/s=15069.2]
Inference: 61%|██████ | 3617/5920 [49:10<16:13, 2.37it/s, tok/s=15069.2]
Inference: 61%|██████ | 3617/5920 [49:10<16:13, 2.37it/s, tok/s=15068.6]
Inference: 61%|██████ | 3618/5920 [49:10<16:12, 2.37it/s, tok/s=15068.6]
Inference: 61%|██████ | 3618/5920 [49:10<16:12, 2.37it/s, tok/s=15067.0]
Inference: 61%|██████ | 3619/5920 [49:10<13:44, 2.79it/s, tok/s=15067.0]
Inference: 61%|██████ | 3619/5920 [49:10<13:44, 2.79it/s, tok/s=15069.3]
Inference: 61%|██████ | 3620/5920 [49:11<13:02, 2.94it/s, tok/s=15069.3]
Inference: 61%|██████ | 3620/5920 [49:11<13:02, 2.94it/s, tok/s=15069.8]
Inference: 61%|██████ | 3621/5920 [49:11<13:01, 2.94it/s, tok/s=15070.3]
Inference: 61%|██████ | 3622/5920 [49:11<10:01, 3.82it/s, tok/s=15070.3]
Inference: 61%|██████ | 3622/5920 [49:11<10:01, 3.82it/s, tok/s=15069.8]
Inference: 61%|██████ | 3623/5920 [49:11<10:29, 3.65it/s, tok/s=15069.8]
Inference: 61%|██████ | 3623/5920 [49:11<10:29, 3.65it/s, tok/s=15068.9]
Inference: 61%|██████ | 3624/5920 [49:11<10:29, 3.65it/s, tok/s=15069.2]
Inference: 61%|██████ | 3625/5920 [49:12<07:01, 5.45it/s, tok/s=15069.2]
Inference: 61%|██████ | 3625/5920 [49:12<07:01, 5.45it/s, tok/s=15069.8]
Inference: 61%|██████▏ | 3626/5920 [49:12<09:13, 4.14it/s, tok/s=15069.8]
Inference: 61%|██████▏ | 3626/5920 [49:12<09:13, 4.14it/s, tok/s=15068.5]
Inference: 61%|██████▏ | 3627/5920 [49:12<08:12, 4.66it/s, tok/s=15068.5]
Inference: 61%|██████▏ | 3627/5920 [49:12<08:12, 4.66it/s, tok/s=15068.5]
Inference: 61%|██████▏ | 3628/5920 [49:12<09:23, 4.07it/s, tok/s=15068.5]
Inference: 61%|██████▏ | 3628/5920 [49:12<09:23, 4.07it/s, tok/s=15068.7]
Inference: 61%|██████▏ | 3629/5920 [49:13<12:56, 2.95it/s, tok/s=15068.7]
Inference: 61%|██████▏ | 3629/5920 [49:13<12:56, 2.95it/s, tok/s=15068.5]
Inference: 61%|██████▏ | 3630/5920 [49:14<15:00, 2.54it/s, tok/s=15068.5]
Inference: 61%|██████▏ | 3630/5920 [49:14<15:00, 2.54it/s, tok/s=15066.3]
Inference: 61%|██████▏ | 3631/5920 [49:14<16:01, 2.38it/s, tok/s=15066.3]
Inference: 61%|██████▏ | 3631/5920 [49:14<16:01, 2.38it/s, tok/s=15065.2]
Inference: 61%|██████▏ | 3632/5920 [49:15<21:49, 1.75it/s, tok/s=15065.2]
Inference: 61%|██████▏ | 3632/5920 [49:15<21:49, 1.75it/s, tok/s=15061.8]
Inference: 61%|██████▏ | 3633/5920 [49:15<18:08, 2.10it/s, tok/s=15061.8]
Inference: 61%|██████▏ | 3633/5920 [49:15<18:08, 2.10it/s, tok/s=15061.9]
Inference: 61%|██████▏ | 3634/5920 [49:16<20:48, 1.83it/s, tok/s=15061.9]
Inference: 61%|██████▏ | 3634/5920 [49:16<20:48, 1.83it/s, tok/s=15058.7]
Inference: 61%|██████▏ | 3635/5920 [49:17<27:22, 1.39it/s, tok/s=15058.7]
Inference: 61%|██████▏ | 3635/5920 [49:17<27:22, 1.39it/s, tok/s=15053.8]
Inference: 61%|██████▏ | 3636/5920 [49:17<23:27, 1.62it/s, tok/s=15053.8]
Inference: 61%|██████▏ | 3636/5920 [49:17<23:27, 1.62it/s, tok/s=15060.7]
Inference: 61%|██████▏ | 3637/5920 [49:18<19:13, 1.98it/s, tok/s=15060.7]
Inference: 61%|██████▏ | 3637/5920 [49:18<19:13, 1.98it/s, tok/s=15059.8]
Inference: 61%|██████▏ | 3638/5920 [49:18<21:47, 1.75it/s, tok/s=15059.8]
Inference: 61%|██████▏ | 3638/5920 [49:18<21:47, 1.75it/s, tok/s=15061.8]
Inference: 61%|██████▏ | 3639/5920 [49:19<17:08, 2.22it/s, tok/s=15061.8]
Inference: 61%|██████▏ | 3639/5920 [49:19<17:08, 2.22it/s, tok/s=15065.2]
Inference: 61%|██████▏ | 3640/5920 [49:19<15:58, 2.38it/s, tok/s=15065.2]
Inference: 61%|██████▏ | 3640/5920 [49:19<15:58, 2.38it/s, tok/s=15064.0]
Inference: 62%|██████▏ | 3641/5920 [49:20<19:47, 1.92it/s, tok/s=15064.0]
Inference: 62%|██████▏ | 3641/5920 [49:20<19:47, 1.92it/s, tok/s=15061.1]
Inference: 62%|██████▏ | 3642/5920 [49:20<15:20, 2.48it/s, tok/s=15061.1]
Inference: 62%|██████▏ | 3642/5920 [49:20<15:20, 2.48it/s, tok/s=15061.8]
Inference: 62%|██████▏ | 3643/5920 [49:21<18:44, 2.02it/s, tok/s=15061.8]
Inference: 62%|██████▏ | 3643/5920 [49:21<18:44, 2.02it/s, tok/s=15065.6]
Inference: 62%|██████▏ | 3644/5920 [49:21<18:44, 2.02it/s, tok/s=15069.6]
Inference: 62%|██████▏ | 3645/5920 [49:21<11:04, 3.42it/s, tok/s=15069.6]
Inference: 62%|██████▏ | 3645/5920 [49:21<11:04, 3.42it/s, tok/s=15069.8]
Inference: 62%|██████▏ | 3646/5920 [49:22<25:00, 1.52it/s, tok/s=15069.8]
Inference: 62%|██████▏ | 3646/5920 [49:22<25:00, 1.52it/s, tok/s=15061.4]
Inference: 62%|██████▏ | 3647/5920 [49:22<25:00, 1.52it/s, tok/s=15061.8]
Inference: 62%|██████▏ | 3648/5920 [49:23<15:50, 2.39it/s, tok/s=15061.8]
Inference: 62%|██████▏ | 3648/5920 [49:23<15:50, 2.39it/s, tok/s=15062.5]
Inference: 62%|██████▏ | 3649/5920 [49:23<15:50, 2.39it/s, tok/s=15063.2]
Inference: 62%|██████▏ | 3650/5920 [49:23<10:42, 3.53it/s, tok/s=15063.2]
Inference: 62%|██████▏ | 3650/5920 [49:23<10:42, 3.53it/s, tok/s=15063.3]
Inference: 62%|██████▏ | 3651/5920 [49:23<10:42, 3.53it/s, tok/s=15064.5]
Inference: 62%|██████▏ | 3652/5920 [49:23<10:41, 3.53it/s, tok/s=15066.1]
Inference: 62%|██████▏ | 3653/5920 [49:23<07:47, 4.85it/s, tok/s=15066.1]
Inference: 62%|██████▏ | 3653/5920 [49:23<07:47, 4.85it/s, tok/s=15066.0]
Inference: 62%|██████▏ | 3654/5920 [49:23<07:46, 4.85it/s, tok/s=15066.0]
Inference: 62%|██████▏ | 3655/5920 [49:24<09:27, 3.99it/s, tok/s=15066.0]
Inference: 62%|██████▏ | 3655/5920 [49:24<09:27, 3.99it/s, tok/s=15068.6]
Inference: 62%|██████▏ | 3656/5920 [49:24<08:28, 4.45it/s, tok/s=15068.6]
Inference: 62%|██████▏ | 3656/5920 [49:24<08:28, 4.45it/s, tok/s=15068.7]
Inference: 62%|██████▏ | 3657/5920 [49:24<08:18, 4.54it/s, tok/s=15068.7]
Inference: 62%|██████▏ | 3657/5920 [49:24<08:18, 4.54it/s, tok/s=15076.8]
Inference: 62%|██████▏ | 3658/5920 [49:24<08:40, 4.35it/s, tok/s=15076.8]
Inference: 62%|██████▏ | 3658/5920 [49:24<08:40, 4.35it/s, tok/s=15076.3]
Inference: 62%|██████▏ | 3659/5920 [49:25<12:39, 2.98it/s, tok/s=15076.3]
Inference: 62%|██████▏ | 3659/5920 [49:25<12:39, 2.98it/s, tok/s=15073.5]
Inference: 62%|██████▏ | 3660/5920 [49:25<12:33, 3.00it/s, tok/s=15073.5]
Inference: 62%|██████▏ | 3660/5920 [49:25<12:33, 3.00it/s, tok/s=15072.7]
Inference: 62%|██████▏ | 3661/5920 [49:25<12:33, 3.00it/s, tok/s=15072.9]
Inference: 62%|██████▏ | 3662/5920 [49:26<12:03, 3.12it/s, tok/s=15072.9]
Inference: 62%|██████▏ | 3662/5920 [49:26<12:03, 3.12it/s, tok/s=15070.9]
Inference: 62%|██████▏ | 3663/5920 [49:26<11:46, 3.19it/s, tok/s=15070.9]
Inference: 62%|██████▏ | 3663/5920 [49:26<11:46, 3.19it/s, tok/s=15077.7]
Inference: 62%|██████▏ | 3664/5920 [49:27<12:25, 3.02it/s, tok/s=15077.7]
Inference: 62%|██████▏ | 3664/5920 [49:27<12:25, 3.02it/s, tok/s=15076.3]
Inference: 62%|██████▏ | 3665/5920 [49:27<12:25, 3.02it/s, tok/s=15080.5]
Inference: 62%|██████▏ | 3666/5920 [49:27<08:52, 4.24it/s, tok/s=15080.5]
Inference: 62%|██████▏ | 3666/5920 [49:27<08:52, 4.24it/s, tok/s=15081.2]
Inference: 62%|██████▏ | 3667/5920 [49:27<08:51, 4.24it/s, tok/s=15081.2]
Inference: 62%|██████▏ | 3668/5920 [49:28<13:09, 2.85it/s, tok/s=15081.2]
Inference: 62%|██████▏ | 3668/5920 [49:28<13:09, 2.85it/s, tok/s=15079.0]
Inference: 62%|██████▏ | 3669/5920 [49:28<13:24, 2.80it/s, tok/s=15079.0]
Inference: 62%|██████▏ | 3669/5920 [49:28<13:24, 2.80it/s, tok/s=15077.6]
Inference: 62%|██████▏ | 3670/5920 [49:28<11:52, 3.16it/s, tok/s=15077.6]
Inference: 62%|██████▏ | 3670/5920 [49:28<11:52, 3.16it/s, tok/s=15077.9]
Inference: 62%|██████▏ | 3671/5920 [49:29<11:48, 3.17it/s, tok/s=15077.9]
Inference: 62%|██████▏ | 3671/5920 [49:29<11:48, 3.17it/s, tok/s=15077.5]
Inference: 62%|██████▏ | 3672/5920 [49:29<09:52, 3.79it/s, tok/s=15077.5]
Inference: 62%|██████▏ | 3672/5920 [49:29<09:52, 3.79it/s, tok/s=15078.6]
Inference: 62%|██████▏ | 3673/5920 [49:29<08:13, 4.55it/s, tok/s=15078.6]
Inference: 62%|██████▏ | 3673/5920 [49:29<08:13, 4.55it/s, tok/s=15078.7]
Inference: 62%|██████▏ | 3674/5920 [49:30<14:45, 2.54it/s, tok/s=15078.7]
Inference: 62%|██████▏ | 3674/5920 [49:30<14:45, 2.54it/s, tok/s=15075.0]
Inference: 62%|██████▏ | 3675/5920 [49:31<20:58, 1.78it/s, tok/s=15075.0]
Inference: 62%|██████▏ | 3675/5920 [49:31<20:58, 1.78it/s, tok/s=15070.8]
Inference: 62%|██████▏ | 3676/5920 [49:31<20:57, 1.78it/s, tok/s=15071.3]
Inference: 62%|██████▏ | 3677/5920 [49:31<13:06, 2.85it/s, tok/s=15071.3]
Inference: 62%|██████▏ | 3677/5920 [49:31<13:06, 2.85it/s, tok/s=15070.9]
Inference: 62%|██████▏ | 3678/5920 [49:31<14:08, 2.64it/s, tok/s=15070.9]
Inference: 62%|██████▏ | 3678/5920 [49:31<14:08, 2.64it/s, tok/s=15069.0]
Inference: 62%|██████▏ | 3679/5920 [49:32<13:59, 2.67it/s, tok/s=15069.0]
Inference: 62%|██████▏ | 3679/5920 [49:32<13:59, 2.67it/s, tok/s=15067.8]
Inference: 62%|██████▏ | 3680/5920 [49:32<13:58, 2.67it/s, tok/s=15070.9]
Inference: 62%|██████▏ | 3681/5920 [49:32<09:44, 3.83it/s, tok/s=15070.9]
Inference: 62%|██████▏ | 3681/5920 [49:32<09:44, 3.83it/s, tok/s=15070.8]
Inference: 62%|██████▏ | 3682/5920 [49:32<09:50, 3.79it/s, tok/s=15070.8]
Inference: 62%|██████▏ | 3682/5920 [49:32<09:50, 3.79it/s, tok/s=15070.3]
Inference: 62%|██████▏ | 3683/5920 [49:33<09:47, 3.81it/s, tok/s=15070.3]
Inference: 62%|██████▏ | 3683/5920 [49:33<09:47, 3.81it/s, tok/s=15069.8]
Inference: 62%|██████▏ | 3684/5920 [49:33<13:55, 2.68it/s, tok/s=15069.8]
Inference: 62%|██████▏ | 3684/5920 [49:33<13:55, 2.68it/s, tok/s=15072.0]
Inference: 62%|██████▏ | 3685/5920 [49:33<13:55, 2.68it/s, tok/s=15079.2]
Inference: 62%|██████▏ | 3686/5920 [49:33<08:56, 4.16it/s, tok/s=15079.2]
Inference: 62%|██████▏ | 3686/5920 [49:33<08:56, 4.16it/s, tok/s=15080.0]
Inference: 62%|██████▏ | 3687/5920 [49:33<08:56, 4.16it/s, tok/s=15089.6]
Inference: 62%|██████▏ | 3688/5920 [49:34<09:17, 4.01it/s, tok/s=15089.6]
Inference: 62%|██████▏ | 3688/5920 [49:34<09:17, 4.01it/s, tok/s=15088.0]
Inference: 62%|██████▏ | 3689/5920 [49:34<10:00, 3.71it/s, tok/s=15088.0]
Inference: 62%|██████▏ | 3689/5920 [49:34<10:00, 3.71it/s, tok/s=15086.7]
Inference: 62%|██████▏ | 3690/5920 [49:34<08:36, 4.32it/s, tok/s=15086.7]
Inference: 62%|██████▏ | 3690/5920 [49:34<08:36, 4.32it/s, tok/s=15087.0]
Inference: 62%|██████▏ | 3691/5920 [49:34<08:36, 4.32it/s, tok/s=15087.3]
Inference: 62%|██████▏ | 3692/5920 [49:34<08:36, 4.32it/s, tok/s=15087.8]
Inference: 62%|██████▏ | 3693/5920 [49:35<10:40, 3.48it/s, tok/s=15087.8]
Inference: 62%|██████▏ | 3693/5920 [49:35<10:40, 3.48it/s, tok/s=15085.4]
Inference: 62%|██████▏ | 3694/5920 [49:36<10:35, 3.51it/s, tok/s=15085.4]
Inference: 62%|██████▏ | 3694/5920 [49:36<10:35, 3.51it/s, tok/s=15084.7]
Inference: 62%|██████▏ | 3695/5920 [49:36<12:07, 3.06it/s, tok/s=15084.7]
Inference: 62%|██████▏ | 3695/5920 [49:36<12:07, 3.06it/s, tok/s=15083.2]
Inference: 62%|██████▏ | 3696/5920 [49:36<10:41, 3.47it/s, tok/s=15083.2]
Inference: 62%|██████▏ | 3696/5920 [49:36<10:41, 3.47it/s, tok/s=15089.0]
Inference: 62%|██████▏ | 3697/5920 [49:37<15:32, 2.38it/s, tok/s=15089.0]
Inference: 62%|██████▏ | 3697/5920 [49:37<15:32, 2.38it/s, tok/s=15085.2]
Inference: 62%|██████▏ | 3698/5920 [49:37<15:32, 2.38it/s, tok/s=15085.8]
Inference: 62%|██████▏ | 3699/5920 [49:38<15:04, 2.46it/s, tok/s=15085.8]
Inference: 62%|██████▏ | 3699/5920 [49:38<15:04, 2.46it/s, tok/s=15083.0]
Inference: 62%|██████▎ | 3700/5920 [49:38<16:19, 2.27it/s, tok/s=15083.0]
Inference: 62%|██████▎ | 3700/5920 [49:38<16:19, 2.27it/s, tok/s=15081.2]
Inference: 63%|██████▎ | 3701/5920 [49:39<21:20, 1.73it/s, tok/s=15081.2]
Inference: 63%|██████▎ | 3701/5920 [49:39<21:20, 1.73it/s, tok/s=15076.8]
Inference: 63%|██████▎ | 3702/5920 [49:39<21:20, 1.73it/s, tok/s=15077.1]
Inference: 63%|██████▎ | 3703/5920 [49:40<15:40, 2.36it/s, tok/s=15077.1]
Inference: 63%|██████▎ | 3703/5920 [49:40<15:40, 2.36it/s, tok/s=15076.2]
Inference: 63%|██████▎ | 3704/5920 [49:40<17:05, 2.16it/s, tok/s=15076.2]
Inference: 63%|██████▎ | 3704/5920 [49:40<17:05, 2.16it/s, tok/s=15073.9]
Inference: 63%|██████▎ | 3705/5920 [49:40<17:05, 2.16it/s, tok/s=15074.0]
Inference: 63%|██████▎ | 3706/5920 [49:41<11:17, 3.27it/s, tok/s=15074.0]
Inference: 63%|██████▎ | 3706/5920 [49:41<11:17, 3.27it/s, tok/s=15074.9]
Inference: 63%|██████▎ | 3707/5920 [49:41<11:27, 3.22it/s, tok/s=15074.9]
Inference: 63%|██████▎ | 3707/5920 [49:41<11:27, 3.22it/s, tok/s=15074.1]
Inference: 63%|██████▎ | 3708/5920 [49:41<12:47, 2.88it/s, tok/s=15074.1]
Inference: 63%|██████▎ | 3708/5920 [49:41<12:47, 2.88it/s, tok/s=15072.7]
Inference: 63%|██████▎ | 3709/5920 [49:42<11:22, 3.24it/s, tok/s=15072.7]
Inference: 63%|██████▎ | 3709/5920 [49:42<11:22, 3.24it/s, tok/s=15073.1]
Inference: 63%|██████▎ | 3710/5920 [49:42<12:50, 2.87it/s, tok/s=15073.1]
Inference: 63%|██████▎ | 3710/5920 [49:42<12:50, 2.87it/s, tok/s=15071.4]
Inference: 63%|██████▎ | 3711/5920 [49:42<14:10, 2.60it/s, tok/s=15071.4]
Inference: 63%|██████▎ | 3711/5920 [49:42<14:10, 2.60it/s, tok/s=15073.7]
Inference: 63%|██████▎ | 3712/5920 [49:43<16:09, 2.28it/s, tok/s=15073.7]
Inference: 63%|██████▎ | 3712/5920 [49:43<16:09, 2.28it/s, tok/s=15071.6]
Inference: 63%|██████▎ | 3713/5920 [49:43<12:41, 2.90it/s, tok/s=15071.6]
Inference: 63%|██████▎ | 3713/5920 [49:43<12:41, 2.90it/s, tok/s=15078.4]
Inference: 63%|██████▎ | 3714/5920 [49:44<12:53, 2.85it/s, tok/s=15078.4]
Inference: 63%|██████▎ | 3714/5920 [49:44<12:53, 2.85it/s, tok/s=15077.6]
Inference: 63%|██████▎ | 3715/5920 [49:44<10:11, 3.61it/s, tok/s=15077.6]
Inference: 63%|██████▎ | 3715/5920 [49:44<10:11, 3.61it/s, tok/s=15086.1]
Inference: 63%|██████▎ | 3716/5920 [49:44<10:11, 3.61it/s, tok/s=15088.3]
Inference: 63%|██████▎ | 3717/5920 [49:44<10:15, 3.58it/s, tok/s=15088.3]
Inference: 63%|██████▎ | 3717/5920 [49:44<10:15, 3.58it/s, tok/s=15086.6]
Inference: 63%|██████▎ | 3718/5920 [49:44<09:46, 3.75it/s, tok/s=15086.6]
Inference: 63%|██████▎ | 3718/5920 [49:44<09:46, 3.75it/s, tok/s=15086.5]
Inference: 63%|██████▎ | 3719/5920 [49:44<09:46, 3.75it/s, tok/s=15086.5]
Inference: 63%|██████▎ | 3720/5920 [49:45<07:24, 4.95it/s, tok/s=15086.5]
Inference: 63%|██████▎ | 3720/5920 [49:45<07:24, 4.95it/s, tok/s=15094.4]
Inference: 63%|██████▎ | 3721/5920 [49:45<07:24, 4.95it/s, tok/s=15094.4]
Inference: 63%|██████▎ | 3721/5920 [49:45<07:24, 4.95it/s, tok/s=15094.3]
Inference: 63%|██████▎ | 3722/5920 [49:45<07:24, 4.94it/s, tok/s=15094.3]
Inference: 63%|██████▎ | 3722/5920 [49:45<07:24, 4.94it/s, tok/s=15094.4]
Inference: 63%|██████▎ | 3723/5920 [49:45<08:47, 4.17it/s, tok/s=15094.4]
Inference: 63%|██████▎ | 3723/5920 [49:45<08:47, 4.17it/s, tok/s=15093.2]
Inference: 63%|██████▎ | 3724/5920 [49:46<14:37, 2.50it/s, tok/s=15093.2]
Inference: 63%|██████▎ | 3724/5920 [49:46<14:37, 2.50it/s, tok/s=15089.5]
Inference: 63%|██████▎ | 3725/5920 [49:46<14:37, 2.50it/s, tok/s=15090.6]
Inference: 63%|██████▎ | 3726/5920 [49:47<12:39, 2.89it/s, tok/s=15090.6]
Inference: 63%|██████▎ | 3726/5920 [49:47<12:39, 2.89it/s, tok/s=15090.0]
Inference: 63%|██████▎ | 3727/5920 [49:47<12:39, 2.89it/s, tok/s=15099.4]
Inference: 63%|██████▎ | 3728/5920 [49:47<09:11, 3.98it/s, tok/s=15099.4]
Inference: 63%|██████▎ | 3728/5920 [49:47<09:11, 3.98it/s, tok/s=15099.7]
Inference: 63%|██████▎ | 3729/5920 [49:47<09:10, 3.98it/s, tok/s=15099.9]
Inference: 63%|██████▎ | 3730/5920 [49:47<06:45, 5.40it/s, tok/s=15099.9]
Inference: 63%|██████▎ | 3730/5920 [49:47<06:45, 5.40it/s, tok/s=15100.5]
Inference: 63%|██████▎ | 3731/5920 [49:47<06:45, 5.40it/s, tok/s=15100.8]
Inference: 63%|██████▎ | 3732/5920 [49:48<09:03, 4.02it/s, tok/s=15100.8]
Inference: 63%|██████▎ | 3732/5920 [49:48<09:03, 4.02it/s, tok/s=15097.9]
Inference: 63%|██████▎ | 3733/5920 [49:48<09:03, 4.02it/s, tok/s=15099.1]
Inference: 63%|██████▎ | 3734/5920 [49:48<09:26, 3.86it/s, tok/s=15099.1]
Inference: 63%|██████▎ | 3734/5920 [49:48<09:26, 3.86it/s, tok/s=15097.7]
Inference: 63%|██████▎ | 3735/5920 [49:49<11:43, 3.10it/s, tok/s=15097.7]
Inference: 63%|██████▎ | 3735/5920 [49:49<11:43, 3.10it/s, tok/s=15102.5]
Inference: 63%|██████▎ | 3736/5920 [49:49<11:43, 3.10it/s, tok/s=15103.4]
Inference: 63%|██████▎ | 3737/5920 [49:49<08:43, 4.17it/s, tok/s=15103.4]
Inference: 63%|██████▎ | 3737/5920 [49:49<08:43, 4.17it/s, tok/s=15103.6]
Inference: 63%|██████▎ | 3738/5920 [49:49<09:15, 3.93it/s, tok/s=15103.6]
Inference: 63%|██████▎ | 3738/5920 [49:49<09:15, 3.93it/s, tok/s=15102.5]
Inference: 63%|██████▎ | 3739/5920 [49:50<09:01, 4.03it/s, tok/s=15102.5]
Inference: 63%|██████▎ | 3739/5920 [49:50<09:01, 4.03it/s, tok/s=15103.8]
Inference: 63%|██████▎ | 3740/5920 [49:50<07:45, 4.68it/s, tok/s=15103.8]
Inference: 63%|██████▎ | 3740/5920 [49:50<07:45, 4.68it/s, tok/s=15103.8]
Inference: 63%|██████▎ | 3741/5920 [49:50<11:13, 3.23it/s, tok/s=15103.8]
Inference: 63%|██████▎ | 3741/5920 [49:50<11:13, 3.23it/s, tok/s=15104.6]
Inference: 63%|██████▎ | 3742/5920 [49:51<10:42, 3.39it/s, tok/s=15104.6]
Inference: 63%|██████▎ | 3742/5920 [49:51<10:42, 3.39it/s, tok/s=15104.0]
Inference: 63%|██████▎ | 3743/5920 [49:51<10:42, 3.39it/s, tok/s=15104.6]
Inference: 63%|██████▎ | 3744/5920 [49:52<16:27, 2.20it/s, tok/s=15104.6]
Inference: 63%|██████▎ | 3744/5920 [49:52<16:27, 2.20it/s, tok/s=15107.9]
Inference: 63%|██████▎ | 3745/5920 [49:52<16:27, 2.20it/s, tok/s=15111.1]
Inference: 63%|██████▎ | 3746/5920 [49:52<12:09, 2.98it/s, tok/s=15111.1]
Inference: 63%|██████▎ | 3746/5920 [49:52<12:09, 2.98it/s, tok/s=15111.7]
Inference: 63%|██████▎ | 3747/5920 [49:52<11:11, 3.24it/s, tok/s=15111.7]
Inference: 63%|██████▎ | 3747/5920 [49:52<11:11, 3.24it/s, tok/s=15120.7]
Inference: 63%|██████▎ | 3748/5920 [49:53<11:27, 3.16it/s, tok/s=15120.7]
Inference: 63%|██████▎ | 3748/5920 [49:53<11:27, 3.16it/s, tok/s=15119.8]
Inference: 63%|██████▎ | 3749/5920 [49:53<10:34, 3.42it/s, tok/s=15119.8]
Inference: 63%|██████▎ | 3749/5920 [49:53<10:34, 3.42it/s, tok/s=15120.0]
Inference: 63%|██████▎ | 3750/5920 [49:53<10:33, 3.42it/s, tok/s=15120.9]
Inference: 63%|██████▎ | 3751/5920 [49:54<10:11, 3.54it/s, tok/s=15120.9]
Inference: 63%|██████▎ | 3751/5920 [49:54<10:11, 3.54it/s, tok/s=15128.1]
Inference: 63%|██████▎ | 3752/5920 [49:54<12:11, 2.96it/s, tok/s=15128.1]
Inference: 63%|██████▎ | 3752/5920 [49:54<12:11, 2.96it/s, tok/s=15135.5]
Inference: 63%|██████▎ | 3753/5920 [49:54<10:54, 3.31it/s, tok/s=15135.5]
Inference: 63%|██████▎ | 3753/5920 [49:54<10:54, 3.31it/s, tok/s=15135.1]
Inference: 63%|██████▎ | 3754/5920 [49:54<09:39, 3.74it/s, tok/s=15135.1]
Inference: 63%|██████▎ | 3754/5920 [49:54<09:39, 3.74it/s, tok/s=15136.2]
Inference: 63%|██████▎ | 3755/5920 [49:55<11:17, 3.20it/s, tok/s=15136.2]
Inference: 63%|██████▎ | 3755/5920 [49:55<11:17, 3.20it/s, tok/s=15134.9]
Inference: 63%|██████▎ | 3756/5920 [49:55<11:17, 3.20it/s, tok/s=15141.0]
Inference: 63%|██████▎ | 3757/5920 [49:55<11:16, 3.20it/s, tok/s=15141.7]
Inference: 63%|██████▎ | 3758/5920 [49:55<06:02, 5.97it/s, tok/s=15141.7]
Inference: 63%|██████▎ | 3758/5920 [49:55<06:02, 5.97it/s, tok/s=15142.3]
Inference: 63%|██████▎ | 3759/5920 [49:55<06:01, 5.97it/s, tok/s=15148.0]
Inference: 64%|██████▎ | 3760/5920 [49:56<08:21, 4.30it/s, tok/s=15148.0]
Inference: 64%|██████▎ | 3760/5920 [49:56<08:21, 4.30it/s, tok/s=15150.7]
Inference: 64%|██████▎ | 3761/5920 [49:56<08:59, 4.00it/s, tok/s=15150.7]
Inference: 64%|██████▎ | 3761/5920 [49:56<08:59, 4.00it/s, tok/s=15155.7]
Inference: 64%|██████▎ | 3762/5920 [49:56<08:59, 4.00it/s, tok/s=15157.9]
Inference: 64%|██████▎ | 3763/5920 [49:56<08:58, 4.00it/s, tok/s=15158.1]
Inference: 64%|██████▎ | 3764/5920 [49:56<05:51, 6.13it/s, tok/s=15158.1]
Inference: 64%|██████▎ | 3764/5920 [49:56<05:51, 6.13it/s, tok/s=15158.3]
Inference: 64%|██████▎ | 3765/5920 [49:56<05:51, 6.13it/s, tok/s=15159.4]
Inference: 64%|██████▎ | 3766/5920 [49:57<07:12, 4.98it/s, tok/s=15159.4]
Inference: 64%|██████▎ | 3766/5920 [49:57<07:12, 4.98it/s, tok/s=15161.9]
Inference: 64%|██████▎ | 3767/5920 [49:57<06:54, 5.20it/s, tok/s=15161.9]
Inference: 64%|██████▎ | 3767/5920 [49:57<06:54, 5.20it/s, tok/s=15161.9]
Inference: 64%|██████▎ | 3768/5920 [49:57<09:10, 3.91it/s, tok/s=15161.9]
Inference: 64%|██████▎ | 3768/5920 [49:57<09:10, 3.91it/s, tok/s=15165.6]
Inference: 64%|██████▎ | 3769/5920 [49:58<13:04, 2.74it/s, tok/s=15165.6]
Inference: 64%|██████▎ | 3769/5920 [49:58<13:04, 2.74it/s, tok/s=15163.7]
Inference: 64%|██████▎ | 3770/5920 [49:59<14:41, 2.44it/s, tok/s=15163.7]
Inference: 64%|██████▎ | 3770/5920 [49:59<14:41, 2.44it/s, tok/s=15161.9]
Inference: 64%|██████▎ | 3771/5920 [49:59<15:31, 2.31it/s, tok/s=15161.9]
Inference: 64%|██████▎ | 3771/5920 [49:59<15:31, 2.31it/s, tok/s=15160.2]
Inference: 64%|██████▎ | 3772/5920 [49:59<15:31, 2.31it/s, tok/s=15160.6]
Inference: 64%|██████▎ | 3773/5920 [50:00<16:22, 2.18it/s, tok/s=15160.6]
Inference: 64%|██████▎ | 3773/5920 [50:00<16:22, 2.18it/s, tok/s=15158.5]
Inference: 64%|██████▍ | 3774/5920 [50:01<16:07, 2.22it/s, tok/s=15158.5]
Inference: 64%|██████▍ | 3774/5920 [50:01<16:07, 2.22it/s, tok/s=15157.2]
Inference: 64%|██████▍ | 3775/5920 [50:01<15:30, 2.31it/s, tok/s=15157.2]
Inference: 64%|██████▍ | 3775/5920 [50:01<15:30, 2.31it/s, tok/s=15155.7]
Inference: 64%|██████▍ | 3776/5920 [50:02<16:26, 2.17it/s, tok/s=15155.7]
Inference: 64%|██████▍ | 3776/5920 [50:02<16:26, 2.17it/s, tok/s=15154.9]
Inference: 64%|██████▍ | 3777/5920 [50:02<13:29, 2.65it/s, tok/s=15154.9]
Inference: 64%|██████▍ | 3777/5920 [50:02<13:29, 2.65it/s, tok/s=15154.7]
Inference: 64%|██████▍ | 3778/5920 [50:02<11:45, 3.04it/s, tok/s=15154.7]
Inference: 64%|██████▍ | 3778/5920 [50:02<11:45, 3.04it/s, tok/s=15154.7]
Inference: 64%|██████▍ | 3779/5920 [50:02<10:40, 3.34it/s, tok/s=15154.7]
Inference: 64%|██████▍ | 3779/5920 [50:02<10:40, 3.34it/s, tok/s=15157.2]
Inference: 64%|██████▍ | 3780/5920 [50:02<08:37, 4.13it/s, tok/s=15157.2]
Inference: 64%|██████▍ | 3780/5920 [50:02<08:37, 4.13it/s, tok/s=15157.7]
Inference: 64%|██████▍ | 3781/5920 [50:02<07:15, 4.92it/s, tok/s=15157.7]
Inference: 64%|██████▍ | 3781/5920 [50:02<07:15, 4.92it/s, tok/s=15167.3]
Inference: 64%|██████▍ | 3782/5920 [50:03<09:02, 3.94it/s, tok/s=15167.3]
Inference: 64%|██████▍ | 3782/5920 [50:03<09:02, 3.94it/s, tok/s=15166.4]
Inference: 64%|██████▍ | 3783/5920 [50:03<10:01, 3.55it/s, tok/s=15166.4]
Inference: 64%|██████▍ | 3783/5920 [50:03<10:01, 3.55it/s, tok/s=15166.6]
Inference: 64%|██████▍ | 3784/5920 [50:04<12:35, 2.83it/s, tok/s=15166.6]
Inference: 64%|██████▍ | 3784/5920 [50:04<12:35, 2.83it/s, tok/s=15164.8]
Inference: 64%|██████▍ | 3785/5920 [50:05<18:41, 1.90it/s, tok/s=15164.8]
Inference: 64%|██████▍ | 3785/5920 [50:05<18:41, 1.90it/s, tok/s=15160.9]
Inference: 64%|██████▍ | 3786/5920 [50:05<14:45, 2.41it/s, tok/s=15160.9]
Inference: 64%|██████▍ | 3786/5920 [50:05<14:45, 2.41it/s, tok/s=15168.2]
Inference: 64%|██████▍ | 3787/5920 [50:05<11:35, 3.07it/s, tok/s=15168.2]
Inference: 64%|██████▍ | 3787/5920 [50:05<11:35, 3.07it/s, tok/s=15168.3]
Inference: 64%|██████▍ | 3788/5920 [50:05<11:34, 3.07it/s, tok/s=15168.7]
Inference: 64%|██████▍ | 3789/5920 [50:05<09:08, 3.89it/s, tok/s=15168.7]
Inference: 64%|██████▍ | 3789/5920 [50:05<09:08, 3.89it/s, tok/s=15168.0]
Inference: 64%|██████▍ | 3790/5920 [50:05<08:20, 4.26it/s, tok/s=15168.0]
Inference: 64%|██████▍ | 3790/5920 [50:05<08:20, 4.26it/s, tok/s=15167.8]
Inference: 64%|██████▍ | 3791/5920 [50:05<08:19, 4.26it/s, tok/s=15168.7]
Inference: 64%|██████▍ | 3792/5920 [50:05<08:19, 4.26it/s, tok/s=15169.0]
Inference: 64%|██████▍ | 3793/5920 [50:05<08:19, 4.26it/s, tok/s=15171.3]
Inference: 64%|██████▍ | 3794/5920 [50:06<04:13, 8.40it/s, tok/s=15171.3]
Inference: 64%|██████▍ | 3794/5920 [50:06<04:13, 8.40it/s, tok/s=15171.3]
Inference: 64%|██████▍ | 3795/5920 [50:06<04:13, 8.40it/s, tok/s=15175.8]
Inference: 64%|██████▍ | 3796/5920 [50:06<06:14, 5.67it/s, tok/s=15175.8]
Inference: 64%|██████▍ | 3796/5920 [50:06<06:14, 5.67it/s, tok/s=15174.4]
Inference: 64%|██████▍ | 3797/5920 [50:07<11:06, 3.19it/s, tok/s=15174.4]
Inference: 64%|██████▍ | 3797/5920 [50:07<11:06, 3.19it/s, tok/s=15172.9]
Inference: 64%|██████▍ | 3798/5920 [50:07<09:48, 3.60it/s, tok/s=15172.9]
Inference: 64%|██████▍ | 3798/5920 [50:07<09:48, 3.60it/s, tok/s=15174.2]
Inference: 64%|██████▍ | 3799/5920 [50:08<12:36, 2.80it/s, tok/s=15174.2]
Inference: 64%|██████▍ | 3799/5920 [50:08<12:36, 2.80it/s, tok/s=15172.1]
Inference: 64%|██████▍ | 3800/5920 [50:08<12:57, 2.73it/s, tok/s=15172.1]
Inference: 64%|██████▍ | 3800/5920 [50:08<12:57, 2.73it/s, tok/s=15172.4]
Inference: 64%|██████▍ | 3801/5920 [50:08<11:16, 3.13it/s, tok/s=15172.4]
Inference: 64%|██████▍ | 3801/5920 [50:08<11:16, 3.13it/s, tok/s=15172.2]
Inference: 64%|██████▍ | 3802/5920 [50:09<10:40, 3.31it/s, tok/s=15172.2]
Inference: 64%|██████▍ | 3802/5920 [50:09<10:40, 3.31it/s, tok/s=15171.6]
Inference: 64%|██████▍ | 3803/5920 [50:10<20:17, 1.74it/s, tok/s=15171.6]
Inference: 64%|██████▍ | 3803/5920 [50:10<20:17, 1.74it/s, tok/s=15175.3]
Inference: 64%|██████▍ | 3804/5920 [50:10<20:17, 1.74it/s, tok/s=15185.5]
Inference: 64%|██████▍ | 3805/5920 [50:10<12:45, 2.76it/s, tok/s=15185.5]
Inference: 64%|██████▍ | 3805/5920 [50:10<12:45, 2.76it/s, tok/s=15185.5]
Inference: 64%|██████▍ | 3806/5920 [50:11<12:33, 2.81it/s, tok/s=15185.5]
Inference: 64%|██████▍ | 3806/5920 [50:11<12:33, 2.81it/s, tok/s=15187.2]
Inference: 64%|██████▍ | 3807/5920 [50:11<13:03, 2.70it/s, tok/s=15187.2]
Inference: 64%|██████▍ | 3807/5920 [50:11<13:03, 2.70it/s, tok/s=15186.0]
Inference: 64%|██████▍ | 3808/5920 [50:11<14:21, 2.45it/s, tok/s=15186.0]
Inference: 64%|██████▍ | 3808/5920 [50:11<14:21, 2.45it/s, tok/s=15184.0]
Inference: 64%|██████▍ | 3809/5920 [50:12<19:15, 1.83it/s, tok/s=15184.0]
Inference: 64%|██████▍ | 3809/5920 [50:12<19:15, 1.83it/s, tok/s=15181.0]
Inference: 64%|██████▍ | 3810/5920 [50:12<14:59, 2.34it/s, tok/s=15181.0]
Inference: 64%|██████▍ | 3810/5920 [50:12<14:59, 2.34it/s, tok/s=15182.0]
Inference: 64%|██████▍ | 3811/5920 [50:13<14:59, 2.34it/s, tok/s=15183.1]
Inference: 64%|██████▍ | 3812/5920 [50:13<10:40, 3.29it/s, tok/s=15183.1]
Inference: 64%|██████▍ | 3812/5920 [50:13<10:40, 3.29it/s, tok/s=15182.5]
Inference: 64%|██████▍ | 3813/5920 [50:13<13:29, 2.60it/s, tok/s=15182.5]
Inference: 64%|██████▍ | 3813/5920 [50:13<13:29, 2.60it/s, tok/s=15180.6]
Inference: 64%|██████▍ | 3814/5920 [50:14<13:29, 2.60it/s, tok/s=15181.3]
Inference: 64%|██████▍ | 3815/5920 [50:14<09:34, 3.66it/s, tok/s=15181.3]
Inference: 64%|██████▍ | 3815/5920 [50:14<09:34, 3.66it/s, tok/s=15181.2]
Inference: 64%|██████▍ | 3816/5920 [50:14<08:31, 4.11it/s, tok/s=15181.2]
Inference: 64%|██████▍ | 3816/5920 [50:14<08:31, 4.11it/s, tok/s=15181.1]
Inference: 64%|██████▍ | 3817/5920 [50:14<08:08, 4.30it/s, tok/s=15181.1]
Inference: 64%|██████▍ | 3817/5920 [50:14<08:08, 4.30it/s, tok/s=15181.4]
Inference: 64%|██████▍ | 3818/5920 [50:14<08:15, 4.24it/s, tok/s=15181.4]
Inference: 64%|██████▍ | 3818/5920 [50:14<08:15, 4.24it/s, tok/s=15180.9]
Inference: 65%|██████▍ | 3819/5920 [50:15<11:28, 3.05it/s, tok/s=15180.9]
Inference: 65%|██████▍ | 3819/5920 [50:15<11:28, 3.05it/s, tok/s=15179.2]
Inference: 65%|██████▍ | 3820/5920 [50:15<11:28, 3.05it/s, tok/s=15180.1]
Inference: 65%|██████▍ | 3821/5920 [50:15<10:29, 3.34it/s, tok/s=15180.1]
Inference: 65%|██████▍ | 3821/5920 [50:15<10:29, 3.34it/s, tok/s=15178.3]
Inference: 65%|██████▍ | 3822/5920 [50:15<10:29, 3.34it/s, tok/s=15184.1]
Inference: 65%|██████▍ | 3823/5920 [50:16<08:40, 4.03it/s, tok/s=15184.1]
Inference: 65%|██████▍ | 3823/5920 [50:16<08:40, 4.03it/s, tok/s=15183.8]
Inference: 65%|██████▍ | 3824/5920 [50:16<10:50, 3.22it/s, tok/s=15183.8]
Inference: 65%|██████▍ | 3824/5920 [50:16<10:50, 3.22it/s, tok/s=15182.1]
Inference: 65%|██████▍ | 3825/5920 [50:16<10:49, 3.22it/s, tok/s=15184.8]
Inference: 65%|██████▍ | 3826/5920 [50:16<07:33, 4.62it/s, tok/s=15184.8]
Inference: 65%|██████▍ | 3826/5920 [50:16<07:33, 4.62it/s, tok/s=15194.4]
Inference: 65%|██████▍ | 3827/5920 [50:17<14:23, 2.42it/s, tok/s=15194.4]
Inference: 65%|██████▍ | 3827/5920 [50:17<14:23, 2.42it/s, tok/s=15194.2]
Inference: 65%|██████▍ | 3828/5920 [50:17<14:23, 2.42it/s, tok/s=15195.0]
Inference: 65%|██████▍ | 3829/5920 [50:18<14:13, 2.45it/s, tok/s=15195.0]
Inference: 65%|██████▍ | 3829/5920 [50:18<14:13, 2.45it/s, tok/s=15194.2]
Inference: 65%|██████▍ | 3830/5920 [50:18<11:57, 2.91it/s, tok/s=15194.2]
Inference: 65%|██████▍ | 3830/5920 [50:18<11:57, 2.91it/s, tok/s=15194.7]
Inference: 65%|██████▍ | 3831/5920 [50:18<10:01, 3.48it/s, tok/s=15194.7]
Inference: 65%|██████▍ | 3831/5920 [50:18<10:01, 3.48it/s, tok/s=15195.5]
Inference: 65%|██████▍ | 3832/5920 [50:19<10:00, 3.48it/s, tok/s=15195.6]
Inference: 65%|██████▍ | 3833/5920 [50:19<10:00, 3.48it/s, tok/s=15196.5]
Inference: 65%|██████▍ | 3834/5920 [50:19<05:49, 5.97it/s, tok/s=15196.5]
Inference: 65%|██████▍ | 3834/5920 [50:19<05:49, 5.97it/s, tok/s=15197.0]
Inference: 65%|██████▍ | 3835/5920 [50:19<05:49, 5.97it/s, tok/s=15199.0]
Inference: 65%|██████▍ | 3836/5920 [50:19<05:34, 6.23it/s, tok/s=15199.0]
Inference: 65%|██████▍ | 3836/5920 [50:19<05:34, 6.23it/s, tok/s=15199.2]
Inference: 65%|██████▍ | 3837/5920 [50:19<05:34, 6.23it/s, tok/s=15202.3]
Inference: 65%|██████▍ | 3838/5920 [50:19<05:48, 5.97it/s, tok/s=15202.3]
Inference: 65%|██████▍ | 3838/5920 [50:19<05:48, 5.97it/s, tok/s=15201.9]
Inference: 65%|██████▍ | 3839/5920 [50:19<05:24, 6.42it/s, tok/s=15201.9]
Inference: 65%|██████▍ | 3839/5920 [50:19<05:24, 6.42it/s, tok/s=15202.6]
Inference: 65%|██████▍ | 3840/5920 [50:20<06:04, 5.71it/s, tok/s=15202.6]
Inference: 65%|██████▍ | 3840/5920 [50:20<06:04, 5.71it/s, tok/s=15206.5]
Inference: 65%|██████▍ | 3841/5920 [50:20<07:04, 4.90it/s, tok/s=15206.5]
Inference: 65%|██████▍ | 3841/5920 [50:20<07:04, 4.90it/s, tok/s=15206.2]
Inference: 65%|██████▍ | 3842/5920 [50:21<14:39, 2.36it/s, tok/s=15206.2]
Inference: 65%|██████▍ | 3842/5920 [50:21<14:39, 2.36it/s, tok/s=15201.1]
Inference: 65%|██████▍ | 3843/5920 [50:22<15:24, 2.25it/s, tok/s=15201.1]
Inference: 65%|██████▍ | 3843/5920 [50:22<15:24, 2.25it/s, tok/s=15200.3]
Inference: 65%|██████▍ | 3844/5920 [50:22<15:24, 2.25it/s, tok/s=15200.7]
Inference: 65%|██████▍ | 3845/5920 [50:22<12:00, 2.88it/s, tok/s=15200.7]
Inference: 65%|██████▍ | 3845/5920 [50:22<12:00, 2.88it/s, tok/s=15200.7]
Inference: 65%|██████▍ | 3846/5920 [50:22<11:22, 3.04it/s, tok/s=15200.7]
Inference: 65%|██████▍ | 3846/5920 [50:22<11:22, 3.04it/s, tok/s=15200.9]
Inference: 65%|██████▍ | 3847/5920 [50:23<11:23, 3.03it/s, tok/s=15200.9]
Inference: 65%|██████▍ | 3847/5920 [50:23<11:23, 3.03it/s, tok/s=15200.0]
Inference: 65%|██████▌ | 3848/5920 [50:23<09:35, 3.60it/s, tok/s=15200.0]
Inference: 65%|██████▌ | 3848/5920 [50:23<09:35, 3.60it/s, tok/s=15199.9]
Inference: 65%|██████▌ | 3849/5920 [50:23<09:35, 3.60it/s, tok/s=15200.6]
Inference: 65%|██████▌ | 3850/5920 [50:23<07:28, 4.61it/s, tok/s=15200.6]
Inference: 65%|██████▌ | 3850/5920 [50:23<07:28, 4.61it/s, tok/s=15200.4]
Inference: 65%|██████▌ | 3851/5920 [50:23<06:57, 4.96it/s, tok/s=15200.4]
Inference: 65%|██████▌ | 3851/5920 [50:23<06:57, 4.96it/s, tok/s=15200.9]
Inference: 65%|██████▌ | 3852/5920 [50:23<08:11, 4.21it/s, tok/s=15200.9]
Inference: 65%|██████▌ | 3852/5920 [50:23<08:11, 4.21it/s, tok/s=15200.1]
Inference: 65%|██████▌ | 3853/5920 [50:24<08:11, 4.21it/s, tok/s=15200.5]
Inference: 65%|██████▌ | 3854/5920 [50:24<06:05, 5.65it/s, tok/s=15200.5]
Inference: 65%|██████▌ | 3854/5920 [50:24<06:05, 5.65it/s, tok/s=15201.0]
Inference: 65%|██████▌ | 3855/5920 [50:24<08:30, 4.04it/s, tok/s=15201.0]
Inference: 65%|██████▌ | 3855/5920 [50:24<08:30, 4.04it/s, tok/s=15199.3]
Inference: 65%|██████▌ | 3856/5920 [50:24<07:37, 4.51it/s, tok/s=15199.3]
Inference: 65%|██████▌ | 3856/5920 [50:24<07:37, 4.51it/s, tok/s=15199.4]
Inference: 65%|██████▌ | 3857/5920 [50:24<07:37, 4.51it/s, tok/s=15200.1]
Inference: 65%|██████▌ | 3858/5920 [50:25<10:53, 3.16it/s, tok/s=15200.1]
Inference: 65%|██████▌ | 3858/5920 [50:25<10:53, 3.16it/s, tok/s=15196.3]
Inference: 65%|██████▌ | 3859/5920 [50:26<11:22, 3.02it/s, tok/s=15196.3]
Inference: 65%|██████▌ | 3859/5920 [50:26<11:22, 3.02it/s, tok/s=15202.5]
Inference: 65%|██████▌ | 3860/5920 [50:26<11:21, 3.02it/s, tok/s=15203.8]
Inference: 65%|██████▌ | 3861/5920 [50:26<08:44, 3.93it/s, tok/s=15203.8]
Inference: 65%|██████▌ | 3861/5920 [50:26<08:44, 3.93it/s, tok/s=15203.4]
Inference: 65%|██████▌ | 3862/5920 [50:26<09:34, 3.58it/s, tok/s=15203.4]
Inference: 65%|██████▌ | 3862/5920 [50:26<09:34, 3.58it/s, tok/s=15202.2]
Inference: 65%|██████▌ | 3863/5920 [50:26<09:34, 3.58it/s, tok/s=15203.6]
Inference: 65%|██████▌ | 3864/5920 [50:27<08:03, 4.25it/s, tok/s=15203.6]
Inference: 65%|██████▌ | 3864/5920 [50:27<08:03, 4.25it/s, tok/s=15202.7]
Inference: 65%|██████▌ | 3865/5920 [50:27<07:07, 4.81it/s, tok/s=15202.7]
Inference: 65%|██████▌ | 3865/5920 [50:27<07:07, 4.81it/s, tok/s=15205.2]
Inference: 65%|██████▌ | 3866/5920 [50:27<06:49, 5.01it/s, tok/s=15205.2]
Inference: 65%|██████▌ | 3866/5920 [50:27<06:49, 5.01it/s, tok/s=15206.3]
Inference: 65%|██████▌ | 3867/5920 [50:27<06:49, 5.01it/s, tok/s=15207.1]
Inference: 65%|██████▌ | 3868/5920 [50:27<05:32, 6.18it/s, tok/s=15207.1]
Inference: 65%|██████▌ | 3868/5920 [50:27<05:32, 6.18it/s, tok/s=15207.9]
Inference: 65%|██████▌ | 3869/5920 [50:27<06:07, 5.58it/s, tok/s=15207.9]
Inference: 65%|██████▌ | 3869/5920 [50:27<06:07, 5.58it/s, tok/s=15207.3]
Inference: 65%|██████▌ | 3870/5920 [50:28<07:32, 4.53it/s, tok/s=15207.3]
Inference: 65%|██████▌ | 3870/5920 [50:28<07:32, 4.53it/s, tok/s=15206.3]
Inference: 65%|██████▌ | 3871/5920 [50:28<07:31, 4.53it/s, tok/s=15207.0]
Inference: 65%|██████▌ | 3872/5920 [50:28<07:31, 4.53it/s, tok/s=15207.8]
Inference: 65%|██████▌ | 3873/5920 [50:28<04:38, 7.34it/s, tok/s=15207.8]
Inference: 65%|██████▌ | 3873/5920 [50:28<04:38, 7.34it/s, tok/s=15207.6]
Inference: 65%|██████▌ | 3874/5920 [50:28<04:38, 7.34it/s, tok/s=15208.7]
Inference: 65%|██████▌ | 3875/5920 [50:28<04:38, 7.34it/s, tok/s=15209.6]
Inference: 65%|██████▌ | 3876/5920 [50:28<05:54, 5.77it/s, tok/s=15209.6]
Inference: 65%|██████▌ | 3876/5920 [50:28<05:54, 5.77it/s, tok/s=15206.8]
Inference: 65%|██████▌ | 3877/5920 [50:29<05:32, 6.14it/s, tok/s=15206.8]
Inference: 65%|██████▌ | 3877/5920 [50:29<05:32, 6.14it/s, tok/s=15206.7]
Inference: 66%|██████▌ | 3878/5920 [50:29<06:30, 5.23it/s, tok/s=15206.7]
Inference: 66%|██████▌ | 3878/5920 [50:29<06:30, 5.23it/s, tok/s=15205.6]
Inference: 66%|██████▌ | 3879/5920 [50:29<06:36, 5.15it/s, tok/s=15205.6]
Inference: 66%|██████▌ | 3879/5920 [50:29<06:36, 5.15it/s, tok/s=15209.4]
Inference: 66%|██████▌ | 3880/5920 [50:29<06:10, 5.51it/s, tok/s=15209.4]
Inference: 66%|██████▌ | 3880/5920 [50:29<06:10, 5.51it/s, tok/s=15209.0]
Inference: 66%|██████▌ | 3881/5920 [50:29<06:09, 5.52it/s, tok/s=15209.0]
Inference: 66%|██████▌ | 3881/5920 [50:29<06:09, 5.52it/s, tok/s=15212.4]
Inference: 66%|██████▌ | 3882/5920 [50:30<06:59, 4.86it/s, tok/s=15212.4]
Inference: 66%|██████▌ | 3882/5920 [50:30<06:59, 4.86it/s, tok/s=15212.8]
Inference: 66%|██████▌ | 3883/5920 [50:30<06:59, 4.86it/s, tok/s=15213.1]
Inference: 66%|██████▌ | 3884/5920 [50:30<07:33, 4.49it/s, tok/s=15213.1]
Inference: 66%|██████▌ | 3884/5920 [50:30<07:33, 4.49it/s, tok/s=15211.5]
Inference: 66%|██████▌ | 3885/5920 [50:30<06:35, 5.15it/s, tok/s=15211.5]
Inference: 66%|██████▌ | 3885/5920 [50:30<06:35, 5.15it/s, tok/s=15219.5]
Inference: 66%|██████▌ | 3886/5920 [50:31<09:56, 3.41it/s, tok/s=15219.5]
Inference: 66%|██████▌ | 3886/5920 [50:31<09:56, 3.41it/s, tok/s=15218.2]
Inference: 66%|██████▌ | 3887/5920 [50:31<12:44, 2.66it/s, tok/s=15218.2]
Inference: 66%|██████▌ | 3887/5920 [50:31<12:44, 2.66it/s, tok/s=15215.9]
Inference: 66%|██████▌ | 3888/5920 [50:32<12:44, 2.66it/s, tok/s=15216.3]
Inference: 66%|██████▌ | 3889/5920 [50:32<08:41, 3.90it/s, tok/s=15216.3]
Inference: 66%|██████▌ | 3889/5920 [50:32<08:41, 3.90it/s, tok/s=15216.2]
Inference: 66%|██████▌ | 3890/5920 [50:32<08:40, 3.90it/s, tok/s=15217.2]
Inference: 66%|██████▌ | 3891/5920 [50:32<06:55, 4.89it/s, tok/s=15217.2]
Inference: 66%|██████▌ | 3891/5920 [50:32<06:55, 4.89it/s, tok/s=15217.7]
Inference: 66%|██████▌ | 3892/5920 [50:32<06:49, 4.95it/s, tok/s=15217.7]
Inference: 66%|██████▌ | 3892/5920 [50:32<06:49, 4.95it/s, tok/s=15217.2]
Inference: 66%|██████▌ | 3893/5920 [50:32<06:44, 5.01it/s, tok/s=15217.2]
Inference: 66%|██████▌ | 3893/5920 [50:32<06:44, 5.01it/s, tok/s=15216.8]
Inference: 66%|██████▌ | 3894/5920 [50:33<07:06, 4.75it/s, tok/s=15216.8]
Inference: 66%|██████▌ | 3894/5920 [50:33<07:06, 4.75it/s, tok/s=15216.3]
Inference: 66%|██████▌ | 3895/5920 [50:33<08:06, 4.17it/s, tok/s=15216.3]
Inference: 66%|██████▌ | 3895/5920 [50:33<08:06, 4.17it/s, tok/s=15215.5]
Inference: 66%|██████▌ | 3896/5920 [50:33<08:05, 4.17it/s, tok/s=15215.8]
Inference: 66%|██████▌ | 3897/5920 [50:33<07:07, 4.74it/s, tok/s=15215.8]
Inference: 66%|██████▌ | 3897/5920 [50:33<07:07, 4.74it/s, tok/s=15215.2]
Inference: 66%|██████▌ | 3898/5920 [50:34<10:55, 3.09it/s, tok/s=15215.2]
Inference: 66%|██████▌ | 3898/5920 [50:34<10:55, 3.09it/s, tok/s=15212.5]
Inference: 66%|██████▌ | 3899/5920 [50:34<10:55, 3.09it/s, tok/s=15215.0]
Inference: 66%|██████▌ | 3900/5920 [50:34<07:20, 4.59it/s, tok/s=15215.0]
Inference: 66%|██████▌ | 3900/5920 [50:34<07:20, 4.59it/s, tok/s=15218.5]
Inference: 66%|██████▌ | 3901/5920 [50:34<06:38, 5.06it/s, tok/s=15218.5]
Inference: 66%|██████▌ | 3901/5920 [50:34<06:38, 5.06it/s, tok/s=15219.3]
Inference: 66%|██████▌ | 3902/5920 [50:35<08:56, 3.76it/s, tok/s=15219.3]
Inference: 66%|██████▌ | 3902/5920 [50:35<08:56, 3.76it/s, tok/s=15217.4]
Inference: 66%|██████▌ | 3903/5920 [50:35<07:59, 4.20it/s, tok/s=15217.4]
Inference: 66%|██████▌ | 3903/5920 [50:35<07:59, 4.20it/s, tok/s=15217.6]
Inference: 66%|██████▌ | 3904/5920 [50:36<16:28, 2.04it/s, tok/s=15217.6]
Inference: 66%|██████▌ | 3904/5920 [50:36<16:28, 2.04it/s, tok/s=15212.5]
Inference: 66%|██████▌ | 3905/5920 [50:36<16:27, 2.04it/s, tok/s=15212.9]
Inference: 66%|██████▌ | 3906/5920 [50:36<12:31, 2.68it/s, tok/s=15212.9]
Inference: 66%|██████▌ | 3906/5920 [50:36<12:31, 2.68it/s, tok/s=15221.1]
Inference: 66%|██████▌ | 3907/5920 [50:37<13:45, 2.44it/s, tok/s=15221.1]
Inference: 66%|██████▌ | 3907/5920 [50:37<13:45, 2.44it/s, tok/s=15224.8]
Inference: 66%|██████▌ | 3908/5920 [50:38<16:17, 2.06it/s, tok/s=15224.8]
Inference: 66%|██████▌ | 3908/5920 [50:38<16:17, 2.06it/s, tok/s=15222.0]
Inference: 66%|██████▌ | 3909/5920 [50:38<16:17, 2.06it/s, tok/s=15225.8]
Inference: 66%|██████▌ | 3910/5920 [50:38<14:59, 2.23it/s, tok/s=15225.8]
Inference: 66%|██████▌ | 3910/5920 [50:38<14:59, 2.23it/s, tok/s=15222.5]
Inference: 66%|██████▌ | 3911/5920 [50:39<12:45, 2.62it/s, tok/s=15222.5]
Inference: 66%|██████▌ | 3911/5920 [50:39<12:45, 2.62it/s, tok/s=15231.3]
Inference: 66%|██████▌ | 3912/5920 [50:39<11:52, 2.82it/s, tok/s=15231.3]
Inference: 66%|██████▌ | 3912/5920 [50:39<11:52, 2.82it/s, tok/s=15230.9]
Inference: 66%|██████▌ | 3913/5920 [50:39<12:09, 2.75it/s, tok/s=15230.9]
Inference: 66%|██████▌ | 3913/5920 [50:39<12:09, 2.75it/s, tok/s=15229.7]
Inference: 66%|██████▌ | 3914/5920 [50:39<12:08, 2.75it/s, tok/s=15230.3]
Inference: 66%|██████▌ | 3915/5920 [50:40<09:03, 3.69it/s, tok/s=15230.3]
Inference: 66%|██████▌ | 3915/5920 [50:40<09:03, 3.69it/s, tok/s=15231.0]
Inference: 66%|██████▌ | 3916/5920 [50:40<07:59, 4.18it/s, tok/s=15231.0]
Inference: 66%|██████▌ | 3916/5920 [50:40<07:59, 4.18it/s, tok/s=15230.6]
Inference: 66%|██████▌ | 3917/5920 [50:40<07:27, 4.48it/s, tok/s=15230.6]
Inference: 66%|██████▌ | 3917/5920 [50:40<07:27, 4.48it/s, tok/s=15230.7]
Inference: 66%|██████▌ | 3918/5920 [50:40<08:27, 3.95it/s, tok/s=15230.7]
Inference: 66%|██████▌ | 3918/5920 [50:40<08:27, 3.95it/s, tok/s=15229.3]
Inference: 66%|██████▌ | 3919/5920 [50:40<07:28, 4.46it/s, tok/s=15229.3]
Inference: 66%|██████▌ | 3919/5920 [50:40<07:28, 4.46it/s, tok/s=15229.0]
Inference: 66%|██████▌ | 3920/5920 [50:41<07:51, 4.25it/s, tok/s=15229.0]
Inference: 66%|██████▌ | 3920/5920 [50:41<07:51, 4.25it/s, tok/s=15228.2]
Inference: 66%|██████▌ | 3921/5920 [50:41<07:50, 4.25it/s, tok/s=15231.8]
Inference: 66%|██████▋ | 3922/5920 [50:41<09:41, 3.44it/s, tok/s=15231.8]
Inference: 66%|██████▋ | 3922/5920 [50:41<09:41, 3.44it/s, tok/s=15230.9]
Inference: 66%|██████▋ | 3923/5920 [50:42<12:17, 2.71it/s, tok/s=15230.9]
Inference: 66%|██████▋ | 3923/5920 [50:42<12:17, 2.71it/s, tok/s=15228.9]
Inference: 66%|██████▋ | 3924/5920 [50:42<12:17, 2.71it/s, tok/s=15229.3]
Inference: 66%|██████▋ | 3925/5920 [50:42<08:25, 3.95it/s, tok/s=15229.3]
Inference: 66%|██████▋ | 3925/5920 [50:42<08:25, 3.95it/s, tok/s=15234.3]
Inference: 66%|██████▋ | 3926/5920 [50:42<07:54, 4.20it/s, tok/s=15234.3]
Inference: 66%|██████▋ | 3926/5920 [50:42<07:54, 4.20it/s, tok/s=15235.0]
Inference: 66%|██████▋ | 3927/5920 [50:42<06:47, 4.89it/s, tok/s=15235.0]
Inference: 66%|██████▋ | 3927/5920 [50:42<06:47, 4.89it/s, tok/s=15236.8]
Inference: 66%|██████▋ | 3928/5920 [50:43<06:30, 5.10it/s, tok/s=15236.8]
Inference: 66%|██████▋ | 3928/5920 [50:43<06:30, 5.10it/s, tok/s=15237.9]
Inference: 66%|██████▋ | 3929/5920 [50:43<07:08, 4.64it/s, tok/s=15237.9]
Inference: 66%|██████▋ | 3929/5920 [50:43<07:08, 4.64it/s, tok/s=15238.3]
Inference: 66%|██████▋ | 3930/5920 [50:43<11:17, 2.94it/s, tok/s=15238.3]
Inference: 66%|██████▋ | 3930/5920 [50:43<11:17, 2.94it/s, tok/s=15236.4]
Inference: 66%|██████▋ | 3931/5920 [50:44<09:55, 3.34it/s, tok/s=15236.4]
Inference: 66%|██████▋ | 3931/5920 [50:44<09:55, 3.34it/s, tok/s=15241.4]
Inference: 66%|██████▋ | 3932/5920 [50:44<09:54, 3.34it/s, tok/s=15243.1]
Inference: 66%|██████▋ | 3933/5920 [50:44<08:15, 4.01it/s, tok/s=15243.1]
Inference: 66%|██████▋ | 3933/5920 [50:44<08:15, 4.01it/s, tok/s=15242.4]
Inference: 66%|██████▋ | 3934/5920 [50:44<08:15, 4.01it/s, tok/s=15243.0]
Inference: 66%|██████▋ | 3935/5920 [50:45<10:12, 3.24it/s, tok/s=15243.0]
Inference: 66%|██████▋ | 3935/5920 [50:45<10:12, 3.24it/s, tok/s=15239.7]
Inference: 66%|██████▋ | 3936/5920 [50:45<10:11, 3.24it/s, tok/s=15239.7]
Inference: 66%|██████▋ | 3936/5920 [50:45<10:11, 3.24it/s, tok/s=15239.7]
Inference: 67%|██████▋ | 3937/5920 [50:45<09:19, 3.54it/s, tok/s=15239.7]
Inference: 67%|██████▋ | 3937/5920 [50:45<09:19, 3.54it/s, tok/s=15248.7]
Inference: 67%|██████▋ | 3938/5920 [50:46<08:42, 3.79it/s, tok/s=15248.7]
Inference: 67%|██████▋ | 3938/5920 [50:46<08:42, 3.79it/s, tok/s=15251.3]
Inference: 67%|██████▋ | 3939/5920 [50:46<08:42, 3.79it/s, tok/s=15259.5]
Inference: 67%|██████▋ | 3940/5920 [50:46<07:03, 4.67it/s, tok/s=15259.5]
Inference: 67%|██████▋ | 3940/5920 [50:46<07:03, 4.67it/s, tok/s=15258.8]
Inference: 67%|██████▋ | 3941/5920 [50:46<06:26, 5.11it/s, tok/s=15258.8]
Inference: 67%|██████▋ | 3941/5920 [50:46<06:26, 5.11it/s, tok/s=15259.8]
Inference: 67%|██████▋ | 3942/5920 [50:46<07:33, 4.36it/s, tok/s=15259.8]
Inference: 67%|██████▋ | 3942/5920 [50:46<07:33, 4.36it/s, tok/s=15260.9]
Inference: 67%|██████▋ | 3943/5920 [50:46<07:33, 4.36it/s, tok/s=15261.5]
Inference: 67%|██████▋ | 3944/5920 [50:47<07:50, 4.20it/s, tok/s=15261.5]
Inference: 67%|██████▋ | 3944/5920 [50:47<07:50, 4.20it/s, tok/s=15269.3]
Inference: 67%|██████▋ | 3945/5920 [50:47<06:58, 4.72it/s, tok/s=15269.3]
Inference: 67%|██████▋ | 3945/5920 [50:47<06:58, 4.72it/s, tok/s=15269.2]
Inference: 67%|██████▋ | 3946/5920 [50:47<06:58, 4.72it/s, tok/s=15269.8]
Inference: 67%|██████▋ | 3947/5920 [50:47<05:08, 6.39it/s, tok/s=15269.8]
Inference: 67%|██████▋ | 3947/5920 [50:47<05:08, 6.39it/s, tok/s=15269.7]
Inference: 67%|██████▋ | 3948/5920 [50:47<05:14, 6.27it/s, tok/s=15269.7]
Inference: 67%|██████▋ | 3948/5920 [50:47<05:14, 6.27it/s, tok/s=15270.4]
Inference: 67%|██████▋ | 3949/5920 [50:47<05:14, 6.27it/s, tok/s=15270.9]
Inference: 67%|██████▋ | 3950/5920 [50:47<05:14, 6.27it/s, tok/s=15271.5]
Inference: 67%|██████▋ | 3951/5920 [50:48<03:58, 8.26it/s, tok/s=15271.5]
Inference: 67%|██████▋ | 3951/5920 [50:48<03:58, 8.26it/s, tok/s=15272.5]
Inference: 67%|██████▋ | 3952/5920 [50:48<03:58, 8.26it/s, tok/s=15272.9]
Inference: 67%|██████▋ | 3953/5920 [50:48<03:53, 8.44it/s, tok/s=15272.9]
Inference: 67%|██████▋ | 3953/5920 [50:48<03:53, 8.44it/s, tok/s=15273.0]
Inference: 67%|██████▋ | 3954/5920 [50:49<08:28, 3.86it/s, tok/s=15273.0]
Inference: 67%|██████▋ | 3954/5920 [50:49<08:28, 3.86it/s, tok/s=15269.5]
Inference: 67%|██████▋ | 3955/5920 [50:49<09:37, 3.40it/s, tok/s=15269.5]
Inference: 67%|██████▋ | 3955/5920 [50:49<09:37, 3.40it/s, tok/s=15269.8]
Inference: 67%|██████▋ | 3956/5920 [50:49<08:19, 3.93it/s, tok/s=15269.8]
Inference: 67%|██████▋ | 3956/5920 [50:49<08:19, 3.93it/s, tok/s=15275.7]
Inference: 67%|██████▋ | 3957/5920 [50:50<09:14, 3.54it/s, tok/s=15275.7]
Inference: 67%|██████▋ | 3957/5920 [50:50<09:14, 3.54it/s, tok/s=15275.3]
Inference: 67%|██████▋ | 3958/5920 [50:50<08:49, 3.71it/s, tok/s=15275.3]
Inference: 67%|██████▋ | 3958/5920 [50:50<08:49, 3.71it/s, tok/s=15274.8]
Inference: 67%|██████▋ | 3959/5920 [50:50<09:06, 3.59it/s, tok/s=15274.8]
Inference: 67%|██████▋ | 3959/5920 [50:50<09:06, 3.59it/s, tok/s=15273.8]
Inference: 67%|██████▋ | 3960/5920 [50:50<08:43, 3.74it/s, tok/s=15273.8]
Inference: 67%|██████▋ | 3960/5920 [50:50<08:43, 3.74it/s, tok/s=15276.7]
Inference: 67%|██████▋ | 3961/5920 [50:51<17:10, 1.90it/s, tok/s=15276.7]
Inference: 67%|██████▋ | 3961/5920 [50:51<17:10, 1.90it/s, tok/s=15271.7]
Inference: 67%|██████▋ | 3962/5920 [50:52<17:09, 1.90it/s, tok/s=15272.5]
Inference: 67%|██████▋ | 3963/5920 [50:52<10:30, 3.11it/s, tok/s=15272.5]
Inference: 67%|██████▋ | 3963/5920 [50:52<10:30, 3.11it/s, tok/s=15272.7]
Inference: 67%|██████▋ | 3964/5920 [50:52<10:29, 3.11it/s, tok/s=15282.3]
Inference: 67%|██████▋ | 3965/5920 [50:52<10:29, 3.11it/s, tok/s=15292.2]
Inference: 67%|██████▋ | 3966/5920 [50:52<10:29, 3.11it/s, tok/s=15292.6]
Inference: 67%|██████▋ | 3967/5920 [50:52<10:28, 3.11it/s, tok/s=15293.3]
Inference: 67%|██████▋ | 3968/5920 [50:52<04:50, 6.73it/s, tok/s=15293.3]
Inference: 67%|██████▋ | 3968/5920 [50:52<04:50, 6.73it/s, tok/s=15292.9]
Inference: 67%|██████▋ | 3969/5920 [50:52<04:49, 6.73it/s, tok/s=15293.1]
Inference: 67%|██████▋ | 3970/5920 [50:52<05:28, 5.93it/s, tok/s=15293.1]
Inference: 67%|██████▋ | 3970/5920 [50:52<05:28, 5.93it/s, tok/s=15291.5]
Inference: 67%|██████▋ | 3971/5920 [50:52<05:30, 5.89it/s, tok/s=15291.5]
Inference: 67%|██████▋ | 3971/5920 [50:52<05:30, 5.89it/s, tok/s=15292.7]
Inference: 67%|██████▋ | 3972/5920 [50:52<05:30, 5.89it/s, tok/s=15302.5]
Inference: 67%|██████▋ | 3973/5920 [50:54<11:27, 2.83it/s, tok/s=15302.5]
Inference: 67%|██████▋ | 3973/5920 [50:54<11:27, 2.83it/s, tok/s=15295.6]
Inference: 67%|██████▋ | 3974/5920 [50:54<11:27, 2.83it/s, tok/s=15296.7]
Inference: 67%|██████▋ | 3975/5920 [50:54<09:47, 3.31it/s, tok/s=15296.7]
Inference: 67%|██████▋ | 3975/5920 [50:54<09:47, 3.31it/s, tok/s=15295.6]
Inference: 67%|██████▋ | 3976/5920 [50:55<11:47, 2.75it/s, tok/s=15295.6]
Inference: 67%|██████▋ | 3976/5920 [50:55<11:47, 2.75it/s, tok/s=15298.1]
Inference: 67%|██████▋ | 3977/5920 [50:55<11:07, 2.91it/s, tok/s=15298.1]
Inference: 67%|██████▋ | 3977/5920 [50:55<11:07, 2.91it/s, tok/s=15297.6]
Inference: 67%|██████▋ | 3978/5920 [50:55<11:07, 2.91it/s, tok/s=15304.9]
Inference: 67%|██████▋ | 3979/5920 [50:56<11:01, 2.93it/s, tok/s=15304.9]
Inference: 67%|██████▋ | 3979/5920 [50:56<11:01, 2.93it/s, tok/s=15302.3]
Inference: 67%|██████▋ | 3980/5920 [50:56<11:32, 2.80it/s, tok/s=15302.3]
Inference: 67%|██████▋ | 3980/5920 [50:56<11:32, 2.80it/s, tok/s=15309.8]
Inference: 67%|██████▋ | 3981/5920 [50:56<11:32, 2.80it/s, tok/s=15311.6]
Inference: 67%|██████▋ | 3982/5920 [50:57<08:14, 3.92it/s, tok/s=15311.6]
Inference: 67%|██████▋ | 3982/5920 [50:57<08:14, 3.92it/s, tok/s=15311.3]
Inference: 67%|██████▋ | 3983/5920 [50:57<08:33, 3.78it/s, tok/s=15311.3]
Inference: 67%|██████▋ | 3983/5920 [50:57<08:33, 3.78it/s, tok/s=15310.1]
Inference: 67%|██████▋ | 3984/5920 [50:57<09:53, 3.26it/s, tok/s=15310.1]
Inference: 67%|██████▋ | 3984/5920 [50:57<09:53, 3.26it/s, tok/s=15308.8]
Inference: 67%|██████▋ | 3985/5920 [50:58<14:52, 2.17it/s, tok/s=15308.8]
Inference: 67%|██████▋ | 3985/5920 [50:58<14:52, 2.17it/s, tok/s=15306.3]
Inference: 67%|██████▋ | 3986/5920 [50:59<14:22, 2.24it/s, tok/s=15306.3]
Inference: 67%|██████▋ | 3986/5920 [50:59<14:22, 2.24it/s, tok/s=15305.2]
Inference: 67%|██████▋ | 3987/5920 [50:59<13:19, 2.42it/s, tok/s=15305.2]
Inference: 67%|██████▋ | 3987/5920 [50:59<13:19, 2.42it/s, tok/s=15305.0]
Inference: 67%|██████▋ | 3988/5920 [50:59<13:18, 2.42it/s, tok/s=15306.4]
Inference: 67%|██████▋ | 3989/5920 [50:59<08:32, 3.77it/s, tok/s=15306.4]
Inference: 67%|██████▋ | 3989/5920 [50:59<08:32, 3.77it/s, tok/s=15315.6]
Inference: 67%|██████▋ | 3990/5920 [50:59<07:19, 4.39it/s, tok/s=15315.6]
Inference: 67%|██████▋ | 3990/5920 [50:59<07:19, 4.39it/s, tok/s=15318.3]
Inference: 67%|██████▋ | 3991/5920 [50:59<07:19, 4.39it/s, tok/s=15319.6]
Inference: 67%|██████▋ | 3992/5920 [50:59<05:25, 5.93it/s, tok/s=15319.6]
Inference: 67%|██████▋ | 3992/5920 [50:59<05:25, 5.93it/s, tok/s=15319.4]
Inference: 67%|██████▋ | 3993/5920 [51:00<11:01, 2.91it/s, tok/s=15319.4]
Inference: 67%|██████▋ | 3993/5920 [51:00<11:01, 2.91it/s, tok/s=15315.2]
Inference: 67%|██████▋ | 3994/5920 [51:01<15:04, 2.13it/s, tok/s=15315.2]
Inference: 67%|██████▋ | 3994/5920 [51:01<15:04, 2.13it/s, tok/s=15312.6]
Inference: 67%|██████▋ | 3995/5920 [51:01<12:02, 2.66it/s, tok/s=15312.6]
Inference: 67%|██████▋ | 3995/5920 [51:01<12:02, 2.66it/s, tok/s=15312.8]
Inference: 68%|██████▊ | 3996/5920 [51:01<12:02, 2.66it/s, tok/s=15313.3]
Inference: 68%|██████▊ | 3997/5920 [51:01<07:39, 4.19it/s, tok/s=15313.3]
Inference: 68%|██████▊ | 3997/5920 [51:01<07:39, 4.19it/s, tok/s=15314.3]
Inference: 68%|██████▊ | 3998/5920 [51:02<07:38, 4.19it/s, tok/s=15314.3]
Inference: 68%|██████▊ | 3999/5920 [51:02<10:23, 3.08it/s, tok/s=15314.3]
Inference: 68%|██████▊ | 3999/5920 [51:02<10:23, 3.08it/s, tok/s=15314.2]
Inference: 68%|██████▊ | 4000/5920 [51:02<09:18, 3.44it/s, tok/s=15314.2]
Inference: 68%|██████▊ | 4000/5920 [51:02<09:18, 3.44it/s, tok/s=15314.9]
Inference: 68%|██████▊ | 4001/5920 [51:03<08:05, 3.95it/s, tok/s=15314.9]
Inference: 68%|██████▊ | 4001/5920 [51:03<08:05, 3.95it/s, tok/s=15315.2]
Inference: 68%|██████▊ | 4002/5920 [51:03<07:10, 4.46it/s, tok/s=15315.2]
Inference: 68%|██████▊ | 4002/5920 [51:03<07:10, 4.46it/s, tok/s=15316.0]
Inference: 68%|██████▊ | 4003/5920 [51:03<07:09, 4.46it/s, tok/s=15318.8]
Inference: 68%|██████▊ | 4004/5920 [51:03<05:43, 5.58it/s, tok/s=15318.8]
Inference: 68%|██████▊ | 4004/5920 [51:03<05:43, 5.58it/s, tok/s=15319.9]
Inference: 68%|██████▊ | 4005/5920 [51:04<09:18, 3.43it/s, tok/s=15319.9]
Inference: 68%|██████▊ | 4005/5920 [51:04<09:18, 3.43it/s, tok/s=15317.7]
Inference: 68%|██████▊ | 4006/5920 [51:04<08:15, 3.86it/s, tok/s=15317.7]
Inference: 68%|██████▊ | 4006/5920 [51:04<08:15, 3.86it/s, tok/s=15317.5]
Inference: 68%|██████▊ | 4007/5920 [51:04<07:54, 4.03it/s, tok/s=15317.5]
Inference: 68%|██████▊ | 4007/5920 [51:04<07:54, 4.03it/s, tok/s=15316.9]
Inference: 68%|██████▊ | 4008/5920 [51:04<07:54, 4.03it/s, tok/s=15319.3]
Inference: 68%|██████▊ | 4009/5920 [51:04<05:13, 6.09it/s, tok/s=15319.3]
Inference: 68%|██████▊ | 4009/5920 [51:04<05:13, 6.09it/s, tok/s=15319.5]
Inference: 68%|██████▊ | 4010/5920 [51:05<08:35, 3.70it/s, tok/s=15319.5]
Inference: 68%|██████▊ | 4010/5920 [51:05<08:35, 3.70it/s, tok/s=15317.9]
Inference: 68%|██████▊ | 4011/5920 [51:05<08:35, 3.70it/s, tok/s=15318.4]
Inference: 68%|██████▊ | 4012/5920 [51:05<07:08, 4.46it/s, tok/s=15318.4]
Inference: 68%|██████▊ | 4012/5920 [51:05<07:08, 4.46it/s, tok/s=15317.5]
Inference: 68%|██████▊ | 4013/5920 [51:05<07:07, 4.46it/s, tok/s=15319.6]
Inference: 68%|██████▊ | 4014/5920 [51:05<07:07, 4.46it/s, tok/s=15320.1]
Inference: 68%|██████▊ | 4015/5920 [51:05<05:11, 6.12it/s, tok/s=15320.1]
Inference: 68%|██████▊ | 4015/5920 [51:05<05:11, 6.12it/s, tok/s=15320.2]
Inference: 68%|██████▊ | 4016/5920 [51:05<05:10, 6.12it/s, tok/s=15320.5]
Inference: 68%|██████▊ | 4017/5920 [51:06<06:04, 5.21it/s, tok/s=15320.5]
Inference: 68%|██████▊ | 4017/5920 [51:06<06:04, 5.21it/s, tok/s=15328.0]
Inference: 68%|██████▊ | 4018/5920 [51:07<09:36, 3.30it/s, tok/s=15328.0]
Inference: 68%|██████▊ | 4018/5920 [51:07<09:36, 3.30it/s, tok/s=15324.7]
Inference: 68%|██████▊ | 4019/5920 [51:07<08:30, 3.72it/s, tok/s=15324.7]
Inference: 68%|██████▊ | 4019/5920 [51:07<08:30, 3.72it/s, tok/s=15333.8]
Inference: 68%|██████▊ | 4020/5920 [51:08<13:30, 2.35it/s, tok/s=15333.8]
Inference: 68%|██████▊ | 4020/5920 [51:08<13:30, 2.35it/s, tok/s=15330.0]
Inference: 68%|██████▊ | 4021/5920 [51:08<12:36, 2.51it/s, tok/s=15330.0]
Inference: 68%|██████▊ | 4021/5920 [51:08<12:36, 2.51it/s, tok/s=15338.3]
Inference: 68%|██████▊ | 4022/5920 [51:08<10:16, 3.08it/s, tok/s=15338.3]
Inference: 68%|██████▊ | 4022/5920 [51:08<10:16, 3.08it/s, tok/s=15341.2]
Inference: 68%|██████▊ | 4023/5920 [51:08<10:15, 3.08it/s, tok/s=15341.5]
Inference: 68%|██████▊ | 4024/5920 [51:08<10:15, 3.08it/s, tok/s=15351.1]
Inference: 68%|██████▊ | 4025/5920 [51:09<07:11, 4.39it/s, tok/s=15351.1]
Inference: 68%|██████▊ | 4025/5920 [51:09<07:11, 4.39it/s, tok/s=15350.5]
Inference: 68%|██████▊ | 4026/5920 [51:09<07:11, 4.39it/s, tok/s=15350.7]
Inference: 68%|██████▊ | 4027/5920 [51:09<05:46, 5.46it/s, tok/s=15350.7]
Inference: 68%|██████▊ | 4027/5920 [51:09<05:46, 5.46it/s, tok/s=15350.9]
Inference: 68%|██████▊ | 4028/5920 [51:09<05:46, 5.46it/s, tok/s=15351.7]
Inference: 68%|██████▊ | 4029/5920 [51:09<05:00, 6.30it/s, tok/s=15351.7]
Inference: 68%|██████▊ | 4029/5920 [51:09<05:00, 6.30it/s, tok/s=15351.5]
Inference: 68%|██████▊ | 4030/5920 [51:09<05:46, 5.46it/s, tok/s=15351.5]
Inference: 68%|██████▊ | 4030/5920 [51:09<05:46, 5.46it/s, tok/s=15350.9]
Inference: 68%|██████▊ | 4031/5920 [51:09<05:51, 5.37it/s, tok/s=15350.9]
Inference: 68%|██████▊ | 4031/5920 [51:09<05:51, 5.37it/s, tok/s=15350.8]
Inference: 68%|██████▊ | 4032/5920 [51:10<06:29, 4.85it/s, tok/s=15350.8]
Inference: 68%|██████▊ | 4032/5920 [51:10<06:29, 4.85it/s, tok/s=15350.2]
Inference: 68%|██████▊ | 4033/5920 [51:10<05:49, 5.40it/s, tok/s=15350.2]
Inference: 68%|██████▊ | 4033/5920 [51:10<05:49, 5.40it/s, tok/s=15349.9]
Inference: 68%|██████▊ | 4034/5920 [51:10<05:49, 5.40it/s, tok/s=15350.1]
Inference: 68%|██████▊ | 4035/5920 [51:10<04:58, 6.31it/s, tok/s=15350.1]
Inference: 68%|██████▊ | 4035/5920 [51:10<04:58, 6.31it/s, tok/s=15350.3]
Inference: 68%|██████▊ | 4036/5920 [51:10<04:58, 6.31it/s, tok/s=15350.4]
Inference: 68%|██████▊ | 4037/5920 [51:10<04:58, 6.31it/s, tok/s=15351.6]
Inference: 68%|██████▊ | 4038/5920 [51:11<04:59, 6.29it/s, tok/s=15351.6]
Inference: 68%|██████▊ | 4038/5920 [51:11<04:59, 6.29it/s, tok/s=15350.0]
Inference: 68%|██████▊ | 4039/5920 [51:11<04:58, 6.29it/s, tok/s=15350.9]
Inference: 68%|██████▊ | 4040/5920 [51:11<04:24, 7.11it/s, tok/s=15350.9]
Inference: 68%|██████▊ | 4040/5920 [51:11<04:24, 7.11it/s, tok/s=15352.0]
Inference: 68%|██████▊ | 4041/5920 [51:11<04:24, 7.11it/s, tok/s=15352.4]
Inference: 68%|██████▊ | 4042/5920 [51:11<03:46, 8.28it/s, tok/s=15352.4]
Inference: 68%|██████▊ | 4042/5920 [51:11<03:46, 8.28it/s, tok/s=15352.2]
Inference: 68%|██████▊ | 4043/5920 [51:12<07:14, 4.32it/s, tok/s=15352.2]
Inference: 68%|██████▊ | 4043/5920 [51:12<07:14, 4.32it/s, tok/s=15349.6]
Inference: 68%|██████▊ | 4044/5920 [51:12<07:16, 4.30it/s, tok/s=15349.6]
Inference: 68%|██████▊ | 4044/5920 [51:12<07:16, 4.30it/s, tok/s=15352.8]
Inference: 68%|██████▊ | 4045/5920 [51:12<07:07, 4.38it/s, tok/s=15352.8]
Inference: 68%|██████▊ | 4045/5920 [51:12<07:07, 4.38it/s, tok/s=15352.4]
Inference: 68%|██████▊ | 4046/5920 [51:12<07:07, 4.38it/s, tok/s=15353.0]
Inference: 68%|██████▊ | 4047/5920 [51:12<06:09, 5.07it/s, tok/s=15353.0]
Inference: 68%|██████▊ | 4047/5920 [51:12<06:09, 5.07it/s, tok/s=15352.2]
Inference: 68%|██████▊ | 4048/5920 [51:13<08:46, 3.55it/s, tok/s=15352.2]
Inference: 68%|██████▊ | 4048/5920 [51:13<08:46, 3.55it/s, tok/s=15351.8]
Inference: 68%|██████▊ | 4049/5920 [51:13<08:46, 3.55it/s, tok/s=15352.3]
Inference: 68%|██████▊ | 4050/5920 [51:13<06:45, 4.61it/s, tok/s=15352.3]
Inference: 68%|██████▊ | 4050/5920 [51:13<06:45, 4.61it/s, tok/s=15352.2]
Inference: 68%|██████▊ | 4051/5920 [51:14<07:20, 4.24it/s, tok/s=15352.2]
Inference: 68%|██████▊ | 4051/5920 [51:14<07:20, 4.24it/s, tok/s=15351.9]
Inference: 68%|██████▊ | 4052/5920 [51:14<08:16, 3.76it/s, tok/s=15351.9]
Inference: 68%|██████▊ | 4052/5920 [51:14<08:16, 3.76it/s, tok/s=15350.7]
Inference: 68%|██████▊ | 4053/5920 [51:14<07:10, 4.34it/s, tok/s=15350.7]
Inference: 68%|██████▊ | 4053/5920 [51:14<07:10, 4.34it/s, tok/s=15351.6]
Inference: 68%|██████▊ | 4054/5920 [51:14<06:29, 4.79it/s, tok/s=15351.6]
Inference: 68%|██████▊ | 4054/5920 [51:14<06:29, 4.79it/s, tok/s=15352.4]
Inference: 68%|██████▊ | 4055/5920 [51:14<06:52, 4.52it/s, tok/s=15352.4]
Inference: 68%|██████▊ | 4055/5920 [51:14<06:52, 4.52it/s, tok/s=15352.8]
Inference: 69%|██████▊ | 4056/5920 [51:15<07:47, 3.99it/s, tok/s=15352.8]
Inference: 69%|██████▊ | 4056/5920 [51:15<07:47, 3.99it/s, tok/s=15352.2]
Inference: 69%|██████▊ | 4057/5920 [51:15<07:11, 4.32it/s, tok/s=15352.2]
Inference: 69%|██████▊ | 4057/5920 [51:15<07:11, 4.32it/s, tok/s=15354.1]
Inference: 69%|██████▊ | 4058/5920 [51:15<07:10, 4.32it/s, tok/s=15355.8]
Inference: 69%|██████▊ | 4059/5920 [51:15<05:25, 5.72it/s, tok/s=15355.8]
Inference: 69%|██████▊ | 4059/5920 [51:15<05:25, 5.72it/s, tok/s=15357.6]
Inference: 69%|██████▊ | 4060/5920 [51:15<06:06, 5.07it/s, tok/s=15357.6]
Inference: 69%|██████▊ | 4060/5920 [51:15<06:06, 5.07it/s, tok/s=15357.2]
Inference: 69%|██████▊ | 4061/5920 [51:15<06:06, 5.07it/s, tok/s=15359.2]
Inference: 69%|██████▊ | 4062/5920 [51:16<04:59, 6.20it/s, tok/s=15359.2]
Inference: 69%|██████▊ | 4062/5920 [51:16<04:59, 6.20it/s, tok/s=15358.5]
Inference: 69%|██████▊ | 4063/5920 [51:16<04:59, 6.20it/s, tok/s=15358.7]
Inference: 69%|██████▊ | 4064/5920 [51:16<05:00, 6.19it/s, tok/s=15358.7]
Inference: 69%|██████▊ | 4064/5920 [51:16<05:00, 6.19it/s, tok/s=15360.2]
Inference: 69%|██████▊ | 4065/5920 [51:16<04:59, 6.19it/s, tok/s=15367.5]
Inference: 69%|██████▊ | 4066/5920 [51:16<04:12, 7.34it/s, tok/s=15367.5]
Inference: 69%|██████▊ | 4066/5920 [51:16<04:12, 7.34it/s, tok/s=15368.9]
Inference: 69%|██████▊ | 4067/5920 [51:16<04:45, 6.48it/s, tok/s=15368.9]
Inference: 69%|██████▊ | 4067/5920 [51:16<04:45, 6.48it/s, tok/s=15375.9]
Inference: 69%|██████▊ | 4068/5920 [51:17<07:19, 4.21it/s, tok/s=15375.9]
Inference: 69%|██████▊ | 4068/5920 [51:17<07:19, 4.21it/s, tok/s=15373.8]
Inference: 69%|██████▊ | 4069/5920 [51:17<07:17, 4.23it/s, tok/s=15373.8]
Inference: 69%|██████▊ | 4069/5920 [51:17<07:17, 4.23it/s, tok/s=15373.3]
Inference: 69%|██████▉ | 4070/5920 [51:17<08:09, 3.78it/s, tok/s=15373.3]
Inference: 69%|██████▉ | 4070/5920 [51:17<08:09, 3.78it/s, tok/s=15373.0]
Inference: 69%|██████▉ | 4071/5920 [51:18<08:08, 3.78it/s, tok/s=15373.5]
Inference: 69%|██████▉ | 4072/5920 [51:18<05:28, 5.63it/s, tok/s=15373.5]
Inference: 69%|██████▉ | 4072/5920 [51:18<05:28, 5.63it/s, tok/s=15374.1]
Inference: 69%|██████▉ | 4073/5920 [51:18<05:36, 5.50it/s, tok/s=15374.1]
Inference: 69%|██████▉ | 4073/5920 [51:18<05:36, 5.50it/s, tok/s=15374.7]
Inference: 69%|██████▉ | 4074/5920 [51:18<05:35, 5.50it/s, tok/s=15375.3]
Inference: 69%|██████▉ | 4075/5920 [51:18<06:40, 4.61it/s, tok/s=15375.3]
Inference: 69%|██████▉ | 4075/5920 [51:18<06:40, 4.61it/s, tok/s=15375.7]
Inference: 69%|██████▉ | 4076/5920 [51:18<06:39, 4.61it/s, tok/s=15384.2]
Inference: 69%|██████▉ | 4077/5920 [51:19<05:28, 5.62it/s, tok/s=15384.2]
Inference: 69%|██████▉ | 4077/5920 [51:19<05:28, 5.62it/s, tok/s=15384.0]
Inference: 69%|██████▉ | 4078/5920 [51:19<06:38, 4.62it/s, tok/s=15384.0]
Inference: 69%|██████▉ | 4078/5920 [51:19<06:38, 4.62it/s, tok/s=15383.0]
Inference: 69%|██████▉ | 4079/5920 [51:19<07:10, 4.28it/s, tok/s=15383.0]
Inference: 69%|██████▉ | 4079/5920 [51:19<07:10, 4.28it/s, tok/s=15385.2]
Inference: 69%|██████▉ | 4080/5920 [51:19<07:10, 4.28it/s, tok/s=15387.7]
Inference: 69%|██████▉ | 4081/5920 [51:20<06:09, 4.97it/s, tok/s=15387.7]
Inference: 69%|██████▉ | 4081/5920 [51:20<06:09, 4.97it/s, tok/s=15386.9]
Inference: 69%|██████▉ | 4082/5920 [51:20<06:14, 4.91it/s, tok/s=15386.9]
Inference: 69%|██████▉ | 4082/5920 [51:20<06:14, 4.91it/s, tok/s=15386.6]
Inference: 69%|██████▉ | 4083/5920 [51:20<06:14, 4.91it/s, tok/s=15387.2]
Inference: 69%|██████▉ | 4084/5920 [51:20<06:13, 4.91it/s, tok/s=15388.1]
Inference: 69%|██████▉ | 4085/5920 [51:20<06:13, 4.91it/s, tok/s=15389.3]
Inference: 69%|██████▉ | 4086/5920 [51:20<03:38, 8.38it/s, tok/s=15389.3]
Inference: 69%|██████▉ | 4086/5920 [51:20<03:38, 8.38it/s, tok/s=15389.8]
Inference: 69%|██████▉ | 4087/5920 [51:20<03:38, 8.38it/s, tok/s=15392.4]
Inference: 69%|██████▉ | 4088/5920 [51:20<03:38, 8.38it/s, tok/s=15394.6]
Inference: 69%|██████▉ | 4089/5920 [51:20<03:38, 8.38it/s, tok/s=15394.7]
Inference: 69%|██████▉ | 4090/5920 [51:20<02:22, 12.80it/s, tok/s=15394.7]
Inference: 69%|██████▉ | 4090/5920 [51:20<02:22, 12.80it/s, tok/s=15395.3]
Inference: 69%|██████▉ | 4091/5920 [51:20<02:22, 12.80it/s, tok/s=15394.9]
Inference: 69%|██████▉ | 4092/5920 [51:20<02:53, 10.56it/s, tok/s=15394.9]
Inference: 69%|██████▉ | 4092/5920 [51:20<02:53, 10.56it/s, tok/s=15395.0]
Inference: 69%|██████▉ | 4093/5920 [51:20<02:53, 10.56it/s, tok/s=15396.5]
Inference: 69%|██████▉ | 4094/5920 [51:20<02:37, 11.57it/s, tok/s=15396.5]
Inference: 69%|██████▉ | 4094/5920 [51:20<02:37, 11.57it/s, tok/s=15398.0]
Inference: 69%|██████▉ | 4095/5920 [51:21<02:37, 11.57it/s, tok/s=15398.6]
Inference: 69%|██████▉ | 4096/5920 [51:21<03:32, 8.58it/s, tok/s=15398.6]
Inference: 69%|██████▉ | 4096/5920 [51:21<03:32, 8.58it/s, tok/s=15398.3]
Inference: 69%|██████▉ | 4097/5920 [51:21<03:32, 8.58it/s, tok/s=15398.5]
Inference: 69%|██████▉ | 4098/5920 [51:21<03:55, 7.75it/s, tok/s=15398.5]
Inference: 69%|██████▉ | 4098/5920 [51:21<03:55, 7.75it/s, tok/s=15400.1]
Inference: 69%|██████▉ | 4099/5920 [51:22<03:54, 7.75it/s, tok/s=15400.3]
Inference: 69%|██████▉ | 4100/5920 [51:22<05:48, 5.22it/s, tok/s=15400.3]
Inference: 69%|██████▉ | 4100/5920 [51:22<05:48, 5.22it/s, tok/s=15407.1]
Inference: 69%|██████▉ | 4101/5920 [51:22<06:04, 4.99it/s, tok/s=15407.1]
Inference: 69%|██████▉ | 4101/5920 [51:22<06:04, 4.99it/s, tok/s=15407.8]
Inference: 69%|██████▉ | 4102/5920 [51:22<06:33, 4.62it/s, tok/s=15407.8]
Inference: 69%|██████▉ | 4102/5920 [51:22<06:33, 4.62it/s, tok/s=15409.7]
Inference: 69%|██████▉ | 4103/5920 [51:23<05:53, 5.14it/s, tok/s=15409.7]
Inference: 69%|██████▉ | 4103/5920 [51:23<05:53, 5.14it/s, tok/s=15410.0]
Inference: 69%|██████▉ | 4104/5920 [51:23<07:11, 4.21it/s, tok/s=15410.0]
Inference: 69%|██████▉ | 4104/5920 [51:23<07:11, 4.21it/s, tok/s=15409.4]
Inference: 69%|██████▉ | 4105/5920 [51:23<07:11, 4.21it/s, tok/s=15411.4]
Inference: 69%|██████▉ | 4106/5920 [51:23<07:29, 4.04it/s, tok/s=15411.4]
Inference: 69%|██████▉ | 4106/5920 [51:23<07:29, 4.04it/s, tok/s=15409.4]
Inference: 69%|██████▉ | 4107/5920 [51:24<06:38, 4.55it/s, tok/s=15409.4]
Inference: 69%|██████▉ | 4107/5920 [51:24<06:38, 4.55it/s, tok/s=15409.2]
Inference: 69%|██████▉ | 4108/5920 [51:24<05:57, 5.06it/s, tok/s=15409.2]
Inference: 69%|██████▉ | 4108/5920 [51:24<05:57, 5.06it/s, tok/s=15409.2]
Inference: 69%|██████▉ | 4109/5920 [51:24<05:54, 5.10it/s, tok/s=15409.2]
Inference: 69%|██████▉ | 4109/5920 [51:24<05:54, 5.10it/s, tok/s=15408.7]
Inference: 69%|██████▉ | 4110/5920 [51:24<07:43, 3.91it/s, tok/s=15408.7]
Inference: 69%|██████▉ | 4110/5920 [51:24<07:43, 3.91it/s, tok/s=15408.1]
Inference: 69%|██████▉ | 4111/5920 [51:24<07:43, 3.91it/s, tok/s=15408.5]
Inference: 69%|██████▉ | 4112/5920 [51:25<07:57, 3.79it/s, tok/s=15408.5]
Inference: 69%|██████▉ | 4112/5920 [51:25<07:57, 3.79it/s, tok/s=15407.8]
Inference: 69%|██████▉ | 4113/5920 [51:25<07:31, 4.00it/s, tok/s=15407.8]
Inference: 69%|██████▉ | 4113/5920 [51:25<07:31, 4.00it/s, tok/s=15408.5]
Inference: 69%|██████▉ | 4114/5920 [51:25<06:55, 4.34it/s, tok/s=15408.5]
Inference: 69%|██████▉ | 4114/5920 [51:25<06:55, 4.34it/s, tok/s=15409.0]
Inference: 70%|██████▉ | 4115/5920 [51:26<11:38, 2.58it/s, tok/s=15409.0]
Inference: 70%|██████▉ | 4115/5920 [51:26<11:38, 2.58it/s, tok/s=15406.0]
Inference: 70%|██████▉ | 4116/5920 [51:26<09:44, 3.09it/s, tok/s=15406.0]
Inference: 70%|██████▉ | 4116/5920 [51:26<09:44, 3.09it/s, tok/s=15413.5]
Inference: 70%|██████▉ | 4117/5920 [51:27<10:08, 2.97it/s, tok/s=15413.5]
Inference: 70%|██████▉ | 4117/5920 [51:27<10:08, 2.97it/s, tok/s=15412.2]
Inference: 70%|██████▉ | 4118/5920 [51:27<08:08, 3.69it/s, tok/s=15412.2]
Inference: 70%|██████▉ | 4118/5920 [51:27<08:08, 3.69it/s, tok/s=15413.2]
Inference: 70%|██████▉ | 4119/5920 [51:27<07:25, 4.04it/s, tok/s=15413.2]
Inference: 70%|██████▉ | 4119/5920 [51:27<07:25, 4.04it/s, tok/s=15413.2]
Inference: 70%|██████▉ | 4120/5920 [51:27<07:25, 4.04it/s, tok/s=15415.2]
Inference: 70%|██████▉ | 4121/5920 [51:27<08:18, 3.61it/s, tok/s=15415.2]
Inference: 70%|██████▉ | 4121/5920 [51:27<08:18, 3.61it/s, tok/s=15413.1]
Inference: 70%|██████▉ | 4122/5920 [51:28<08:58, 3.34it/s, tok/s=15413.1]
Inference: 70%|██████▉ | 4122/5920 [51:28<08:58, 3.34it/s, tok/s=15421.0]
Inference: 70%|██████▉ | 4123/5920 [51:28<08:03, 3.72it/s, tok/s=15421.0]
Inference: 70%|██████▉ | 4123/5920 [51:28<08:03, 3.72it/s, tok/s=15423.9]
Inference: 70%|██████▉ | 4124/5920 [51:28<07:18, 4.09it/s, tok/s=15423.9]
Inference: 70%|██████▉ | 4124/5920 [51:28<07:18, 4.09it/s, tok/s=15425.0]
Inference: 70%|██████▉ | 4125/5920 [51:28<06:58, 4.29it/s, tok/s=15425.0]
Inference: 70%|██████▉ | 4125/5920 [51:28<06:58, 4.29it/s, tok/s=15426.4]
Inference: 70%|██████▉ | 4126/5920 [51:29<06:46, 4.41it/s, tok/s=15426.4]
Inference: 70%|██████▉ | 4126/5920 [51:29<06:46, 4.41it/s, tok/s=15425.9]
Inference: 70%|██████▉ | 4127/5920 [51:29<06:31, 4.58it/s, tok/s=15425.9]
Inference: 70%|██████▉ | 4127/5920 [51:29<06:31, 4.58it/s, tok/s=15425.6]
Inference: 70%|██████▉ | 4128/5920 [51:29<08:00, 3.73it/s, tok/s=15425.6]
Inference: 70%|██████▉ | 4128/5920 [51:29<08:00, 3.73it/s, tok/s=15424.3]
Inference: 70%|██████▉ | 4129/5920 [51:29<08:00, 3.73it/s, tok/s=15424.5]
Inference: 70%|██████▉ | 4130/5920 [51:29<05:53, 5.07it/s, tok/s=15424.5]
Inference: 70%|██████▉ | 4130/5920 [51:29<05:53, 5.07it/s, tok/s=15424.8]
Inference: 70%|██████▉ | 4131/5920 [51:30<07:01, 4.24it/s, tok/s=15424.8]
Inference: 70%|██████▉ | 4131/5920 [51:30<07:01, 4.24it/s, tok/s=15423.5]
Inference: 70%|██████▉ | 4132/5920 [51:30<06:19, 4.71it/s, tok/s=15423.5]
Inference: 70%|██████▉ | 4132/5920 [51:30<06:19, 4.71it/s, tok/s=15427.2]
Inference: 70%|██████▉ | 4133/5920 [51:31<15:02, 1.98it/s, tok/s=15427.2]
Inference: 70%|██████▉ | 4133/5920 [51:31<15:02, 1.98it/s, tok/s=15421.3]
Inference: 70%|██████▉ | 4134/5920 [51:32<13:38, 2.18it/s, tok/s=15421.3]
Inference: 70%|██████▉ | 4134/5920 [51:32<13:38, 2.18it/s, tok/s=15420.5]
Inference: 70%|██████▉ | 4135/5920 [51:32<11:42, 2.54it/s, tok/s=15420.5]
Inference: 70%|██████▉ | 4135/5920 [51:32<11:42, 2.54it/s, tok/s=15420.5]
Inference: 70%|██████▉ | 4136/5920 [51:32<10:19, 2.88it/s, tok/s=15420.5]
Inference: 70%|██████▉ | 4136/5920 [51:32<10:19, 2.88it/s, tok/s=15421.1]
Inference: 70%|██████▉ | 4137/5920 [51:32<09:53, 3.00it/s, tok/s=15421.1]
Inference: 70%|██████▉ | 4137/5920 [51:32<09:53, 3.00it/s, tok/s=15420.1]
Inference: 70%|██████▉ | 4138/5920 [51:33<09:35, 3.09it/s, tok/s=15420.1]
Inference: 70%|██████▉ | 4138/5920 [51:33<09:35, 3.09it/s, tok/s=15420.7]
Inference: 70%|██████▉ | 4139/5920 [51:33<09:48, 3.03it/s, tok/s=15420.7]
Inference: 70%|██████▉ | 4139/5920 [51:33<09:48, 3.03it/s, tok/s=15419.2]
Inference: 70%|██████▉ | 4140/5920 [51:33<08:03, 3.68it/s, tok/s=15419.2]
Inference: 70%|██████▉ | 4140/5920 [51:33<08:03, 3.68it/s, tok/s=15419.5]
Inference: 70%|██████▉ | 4141/5920 [51:33<08:03, 3.68it/s, tok/s=15420.5]
Inference: 70%|██████▉ | 4142/5920 [51:33<06:30, 4.55it/s, tok/s=15420.5]
Inference: 70%|██████▉ | 4142/5920 [51:33<06:30, 4.55it/s, tok/s=15419.3]
Inference: 70%|██████▉ | 4143/5920 [51:33<06:30, 4.55it/s, tok/s=15426.0]
Inference: 70%|███████ | 4144/5920 [51:34<04:51, 6.10it/s, tok/s=15426.0]
Inference: 70%|███████ | 4144/5920 [51:34<04:51, 6.10it/s, tok/s=15426.2]
Inference: 70%|███████ | 4145/5920 [51:34<05:13, 5.66it/s, tok/s=15426.2]
Inference: 70%|███████ | 4145/5920 [51:34<05:13, 5.66it/s, tok/s=15425.7]
Inference: 70%|███████ | 4146/5920 [51:34<05:49, 5.07it/s, tok/s=15425.7]
Inference: 70%|███████ | 4146/5920 [51:34<05:49, 5.07it/s, tok/s=15425.3]
Inference: 70%|███████ | 4147/5920 [51:34<05:49, 5.07it/s, tok/s=15425.2]
Inference: 70%|███████ | 4148/5920 [51:34<04:03, 7.27it/s, tok/s=15425.2]
Inference: 70%|███████ | 4148/5920 [51:34<04:03, 7.27it/s, tok/s=15425.9]
Inference: 70%|███████ | 4149/5920 [51:35<07:54, 3.73it/s, tok/s=15425.9]
Inference: 70%|███████ | 4149/5920 [51:35<07:54, 3.73it/s, tok/s=15423.0]
Inference: 70%|███████ | 4150/5920 [51:35<07:54, 3.73it/s, tok/s=15423.5]
Inference: 70%|███████ | 4151/5920 [51:36<08:53, 3.31it/s, tok/s=15423.5]
Inference: 70%|███████ | 4151/5920 [51:36<08:53, 3.31it/s, tok/s=15421.8]
Inference: 70%|███████ | 4152/5920 [51:36<09:01, 3.27it/s, tok/s=15421.8]
Inference: 70%|███████ | 4152/5920 [51:36<09:01, 3.27it/s, tok/s=15424.0]
Inference: 70%|███████ | 4153/5920 [51:36<09:00, 3.27it/s, tok/s=15425.2]
Inference: 70%|███████ | 4154/5920 [51:36<07:00, 4.20it/s, tok/s=15425.2]
Inference: 70%|███████ | 4154/5920 [51:36<07:00, 4.20it/s, tok/s=15424.6]
Inference: 70%|███████ | 4155/5920 [51:36<06:59, 4.20it/s, tok/s=15425.1]
Inference: 70%|███████ | 4156/5920 [51:37<06:57, 4.23it/s, tok/s=15425.1]
Inference: 70%|███████ | 4156/5920 [51:37<06:57, 4.23it/s, tok/s=15424.1]
Inference: 70%|███████ | 4157/5920 [51:37<06:59, 4.20it/s, tok/s=15424.1]
Inference: 70%|███████ | 4157/5920 [51:37<06:59, 4.20it/s, tok/s=15425.7]
Inference: 70%|███████ | 4158/5920 [51:38<09:40, 3.04it/s, tok/s=15425.7]
Inference: 70%|███████ | 4158/5920 [51:38<09:40, 3.04it/s, tok/s=15423.1]
Inference: 70%|███████ | 4159/5920 [51:38<13:12, 2.22it/s, tok/s=15423.1]
Inference: 70%|███████ | 4159/5920 [51:38<13:12, 2.22it/s, tok/s=15420.4]
Inference: 70%|███████ | 4160/5920 [51:38<13:11, 2.22it/s, tok/s=15420.4]
Inference: 70%|███████ | 4161/5920 [51:39<08:54, 3.29it/s, tok/s=15420.4]
Inference: 70%|███████ | 4161/5920 [51:39<08:54, 3.29it/s, tok/s=15420.4]
Inference: 70%|███████ | 4162/5920 [51:39<08:54, 3.29it/s, tok/s=15422.1]
Inference: 70%|███████ | 4163/5920 [51:39<09:29, 3.09it/s, tok/s=15422.1]
Inference: 70%|███████ | 4163/5920 [51:39<09:29, 3.09it/s, tok/s=15419.9]
Inference: 70%|███████ | 4164/5920 [51:39<09:29, 3.09it/s, tok/s=15426.9]
Inference: 70%|███████ | 4165/5920 [51:40<10:11, 2.87it/s, tok/s=15426.9]
Inference: 70%|███████ | 4165/5920 [51:40<10:11, 2.87it/s, tok/s=15423.7]
Inference: 70%|███████ | 4166/5920 [51:40<08:57, 3.27it/s, tok/s=15423.7]
Inference: 70%|███████ | 4166/5920 [51:40<08:57, 3.27it/s, tok/s=15423.7]
Inference: 70%|███████ | 4167/5920 [51:40<08:42, 3.35it/s, tok/s=15423.7]
Inference: 70%|███████ | 4167/5920 [51:40<08:42, 3.35it/s, tok/s=15422.8]
Inference: 70%|███████ | 4168/5920 [51:41<11:24, 2.56it/s, tok/s=15422.8]
Inference: 70%|███████ | 4168/5920 [51:41<11:24, 2.56it/s, tok/s=15419.8]
Inference: 70%|███████ | 4169/5920 [51:42<11:30, 2.53it/s, tok/s=15419.8]
Inference: 70%|███████ | 4169/5920 [51:42<11:30, 2.53it/s, tok/s=15418.5]
Inference: 70%|███████ | 4170/5920 [51:42<11:30, 2.53it/s, tok/s=15418.7]
Inference: 70%|███████ | 4171/5920 [51:42<07:23, 3.94it/s, tok/s=15418.7]
Inference: 70%|███████ | 4171/5920 [51:42<07:23, 3.94it/s, tok/s=15419.6]
Inference: 70%|███████ | 4172/5920 [51:42<08:26, 3.45it/s, tok/s=15419.6]
Inference: 70%|███████ | 4172/5920 [51:42<08:26, 3.45it/s, tok/s=15427.2]
Inference: 70%|███████ | 4173/5920 [51:42<09:12, 3.16it/s, tok/s=15427.2]
Inference: 70%|███████ | 4173/5920 [51:42<09:12, 3.16it/s, tok/s=15426.5]
Inference: 71%|███████ | 4174/5920 [51:43<08:35, 3.38it/s, tok/s=15426.5]
Inference: 71%|███████ | 4174/5920 [51:43<08:35, 3.38it/s, tok/s=15426.6]
Inference: 71%|███████ | 4175/5920 [51:43<09:59, 2.91it/s, tok/s=15426.6]
Inference: 71%|███████ | 4175/5920 [51:43<09:59, 2.91it/s, tok/s=15426.2]
Inference: 71%|███████ | 4176/5920 [51:44<09:50, 2.95it/s, tok/s=15426.2]
Inference: 71%|███████ | 4176/5920 [51:44<09:50, 2.95it/s, tok/s=15425.3]
Inference: 71%|███████ | 4177/5920 [51:44<09:04, 3.20it/s, tok/s=15425.3]
Inference: 71%|███████ | 4177/5920 [51:44<09:04, 3.20it/s, tok/s=15425.4]
Inference: 71%|███████ | 4178/5920 [51:44<09:44, 2.98it/s, tok/s=15425.4]
Inference: 71%|███████ | 4178/5920 [51:44<09:44, 2.98it/s, tok/s=15424.0]
Inference: 71%|███████ | 4179/5920 [51:44<07:51, 3.70it/s, tok/s=15424.0]
Inference: 71%|███████ | 4179/5920 [51:44<07:51, 3.70it/s, tok/s=15425.2]
Inference: 71%|███████ | 4180/5920 [51:44<07:50, 3.70it/s, tok/s=15425.1]
Inference: 71%|███████ | 4181/5920 [51:45<06:01, 4.81it/s, tok/s=15425.1]
Inference: 71%|███████ | 4181/5920 [51:45<06:01, 4.81it/s, tok/s=15424.8]
Inference: 71%|███████ | 4182/5920 [51:45<05:59, 4.83it/s, tok/s=15424.8]
Inference: 71%|███████ | 4182/5920 [51:45<05:59, 4.83it/s, tok/s=15424.6]
Inference: 71%|███████ | 4183/5920 [51:45<06:18, 4.59it/s, tok/s=15424.6]
Inference: 71%|███████ | 4183/5920 [51:45<06:18, 4.59it/s, tok/s=15427.8]
Inference: 71%|███████ | 4184/5920 [51:45<06:18, 4.59it/s, tok/s=15428.2]
Inference: 71%|███████ | 4185/5920 [51:45<04:14, 6.83it/s, tok/s=15428.2]
Inference: 71%|███████ | 4185/5920 [51:45<04:14, 6.83it/s, tok/s=15430.2]
Inference: 71%|███████ | 4186/5920 [51:45<04:13, 6.85it/s, tok/s=15430.2]
Inference: 71%|███████ | 4186/5920 [51:45<04:13, 6.85it/s, tok/s=15430.2]
Inference: 71%|███████ | 4187/5920 [51:46<05:43, 5.04it/s, tok/s=15430.2]
Inference: 71%|███████ | 4187/5920 [51:46<05:43, 5.04it/s, tok/s=15431.5]
Inference: 71%|███████ | 4188/5920 [51:46<05:13, 5.53it/s, tok/s=15431.5]
Inference: 71%|███████ | 4188/5920 [51:46<05:13, 5.53it/s, tok/s=15432.1]
Inference: 71%|███████ | 4189/5920 [51:46<05:12, 5.53it/s, tok/s=15438.1]
Inference: 71%|███████ | 4190/5920 [51:46<04:52, 5.92it/s, tok/s=15438.1]
Inference: 71%|███████ | 4190/5920 [51:46<04:52, 5.92it/s, tok/s=15437.6]
Inference: 71%|███████ | 4191/5920 [51:46<04:52, 5.92it/s, tok/s=15446.7]
Inference: 71%|███████ | 4192/5920 [51:46<04:52, 5.92it/s, tok/s=15447.0]
Inference: 71%|███████ | 4193/5920 [51:47<04:48, 5.98it/s, tok/s=15447.0]
Inference: 71%|███████ | 4193/5920 [51:47<04:48, 5.98it/s, tok/s=15449.6]
Inference: 71%|███████ | 4194/5920 [51:47<04:48, 5.98it/s, tok/s=15450.0]
Inference: 71%|███████ | 4195/5920 [51:47<04:22, 6.58it/s, tok/s=15450.0]
Inference: 71%|███████ | 4195/5920 [51:47<04:22, 6.58it/s, tok/s=15449.5]
Inference: 71%|███████ | 4196/5920 [51:47<04:22, 6.57it/s, tok/s=15449.5]
Inference: 71%|███████ | 4196/5920 [51:47<04:22, 6.57it/s, tok/s=15449.1]
Inference: 71%|███████ | 4197/5920 [51:47<05:04, 5.65it/s, tok/s=15449.1]
Inference: 71%|███████ | 4197/5920 [51:47<05:04, 5.65it/s, tok/s=15448.1]
Inference: 71%|███████ | 4198/5920 [51:48<07:39, 3.75it/s, tok/s=15448.1]
Inference: 71%|███████ | 4198/5920 [51:48<07:39, 3.75it/s, tok/s=15445.6]
Inference: 71%|███████ | 4199/5920 [51:48<07:38, 3.75it/s, tok/s=15445.6]
Inference: 71%|███████ | 4200/5920 [51:48<06:14, 4.60it/s, tok/s=15445.6]
Inference: 71%|███████ | 4200/5920 [51:48<06:14, 4.60it/s, tok/s=15445.3]
Inference: 71%|███████ | 4201/5920 [51:48<05:35, 5.12it/s, tok/s=15445.3]
Inference: 71%|███████ | 4201/5920 [51:48<05:35, 5.12it/s, tok/s=15445.2]
Inference: 71%|███████ | 4202/5920 [51:49<06:57, 4.12it/s, tok/s=15445.2]
Inference: 71%|███████ | 4202/5920 [51:49<06:57, 4.12it/s, tok/s=15446.0]
Inference: 71%|███████ | 4203/5920 [51:49<05:56, 4.81it/s, tok/s=15446.0]
Inference: 71%|███████ | 4203/5920 [51:49<05:56, 4.81it/s, tok/s=15445.9]
Inference: 71%|███████ | 4204/5920 [51:49<05:29, 5.21it/s, tok/s=15445.9]
Inference: 71%|███████ | 4204/5920 [51:49<05:29, 5.21it/s, tok/s=15445.7]
Inference: 71%|███████ | 4205/5920 [51:49<05:49, 4.91it/s, tok/s=15445.7]
Inference: 71%|███████ | 4205/5920 [51:49<05:49, 4.91it/s, tok/s=15445.0]
Inference: 71%|███████ | 4206/5920 [51:50<11:22, 2.51it/s, tok/s=15445.0]
Inference: 71%|███████ | 4206/5920 [51:50<11:22, 2.51it/s, tok/s=15441.0]
Inference: 71%|███████ | 4207/5920 [51:50<09:22, 3.05it/s, tok/s=15441.0]
Inference: 71%|███████ | 4207/5920 [51:50<09:22, 3.05it/s, tok/s=15441.7]
Inference: 71%|███████ | 4208/5920 [51:50<07:45, 3.68it/s, tok/s=15441.7]
Inference: 71%|███████ | 4208/5920 [51:50<07:45, 3.68it/s, tok/s=15441.4]
Inference: 71%|███████ | 4209/5920 [51:51<07:49, 3.65it/s, tok/s=15441.4]
Inference: 71%|███████ | 4209/5920 [51:51<07:49, 3.65it/s, tok/s=15449.7]
Inference: 71%|███████ | 4210/5920 [51:51<06:52, 4.14it/s, tok/s=15449.7]
Inference: 71%|███████ | 4210/5920 [51:51<06:52, 4.14it/s, tok/s=15450.8]
Inference: 71%|███████ | 4211/5920 [51:51<05:58, 4.76it/s, tok/s=15450.8]
Inference: 71%|███████ | 4211/5920 [51:51<05:58, 4.76it/s, tok/s=15451.0]
Inference: 71%|███████ | 4212/5920 [51:51<08:06, 3.51it/s, tok/s=15451.0]
Inference: 71%|███████ | 4212/5920 [51:51<08:06, 3.51it/s, tok/s=15449.2]
Inference: 71%|███████ | 4213/5920 [51:51<08:06, 3.51it/s, tok/s=15449.4]
Inference: 71%|███████ | 4214/5920 [51:51<08:05, 3.51it/s, tok/s=15451.5]
Inference: 71%|███████ | 4215/5920 [51:51<08:05, 3.51it/s, tok/s=15452.1]
Inference: 71%|███████ | 4216/5920 [51:51<03:58, 7.15it/s, tok/s=15452.1]
Inference: 71%|███████ | 4216/5920 [51:51<03:58, 7.15it/s, tok/s=15453.2]
Inference: 71%|███████ | 4217/5920 [51:52<04:40, 6.07it/s, tok/s=15453.2]
Inference: 71%|███████ | 4217/5920 [51:52<04:40, 6.07it/s, tok/s=15452.4]
Inference: 71%|███████▏ | 4218/5920 [51:52<06:37, 4.29it/s, tok/s=15452.4]
Inference: 71%|███████▏ | 4218/5920 [51:52<06:37, 4.29it/s, tok/s=15450.5]
Inference: 71%|███████▏ | 4219/5920 [51:52<06:36, 4.29it/s, tok/s=15451.0]
Inference: 71%|███████▏ | 4220/5920 [51:53<05:44, 4.94it/s, tok/s=15451.0]
Inference: 71%|███████▏ | 4220/5920 [51:53<05:44, 4.94it/s, tok/s=15450.1]
Inference: 71%|███████▏ | 4221/5920 [51:53<06:28, 4.37it/s, tok/s=15450.1]
Inference: 71%|███████▏ | 4221/5920 [51:53<06:28, 4.37it/s, tok/s=15448.8]
Inference: 71%|███████▏ | 4222/5920 [51:53<06:28, 4.37it/s, tok/s=15450.0]
Inference: 71%|███████▏ | 4223/5920 [51:53<05:31, 5.12it/s, tok/s=15450.0]
Inference: 71%|███████▏ | 4223/5920 [51:53<05:31, 5.12it/s, tok/s=15450.3]
Inference: 71%|███████▏ | 4224/5920 [51:54<09:57, 2.84it/s, tok/s=15450.3]
Inference: 71%|███████▏ | 4224/5920 [51:54<09:57, 2.84it/s, tok/s=15450.1]
Inference: 71%|███████▏ | 4225/5920 [51:54<08:29, 3.33it/s, tok/s=15450.1]
Inference: 71%|███████▏ | 4225/5920 [51:54<08:29, 3.33it/s, tok/s=15451.2]
Inference: 71%|███████▏ | 4226/5920 [51:54<08:28, 3.33it/s, tok/s=15453.0]
Inference: 71%|███████▏ | 4227/5920 [51:54<05:56, 4.75it/s, tok/s=15453.0]
Inference: 71%|███████▏ | 4227/5920 [51:54<05:56, 4.75it/s, tok/s=15453.3]
Inference: 71%|███████▏ | 4228/5920 [51:54<05:56, 4.75it/s, tok/s=15453.8]
Inference: 71%|███████▏ | 4229/5920 [51:55<04:49, 5.84it/s, tok/s=15453.8]
Inference: 71%|███████▏ | 4229/5920 [51:55<04:49, 5.84it/s, tok/s=15453.7]
Inference: 71%|███████▏ | 4230/5920 [51:56<10:55, 2.58it/s, tok/s=15453.7]
Inference: 71%|███████▏ | 4230/5920 [51:56<10:55, 2.58it/s, tok/s=15448.0]
Inference: 71%|███████▏ | 4231/5920 [51:57<13:06, 2.15it/s, tok/s=15448.0]
Inference: 71%|███████▏ | 4231/5920 [51:57<13:06, 2.15it/s, tok/s=15450.3]
Inference: 71%|███████▏ | 4232/5920 [51:57<11:50, 2.37it/s, tok/s=15450.3]
Inference: 71%|███████▏ | 4232/5920 [51:57<11:50, 2.37it/s, tok/s=15449.5]
Inference: 72%|███████▏ | 4233/5920 [51:57<11:24, 2.47it/s, tok/s=15449.5]
Inference: 72%|███████▏ | 4233/5920 [51:57<11:24, 2.47it/s, tok/s=15449.1]
Inference: 72%|███████▏ | 4234/5920 [51:57<09:35, 2.93it/s, tok/s=15449.1]
Inference: 72%|███████▏ | 4234/5920 [51:57<09:35, 2.93it/s, tok/s=15448.6]
Inference: 72%|███████▏ | 4235/5920 [51:58<09:42, 2.89it/s, tok/s=15448.6]
Inference: 72%|███████▏ | 4235/5920 [51:58<09:42, 2.89it/s, tok/s=15447.7]
Inference: 72%|███████▏ | 4236/5920 [51:58<09:05, 3.09it/s, tok/s=15447.7]
Inference: 72%|███████▏ | 4236/5920 [51:58<09:05, 3.09it/s, tok/s=15447.2]
Inference: 72%|███████▏ | 4237/5920 [51:58<07:41, 3.65it/s, tok/s=15447.2]
Inference: 72%|███████▏ | 4237/5920 [51:58<07:41, 3.65it/s, tok/s=15448.8]
Inference: 72%|███████▏ | 4238/5920 [51:58<07:41, 3.65it/s, tok/s=15449.0]
Inference: 72%|███████▏ | 4239/5920 [51:58<05:27, 5.13it/s, tok/s=15449.0]
Inference: 72%|███████▏ | 4239/5920 [51:58<05:27, 5.13it/s, tok/s=15449.2]
Inference: 72%|███████▏ | 4240/5920 [51:58<05:27, 5.13it/s, tok/s=15449.7]
Inference: 72%|███████▏ | 4241/5920 [51:59<07:13, 3.87it/s, tok/s=15449.7]
Inference: 72%|███████▏ | 4241/5920 [51:59<07:13, 3.87it/s, tok/s=15449.6]
Inference: 72%|███████▏ | 4242/5920 [51:59<07:13, 3.87it/s, tok/s=15450.3]
Inference: 72%|███████▏ | 4243/5920 [51:59<05:34, 5.01it/s, tok/s=15450.3]
Inference: 72%|███████▏ | 4243/5920 [51:59<05:34, 5.01it/s, tok/s=15453.9]
Inference: 72%|███████▏ | 4244/5920 [52:00<06:12, 4.50it/s, tok/s=15453.9]
Inference: 72%|███████▏ | 4244/5920 [52:00<06:12, 4.50it/s, tok/s=15453.6]
Inference: 72%|███████▏ | 4245/5920 [52:00<07:12, 3.87it/s, tok/s=15453.6]
Inference: 72%|███████▏ | 4245/5920 [52:00<07:12, 3.87it/s, tok/s=15452.3]
Inference: 72%|███████▏ | 4246/5920 [52:00<06:51, 4.06it/s, tok/s=15452.3]
Inference: 72%|███████▏ | 4246/5920 [52:00<06:51, 4.06it/s, tok/s=15452.1]
Inference: 72%|███████▏ | 4247/5920 [52:00<06:06, 4.56it/s, tok/s=15452.1]
Inference: 72%|███████▏ | 4247/5920 [52:00<06:06, 4.56it/s, tok/s=15451.9]
Inference: 72%|███████▏ | 4248/5920 [52:00<06:06, 4.56it/s, tok/s=15452.3]
Inference: 72%|███████▏ | 4249/5920 [52:00<04:17, 6.50it/s, tok/s=15452.3]
Inference: 72%|███████▏ | 4249/5920 [52:00<04:17, 6.50it/s, tok/s=15452.5]
Inference: 72%|███████▏ | 4250/5920 [52:01<04:54, 5.66it/s, tok/s=15452.5]
Inference: 72%|███████▏ | 4250/5920 [52:01<04:54, 5.66it/s, tok/s=15451.8]
Inference: 72%|███████▏ | 4251/5920 [52:01<04:54, 5.66it/s, tok/s=15452.1]
Inference: 72%|███████▏ | 4252/5920 [52:01<04:54, 5.66it/s, tok/s=15452.5]
Inference: 72%|███████▏ | 4253/5920 [52:01<03:07, 8.88it/s, tok/s=15452.5]
Inference: 72%|███████▏ | 4253/5920 [52:01<03:07, 8.88it/s, tok/s=15452.7]
Inference: 72%|███████▏ | 4254/5920 [52:01<03:07, 8.88it/s, tok/s=15450.8]
Inference: 72%|███████▏ | 4255/5920 [52:02<06:50, 4.06it/s, tok/s=15450.8]
Inference: 72%|███████▏ | 4255/5920 [52:02<06:50, 4.06it/s, tok/s=15448.2]
Inference: 72%|███████▏ | 4256/5920 [52:02<06:50, 4.06it/s, tok/s=15448.7]
Inference: 72%|███████▏ | 4257/5920 [52:03<08:51, 3.13it/s, tok/s=15448.7]
Inference: 72%|███████▏ | 4257/5920 [52:03<08:51, 3.13it/s, tok/s=15444.8]
Inference: 72%|███████▏ | 4258/5920 [52:04<11:10, 2.48it/s, tok/s=15444.8]
Inference: 72%|███████▏ | 4258/5920 [52:04<11:10, 2.48it/s, tok/s=15442.4]
Inference: 72%|███████▏ | 4259/5920 [52:04<11:10, 2.48it/s, tok/s=15442.5]
Inference: 72%|███████▏ | 4260/5920 [52:04<08:49, 3.14it/s, tok/s=15442.5]
Inference: 72%|███████▏ | 4260/5920 [52:04<08:49, 3.14it/s, tok/s=15442.8]
Inference: 72%|███████▏ | 4261/5920 [52:04<07:49, 3.54it/s, tok/s=15442.8]
Inference: 72%|███████▏ | 4261/5920 [52:04<07:49, 3.54it/s, tok/s=15443.5]
Inference: 72%|███████▏ | 4262/5920 [52:04<07:00, 3.95it/s, tok/s=15443.5]
Inference: 72%|███████▏ | 4262/5920 [52:04<07:00, 3.95it/s, tok/s=15446.3]
Inference: 72%|███████▏ | 4263/5920 [52:04<06:59, 3.95it/s, tok/s=15446.6]
Inference: 72%|███████▏ | 4264/5920 [52:04<06:59, 3.95it/s, tok/s=15446.9]
Inference: 72%|███████▏ | 4265/5920 [52:04<04:39, 5.91it/s, tok/s=15446.9]
Inference: 72%|███████▏ | 4265/5920 [52:04<04:39, 5.91it/s, tok/s=15450.2]
Inference: 72%|███████▏ | 4266/5920 [52:04<04:39, 5.91it/s, tok/s=15450.6]
Inference: 72%|███████▏ | 4267/5920 [52:05<04:39, 5.91it/s, tok/s=15450.8]
Inference: 72%|███████▏ | 4268/5920 [52:05<04:21, 6.31it/s, tok/s=15450.8]
Inference: 72%|███████▏ | 4268/5920 [52:05<04:21, 6.31it/s, tok/s=15449.6]
Inference: 72%|███████▏ | 4269/5920 [52:05<04:06, 6.69it/s, tok/s=15449.6]
Inference: 72%|███████▏ | 4269/5920 [52:05<04:06, 6.69it/s, tok/s=15450.0]
Inference: 72%|███████▏ | 4270/5920 [52:05<04:06, 6.69it/s, tok/s=15450.6]
Inference: 72%|███████▏ | 4271/5920 [52:05<03:36, 7.61it/s, tok/s=15450.6]
Inference: 72%|███████▏ | 4271/5920 [52:05<03:36, 7.61it/s, tok/s=15450.7]
Inference: 72%|███████▏ | 4272/5920 [52:06<05:31, 4.98it/s, tok/s=15450.7]
Inference: 72%|███████▏ | 4272/5920 [52:06<05:31, 4.98it/s, tok/s=15449.2]
Inference: 72%|███████▏ | 4273/5920 [52:06<05:30, 4.98it/s, tok/s=15450.0]
Inference: 72%|███████▏ | 4274/5920 [52:06<04:19, 6.34it/s, tok/s=15450.0]
Inference: 72%|███████▏ | 4274/5920 [52:06<04:19, 6.34it/s, tok/s=15451.3]
Inference: 72%|███████▏ | 4275/5920 [52:06<06:38, 4.12it/s, tok/s=15451.3]
Inference: 72%|███████▏ | 4275/5920 [52:06<06:38, 4.12it/s, tok/s=15449.5]
Inference: 72%|███████▏ | 4276/5920 [52:06<06:38, 4.12it/s, tok/s=15450.2]
Inference: 72%|███████▏ | 4277/5920 [52:07<05:51, 4.67it/s, tok/s=15450.2]
Inference: 72%|███████▏ | 4277/5920 [52:07<05:51, 4.67it/s, tok/s=15449.8]
Inference: 72%|███████▏ | 4278/5920 [52:07<06:18, 4.34it/s, tok/s=15449.8]
Inference: 72%|███████▏ | 4278/5920 [52:07<06:18, 4.34it/s, tok/s=15449.0]
Inference: 72%|███████▏ | 4279/5920 [52:07<06:17, 4.35it/s, tok/s=15449.0]
Inference: 72%|███████▏ | 4279/5920 [52:07<06:17, 4.35it/s, tok/s=15449.3]
Inference: 72%|███████▏ | 4280/5920 [52:08<08:30, 3.22it/s, tok/s=15449.3]
Inference: 72%|███████▏ | 4280/5920 [52:08<08:30, 3.22it/s, tok/s=15448.5]
Inference: 72%|███████▏ | 4281/5920 [52:08<07:42, 3.54it/s, tok/s=15448.5]
Inference: 72%|███████▏ | 4281/5920 [52:08<07:42, 3.54it/s, tok/s=15448.7]
Inference: 72%|███████▏ | 4282/5920 [52:08<08:16, 3.30it/s, tok/s=15448.7]
Inference: 72%|███████▏ | 4282/5920 [52:08<08:16, 3.30it/s, tok/s=15447.7]
Inference: 72%|███████▏ | 4283/5920 [52:09<08:30, 3.21it/s, tok/s=15447.7]
Inference: 72%|███████▏ | 4283/5920 [52:09<08:30, 3.21it/s, tok/s=15446.5]
Inference: 72%|███████▏ | 4284/5920 [52:09<08:05, 3.37it/s, tok/s=15446.5]
Inference: 72%|███████▏ | 4284/5920 [52:09<08:05, 3.37it/s, tok/s=15454.9]
Inference: 72%|███████▏ | 4285/5920 [52:09<08:20, 3.27it/s, tok/s=15454.9]
Inference: 72%|███████▏ | 4285/5920 [52:09<08:20, 3.27it/s, tok/s=15462.2]
Inference: 72%|███████▏ | 4286/5920 [52:09<08:20, 3.27it/s, tok/s=15469.5]
Inference: 72%|███████▏ | 4287/5920 [52:09<05:30, 4.94it/s, tok/s=15469.5]
Inference: 72%|███████▏ | 4287/5920 [52:09<05:30, 4.94it/s, tok/s=15469.5]
Inference: 72%|███████▏ | 4288/5920 [52:10<05:19, 5.10it/s, tok/s=15469.5]
Inference: 72%|███████▏ | 4288/5920 [52:10<05:19, 5.10it/s, tok/s=15475.0]
Inference: 72%|███████▏ | 4289/5920 [52:10<04:42, 5.77it/s, tok/s=15475.0]
Inference: 72%|███████▏ | 4289/5920 [52:10<04:42, 5.77it/s, tok/s=15475.9]
Inference: 72%|███████▏ | 4290/5920 [52:10<04:58, 5.46it/s, tok/s=15475.9]
Inference: 72%|███████▏ | 4290/5920 [52:10<04:58, 5.46it/s, tok/s=15475.6]
Inference: 72%|███████▏ | 4291/5920 [52:10<05:45, 4.72it/s, tok/s=15475.6]
Inference: 72%|███████▏ | 4291/5920 [52:10<05:45, 4.72it/s, tok/s=15481.6]
Inference: 72%|███████▎ | 4292/5920 [52:10<05:23, 5.04it/s, tok/s=15481.6]
Inference: 72%|███████▎ | 4292/5920 [52:10<05:23, 5.04it/s, tok/s=15481.5]
Inference: 73%|███████▎ | 4293/5920 [52:10<05:22, 5.04it/s, tok/s=15482.0]
Inference: 73%|███████▎ | 4294/5920 [52:11<04:47, 5.66it/s, tok/s=15482.0]
Inference: 73%|███████▎ | 4294/5920 [52:11<04:47, 5.66it/s, tok/s=15482.1]
Inference: 73%|███████▎ | 4295/5920 [52:11<04:47, 5.66it/s, tok/s=15483.8]
Inference: 73%|███████▎ | 4296/5920 [52:11<03:27, 7.81it/s, tok/s=15483.8]
Inference: 73%|███████▎ | 4296/5920 [52:11<03:27, 7.81it/s, tok/s=15486.1]
Inference: 73%|███████▎ | 4297/5920 [52:11<03:26, 7.84it/s, tok/s=15486.1]
Inference: 73%|███████▎ | 4297/5920 [52:11<03:26, 7.84it/s, tok/s=15493.4]
Inference: 73%|███████▎ | 4298/5920 [52:11<03:26, 7.84it/s, tok/s=15497.4]
Inference: 73%|███████▎ | 4299/5920 [52:11<03:26, 7.84it/s, tok/s=15498.0]
Inference: 73%|███████▎ | 4300/5920 [52:11<02:31, 10.72it/s, tok/s=15498.0]
Inference: 73%|███████▎ | 4300/5920 [52:11<02:31, 10.72it/s, tok/s=15497.8]
Inference: 73%|███████▎ | 4301/5920 [52:11<02:31, 10.72it/s, tok/s=15501.0]
Inference: 73%|███████▎ | 4302/5920 [52:11<03:16, 8.25it/s, tok/s=15501.0]
Inference: 73%|███████▎ | 4302/5920 [52:11<03:16, 8.25it/s, tok/s=15501.8]
Inference: 73%|███████▎ | 4303/5920 [52:12<04:01, 6.69it/s, tok/s=15501.8]
Inference: 73%|███████▎ | 4303/5920 [52:12<04:01, 6.69it/s, tok/s=15504.2]
Inference: 73%|███████▎ | 4304/5920 [52:12<04:30, 5.98it/s, tok/s=15504.2]
Inference: 73%|███████▎ | 4304/5920 [52:12<04:30, 5.98it/s, tok/s=15505.4]
Inference: 73%|███████▎ | 4305/5920 [52:12<05:04, 5.30it/s, tok/s=15505.4]
Inference: 73%|███████▎ | 4305/5920 [52:12<05:04, 5.30it/s, tok/s=15507.0]
Inference: 73%|███████▎ | 4306/5920 [52:12<04:44, 5.68it/s, tok/s=15507.0]
Inference: 73%|███████▎ | 4306/5920 [52:12<04:44, 5.68it/s, tok/s=15507.0]
Inference: 73%|███████▎ | 4307/5920 [52:12<04:43, 5.68it/s, tok/s=15507.5]
Inference: 73%|███████▎ | 4308/5920 [52:12<04:43, 5.68it/s, tok/s=15507.8]
Inference: 73%|███████▎ | 4309/5920 [52:12<04:43, 5.68it/s, tok/s=15509.0]
Inference: 73%|███████▎ | 4310/5920 [52:13<02:36, 10.30it/s, tok/s=15509.0]
Inference: 73%|███████▎ | 4310/5920 [52:13<02:36, 10.30it/s, tok/s=15509.1]
Inference: 73%|███████▎ | 4311/5920 [52:13<02:36, 10.30it/s, tok/s=15509.5]
Inference: 73%|███████▎ | 4312/5920 [52:13<02:19, 11.54it/s, tok/s=15509.5]
Inference: 73%|███████▎ | 4312/5920 [52:13<02:19, 11.54it/s, tok/s=15512.2]
Inference: 73%|███████▎ | 4313/5920 [52:13<02:19, 11.54it/s, tok/s=15513.0]
Inference: 73%|███████▎ | 4314/5920 [52:13<03:00, 8.89it/s, tok/s=15513.0]
Inference: 73%|███████▎ | 4314/5920 [52:13<03:00, 8.89it/s, tok/s=15513.4]
Inference: 73%|███████▎ | 4315/5920 [52:13<03:00, 8.89it/s, tok/s=15513.4]
Inference: 73%|███████▎ | 4316/5920 [52:13<03:37, 7.38it/s, tok/s=15513.4]
Inference: 73%|███████▎ | 4316/5920 [52:13<03:37, 7.38it/s, tok/s=15513.0]
Inference: 73%|███████▎ | 4317/5920 [52:13<03:37, 7.38it/s, tok/s=15513.5]
Inference: 73%|███████▎ | 4318/5920 [52:13<03:37, 7.38it/s, tok/s=15518.2]
Inference: 73%|███████▎ | 4319/5920 [52:14<02:58, 8.99it/s, tok/s=15518.2]
Inference: 73%|███████▎ | 4319/5920 [52:14<02:58, 8.99it/s, tok/s=15527.3]
Inference: 73%|███████▎ | 4320/5920 [52:14<02:57, 8.99it/s, tok/s=15526.2]
Inference: 73%|███████▎ | 4321/5920 [52:15<07:37, 3.50it/s, tok/s=15526.2]
Inference: 73%|███████▎ | 4321/5920 [52:15<07:37, 3.50it/s, tok/s=15530.3]
Inference: 73%|███████▎ | 4322/5920 [52:15<07:37, 3.50it/s, tok/s=15530.8]
Inference: 73%|███████▎ | 4323/5920 [52:15<06:23, 4.16it/s, tok/s=15530.8]
Inference: 73%|███████▎ | 4323/5920 [52:15<06:23, 4.16it/s, tok/s=15530.7]
Inference: 73%|███████▎ | 4324/5920 [52:16<07:11, 3.70it/s, tok/s=15530.7]
Inference: 73%|███████▎ | 4324/5920 [52:16<07:11, 3.70it/s, tok/s=15529.2]
Inference: 73%|███████▎ | 4325/5920 [52:16<06:55, 3.83it/s, tok/s=15529.2]
Inference: 73%|███████▎ | 4325/5920 [52:16<06:55, 3.83it/s, tok/s=15529.0]
Inference: 73%|███████▎ | 4326/5920 [52:16<07:35, 3.50it/s, tok/s=15529.0]
Inference: 73%|███████▎ | 4326/5920 [52:16<07:35, 3.50it/s, tok/s=15527.6]
Inference: 73%|███████▎ | 4327/5920 [52:16<07:35, 3.50it/s, tok/s=15527.6]
Inference: 73%|███████▎ | 4328/5920 [52:17<05:56, 4.47it/s, tok/s=15527.6]
Inference: 73%|███████▎ | 4328/5920 [52:17<05:56, 4.47it/s, tok/s=15528.2]
Inference: 73%|███████▎ | 4329/5920 [52:17<05:56, 4.47it/s, tok/s=15528.4]
Inference: 73%|███████▎ | 4330/5920 [52:17<05:54, 4.48it/s, tok/s=15528.4]
Inference: 73%|███████▎ | 4330/5920 [52:17<05:54, 4.48it/s, tok/s=15528.6]
Inference: 73%|███████▎ | 4331/5920 [52:17<05:54, 4.48it/s, tok/s=15529.4]
Inference: 73%|███████▎ | 4332/5920 [52:17<05:54, 4.48it/s, tok/s=15536.4]
Inference: 73%|███████▎ | 4333/5920 [52:17<04:50, 5.47it/s, tok/s=15536.4]
Inference: 73%|███████▎ | 4333/5920 [52:17<04:50, 5.47it/s, tok/s=15537.1]
Inference: 73%|███████▎ | 4334/5920 [52:18<04:50, 5.47it/s, tok/s=15537.7]
Inference: 73%|███████▎ | 4335/5920 [52:18<04:59, 5.29it/s, tok/s=15537.7]
Inference: 73%|███████▎ | 4335/5920 [52:18<04:59, 5.29it/s, tok/s=15541.4]
Inference: 73%|███████▎ | 4336/5920 [52:18<04:45, 5.55it/s, tok/s=15541.4]
Inference: 73%|███████▎ | 4336/5920 [52:18<04:45, 5.55it/s, tok/s=15542.9]
Inference: 73%|███████▎ | 4337/5920 [52:19<06:35, 4.00it/s, tok/s=15542.9]
Inference: 73%|███████▎ | 4337/5920 [52:19<06:35, 4.00it/s, tok/s=15541.4]
Inference: 73%|███████▎ | 4338/5920 [52:19<05:51, 4.50it/s, tok/s=15541.4]
Inference: 73%|███████▎ | 4338/5920 [52:19<05:51, 4.50it/s, tok/s=15550.4]
Inference: 73%|███████▎ | 4339/5920 [52:19<08:15, 3.19it/s, tok/s=15550.4]
Inference: 73%|███████▎ | 4339/5920 [52:19<08:15, 3.19it/s, tok/s=15549.6]
Inference: 73%|███████▎ | 4340/5920 [52:20<07:53, 3.34it/s, tok/s=15549.6]
Inference: 73%|███████▎ | 4340/5920 [52:20<07:53, 3.34it/s, tok/s=15549.5]
Inference: 73%|███████▎ | 4341/5920 [52:20<08:55, 2.95it/s, tok/s=15549.5]
Inference: 73%|███████▎ | 4341/5920 [52:20<08:55, 2.95it/s, tok/s=15547.7]
Inference: 73%|███████▎ | 4342/5920 [52:20<07:35, 3.46it/s, tok/s=15547.7]
Inference: 73%|███████▎ | 4342/5920 [52:20<07:35, 3.46it/s, tok/s=15547.6]
Inference: 73%|███████▎ | 4343/5920 [52:20<07:35, 3.46it/s, tok/s=15548.7]
Inference: 73%|███████▎ | 4344/5920 [52:20<07:35, 3.46it/s, tok/s=15549.3]
Inference: 73%|███████▎ | 4345/5920 [52:20<07:35, 3.46it/s, tok/s=15550.0]
Inference: 73%|███████▎ | 4346/5920 [52:20<04:11, 6.27it/s, tok/s=15550.0]
Inference: 73%|███████▎ | 4346/5920 [52:20<04:11, 6.27it/s, tok/s=15550.3]
Inference: 73%|███████▎ | 4347/5920 [52:20<04:10, 6.27it/s, tok/s=15557.5]
Inference: 73%|███████▎ | 4348/5920 [52:21<05:32, 4.72it/s, tok/s=15557.5]
Inference: 73%|███████▎ | 4348/5920 [52:21<05:32, 4.72it/s, tok/s=15555.0]
Inference: 73%|███████▎ | 4349/5920 [52:21<05:51, 4.47it/s, tok/s=15555.0]
Inference: 73%|███████▎ | 4349/5920 [52:21<05:51, 4.47it/s, tok/s=15556.2]
Inference: 73%|███████▎ | 4350/5920 [52:21<05:51, 4.47it/s, tok/s=15556.6]
Inference: 73%|███████▎ | 4351/5920 [52:21<04:26, 5.88it/s, tok/s=15556.6]
Inference: 73%|███████▎ | 4351/5920 [52:21<04:26, 5.88it/s, tok/s=15556.8]
Inference: 74%|███████▎ | 4352/5920 [52:22<04:26, 5.88it/s, tok/s=15560.5]
Inference: 74%|███████▎ | 4353/5920 [52:22<04:26, 5.88it/s, tok/s=15563.5]
Inference: 74%|███████▎ | 4354/5920 [52:22<04:21, 6.00it/s, tok/s=15563.5]
Inference: 74%|███████▎ | 4354/5920 [52:22<04:21, 6.00it/s, tok/s=15568.7]
Inference: 74%|███████▎ | 4355/5920 [52:22<04:20, 6.00it/s, tok/s=15569.1]
Inference: 74%|███████▎ | 4356/5920 [52:22<03:29, 7.47it/s, tok/s=15569.1]
Inference: 74%|███████▎ | 4356/5920 [52:22<03:29, 7.47it/s, tok/s=15570.1]
Inference: 74%|███████▎ | 4357/5920 [52:22<03:29, 7.47it/s, tok/s=15575.5]
Inference: 74%|███████▎ | 4358/5920 [52:22<03:04, 8.46it/s, tok/s=15575.5]
Inference: 74%|███████▎ | 4358/5920 [52:22<03:04, 8.46it/s, tok/s=15576.4]
Inference: 74%|███████▎ | 4359/5920 [52:22<03:04, 8.46it/s, tok/s=15576.5]
Inference: 74%|███████▎ | 4360/5920 [52:22<02:51, 9.09it/s, tok/s=15576.5]
Inference: 74%|███████▎ | 4360/5920 [52:22<02:51, 9.09it/s, tok/s=15579.0]
Inference: 74%|███████▎ | 4361/5920 [52:23<02:51, 9.09it/s, tok/s=15579.9]
Inference: 74%|███████▎ | 4362/5920 [52:23<03:59, 6.49it/s, tok/s=15579.9]
Inference: 74%|███████▎ | 4362/5920 [52:23<03:59, 6.49it/s, tok/s=15578.6]
Inference: 74%|███████▎ | 4363/5920 [52:23<03:59, 6.49it/s, tok/s=15582.4]
Inference: 74%|███████▎ | 4364/5920 [52:23<04:01, 6.44it/s, tok/s=15582.4]
Inference: 74%|███████▎ | 4364/5920 [52:23<04:01, 6.44it/s, tok/s=15581.4]
Inference: 74%|███████▎ | 4365/5920 [52:23<04:01, 6.44it/s, tok/s=15589.8]
Inference: 74%|███████▍ | 4366/5920 [52:23<04:01, 6.44it/s, tok/s=15590.5]
Inference: 74%|███████▍ | 4367/5920 [52:24<03:29, 7.41it/s, tok/s=15590.5]
Inference: 74%|███████▍ | 4367/5920 [52:24<03:29, 7.41it/s, tok/s=15590.3]
Inference: 74%|███████▍ | 4368/5920 [52:24<05:04, 5.10it/s, tok/s=15590.3]
Inference: 74%|███████▍ | 4368/5920 [52:24<05:04, 5.10it/s, tok/s=15588.7]
Inference: 74%|███████▍ | 4369/5920 [52:24<04:59, 5.18it/s, tok/s=15588.7]
Inference: 74%|███████▍ | 4369/5920 [52:24<04:59, 5.18it/s, tok/s=15594.4]
Inference: 74%|███████▍ | 4370/5920 [52:25<06:37, 3.90it/s, tok/s=15594.4]
Inference: 74%|███████▍ | 4370/5920 [52:25<06:37, 3.90it/s, tok/s=15593.0]
Inference: 74%|███████▍ | 4371/5920 [52:25<06:37, 3.90it/s, tok/s=15593.7]
Inference: 74%|███████▍ | 4372/5920 [52:25<04:59, 5.17it/s, tok/s=15593.7]
Inference: 74%|███████▍ | 4372/5920 [52:25<04:59, 5.17it/s, tok/s=15594.1]
Inference: 74%|███████▍ | 4373/5920 [52:26<07:25, 3.47it/s, tok/s=15594.1]
Inference: 74%|███████▍ | 4373/5920 [52:26<07:25, 3.47it/s, tok/s=15593.4]
Inference: 74%|███████▍ | 4374/5920 [52:26<09:25, 2.73it/s, tok/s=15593.4]
Inference: 74%|███████▍ | 4374/5920 [52:26<09:25, 2.73it/s, tok/s=15598.5]
Inference: 74%|███████▍ | 4375/5920 [52:27<10:02, 2.56it/s, tok/s=15598.5]
Inference: 74%|███████▍ | 4375/5920 [52:27<10:02, 2.56it/s, tok/s=15598.4]
Inference: 74%|███████▍ | 4376/5920 [52:27<08:34, 3.00it/s, tok/s=15598.4]
Inference: 74%|███████▍ | 4376/5920 [52:27<08:34, 3.00it/s, tok/s=15598.8]
Inference: 74%|███████▍ | 4377/5920 [52:27<08:34, 3.00it/s, tok/s=15599.3]
Inference: 74%|███████▍ | 4378/5920 [52:27<06:32, 3.93it/s, tok/s=15599.3]
Inference: 74%|███████▍ | 4378/5920 [52:27<06:32, 3.93it/s, tok/s=15600.0]
Inference: 74%|███████▍ | 4379/5920 [52:27<06:46, 3.79it/s, tok/s=15600.0]
Inference: 74%|███████▍ | 4379/5920 [52:27<06:46, 3.79it/s, tok/s=15599.8]
Inference: 74%|███████▍ | 4380/5920 [52:27<06:46, 3.79it/s, tok/s=15600.3]
Inference: 74%|███████▍ | 4381/5920 [52:29<10:05, 2.54it/s, tok/s=15600.3]
Inference: 74%|███████▍ | 4381/5920 [52:29<10:05, 2.54it/s, tok/s=15598.9]
Inference: 74%|███████▍ | 4382/5920 [52:29<08:43, 2.94it/s, tok/s=15598.9]
Inference: 74%|███████▍ | 4382/5920 [52:29<08:43, 2.94it/s, tok/s=15607.7]
Inference: 74%|███████▍ | 4383/5920 [52:30<11:53, 2.16it/s, tok/s=15607.7]
Inference: 74%|███████▍ | 4383/5920 [52:30<11:53, 2.16it/s, tok/s=15613.0]
Inference: 74%|███████▍ | 4384/5920 [52:30<11:52, 2.16it/s, tok/s=15613.4]
Inference: 74%|███████▍ | 4385/5920 [52:30<07:41, 3.33it/s, tok/s=15613.4]
Inference: 74%|███████▍ | 4385/5920 [52:30<07:41, 3.33it/s, tok/s=15613.9]
Inference: 74%|███████▍ | 4386/5920 [52:30<07:40, 3.33it/s, tok/s=15614.6]
Inference: 74%|███████▍ | 4387/5920 [52:30<06:56, 3.68it/s, tok/s=15614.6]
Inference: 74%|███████▍ | 4387/5920 [52:30<06:56, 3.68it/s, tok/s=15613.0]
Inference: 74%|███████▍ | 4388/5920 [52:30<06:56, 3.68it/s, tok/s=15613.4]
Inference: 74%|███████▍ | 4389/5920 [52:30<05:27, 4.67it/s, tok/s=15613.4]
Inference: 74%|███████▍ | 4389/5920 [52:30<05:27, 4.67it/s, tok/s=15612.8]
Inference: 74%|███████▍ | 4390/5920 [52:31<04:54, 5.20it/s, tok/s=15612.8]
Inference: 74%|███████▍ | 4390/5920 [52:31<04:54, 5.20it/s, tok/s=15613.2]
Inference: 74%|███████▍ | 4391/5920 [52:31<04:53, 5.20it/s, tok/s=15618.2]
Inference: 74%|███████▍ | 4392/5920 [52:31<03:39, 6.96it/s, tok/s=15618.2]
Inference: 74%|███████▍ | 4392/5920 [52:31<03:39, 6.96it/s, tok/s=15618.6]
Inference: 74%|███████▍ | 4393/5920 [52:31<03:39, 6.96it/s, tok/s=15619.5]
Inference: 74%|███████▍ | 4394/5920 [52:32<06:14, 4.07it/s, tok/s=15619.5]
Inference: 74%|███████▍ | 4394/5920 [52:32<06:14, 4.07it/s, tok/s=15616.2]
Inference: 74%|███████▍ | 4395/5920 [52:32<05:39, 4.49it/s, tok/s=15616.2]
Inference: 74%|███████▍ | 4395/5920 [52:32<05:39, 4.49it/s, tok/s=15616.8]
Inference: 74%|███████▍ | 4396/5920 [52:32<05:31, 4.59it/s, tok/s=15616.8]
Inference: 74%|███████▍ | 4396/5920 [52:32<05:31, 4.59it/s, tok/s=15625.4]
Inference: 74%|███████▍ | 4397/5920 [52:32<05:31, 4.59it/s, tok/s=15626.2]
Inference: 74%|███████▍ | 4398/5920 [52:32<04:13, 6.00it/s, tok/s=15626.2]
Inference: 74%|███████▍ | 4398/5920 [52:32<04:13, 6.00it/s, tok/s=15626.8]
Inference: 74%|███████▍ | 4399/5920 [52:32<04:13, 6.00it/s, tok/s=15628.7]
Inference: 74%|███████▍ | 4400/5920 [52:32<03:30, 7.23it/s, tok/s=15628.7]
Inference: 74%|███████▍ | 4400/5920 [52:32<03:30, 7.23it/s, tok/s=15629.1]
Inference: 74%|███████▍ | 4401/5920 [52:32<03:44, 6.77it/s, tok/s=15629.1]
Inference: 74%|███████▍ | 4401/5920 [52:32<03:44, 6.77it/s, tok/s=15629.6]
Inference: 74%|███████▍ | 4402/5920 [52:33<07:45, 3.26it/s, tok/s=15629.6]
Inference: 74%|███████▍ | 4402/5920 [52:33<07:45, 3.26it/s, tok/s=15630.1]
Inference: 74%|███████▍ | 4403/5920 [52:34<08:25, 3.00it/s, tok/s=15630.1]
Inference: 74%|███████▍ | 4403/5920 [52:34<08:25, 3.00it/s, tok/s=15637.6]
Inference: 74%|███████▍ | 4404/5920 [52:34<08:25, 3.00it/s, tok/s=15638.5]
Inference: 74%|███████▍ | 4405/5920 [52:34<08:25, 3.00it/s, tok/s=15647.3]
Inference: 74%|███████▍ | 4406/5920 [52:34<06:00, 4.20it/s, tok/s=15647.3]
Inference: 74%|███████▍ | 4406/5920 [52:34<06:00, 4.20it/s, tok/s=15648.2]
Inference: 74%|███████▍ | 4407/5920 [52:35<08:56, 2.82it/s, tok/s=15648.2]
Inference: 74%|███████▍ | 4407/5920 [52:35<08:56, 2.82it/s, tok/s=15644.9]
Inference: 74%|███████▍ | 4408/5920 [52:35<08:30, 2.96it/s, tok/s=15644.9]
Inference: 74%|███████▍ | 4408/5920 [52:35<08:30, 2.96it/s, tok/s=15644.3]
Inference: 74%|███████▍ | 4409/5920 [52:35<07:25, 3.39it/s, tok/s=15644.3]
Inference: 74%|███████▍ | 4409/5920 [52:35<07:25, 3.39it/s, tok/s=15645.8]
Inference: 74%|███████▍ | 4410/5920 [52:35<07:25, 3.39it/s, tok/s=15648.2]
Inference: 75%|███████▍ | 4411/5920 [52:36<06:19, 3.98it/s, tok/s=15648.2]
Inference: 75%|███████▍ | 4411/5920 [52:36<06:19, 3.98it/s, tok/s=15648.5]
Inference: 75%|███████▍ | 4412/5920 [52:36<06:47, 3.70it/s, tok/s=15648.5]
Inference: 75%|███████▍ | 4412/5920 [52:36<06:47, 3.70it/s, tok/s=15648.0]
Inference: 75%|███████▍ | 4413/5920 [52:36<06:53, 3.65it/s, tok/s=15648.0]
Inference: 75%|███████▍ | 4413/5920 [52:36<06:53, 3.65it/s, tok/s=15647.1]
Inference: 75%|███████▍ | 4414/5920 [52:36<06:53, 3.65it/s, tok/s=15649.5]
Inference: 75%|███████▍ | 4415/5920 [52:36<06:52, 3.65it/s, tok/s=15651.8]
Inference: 75%|███████▍ | 4416/5920 [52:36<06:52, 3.65it/s, tok/s=15652.2]
Inference: 75%|███████▍ | 4417/5920 [52:38<07:53, 3.17it/s, tok/s=15652.2]
Inference: 75%|███████▍ | 4417/5920 [52:38<07:53, 3.17it/s, tok/s=15646.2]
Inference: 75%|███████▍ | 4418/5920 [52:38<07:01, 3.56it/s, tok/s=15646.2]
Inference: 75%|███████▍ | 4418/5920 [52:38<07:01, 3.56it/s, tok/s=15647.9]
Inference: 75%|███████▍ | 4419/5920 [52:38<07:52, 3.18it/s, tok/s=15647.9]
Inference: 75%|███████▍ | 4419/5920 [52:38<07:52, 3.18it/s, tok/s=15646.3]
Inference: 75%|███████▍ | 4420/5920 [52:38<07:13, 3.46it/s, tok/s=15646.3]
Inference: 75%|███████▍ | 4420/5920 [52:38<07:13, 3.46it/s, tok/s=15645.5]
Inference: 75%|███████▍ | 4421/5920 [52:39<07:13, 3.46it/s, tok/s=15647.6]
Inference: 75%|███████▍ | 4422/5920 [52:39<04:57, 5.03it/s, tok/s=15647.6]
Inference: 75%|███████▍ | 4422/5920 [52:39<04:57, 5.03it/s, tok/s=15647.6]
Inference: 75%|███████▍ | 4423/5920 [52:39<04:44, 5.26it/s, tok/s=15647.6]
Inference: 75%|███████▍ | 4423/5920 [52:39<04:44, 5.26it/s, tok/s=15648.9]
Inference: 75%|███████▍ | 4424/5920 [52:39<04:44, 5.26it/s, tok/s=15655.3]
Inference: 75%|███████▍ | 4425/5920 [52:39<04:43, 5.26it/s, tok/s=15659.3]
Inference: 75%|███████▍ | 4426/5920 [52:39<04:17, 5.80it/s, tok/s=15659.3]
Inference: 75%|███████▍ | 4426/5920 [52:39<04:17, 5.80it/s, tok/s=15660.9]
Inference: 75%|███████▍ | 4427/5920 [52:39<04:17, 5.80it/s, tok/s=15662.0]
Inference: 75%|███████▍ | 4428/5920 [52:39<03:35, 6.93it/s, tok/s=15662.0]
Inference: 75%|███████▍ | 4428/5920 [52:39<03:35, 6.93it/s, tok/s=15668.5]
Inference: 75%|███████▍ | 4429/5920 [52:39<03:35, 6.93it/s, tok/s=15668.4]
Inference: 75%|███████▍ | 4430/5920 [52:40<03:32, 7.02it/s, tok/s=15668.4]
Inference: 75%|███████▍ | 4430/5920 [52:40<03:32, 7.02it/s, tok/s=15669.6]
Inference: 75%|███████▍ | 4431/5920 [52:40<03:31, 7.05it/s, tok/s=15669.6]
Inference: 75%|███████▍ | 4431/5920 [52:40<03:31, 7.05it/s, tok/s=15669.3]
Inference: 75%|███████▍ | 4432/5920 [52:40<03:30, 7.05it/s, tok/s=15671.9]
Inference: 75%|███████▍ | 4433/5920 [52:40<02:56, 8.41it/s, tok/s=15671.9]
Inference: 75%|███████▍ | 4433/5920 [52:40<02:56, 8.41it/s, tok/s=15671.9]
Inference: 75%|███████▍ | 4434/5920 [52:40<04:43, 5.25it/s, tok/s=15671.9]
Inference: 75%|███████▍ | 4434/5920 [52:40<04:43, 5.25it/s, tok/s=15672.4]
Inference: 75%|███████▍ | 4435/5920 [52:41<06:01, 4.11it/s, tok/s=15672.4]
Inference: 75%|███████▍ | 4435/5920 [52:41<06:01, 4.11it/s, tok/s=15671.2]
Inference: 75%|███████▍ | 4436/5920 [52:41<06:01, 4.11it/s, tok/s=15678.1]
Inference: 75%|███████▍ | 4437/5920 [52:41<05:01, 4.92it/s, tok/s=15678.1]
Inference: 75%|███████▍ | 4437/5920 [52:41<05:01, 4.92it/s, tok/s=15678.9]
Inference: 75%|███████▍ | 4438/5920 [52:41<04:45, 5.19it/s, tok/s=15678.9]
Inference: 75%|███████▍ | 4438/5920 [52:41<04:45, 5.19it/s, tok/s=15679.7]
Inference: 75%|███████▍ | 4439/5920 [52:41<04:45, 5.19it/s, tok/s=15679.8]
Inference: 75%|███████▌ | 4440/5920 [52:42<04:13, 5.83it/s, tok/s=15679.8]
Inference: 75%|███████▌ | 4440/5920 [52:42<04:13, 5.83it/s, tok/s=15686.1]
Inference: 75%|███████▌ | 4441/5920 [52:42<04:23, 5.61it/s, tok/s=15686.1]
Inference: 75%|███████▌ | 4441/5920 [52:42<04:23, 5.61it/s, tok/s=15687.2]
Inference: 75%|███████▌ | 4442/5920 [52:42<04:11, 5.88it/s, tok/s=15687.2]
Inference: 75%|███████▌ | 4442/5920 [52:42<04:11, 5.88it/s, tok/s=15687.8]
Inference: 75%|███████▌ | 4443/5920 [52:42<04:11, 5.88it/s, tok/s=15689.3]
Inference: 75%|███████▌ | 4444/5920 [52:42<04:11, 5.88it/s, tok/s=15690.7]
Inference: 75%|███████▌ | 4445/5920 [52:42<02:35, 9.50it/s, tok/s=15690.7]
Inference: 75%|███████▌ | 4445/5920 [52:42<02:35, 9.50it/s, tok/s=15696.4]
Inference: 75%|███████▌ | 4446/5920 [52:42<02:35, 9.50it/s, tok/s=15696.6]
Inference: 75%|███████▌ | 4447/5920 [52:43<04:02, 6.09it/s, tok/s=15696.6]
Inference: 75%|███████▌ | 4447/5920 [52:43<04:02, 6.09it/s, tok/s=15696.1]
Inference: 75%|███████▌ | 4448/5920 [52:43<05:26, 4.51it/s, tok/s=15696.1]
Inference: 75%|███████▌ | 4448/5920 [52:43<05:26, 4.51it/s, tok/s=15702.1]
Inference: 75%|███████▌ | 4449/5920 [52:43<05:26, 4.51it/s, tok/s=15710.9]
Inference: 75%|███████▌ | 4450/5920 [52:43<05:25, 4.51it/s, tok/s=15711.5]
Inference: 75%|███████▌ | 4451/5920 [52:43<05:25, 4.51it/s, tok/s=15711.8]
Inference: 75%|███████▌ | 4452/5920 [52:43<05:25, 4.51it/s, tok/s=15721.3]
Inference: 75%|███████▌ | 4453/5920 [52:43<02:39, 9.20it/s, tok/s=15721.3]
Inference: 75%|███████▌ | 4453/5920 [52:43<02:39, 9.20it/s, tok/s=15728.9]
Inference: 75%|███████▌ | 4454/5920 [52:43<02:39, 9.20it/s, tok/s=15729.3]
Inference: 75%|███████▌ | 4455/5920 [52:44<03:02, 8.05it/s, tok/s=15729.3]
Inference: 75%|███████▌ | 4455/5920 [52:44<03:02, 8.05it/s, tok/s=15733.7]
Inference: 75%|███████▌ | 4456/5920 [52:44<03:01, 8.05it/s, tok/s=15733.1]
Inference: 75%|███████▌ | 4457/5920 [52:44<04:49, 5.05it/s, tok/s=15733.1]
Inference: 75%|███████▌ | 4457/5920 [52:44<04:49, 5.05it/s, tok/s=15731.2]
Inference: 75%|███████▌ | 4458/5920 [52:45<04:49, 5.05it/s, tok/s=15731.8]
Inference: 75%|███████▌ | 4459/5920 [52:45<04:55, 4.94it/s, tok/s=15731.8]
Inference: 75%|███████▌ | 4459/5920 [52:45<04:55, 4.94it/s, tok/s=15735.0]
Inference: 75%|███████▌ | 4460/5920 [52:45<04:55, 4.94it/s, tok/s=15735.3]
Inference: 75%|███████▌ | 4461/5920 [52:45<05:43, 4.25it/s, tok/s=15735.3]
Inference: 75%|███████▌ | 4461/5920 [52:45<05:43, 4.25it/s, tok/s=15733.2]
Inference: 75%|███████▌ | 4462/5920 [52:46<05:43, 4.25it/s, tok/s=15733.4]
Inference: 75%|███████▌ | 4463/5920 [52:46<05:17, 4.59it/s, tok/s=15733.4]
Inference: 75%|███████▌ | 4463/5920 [52:46<05:17, 4.59it/s, tok/s=15732.5]
Inference: 75%|███████▌ | 4464/5920 [52:46<07:11, 3.38it/s, tok/s=15732.5]
Inference: 75%|███████▌ | 4464/5920 [52:46<07:11, 3.38it/s, tok/s=15732.6]
Inference: 75%|███████▌ | 4465/5920 [52:47<08:26, 2.87it/s, tok/s=15732.6]
Inference: 75%|███████▌ | 4465/5920 [52:47<08:26, 2.87it/s, tok/s=15730.7]
Inference: 75%|███████▌ | 4466/5920 [52:47<07:08, 3.39it/s, tok/s=15730.7]
Inference: 75%|███████▌ | 4466/5920 [52:47<07:08, 3.39it/s, tok/s=15731.4]
Inference: 75%|███████▌ | 4467/5920 [52:47<07:08, 3.39it/s, tok/s=15736.8]
Inference: 75%|███████▌ | 4468/5920 [52:47<05:05, 4.75it/s, tok/s=15736.8]
Inference: 75%|███████▌ | 4468/5920 [52:47<05:05, 4.75it/s, tok/s=15739.9]
Inference: 75%|███████▌ | 4469/5920 [52:48<07:42, 3.14it/s, tok/s=15739.9]
Inference: 75%|███████▌ | 4469/5920 [52:48<07:42, 3.14it/s, tok/s=15739.1]
Inference: 76%|███████▌ | 4470/5920 [52:48<07:06, 3.40it/s, tok/s=15739.1]
Inference: 76%|███████▌ | 4470/5920 [52:48<07:06, 3.40it/s, tok/s=15745.8]
Inference: 76%|███████▌ | 4471/5920 [52:48<07:05, 3.40it/s, tok/s=15746.8]
Inference: 76%|███████▌ | 4472/5920 [52:49<06:11, 3.90it/s, tok/s=15746.8]
Inference: 76%|███████▌ | 4472/5920 [52:49<06:11, 3.90it/s, tok/s=15752.2]
Inference: 76%|███████▌ | 4473/5920 [52:49<06:11, 3.90it/s, tok/s=15752.5]
Inference: 76%|███████▌ | 4474/5920 [52:49<06:11, 3.90it/s, tok/s=15753.3]
Inference: 76%|███████▌ | 4475/5920 [52:49<04:13, 5.71it/s, tok/s=15753.3]
Inference: 76%|███████▌ | 4475/5920 [52:49<04:13, 5.71it/s, tok/s=15753.6]
Inference: 76%|███████▌ | 4476/5920 [52:49<04:39, 5.17it/s, tok/s=15753.6]
Inference: 76%|███████▌ | 4476/5920 [52:49<04:39, 5.17it/s, tok/s=15753.7]
Inference: 76%|███████▌ | 4477/5920 [52:49<04:23, 5.48it/s, tok/s=15753.7]
Inference: 76%|███████▌ | 4477/5920 [52:49<04:23, 5.48it/s, tok/s=15753.9]
Inference: 76%|███████▌ | 4478/5920 [52:49<04:33, 5.26it/s, tok/s=15753.9]
Inference: 76%|███████▌ | 4478/5920 [52:49<04:33, 5.26it/s, tok/s=15753.5]
Inference: 76%|███████▌ | 4479/5920 [52:50<04:33, 5.26it/s, tok/s=15762.8]
Inference: 76%|███████▌ | 4480/5920 [52:50<03:51, 6.23it/s, tok/s=15762.8]
Inference: 76%|███████▌ | 4480/5920 [52:50<03:51, 6.23it/s, tok/s=15763.3]
Inference: 76%|███████▌ | 4481/5920 [52:50<03:50, 6.24it/s, tok/s=15763.3]
Inference: 76%|███████▌ | 4481/5920 [52:50<03:50, 6.24it/s, tok/s=15772.0]
Inference: 76%|███████▌ | 4482/5920 [52:50<05:42, 4.20it/s, tok/s=15772.0]
Inference: 76%|███████▌ | 4482/5920 [52:50<05:42, 4.20it/s, tok/s=15770.5]
Inference: 76%|███████▌ | 4483/5920 [52:51<06:13, 3.85it/s, tok/s=15770.5]
Inference: 76%|███████▌ | 4483/5920 [52:51<06:13, 3.85it/s, tok/s=15769.8]
Inference: 76%|███████▌ | 4484/5920 [52:51<06:59, 3.42it/s, tok/s=15769.8]
Inference: 76%|███████▌ | 4484/5920 [52:51<06:59, 3.42it/s, tok/s=15777.1]
Inference: 76%|███████▌ | 4485/5920 [52:51<06:59, 3.42it/s, tok/s=15780.0]
Inference: 76%|███████▌ | 4486/5920 [52:51<04:59, 4.79it/s, tok/s=15780.0]
Inference: 76%|███████▌ | 4486/5920 [52:51<04:59, 4.79it/s, tok/s=15780.1]
Inference: 76%|███████▌ | 4487/5920 [52:51<05:07, 4.67it/s, tok/s=15780.1]
Inference: 76%|███████▌ | 4487/5920 [52:51<05:07, 4.67it/s, tok/s=15782.0]
Inference: 76%|███████▌ | 4488/5920 [52:52<08:28, 2.82it/s, tok/s=15782.0]
Inference: 76%|███████▌ | 4488/5920 [52:52<08:28, 2.82it/s, tok/s=15779.0]
Inference: 76%|███████▌ | 4489/5920 [52:54<14:10, 1.68it/s, tok/s=15779.0]
Inference: 76%|███████▌ | 4489/5920 [52:54<14:10, 1.68it/s, tok/s=15773.5]
Inference: 76%|███████▌ | 4490/5920 [52:54<13:57, 1.71it/s, tok/s=15773.5]
Inference: 76%|███████▌ | 4490/5920 [52:54<13:57, 1.71it/s, tok/s=15771.4]
Inference: 76%|███████▌ | 4491/5920 [52:54<11:05, 2.15it/s, tok/s=15771.4]
Inference: 76%|███████▌ | 4491/5920 [52:54<11:05, 2.15it/s, tok/s=15771.3]
Inference: 76%|███████▌ | 4492/5920 [52:54<11:05, 2.15it/s, tok/s=15771.6]
Inference: 76%|███████▌ | 4493/5920 [52:54<06:47, 3.50it/s, tok/s=15771.6]
Inference: 76%|███████▌ | 4493/5920 [52:54<06:47, 3.50it/s, tok/s=15780.8]
Inference: 76%|███████▌ | 4494/5920 [52:55<05:53, 4.03it/s, tok/s=15780.8]
Inference: 76%|███████▌ | 4494/5920 [52:55<05:53, 4.03it/s, tok/s=15788.8]
Inference: 76%|███████▌ | 4495/5920 [52:55<05:53, 4.03it/s, tok/s=15788.8]
Inference: 76%|███████▌ | 4495/5920 [52:55<05:53, 4.03it/s, tok/s=15788.8]
Inference: 76%|███████▌ | 4496/5920 [52:55<06:03, 3.91it/s, tok/s=15788.8]
Inference: 76%|███████▌ | 4496/5920 [52:55<06:03, 3.91it/s, tok/s=15787.9]
Inference: 76%|███████▌ | 4497/5920 [52:55<05:23, 4.40it/s, tok/s=15787.9]
Inference: 76%|███████▌ | 4497/5920 [52:55<05:23, 4.40it/s, tok/s=15790.3]
Inference: 76%|███████▌ | 4498/5920 [52:55<05:22, 4.40it/s, tok/s=15791.0]
Inference: 76%|███████▌ | 4499/5920 [52:55<03:52, 6.11it/s, tok/s=15791.0]
Inference: 76%|███████▌ | 4499/5920 [52:55<03:52, 6.11it/s, tok/s=15792.2]
Inference: 76%|███████▌ | 4500/5920 [52:56<03:56, 6.01it/s, tok/s=15792.2]
Inference: 76%|███████▌ | 4500/5920 [52:56<03:56, 6.01it/s, tok/s=15791.8]
Inference: 76%|███████▌ | 4501/5920 [52:56<04:30, 5.24it/s, tok/s=15791.8]
Inference: 76%|███████▌ | 4501/5920 [52:56<04:30, 5.24it/s, tok/s=15793.4]
Inference: 76%|███████▌ | 4502/5920 [52:56<04:02, 5.85it/s, tok/s=15793.4]
Inference: 76%|███████▌ | 4502/5920 [52:56<04:02, 5.85it/s, tok/s=15794.6]
Inference: 76%|███████▌ | 4503/5920 [52:56<04:12, 5.60it/s, tok/s=15794.6]
Inference: 76%|███████▌ | 4503/5920 [52:56<04:12, 5.60it/s, tok/s=15795.0]
Inference: 76%|███████▌ | 4504/5920 [52:56<03:52, 6.09it/s, tok/s=15795.0]
Inference: 76%|███████▌ | 4504/5920 [52:56<03:52, 6.09it/s, tok/s=15796.0]
Inference: 76%|███████▌ | 4505/5920 [52:56<04:00, 5.87it/s, tok/s=15796.0]
Inference: 76%|███████▌ | 4505/5920 [52:56<04:00, 5.87it/s, tok/s=15804.2]
Inference: 76%|███████▌ | 4506/5920 [52:57<06:09, 3.82it/s, tok/s=15804.2]
Inference: 76%|███████▌ | 4506/5920 [52:57<06:09, 3.82it/s, tok/s=15803.0]
Inference: 76%|███████▌ | 4507/5920 [52:57<06:09, 3.82it/s, tok/s=15807.9]
Inference: 76%|███████▌ | 4508/5920 [52:57<04:50, 4.87it/s, tok/s=15807.9]
Inference: 76%|███████▌ | 4508/5920 [52:57<04:50, 4.87it/s, tok/s=15807.6]
Inference: 76%|███████▌ | 4509/5920 [52:57<04:49, 4.87it/s, tok/s=15816.8]
Inference: 76%|███████▌ | 4510/5920 [52:58<04:53, 4.80it/s, tok/s=15816.8]
Inference: 76%|███████▌ | 4510/5920 [52:58<04:53, 4.80it/s, tok/s=15816.0]
Inference: 76%|███████▌ | 4511/5920 [52:58<04:53, 4.80it/s, tok/s=15817.1]
Inference: 76%|███████▌ | 4512/5920 [52:58<05:16, 4.44it/s, tok/s=15817.1]
Inference: 76%|███████▌ | 4512/5920 [52:58<05:16, 4.44it/s, tok/s=15816.1]
Inference: 76%|███████▌ | 4513/5920 [52:58<05:13, 4.49it/s, tok/s=15816.1]
Inference: 76%|███████▌ | 4513/5920 [52:58<05:13, 4.49it/s, tok/s=15817.3]
Inference: 76%|███████▋ | 4514/5920 [52:59<07:05, 3.30it/s, tok/s=15817.3]
Inference: 76%|███████▋ | 4514/5920 [52:59<07:05, 3.30it/s, tok/s=15816.3]
Inference: 76%|███████▋ | 4515/5920 [52:59<07:05, 3.30it/s, tok/s=15817.7]
Inference: 76%|███████▋ | 4516/5920 [52:59<05:34, 4.20it/s, tok/s=15817.7]
Inference: 76%|███████▋ | 4516/5920 [52:59<05:34, 4.20it/s, tok/s=15818.2]
Inference: 76%|███████▋ | 4517/5920 [52:59<05:34, 4.20it/s, tok/s=15820.5]
Inference: 76%|███████▋ | 4518/5920 [52:59<04:39, 5.02it/s, tok/s=15820.5]
Inference: 76%|███████▋ | 4518/5920 [52:59<04:39, 5.02it/s, tok/s=15820.8]
Inference: 76%|███████▋ | 4519/5920 [53:00<05:30, 4.24it/s, tok/s=15820.8]
Inference: 76%|███████▋ | 4519/5920 [53:00<05:30, 4.24it/s, tok/s=15819.9]
Inference: 76%|███████▋ | 4520/5920 [53:00<05:53, 3.96it/s, tok/s=15819.9]
Inference: 76%|███████▋ | 4520/5920 [53:00<05:53, 3.96it/s, tok/s=15826.3]
Inference: 76%|███████▋ | 4521/5920 [53:01<06:54, 3.38it/s, tok/s=15826.3]
Inference: 76%|███████▋ | 4521/5920 [53:01<06:54, 3.38it/s, tok/s=15824.9]
Inference: 76%|███████▋ | 4522/5920 [53:02<12:47, 1.82it/s, tok/s=15824.9]
Inference: 76%|███████▋ | 4522/5920 [53:02<12:47, 1.82it/s, tok/s=15819.6]
Inference: 76%|███████▋ | 4523/5920 [53:02<12:45, 1.82it/s, tok/s=15819.6]
Inference: 76%|███████▋ | 4523/5920 [53:02<12:45, 1.82it/s, tok/s=15818.2]
Inference: 76%|███████▋ | 4524/5920 [53:02<12:45, 1.82it/s, tok/s=15818.7]
Inference: 76%|███████▋ | 4525/5920 [53:03<11:36, 2.00it/s, tok/s=15818.7]
Inference: 76%|███████▋ | 4525/5920 [53:03<11:36, 2.00it/s, tok/s=15818.7]
Inference: 76%|███████▋ | 4526/5920 [53:04<10:23, 2.23it/s, tok/s=15818.7]
Inference: 76%|███████▋ | 4526/5920 [53:04<10:23, 2.23it/s, tok/s=15818.4]
Inference: 76%|███████▋ | 4527/5920 [53:04<10:23, 2.23it/s, tok/s=15821.8]
Inference: 76%|███████▋ | 4528/5920 [53:06<15:42, 1.48it/s, tok/s=15821.8]
Inference: 76%|███████▋ | 4528/5920 [53:06<15:42, 1.48it/s, tok/s=15812.9]
Inference: 77%|███████▋ | 4529/5920 [53:06<12:51, 1.80it/s, tok/s=15812.9]
Inference: 77%|███████▋ | 4529/5920 [53:06<12:51, 1.80it/s, tok/s=15814.4]
Inference: 77%|███████▋ | 4530/5920 [53:06<12:51, 1.80it/s, tok/s=15815.3]
Inference: 77%|███████▋ | 4531/5920 [53:06<08:38, 2.68it/s, tok/s=15815.3]
Inference: 77%|███████▋ | 4531/5920 [53:06<08:38, 2.68it/s, tok/s=15815.4]
Inference: 77%|███████▋ | 4532/5920 [53:06<09:49, 2.35it/s, tok/s=15815.4]
Inference: 77%|███████▋ | 4532/5920 [53:06<09:49, 2.35it/s, tok/s=15813.5]
Inference: 77%|███████▋ | 4533/5920 [53:07<09:26, 2.45it/s, tok/s=15813.5]
Inference: 77%|███████▋ | 4533/5920 [53:07<09:26, 2.45it/s, tok/s=15812.5]
Inference: 77%|███████▋ | 4534/5920 [53:08<14:35, 1.58it/s, tok/s=15812.5]
Inference: 77%|███████▋ | 4534/5920 [53:08<14:35, 1.58it/s, tok/s=15807.5]
Inference: 77%|███████▋ | 4535/5920 [53:08<11:57, 1.93it/s, tok/s=15807.5]
Inference: 77%|███████▋ | 4535/5920 [53:08<11:57, 1.93it/s, tok/s=15808.0]
Inference: 77%|███████▋ | 4536/5920 [53:08<11:56, 1.93it/s, tok/s=15808.9]
Inference: 77%|███████▋ | 4537/5920 [53:09<08:41, 2.65it/s, tok/s=15808.9]
Inference: 77%|███████▋ | 4537/5920 [53:09<08:41, 2.65it/s, tok/s=15808.9]
Inference: 77%|███████▋ | 4538/5920 [53:09<08:41, 2.65it/s, tok/s=15810.3]
Inference: 77%|███████▋ | 4539/5920 [53:09<06:17, 3.66it/s, tok/s=15810.3]
Inference: 77%|███████▋ | 4539/5920 [53:09<06:17, 3.66it/s, tok/s=15810.6]
Inference: 77%|███████▋ | 4540/5920 [53:09<06:05, 3.78it/s, tok/s=15810.6]
Inference: 77%|███████▋ | 4540/5920 [53:09<06:05, 3.78it/s, tok/s=15810.6]
Inference: 77%|███████▋ | 4541/5920 [53:09<06:05, 3.78it/s, tok/s=15811.6]
Inference: 77%|███████▋ | 4542/5920 [53:09<04:24, 5.21it/s, tok/s=15811.6]
Inference: 77%|███████▋ | 4542/5920 [53:09<04:24, 5.21it/s, tok/s=15812.6]
Inference: 77%|███████▋ | 4543/5920 [53:10<05:07, 4.48it/s, tok/s=15812.6]
Inference: 77%|███████▋ | 4543/5920 [53:10<05:07, 4.48it/s, tok/s=15811.6]
Inference: 77%|███████▋ | 4544/5920 [53:10<05:27, 4.20it/s, tok/s=15811.6]
Inference: 77%|███████▋ | 4544/5920 [53:10<05:27, 4.20it/s, tok/s=15810.9]
Inference: 77%|███████▋ | 4545/5920 [53:10<06:44, 3.40it/s, tok/s=15810.9]
Inference: 77%|███████▋ | 4545/5920 [53:10<06:44, 3.40it/s, tok/s=15809.7]
Inference: 77%|███████▋ | 4546/5920 [53:10<06:44, 3.40it/s, tok/s=15813.9]
Inference: 77%|███████▋ | 4547/5920 [53:11<04:59, 4.58it/s, tok/s=15813.9]
Inference: 77%|███████▋ | 4547/5920 [53:11<04:59, 4.58it/s, tok/s=15814.5]
Inference: 77%|███████▋ | 4548/5920 [53:11<04:59, 4.58it/s, tok/s=15816.6]
Inference: 77%|███████▋ | 4549/5920 [53:11<04:59, 4.58it/s, tok/s=15817.2]
Inference: 77%|███████▋ | 4550/5920 [53:11<03:00, 7.57it/s, tok/s=15817.2]
Inference: 77%|███████▋ | 4550/5920 [53:11<03:00, 7.57it/s, tok/s=15817.5]
Inference: 77%|███████▋ | 4551/5920 [53:11<03:00, 7.57it/s, tok/s=15820.3]
Inference: 77%|███████▋ | 4552/5920 [53:11<02:28, 9.19it/s, tok/s=15820.3]
Inference: 77%|███████▋ | 4552/5920 [53:11<02:28, 9.19it/s, tok/s=15821.0]
Inference: 77%|███████▋ | 4553/5920 [53:11<02:28, 9.19it/s, tok/s=15821.5]
Inference: 77%|███████▋ | 4554/5920 [53:11<03:22, 6.76it/s, tok/s=15821.5]
Inference: 77%|███████▋ | 4554/5920 [53:11<03:22, 6.76it/s, tok/s=15822.1]
Inference: 77%|███████▋ | 4555/5920 [53:12<03:21, 6.76it/s, tok/s=15822.6]
Inference: 77%|███████▋ | 4556/5920 [53:12<03:31, 6.46it/s, tok/s=15822.6]
Inference: 77%|███████▋ | 4556/5920 [53:12<03:31, 6.46it/s, tok/s=15824.7]
Inference: 77%|███████▋ | 4557/5920 [53:12<04:52, 4.66it/s, tok/s=15824.7]
Inference: 77%|███████▋ | 4557/5920 [53:12<04:52, 4.66it/s, tok/s=15823.4]
Inference: 77%|███████▋ | 4558/5920 [53:12<04:52, 4.66it/s, tok/s=15824.9]
Inference: 77%|███████▋ | 4559/5920 [53:12<03:38, 6.23it/s, tok/s=15824.9]
Inference: 77%|███████▋ | 4559/5920 [53:12<03:38, 6.23it/s, tok/s=15825.4]
Inference: 77%|███████▋ | 4560/5920 [53:13<03:38, 6.23it/s, tok/s=15824.8]
Inference: 77%|███████▋ | 4561/5920 [53:13<05:40, 3.99it/s, tok/s=15824.8]
Inference: 77%|███████▋ | 4561/5920 [53:13<05:40, 3.99it/s, tok/s=15823.4]
Inference: 77%|███████▋ | 4562/5920 [53:13<05:40, 3.99it/s, tok/s=15824.1]
Inference: 77%|███████▋ | 4563/5920 [53:13<05:40, 3.99it/s, tok/s=15827.4]
Inference: 77%|███████▋ | 4564/5920 [53:13<03:45, 6.01it/s, tok/s=15827.4]
Inference: 77%|███████▋ | 4564/5920 [53:13<03:45, 6.01it/s, tok/s=15828.2]
Inference: 77%|███████▋ | 4565/5920 [53:13<03:45, 6.01it/s, tok/s=15832.5]
Inference: 77%|███████▋ | 4566/5920 [53:14<05:23, 4.18it/s, tok/s=15832.5]
Inference: 77%|███████▋ | 4566/5920 [53:14<05:23, 4.18it/s, tok/s=15829.6]
Inference: 77%|███████▋ | 4567/5920 [53:14<04:59, 4.51it/s, tok/s=15829.6]
Inference: 77%|███████▋ | 4567/5920 [53:14<04:59, 4.51it/s, tok/s=15830.9]
Inference: 77%|███████▋ | 4568/5920 [53:15<08:45, 2.57it/s, tok/s=15830.9]
Inference: 77%|███████▋ | 4568/5920 [53:15<08:45, 2.57it/s, tok/s=15835.0]
Inference: 77%|███████▋ | 4569/5920 [53:16<10:16, 2.19it/s, tok/s=15835.0]
Inference: 77%|███████▋ | 4569/5920 [53:16<10:16, 2.19it/s, tok/s=15831.8]
Inference: 77%|███████▋ | 4570/5920 [53:16<08:48, 2.56it/s, tok/s=15831.8]
Inference: 77%|███████▋ | 4570/5920 [53:16<08:48, 2.56it/s, tok/s=15832.0]
Inference: 77%|███████▋ | 4571/5920 [53:16<07:42, 2.92it/s, tok/s=15832.0]
Inference: 77%|███████▋ | 4571/5920 [53:16<07:42, 2.92it/s, tok/s=15835.2]
Inference: 77%|███████▋ | 4572/5920 [53:17<09:55, 2.26it/s, tok/s=15835.2]
Inference: 77%|███████▋ | 4572/5920 [53:17<09:55, 2.26it/s, tok/s=15834.8]
Inference: 77%|███████▋ | 4573/5920 [53:17<08:57, 2.50it/s, tok/s=15834.8]
Inference: 77%|███████▋ | 4573/5920 [53:17<08:57, 2.50it/s, tok/s=15834.2]
Inference: 77%|███████▋ | 4574/5920 [53:18<07:14, 3.10it/s, tok/s=15834.2]
Inference: 77%|███████▋ | 4574/5920 [53:18<07:14, 3.10it/s, tok/s=15834.9]
Inference: 77%|███████▋ | 4575/5920 [53:18<07:14, 3.10it/s, tok/s=15836.3]
Inference: 77%|███████▋ | 4576/5920 [53:18<07:07, 3.15it/s, tok/s=15836.3]
Inference: 77%|███████▋ | 4576/5920 [53:18<07:07, 3.15it/s, tok/s=15834.2]
Inference: 77%|███████▋ | 4577/5920 [53:18<07:06, 3.15it/s, tok/s=15834.9]
Inference: 77%|███████▋ | 4578/5920 [53:19<07:19, 3.05it/s, tok/s=15834.9]
Inference: 77%|███████▋ | 4578/5920 [53:19<07:19, 3.05it/s, tok/s=15834.2]
Inference: 77%|███████▋ | 4579/5920 [53:19<06:29, 3.44it/s, tok/s=15834.2]
Inference: 77%|███████▋ | 4579/5920 [53:19<06:29, 3.44it/s, tok/s=15834.6]
Inference: 77%|███████▋ | 4580/5920 [53:20<07:48, 2.86it/s, tok/s=15834.6]
Inference: 77%|███████▋ | 4580/5920 [53:20<07:48, 2.86it/s, tok/s=15833.0]
Inference: 77%|███████▋ | 4581/5920 [53:20<06:26, 3.47it/s, tok/s=15833.0]
Inference: 77%|███████▋ | 4581/5920 [53:20<06:26, 3.47it/s, tok/s=15833.4]
Inference: 77%|███████▋ | 4582/5920 [53:20<05:19, 4.18it/s, tok/s=15833.4]
Inference: 77%|███████▋ | 4582/5920 [53:20<05:19, 4.18it/s, tok/s=15833.6]
Inference: 77%|███████▋ | 4583/5920 [53:20<05:19, 4.18it/s, tok/s=15833.9]
Inference: 77%|███████▋ | 4584/5920 [53:20<05:51, 3.80it/s, tok/s=15833.9]
Inference: 77%|███████▋ | 4584/5920 [53:20<05:51, 3.80it/s, tok/s=15837.0]
Inference: 77%|███████▋ | 4585/5920 [53:21<07:39, 2.90it/s, tok/s=15837.0]
Inference: 77%|███████▋ | 4585/5920 [53:21<07:39, 2.90it/s, tok/s=15835.1]
Inference: 77%|███████▋ | 4586/5920 [53:21<07:58, 2.79it/s, tok/s=15835.1]
Inference: 77%|███████▋ | 4586/5920 [53:21<07:58, 2.79it/s, tok/s=15835.6]
Inference: 77%|███████▋ | 4587/5920 [53:21<07:58, 2.79it/s, tok/s=15836.9]
Inference: 78%|███████▊ | 4588/5920 [53:22<06:39, 3.33it/s, tok/s=15836.9]
Inference: 78%|███████▊ | 4588/5920 [53:22<06:39, 3.33it/s, tok/s=15836.8]
Inference: 78%|███████▊ | 4589/5920 [53:25<22:10, 1.00it/s, tok/s=15836.8]
Inference: 78%|███████▊ | 4589/5920 [53:25<22:10, 1.00it/s, tok/s=15821.5]
Inference: 78%|███████▊ | 4590/5920 [53:25<22:09, 1.00it/s, tok/s=15823.6]
Inference: 78%|███████▊ | 4591/5920 [53:26<15:13, 1.46it/s, tok/s=15823.6]
Inference: 78%|███████▊ | 4591/5920 [53:26<15:13, 1.46it/s, tok/s=15822.3]
Inference: 78%|███████▊ | 4592/5920 [53:26<12:28, 1.77it/s, tok/s=15822.3]
Inference: 78%|███████▊ | 4592/5920 [53:26<12:28, 1.77it/s, tok/s=15831.1]
Inference: 78%|███████▊ | 4593/5920 [53:26<13:29, 1.64it/s, tok/s=15831.1]
Inference: 78%|███████▊ | 4593/5920 [53:26<13:29, 1.64it/s, tok/s=15828.7]
Inference: 78%|███████▊ | 4594/5920 [53:27<12:41, 1.74it/s, tok/s=15828.7]
Inference: 78%|███████▊ | 4594/5920 [53:27<12:41, 1.74it/s, tok/s=15827.4]
Inference: 78%|███████▊ | 4595/5920 [53:27<10:35, 2.08it/s, tok/s=15827.4]
Inference: 78%|███████▊ | 4595/5920 [53:27<10:35, 2.08it/s, tok/s=15827.6]
Inference: 78%|███████▊ | 4596/5920 [53:27<08:58, 2.46it/s, tok/s=15827.6]
Inference: 78%|███████▊ | 4596/5920 [53:27<08:58, 2.46it/s, tok/s=15829.7]
Inference: 78%|███████▊ | 4597/5920 [53:28<11:23, 1.94it/s, tok/s=15829.7]
Inference: 78%|███████▊ | 4597/5920 [53:28<11:23, 1.94it/s, tok/s=15826.3]
Inference: 78%|███████▊ | 4598/5920 [53:28<11:22, 1.94it/s, tok/s=15827.7]
Inference: 78%|███████▊ | 4599/5920 [53:29<08:08, 2.70it/s, tok/s=15827.7]
Inference: 78%|███████▊ | 4599/5920 [53:29<08:08, 2.70it/s, tok/s=15826.8]
Inference: 78%|███████▊ | 4600/5920 [53:29<07:25, 2.96it/s, tok/s=15826.8]
Inference: 78%|███████▊ | 4600/5920 [53:29<07:25, 2.96it/s, tok/s=15827.4]
Inference: 78%|███████▊ | 4601/5920 [53:29<07:24, 2.96it/s, tok/s=15827.8]
Inference: 78%|███████▊ | 4602/5920 [53:29<07:24, 2.96it/s, tok/s=15829.5]
Inference: 78%|███████▊ | 4603/5920 [53:29<07:24, 2.96it/s, tok/s=15830.0]
Inference: 78%|███████▊ | 4604/5920 [53:29<07:23, 2.96it/s, tok/s=15831.9]
Inference: 78%|███████▊ | 4605/5920 [53:29<03:46, 5.80it/s, tok/s=15831.9]
Inference: 78%|███████▊ | 4605/5920 [53:29<03:46, 5.80it/s, tok/s=15832.3]
Inference: 78%|███████▊ | 4606/5920 [53:29<03:35, 6.10it/s, tok/s=15832.3]
Inference: 78%|███████▊ | 4606/5920 [53:29<03:35, 6.10it/s, tok/s=15832.0]
Inference: 78%|███████▊ | 4607/5920 [53:29<03:35, 6.10it/s, tok/s=15833.1]
Inference: 78%|███████▊ | 4608/5920 [53:29<03:03, 7.15it/s, tok/s=15833.1]
Inference: 78%|███████▊ | 4608/5920 [53:29<03:03, 7.15it/s, tok/s=15842.1]
Inference: 78%|███████▊ | 4609/5920 [53:29<03:03, 7.15it/s, tok/s=15843.0]
Inference: 78%|███████▊ | 4610/5920 [53:30<02:51, 7.64it/s, tok/s=15843.0]
Inference: 78%|███████▊ | 4610/5920 [53:30<02:51, 7.64it/s, tok/s=15842.7]
Inference: 78%|███████▊ | 4611/5920 [53:30<04:09, 5.24it/s, tok/s=15842.7]
Inference: 78%|███████▊ | 4611/5920 [53:30<04:09, 5.24it/s, tok/s=15841.3]
Inference: 78%|███████▊ | 4612/5920 [53:30<04:28, 4.87it/s, tok/s=15841.3]
Inference: 78%|███████▊ | 4612/5920 [53:30<04:28, 4.87it/s, tok/s=15840.6]
Inference: 78%|███████▊ | 4613/5920 [53:30<04:28, 4.87it/s, tok/s=15840.6]
Inference: 78%|███████▊ | 4614/5920 [53:31<03:20, 6.53it/s, tok/s=15840.6]
Inference: 78%|███████▊ | 4614/5920 [53:31<03:20, 6.53it/s, tok/s=15841.9]
Inference: 78%|███████▊ | 4615/5920 [53:31<03:47, 5.74it/s, tok/s=15841.9]
Inference: 78%|███████▊ | 4615/5920 [53:31<03:47, 5.74it/s, tok/s=15849.5]
Inference: 78%|███████▊ | 4616/5920 [53:31<03:47, 5.74it/s, tok/s=15850.4]
Inference: 78%|███████▊ | 4617/5920 [53:31<03:50, 5.65it/s, tok/s=15850.4]
Inference: 78%|███████▊ | 4617/5920 [53:31<03:50, 5.65it/s, tok/s=15849.6]
Inference: 78%|███████▊ | 4618/5920 [53:31<03:50, 5.65it/s, tok/s=15851.3]
Inference: 78%|███████▊ | 4619/5920 [53:32<06:40, 3.25it/s, tok/s=15851.3]
Inference: 78%|███████▊ | 4619/5920 [53:32<06:40, 3.25it/s, tok/s=15849.4]
Inference: 78%|███████▊ | 4620/5920 [53:33<09:37, 2.25it/s, tok/s=15849.4]
Inference: 78%|███████▊ | 4620/5920 [53:33<09:37, 2.25it/s, tok/s=15846.9]
Inference: 78%|███████▊ | 4621/5920 [53:34<12:20, 1.75it/s, tok/s=15846.9]
Inference: 78%|███████▊ | 4621/5920 [53:34<12:20, 1.75it/s, tok/s=15842.7]
Inference: 78%|███████▊ | 4622/5920 [53:35<11:53, 1.82it/s, tok/s=15842.7]
Inference: 78%|███████▊ | 4622/5920 [53:35<11:53, 1.82it/s, tok/s=15842.9]
Inference: 78%|███████▊ | 4623/5920 [53:35<11:52, 1.82it/s, tok/s=15843.8]
Inference: 78%|███████▊ | 4624/5920 [53:35<09:52, 2.19it/s, tok/s=15843.8]
Inference: 78%|███████▊ | 4624/5920 [53:35<09:52, 2.19it/s, tok/s=15841.8]
Inference: 78%|███████▊ | 4625/5920 [53:36<08:20, 2.59it/s, tok/s=15841.8]
Inference: 78%|███████▊ | 4625/5920 [53:36<08:20, 2.59it/s, tok/s=15842.9]
Inference: 78%|███████▊ | 4626/5920 [53:36<07:56, 2.72it/s, tok/s=15842.9]
Inference: 78%|███████▊ | 4626/5920 [53:36<07:56, 2.72it/s, tok/s=15844.8]
Inference: 78%|███████▊ | 4627/5920 [53:36<07:55, 2.72it/s, tok/s=15846.2]
Inference: 78%|███████▊ | 4628/5920 [53:36<05:12, 4.14it/s, tok/s=15846.2]
Inference: 78%|███████▊ | 4628/5920 [53:36<05:12, 4.14it/s, tok/s=15848.2]
Inference: 78%|███████▊ | 4629/5920 [53:36<05:12, 4.14it/s, tok/s=15849.3]
Inference: 78%|███████▊ | 4630/5920 [53:36<04:14, 5.07it/s, tok/s=15849.3]
Inference: 78%|███████▊ | 4630/5920 [53:36<04:14, 5.07it/s, tok/s=15857.8]
Inference: 78%|███████▊ | 4631/5920 [53:37<04:46, 4.50it/s, tok/s=15857.8]
Inference: 78%|███████▊ | 4631/5920 [53:37<04:46, 4.50it/s, tok/s=15856.6]
Inference: 78%|███████▊ | 4632/5920 [53:37<04:46, 4.50it/s, tok/s=15857.9]
Inference: 78%|███████▊ | 4633/5920 [53:37<04:49, 4.44it/s, tok/s=15857.9]
Inference: 78%|███████▊ | 4633/5920 [53:37<04:49, 4.44it/s, tok/s=15857.0]
Inference: 78%|███████▊ | 4634/5920 [53:37<05:20, 4.01it/s, tok/s=15857.0]
Inference: 78%|███████▊ | 4634/5920 [53:37<05:20, 4.01it/s, tok/s=15856.3]
Inference: 78%|███████▊ | 4635/5920 [53:37<05:20, 4.01it/s, tok/s=15864.6]
Inference: 78%|███████▊ | 4636/5920 [53:38<04:40, 4.58it/s, tok/s=15864.6]
Inference: 78%|███████▊ | 4636/5920 [53:38<04:40, 4.58it/s, tok/s=15867.7]
Inference: 78%|███████▊ | 4637/5920 [53:38<04:40, 4.58it/s, tok/s=15868.4]
Inference: 78%|███████▊ | 4638/5920 [53:38<05:35, 3.82it/s, tok/s=15868.4]
Inference: 78%|███████▊ | 4638/5920 [53:38<05:35, 3.82it/s, tok/s=15867.1]
Inference: 78%|███████▊ | 4639/5920 [53:38<05:12, 4.10it/s, tok/s=15867.1]
Inference: 78%|███████▊ | 4639/5920 [53:38<05:12, 4.10it/s, tok/s=15866.8]
Inference: 78%|███████▊ | 4640/5920 [53:39<04:34, 4.65it/s, tok/s=15866.8]
Inference: 78%|███████▊ | 4640/5920 [53:39<04:34, 4.65it/s, tok/s=15868.2]
Inference: 78%|███████▊ | 4641/5920 [53:39<04:08, 5.16it/s, tok/s=15868.2]
Inference: 78%|███████▊ | 4641/5920 [53:39<04:08, 5.16it/s, tok/s=15868.9]
Inference: 78%|███████▊ | 4642/5920 [53:39<04:07, 5.16it/s, tok/s=15871.8]
Inference: 78%|███████▊ | 4643/5920 [53:39<02:55, 7.27it/s, tok/s=15871.8]
Inference: 78%|███████▊ | 4643/5920 [53:39<02:55, 7.27it/s, tok/s=15873.3]
Inference: 78%|███████▊ | 4644/5920 [53:39<02:55, 7.27it/s, tok/s=15871.4]
Inference: 78%|███████▊ | 4645/5920 [53:39<04:05, 5.19it/s, tok/s=15871.4]
Inference: 78%|███████▊ | 4645/5920 [53:39<04:05, 5.19it/s, tok/s=15873.7]
Inference: 78%|███████▊ | 4646/5920 [53:39<04:05, 5.19it/s, tok/s=15874.1]
Inference: 78%|███████▊ | 4647/5920 [53:40<03:15, 6.52it/s, tok/s=15874.1]
Inference: 78%|███████▊ | 4647/5920 [53:40<03:15, 6.52it/s, tok/s=15874.8]
Inference: 79%|███████▊ | 4648/5920 [53:40<04:13, 5.03it/s, tok/s=15874.8]
Inference: 79%|███████▊ | 4648/5920 [53:40<04:13, 5.03it/s, tok/s=15878.4]
Inference: 79%|███████▊ | 4649/5920 [53:41<06:41, 3.17it/s, tok/s=15878.4]
Inference: 79%|███████▊ | 4649/5920 [53:41<06:41, 3.17it/s, tok/s=15876.7]
Inference: 79%|███████▊ | 4650/5920 [53:41<06:54, 3.06it/s, tok/s=15876.7]
Inference: 79%|███████▊ | 4650/5920 [53:41<06:54, 3.06it/s, tok/s=15877.6]
Inference: 79%|███████▊ | 4651/5920 [53:41<06:54, 3.06it/s, tok/s=15879.6]
Inference: 79%|███████▊ | 4652/5920 [53:41<04:52, 4.34it/s, tok/s=15879.6]
Inference: 79%|███████▊ | 4652/5920 [53:41<04:52, 4.34it/s, tok/s=15879.8]
Inference: 79%|███████▊ | 4653/5920 [53:42<05:22, 3.92it/s, tok/s=15879.8]
Inference: 79%|███████▊ | 4653/5920 [53:42<05:22, 3.92it/s, tok/s=15878.8]
Inference: 79%|███████▊ | 4654/5920 [53:42<07:03, 2.99it/s, tok/s=15878.8]
Inference: 79%|███████▊ | 4654/5920 [53:42<07:03, 2.99it/s, tok/s=15876.8]
Inference: 79%|███████▊ | 4655/5920 [53:43<08:05, 2.61it/s, tok/s=15876.8]
Inference: 79%|███████▊ | 4655/5920 [53:43<08:05, 2.61it/s, tok/s=15878.0]
Inference: 79%|███████▊ | 4656/5920 [53:43<08:55, 2.36it/s, tok/s=15878.0]
Inference: 79%|███████▊ | 4656/5920 [53:43<08:55, 2.36it/s, tok/s=15875.7]
Inference: 79%|███████▊ | 4657/5920 [53:43<08:55, 2.36it/s, tok/s=15876.0]
Inference: 79%|███████▊ | 4658/5920 [53:44<07:51, 2.67it/s, tok/s=15876.0]
Inference: 79%|███████▊ | 4658/5920 [53:44<07:51, 2.67it/s, tok/s=15874.6]
Inference: 79%|███████▊ | 4659/5920 [53:44<07:32, 2.78it/s, tok/s=15874.6]
Inference: 79%|███████▊ | 4659/5920 [53:44<07:32, 2.78it/s, tok/s=15877.6]
Inference: 79%|███████▊ | 4660/5920 [53:44<07:06, 2.95it/s, tok/s=15877.6]
Inference: 79%|███████▊ | 4660/5920 [53:44<07:06, 2.95it/s, tok/s=15876.6]
Inference: 79%|███████▊ | 4661/5920 [53:45<06:14, 3.36it/s, tok/s=15876.6]
Inference: 79%|███████▊ | 4661/5920 [53:45<06:14, 3.36it/s, tok/s=15884.7]
Inference: 79%|███████▉ | 4662/5920 [53:45<06:14, 3.36it/s, tok/s=15886.8]
Inference: 79%|███████▉ | 4663/5920 [53:45<06:13, 3.36it/s, tok/s=15889.5]
Inference: 79%|███████▉ | 4664/5920 [53:45<03:57, 5.29it/s, tok/s=15889.5]
Inference: 79%|███████▉ | 4664/5920 [53:45<03:57, 5.29it/s, tok/s=15888.8]
Inference: 79%|███████▉ | 4665/5920 [53:45<03:57, 5.29it/s, tok/s=15888.8]
Inference: 79%|███████▉ | 4665/5920 [53:45<03:57, 5.29it/s, tok/s=15890.1]
Inference: 79%|███████▉ | 4666/5920 [53:45<04:15, 4.91it/s, tok/s=15890.1]
Inference: 79%|███████▉ | 4666/5920 [53:45<04:15, 4.91it/s, tok/s=15890.1]
Inference: 79%|███████▉ | 4667/5920 [53:46<05:05, 4.10it/s, tok/s=15890.1]
Inference: 79%|███████▉ | 4667/5920 [53:46<05:05, 4.10it/s, tok/s=15889.4]
Inference: 79%|███████▉ | 4668/5920 [53:46<05:01, 4.15it/s, tok/s=15889.4]
Inference: 79%|███████▉ | 4668/5920 [53:46<05:01, 4.15it/s, tok/s=15888.9]
Inference: 79%|███████▉ | 4669/5920 [53:46<05:13, 3.99it/s, tok/s=15888.9]
Inference: 79%|███████▉ | 4669/5920 [53:46<05:13, 3.99it/s, tok/s=15890.6]
Inference: 79%|███████▉ | 4670/5920 [53:46<05:13, 3.99it/s, tok/s=15891.2]
Inference: 79%|███████▉ | 4671/5920 [53:47<06:22, 3.26it/s, tok/s=15891.2]
Inference: 79%|███████▉ | 4671/5920 [53:47<06:22, 3.26it/s, tok/s=15888.4]
Inference: 79%|███████▉ | 4672/5920 [53:47<06:22, 3.26it/s, tok/s=15888.7]
Inference: 79%|███████▉ | 4673/5920 [53:47<04:37, 4.49it/s, tok/s=15888.7]
Inference: 79%|███████▉ | 4673/5920 [53:47<04:37, 4.49it/s, tok/s=15889.2]
Inference: 79%|███████▉ | 4674/5920 [53:47<04:05, 5.07it/s, tok/s=15889.2]
Inference: 79%|███████▉ | 4674/5920 [53:47<04:05, 5.07it/s, tok/s=15898.0]
Inference: 79%|███████▉ | 4675/5920 [53:47<04:05, 5.07it/s, tok/s=15898.4]
Inference: 79%|███████▉ | 4676/5920 [53:48<04:14, 4.90it/s, tok/s=15898.4]
Inference: 79%|███████▉ | 4676/5920 [53:48<04:14, 4.90it/s, tok/s=15897.9]
Inference: 79%|███████▉ | 4677/5920 [53:48<04:36, 4.49it/s, tok/s=15897.9]
Inference: 79%|███████▉ | 4677/5920 [53:48<04:36, 4.49it/s, tok/s=15897.3]
Inference: 79%|███████▉ | 4678/5920 [53:48<04:36, 4.49it/s, tok/s=15900.3]
Inference: 79%|███████▉ | 4679/5920 [53:48<03:49, 5.41it/s, tok/s=15900.3]
Inference: 79%|███████▉ | 4679/5920 [53:48<03:49, 5.41it/s, tok/s=15903.2]
Inference: 79%|███████▉ | 4680/5920 [53:48<03:27, 5.99it/s, tok/s=15903.2]
Inference: 79%|███████▉ | 4680/5920 [53:48<03:27, 5.99it/s, tok/s=15903.7]
Inference: 79%|███████▉ | 4681/5920 [53:49<04:32, 4.55it/s, tok/s=15903.7]
Inference: 79%|███████▉ | 4681/5920 [53:49<04:32, 4.55it/s, tok/s=15902.3]
Inference: 79%|███████▉ | 4682/5920 [53:49<04:32, 4.55it/s, tok/s=15902.4]
Inference: 79%|███████▉ | 4683/5920 [53:49<04:29, 4.58it/s, tok/s=15902.4]
Inference: 79%|███████▉ | 4683/5920 [53:49<04:29, 4.58it/s, tok/s=15901.1]
Inference: 79%|███████▉ | 4684/5920 [53:49<05:06, 4.03it/s, tok/s=15901.1]
Inference: 79%|███████▉ | 4684/5920 [53:49<05:06, 4.03it/s, tok/s=15901.8]
Inference: 79%|███████▉ | 4685/5920 [53:50<08:18, 2.48it/s, tok/s=15901.8]
Inference: 79%|███████▉ | 4685/5920 [53:50<08:18, 2.48it/s, tok/s=15897.8]
Inference: 79%|███████▉ | 4686/5920 [53:51<07:32, 2.73it/s, tok/s=15897.8]
Inference: 79%|███████▉ | 4686/5920 [53:51<07:32, 2.73it/s, tok/s=15898.3]
Inference: 79%|███████▉ | 4687/5920 [53:52<11:01, 1.86it/s, tok/s=15898.3]
Inference: 79%|███████▉ | 4687/5920 [53:52<11:01, 1.86it/s, tok/s=15893.7]
Inference: 79%|███████▉ | 4688/5920 [53:52<09:11, 2.23it/s, tok/s=15893.7]
Inference: 79%|███████▉ | 4688/5920 [53:52<09:11, 2.23it/s, tok/s=15893.0]
Inference: 79%|███████▉ | 4689/5920 [53:52<08:12, 2.50it/s, tok/s=15893.0]
Inference: 79%|███████▉ | 4689/5920 [53:52<08:12, 2.50it/s, tok/s=15892.3]
Inference: 79%|███████▉ | 4690/5920 [53:52<08:12, 2.50it/s, tok/s=15893.7]
Inference: 79%|███████▉ | 4691/5920 [53:52<08:11, 2.50it/s, tok/s=15894.1]
Inference: 79%|███████▉ | 4692/5920 [53:52<04:10, 4.90it/s, tok/s=15894.1]
Inference: 79%|███████▉ | 4692/5920 [53:52<04:10, 4.90it/s, tok/s=15895.8]
Inference: 79%|███████▉ | 4693/5920 [53:52<04:11, 4.87it/s, tok/s=15895.8]
Inference: 79%|███████▉ | 4693/5920 [53:52<04:11, 4.87it/s, tok/s=15895.2]
Inference: 79%|███████▉ | 4694/5920 [53:53<04:11, 4.87it/s, tok/s=15898.4]
Inference: 79%|███████▉ | 4695/5920 [53:53<03:08, 6.49it/s, tok/s=15898.4]
Inference: 79%|███████▉ | 4695/5920 [53:53<03:08, 6.49it/s, tok/s=15898.4]
Inference: 79%|███████▉ | 4696/5920 [53:53<03:08, 6.49it/s, tok/s=15897.7]
Inference: 79%|███████▉ | 4697/5920 [53:53<03:13, 6.31it/s, tok/s=15897.7]
Inference: 79%|███████▉ | 4697/5920 [53:53<03:13, 6.31it/s, tok/s=15897.8]
Inference: 79%|███████▉ | 4698/5920 [53:53<04:11, 4.86it/s, tok/s=15897.8]
Inference: 79%|███████▉ | 4698/5920 [53:53<04:11, 4.86it/s, tok/s=15896.5]
Inference: 79%|███████▉ | 4699/5920 [53:54<04:22, 4.65it/s, tok/s=15896.5]
Inference: 79%|███████▉ | 4699/5920 [53:54<04:22, 4.65it/s, tok/s=15897.8]
Inference: 79%|███████▉ | 4700/5920 [53:54<04:22, 4.65it/s, tok/s=15898.5]
Inference: 79%|███████▉ | 4701/5920 [53:54<06:01, 3.38it/s, tok/s=15898.5]
Inference: 79%|███████▉ | 4701/5920 [53:54<06:01, 3.38it/s, tok/s=15898.4]
Inference: 79%|███████▉ | 4702/5920 [53:55<05:15, 3.86it/s, tok/s=15898.4]
Inference: 79%|███████▉ | 4702/5920 [53:55<05:15, 3.86it/s, tok/s=15900.1]
Inference: 79%|███████▉ | 4703/5920 [53:55<05:40, 3.57it/s, tok/s=15900.1]
Inference: 79%|███████▉ | 4703/5920 [53:55<05:40, 3.57it/s, tok/s=15898.7]
Inference: 79%|███████▉ | 4704/5920 [53:55<05:40, 3.57it/s, tok/s=15900.0]
Inference: 79%|███████▉ | 4705/5920 [53:55<04:07, 4.92it/s, tok/s=15900.0]
Inference: 79%|███████▉ | 4705/5920 [53:55<04:07, 4.92it/s, tok/s=15899.8]
Inference: 79%|███████▉ | 4706/5920 [53:55<04:03, 4.98it/s, tok/s=15899.8]
Inference: 79%|███████▉ | 4706/5920 [53:55<04:03, 4.98it/s, tok/s=15900.2]
Inference: 80%|███████▉ | 4707/5920 [53:56<05:23, 3.75it/s, tok/s=15900.2]
Inference: 80%|███████▉ | 4707/5920 [53:56<05:23, 3.75it/s, tok/s=15898.3]
Inference: 80%|███████▉ | 4708/5920 [53:56<04:36, 4.39it/s, tok/s=15898.3]
Inference: 80%|███████▉ | 4708/5920 [53:56<04:36, 4.39it/s, tok/s=15899.3]
Inference: 80%|███████▉ | 4709/5920 [53:57<07:37, 2.65it/s, tok/s=15899.3]
Inference: 80%|███████▉ | 4709/5920 [53:57<07:37, 2.65it/s, tok/s=15895.8]
Inference: 80%|███████▉ | 4710/5920 [53:57<06:06, 3.30it/s, tok/s=15895.8]
Inference: 80%|███████▉ | 4710/5920 [53:57<06:06, 3.30it/s, tok/s=15898.3]
Inference: 80%|███████▉ | 4711/5920 [53:57<05:47, 3.48it/s, tok/s=15898.3]
Inference: 80%|███████▉ | 4711/5920 [53:57<05:47, 3.48it/s, tok/s=15898.5]
Inference: 80%|███████▉ | 4712/5920 [53:57<05:46, 3.48it/s, tok/s=15899.3]
Inference: 80%|███████▉ | 4713/5920 [53:58<05:30, 3.66it/s, tok/s=15899.3]
Inference: 80%|███████▉ | 4713/5920 [53:58<05:30, 3.66it/s, tok/s=15897.8]
Inference: 80%|███████▉ | 4714/5920 [53:58<04:40, 4.29it/s, tok/s=15897.8]
Inference: 80%|███████▉ | 4714/5920 [53:58<04:40, 4.29it/s, tok/s=15898.8]
Inference: 80%|███████▉ | 4715/5920 [53:58<07:01, 2.86it/s, tok/s=15898.8]
Inference: 80%|███████▉ | 4715/5920 [53:58<07:01, 2.86it/s, tok/s=15897.2]
Inference: 80%|███████▉ | 4716/5920 [53:58<07:01, 2.86it/s, tok/s=15899.8]
Inference: 80%|███████▉ | 4717/5920 [53:58<07:00, 2.86it/s, tok/s=15901.0]
Inference: 80%|███████▉ | 4718/5920 [53:58<03:49, 5.23it/s, tok/s=15901.0]
Inference: 80%|███████▉ | 4718/5920 [53:58<03:49, 5.23it/s, tok/s=15910.0]
Inference: 80%|███████▉ | 4719/5920 [53:59<03:49, 5.23it/s, tok/s=15917.5]
Inference: 80%|███████▉ | 4720/5920 [53:59<03:50, 5.20it/s, tok/s=15917.5]
Inference: 80%|███████▉ | 4720/5920 [53:59<03:50, 5.20it/s, tok/s=15916.4]
Inference: 80%|███████▉ | 4721/5920 [53:59<03:33, 5.61it/s, tok/s=15916.4]
Inference: 80%|███████▉ | 4721/5920 [53:59<03:33, 5.61it/s, tok/s=15917.1]
Inference: 80%|███████▉ | 4722/5920 [53:59<03:19, 6.02it/s, tok/s=15917.1]
Inference: 80%|███████▉ | 4722/5920 [53:59<03:19, 6.02it/s, tok/s=15917.2]
Inference: 80%|███████▉ | 4723/5920 [54:00<04:53, 4.08it/s, tok/s=15917.2]
Inference: 80%|███████▉ | 4723/5920 [54:00<04:53, 4.08it/s, tok/s=15915.6]
Inference: 80%|███████▉ | 4724/5920 [54:00<04:53, 4.08it/s, tok/s=15916.2]
Inference: 80%|███████▉ | 4725/5920 [54:00<03:32, 5.62it/s, tok/s=15916.2]
Inference: 80%|███████▉ | 4725/5920 [54:00<03:32, 5.62it/s, tok/s=15918.9]
Inference: 80%|███████▉ | 4726/5920 [54:00<05:16, 3.78it/s, tok/s=15918.9]
Inference: 80%|███████▉ | 4726/5920 [54:00<05:16, 3.78it/s, tok/s=15918.0]
Inference: 80%|███████▉ | 4727/5920 [54:01<06:11, 3.22it/s, tok/s=15918.0]
Inference: 80%|███████▉ | 4727/5920 [54:01<06:11, 3.22it/s, tok/s=15925.1]
Inference: 80%|███████▉ | 4728/5920 [54:01<06:47, 2.92it/s, tok/s=15925.1]
Inference: 80%|███████▉ | 4728/5920 [54:01<06:47, 2.92it/s, tok/s=15923.5]
Inference: 80%|███████▉ | 4729/5920 [54:02<09:15, 2.14it/s, tok/s=15923.5]
Inference: 80%|███████▉ | 4729/5920 [54:02<09:15, 2.14it/s, tok/s=15920.8]
Inference: 80%|███████▉ | 4730/5920 [54:02<09:15, 2.14it/s, tok/s=15929.8]
Inference: 80%|███████▉ | 4731/5920 [54:02<06:56, 2.85it/s, tok/s=15929.8]
Inference: 80%|███████▉ | 4731/5920 [54:02<06:56, 2.85it/s, tok/s=15933.0]
Inference: 80%|███████▉ | 4732/5920 [54:03<07:49, 2.53it/s, tok/s=15933.0]
Inference: 80%|███████▉ | 4732/5920 [54:03<07:49, 2.53it/s, tok/s=15931.5]
Inference: 80%|███████▉ | 4733/5920 [54:03<07:01, 2.82it/s, tok/s=15931.5]
Inference: 80%|███████▉ | 4733/5920 [54:03<07:01, 2.82it/s, tok/s=15932.3]
Inference: 80%|███████▉ | 4734/5920 [54:03<07:01, 2.82it/s, tok/s=15941.0]
Inference: 80%|███████▉ | 4735/5920 [54:04<05:22, 3.67it/s, tok/s=15941.0]
Inference: 80%|███████▉ | 4735/5920 [54:04<05:22, 3.67it/s, tok/s=15941.6]
Inference: 80%|████████ | 4736/5920 [54:04<05:41, 3.47it/s, tok/s=15941.6]
Inference: 80%|████████ | 4736/5920 [54:04<05:41, 3.47it/s, tok/s=15942.4]
Inference: 80%|████████ | 4737/5920 [54:04<07:19, 2.69it/s, tok/s=15942.4]
Inference: 80%|████████ | 4737/5920 [54:04<07:19, 2.69it/s, tok/s=15939.8]
Inference: 80%|████████ | 4738/5920 [54:05<07:18, 2.69it/s, tok/s=15940.5]
Inference: 80%|████████ | 4739/5920 [54:05<04:50, 4.07it/s, tok/s=15940.5]
Inference: 80%|████████ | 4739/5920 [54:05<04:50, 4.07it/s, tok/s=15940.5]
Inference: 80%|████████ | 4740/5920 [54:05<04:40, 4.20it/s, tok/s=15940.5]
Inference: 80%|████████ | 4740/5920 [54:05<04:40, 4.20it/s, tok/s=15940.4]
Inference: 80%|████████ | 4741/5920 [54:05<04:40, 4.20it/s, tok/s=15941.3]
Inference: 80%|████████ | 4742/5920 [54:05<04:39, 4.22it/s, tok/s=15941.3]
Inference: 80%|████████ | 4742/5920 [54:05<04:39, 4.22it/s, tok/s=15939.8]
Inference: 80%|████████ | 4743/5920 [54:05<04:11, 4.68it/s, tok/s=15939.8]
Inference: 80%|████████ | 4743/5920 [54:05<04:11, 4.68it/s, tok/s=15939.7]
Inference: 80%|████████ | 4744/5920 [54:06<04:11, 4.68it/s, tok/s=15943.7]
Inference: 80%|████████ | 4745/5920 [54:06<04:11, 4.68it/s, tok/s=15944.2]
Inference: 80%|████████ | 4746/5920 [54:06<03:29, 5.60it/s, tok/s=15944.2]
Inference: 80%|████████ | 4746/5920 [54:06<03:29, 5.60it/s, tok/s=15946.8]
Inference: 80%|████████ | 4747/5920 [54:06<03:26, 5.69it/s, tok/s=15946.8]
Inference: 80%|████████ | 4747/5920 [54:06<03:26, 5.69it/s, tok/s=15946.7]
Inference: 80%|████████ | 4748/5920 [54:06<04:06, 4.75it/s, tok/s=15946.7]
Inference: 80%|████████ | 4748/5920 [54:06<04:06, 4.75it/s, tok/s=15947.3]
Inference: 80%|████████ | 4749/5920 [54:07<04:13, 4.61it/s, tok/s=15947.3]
Inference: 80%|████████ | 4749/5920 [54:07<04:13, 4.61it/s, tok/s=15951.5]
Inference: 80%|████████ | 4750/5920 [54:07<04:30, 4.32it/s, tok/s=15951.5]
Inference: 80%|████████ | 4750/5920 [54:07<04:30, 4.32it/s, tok/s=15956.5]
Inference: 80%|████████ | 4751/5920 [54:07<04:30, 4.32it/s, tok/s=15956.9]
Inference: 80%|████████ | 4752/5920 [54:07<04:06, 4.73it/s, tok/s=15956.9]
Inference: 80%|████████ | 4752/5920 [54:07<04:06, 4.73it/s, tok/s=15957.2]
Inference: 80%|████████ | 4753/5920 [54:08<05:40, 3.43it/s, tok/s=15957.2]
Inference: 80%|████████ | 4753/5920 [54:08<05:40, 3.43it/s, tok/s=15956.8]
Inference: 80%|████████ | 4754/5920 [54:08<04:48, 4.05it/s, tok/s=15956.8]
Inference: 80%|████████ | 4754/5920 [54:08<04:48, 4.05it/s, tok/s=15958.3]
Inference: 80%|████████ | 4755/5920 [54:08<04:33, 4.26it/s, tok/s=15958.3]
Inference: 80%|████████ | 4755/5920 [54:08<04:33, 4.26it/s, tok/s=15959.3]
Inference: 80%|████████ | 4756/5920 [54:08<04:33, 4.26it/s, tok/s=15960.0]
Inference: 80%|████████ | 4757/5920 [54:09<07:21, 2.63it/s, tok/s=15960.0]
Inference: 80%|████████ | 4757/5920 [54:09<07:21, 2.63it/s, tok/s=15955.3]
Inference: 80%|████████ | 4758/5920 [54:09<07:21, 2.63it/s, tok/s=15955.6]
Inference: 80%|████████ | 4759/5920 [54:09<04:59, 3.87it/s, tok/s=15955.6]
Inference: 80%|████████ | 4759/5920 [54:09<04:59, 3.87it/s, tok/s=15955.9]
Inference: 80%|████████ | 4760/5920 [54:09<04:59, 3.87it/s, tok/s=15958.2]
Inference: 80%|████████ | 4761/5920 [54:09<04:59, 3.87it/s, tok/s=15959.1]
Inference: 80%|████████ | 4762/5920 [54:10<04:21, 4.44it/s, tok/s=15959.1]
Inference: 80%|████████ | 4762/5920 [54:10<04:21, 4.44it/s, tok/s=15957.4]
Inference: 80%|████████ | 4763/5920 [54:10<04:20, 4.44it/s, tok/s=15958.6]
Inference: 80%|████████ | 4764/5920 [54:10<03:27, 5.58it/s, tok/s=15958.6]
Inference: 80%|████████ | 4764/5920 [54:10<03:27, 5.58it/s, tok/s=15960.7]
Inference: 80%|████████ | 4765/5920 [54:10<03:52, 4.98it/s, tok/s=15960.7]
Inference: 80%|████████ | 4765/5920 [54:10<03:52, 4.98it/s, tok/s=15959.8]
Inference: 81%|████████ | 4766/5920 [54:11<04:04, 4.72it/s, tok/s=15959.8]
Inference: 81%|████████ | 4766/5920 [54:11<04:04, 4.72it/s, tok/s=15961.6]
Inference: 81%|████████ | 4767/5920 [54:11<06:11, 3.11it/s, tok/s=15961.6]
Inference: 81%|████████ | 4767/5920 [54:11<06:11, 3.11it/s, tok/s=15958.9]
Inference: 81%|████████ | 4768/5920 [54:11<06:10, 3.11it/s, tok/s=15960.3]
Inference: 81%|████████ | 4769/5920 [54:12<05:21, 3.58it/s, tok/s=15960.3]
Inference: 81%|████████ | 4769/5920 [54:12<05:21, 3.58it/s, tok/s=15959.2]
Inference: 81%|████████ | 4770/5920 [54:12<06:03, 3.16it/s, tok/s=15959.2]
Inference: 81%|████████ | 4770/5920 [54:12<06:03, 3.16it/s, tok/s=15958.1]
Inference: 81%|████████ | 4771/5920 [54:12<05:36, 3.42it/s, tok/s=15958.1]
Inference: 81%|████████ | 4771/5920 [54:12<05:36, 3.42it/s, tok/s=15957.5]
Inference: 81%|████████ | 4772/5920 [54:13<05:26, 3.52it/s, tok/s=15957.5]
Inference: 81%|████████ | 4772/5920 [54:13<05:26, 3.52it/s, tok/s=15959.0]
Inference: 81%|████████ | 4773/5920 [54:13<06:10, 3.10it/s, tok/s=15959.0]
Inference: 81%|████████ | 4773/5920 [54:13<06:10, 3.10it/s, tok/s=15957.6]
Inference: 81%|████████ | 4774/5920 [54:13<05:58, 3.20it/s, tok/s=15957.6]
Inference: 81%|████████ | 4774/5920 [54:13<05:58, 3.20it/s, tok/s=15958.7]
Inference: 81%|████████ | 4775/5920 [54:14<05:26, 3.51it/s, tok/s=15958.7]
Inference: 81%|████████ | 4775/5920 [54:14<05:26, 3.51it/s, tok/s=15958.6]
Inference: 81%|████████ | 4776/5920 [54:14<04:40, 4.08it/s, tok/s=15958.6]
Inference: 81%|████████ | 4776/5920 [54:14<04:40, 4.08it/s, tok/s=15959.0]
Inference: 81%|████████ | 4777/5920 [54:14<04:50, 3.93it/s, tok/s=15959.0]
Inference: 81%|████████ | 4777/5920 [54:14<04:50, 3.93it/s, tok/s=15958.8]
Inference: 81%|████████ | 4778/5920 [54:14<04:22, 4.35it/s, tok/s=15958.8]
Inference: 81%|████████ | 4778/5920 [54:14<04:22, 4.35it/s, tok/s=15958.5]
Inference: 81%|████████ | 4779/5920 [54:15<06:06, 3.12it/s, tok/s=15958.5]
Inference: 81%|████████ | 4779/5920 [54:15<06:06, 3.12it/s, tok/s=15957.7]
Inference: 81%|████████ | 4780/5920 [54:15<08:22, 2.27it/s, tok/s=15957.7]
Inference: 81%|████████ | 4780/5920 [54:15<08:22, 2.27it/s, tok/s=15955.4]
Inference: 81%|████████ | 4781/5920 [54:16<07:01, 2.70it/s, tok/s=15955.4]
Inference: 81%|████████ | 4781/5920 [54:16<07:01, 2.70it/s, tok/s=15954.8]
Inference: 81%|████████ | 4782/5920 [54:16<06:35, 2.87it/s, tok/s=15954.8]
Inference: 81%|████████ | 4782/5920 [54:16<06:35, 2.87it/s, tok/s=15954.0]
Inference: 81%|████████ | 4783/5920 [54:16<06:29, 2.92it/s, tok/s=15954.0]
Inference: 81%|████████ | 4783/5920 [54:16<06:29, 2.92it/s, tok/s=15953.2]
Inference: 81%|████████ | 4784/5920 [54:17<06:19, 2.99it/s, tok/s=15953.2]
Inference: 81%|████████ | 4784/5920 [54:17<06:19, 2.99it/s, tok/s=15955.2]
Inference: 81%|████████ | 4785/5920 [54:17<05:41, 3.32it/s, tok/s=15955.2]
Inference: 81%|████████ | 4785/5920 [54:17<05:41, 3.32it/s, tok/s=15954.5]
Inference: 81%|████████ | 4786/5920 [54:17<05:41, 3.32it/s, tok/s=15957.1]
Inference: 81%|████████ | 4787/5920 [54:17<03:56, 4.79it/s, tok/s=15957.1]
Inference: 81%|████████ | 4787/5920 [54:17<03:56, 4.79it/s, tok/s=15959.0]
Inference: 81%|████████ | 4788/5920 [54:17<04:33, 4.14it/s, tok/s=15959.0]
Inference: 81%|████████ | 4788/5920 [54:17<04:33, 4.14it/s, tok/s=15958.4]
Inference: 81%|████████ | 4789/5920 [54:18<04:01, 4.69it/s, tok/s=15958.4]
Inference: 81%|████████ | 4789/5920 [54:18<04:01, 4.69it/s, tok/s=15958.8]
Inference: 81%|████████ | 4790/5920 [54:18<04:14, 4.44it/s, tok/s=15958.8]
Inference: 81%|████████ | 4790/5920 [54:18<04:14, 4.44it/s, tok/s=15957.9]
Inference: 81%|████████ | 4791/5920 [54:18<04:14, 4.44it/s, tok/s=15958.3]
Inference: 81%|████████ | 4792/5920 [54:19<05:36, 3.35it/s, tok/s=15958.3]
Inference: 81%|████████ | 4792/5920 [54:19<05:36, 3.35it/s, tok/s=15963.4]
Inference: 81%|████████ | 4793/5920 [54:19<05:51, 3.20it/s, tok/s=15963.4]
Inference: 81%|████████ | 4793/5920 [54:19<05:51, 3.20it/s, tok/s=15962.9]
Inference: 81%|████████ | 4794/5920 [54:20<07:54, 2.37it/s, tok/s=15962.9]
Inference: 81%|████████ | 4794/5920 [54:20<07:54, 2.37it/s, tok/s=15968.5]
Inference: 81%|████████ | 4795/5920 [54:20<07:54, 2.37it/s, tok/s=15970.4]
Inference: 81%|████████ | 4796/5920 [54:20<07:53, 2.37it/s, tok/s=15970.6]
Inference: 81%|████████ | 4797/5920 [54:20<07:53, 2.37it/s, tok/s=15970.9]
Inference: 81%|████████ | 4798/5920 [54:20<04:31, 4.14it/s, tok/s=15970.9]
Inference: 81%|████████ | 4798/5920 [54:20<04:31, 4.14it/s, tok/s=15973.8]
Inference: 81%|████████ | 4799/5920 [54:20<04:30, 4.14it/s, tok/s=15974.9]
Inference: 81%|████████ | 4800/5920 [54:20<04:30, 4.14it/s, tok/s=15978.1]
Inference: 81%|████████ | 4801/5920 [54:20<03:28, 5.36it/s, tok/s=15978.1]
Inference: 81%|████████ | 4801/5920 [54:20<03:28, 5.36it/s, tok/s=15977.7]
Inference: 81%|████████ | 4802/5920 [54:21<03:38, 5.11it/s, tok/s=15977.7]
Inference: 81%|████████ | 4802/5920 [54:21<03:38, 5.11it/s, tok/s=15976.9]
Inference: 81%|████████ | 4803/5920 [54:21<03:38, 5.11it/s, tok/s=15977.5]
Inference: 81%|████████ | 4804/5920 [54:21<03:38, 5.11it/s, tok/s=15979.1]
Inference: 81%|████████ | 4805/5920 [54:21<02:32, 7.30it/s, tok/s=15979.1]
Inference: 81%|████████ | 4805/5920 [54:21<02:32, 7.30it/s, tok/s=15979.8]
Inference: 81%|████████ | 4806/5920 [54:21<02:32, 7.30it/s, tok/s=15977.0]
Inference: 81%|████████ | 4807/5920 [54:22<03:55, 4.72it/s, tok/s=15977.0]
Inference: 81%|████████ | 4807/5920 [54:22<03:55, 4.72it/s, tok/s=15981.3]
Inference: 81%|████████ | 4808/5920 [54:22<03:55, 4.72it/s, tok/s=15982.5]
Inference: 81%|████████ | 4809/5920 [54:22<03:08, 5.89it/s, tok/s=15982.5]
Inference: 81%|████████ | 4809/5920 [54:22<03:08, 5.89it/s, tok/s=15986.5]
Inference: 81%|████████▏ | 4810/5920 [54:22<03:08, 5.89it/s, tok/s=15985.2]
Inference: 81%|████████▏ | 4811/5920 [54:22<04:10, 4.43it/s, tok/s=15985.2]
Inference: 81%|████████▏ | 4811/5920 [54:22<04:10, 4.43it/s, tok/s=15984.6]
Inference: 81%|████████▏ | 4812/5920 [54:23<04:09, 4.43it/s, tok/s=15985.0]
Inference: 81%|████████▏ | 4813/5920 [54:23<03:16, 5.63it/s, tok/s=15985.0]
Inference: 81%|████████▏ | 4813/5920 [54:23<03:16, 5.63it/s, tok/s=15985.7]
Inference: 81%|████████▏ | 4814/5920 [54:23<03:16, 5.63it/s, tok/s=15988.0]
Inference: 81%|████████▏ | 4815/5920 [54:24<06:36, 2.79it/s, tok/s=15988.0]
Inference: 81%|████████▏ | 4815/5920 [54:24<06:36, 2.79it/s, tok/s=15982.1]
Inference: 81%|████████▏ | 4816/5920 [54:24<05:58, 3.08it/s, tok/s=15982.1]
Inference: 81%|████████▏ | 4816/5920 [54:24<05:58, 3.08it/s, tok/s=15982.0]
Inference: 81%|████████▏ | 4817/5920 [54:25<05:40, 3.24it/s, tok/s=15982.0]
Inference: 81%|████████▏ | 4817/5920 [54:25<05:40, 3.24it/s, tok/s=15984.5]
Inference: 81%|████████▏ | 4818/5920 [54:25<05:02, 3.64it/s, tok/s=15984.5]
Inference: 81%|████████▏ | 4818/5920 [54:25<05:02, 3.64it/s, tok/s=15985.0]
Inference: 81%|████████▏ | 4819/5920 [54:25<05:02, 3.64it/s, tok/s=15985.8]
Inference: 81%|████████▏ | 4820/5920 [54:25<05:01, 3.64it/s, tok/s=15986.2]
Inference: 81%|████████▏ | 4821/5920 [54:25<03:27, 5.30it/s, tok/s=15986.2]
Inference: 81%|████████▏ | 4821/5920 [54:25<03:27, 5.30it/s, tok/s=15986.8]
Inference: 81%|████████▏ | 4822/5920 [54:25<03:09, 5.80it/s, tok/s=15986.8]
Inference: 81%|████████▏ | 4822/5920 [54:25<03:09, 5.80it/s, tok/s=15994.9]
Inference: 81%|████████▏ | 4823/5920 [54:25<03:45, 4.87it/s, tok/s=15994.9]
Inference: 81%|████████▏ | 4823/5920 [54:25<03:45, 4.87it/s, tok/s=15998.3]
Inference: 81%|████████▏ | 4824/5920 [54:26<03:45, 4.87it/s, tok/s=15999.1]
Inference: 82%|████████▏ | 4825/5920 [54:26<03:12, 5.70it/s, tok/s=15999.1]
Inference: 82%|████████▏ | 4825/5920 [54:26<03:12, 5.70it/s, tok/s=15999.3]
Inference: 82%|████████▏ | 4826/5920 [54:26<03:11, 5.70it/s, tok/s=15999.6]
Inference: 82%|████████▏ | 4827/5920 [54:26<02:46, 6.56it/s, tok/s=15999.6]
Inference: 82%|████████▏ | 4827/5920 [54:26<02:46, 6.56it/s, tok/s=16000.0]
Inference: 82%|████████▏ | 4828/5920 [54:26<02:40, 6.81it/s, tok/s=16000.0]
Inference: 82%|████████▏ | 4828/5920 [54:26<02:40, 6.81it/s, tok/s=16008.6]
Inference: 82%|████████▏ | 4829/5920 [54:26<02:36, 6.96it/s, tok/s=16008.6]
Inference: 82%|████████▏ | 4829/5920 [54:26<02:36, 6.96it/s, tok/s=16012.0]
Inference: 82%|████████▏ | 4830/5920 [54:26<02:36, 6.96it/s, tok/s=16015.1]
Inference: 82%|████████▏ | 4831/5920 [54:26<02:36, 6.96it/s, tok/s=16015.6]
Inference: 82%|████████▏ | 4832/5920 [54:26<02:36, 6.96it/s, tok/s=16018.0]
Inference: 82%|████████▏ | 4833/5920 [54:26<01:50, 9.88it/s, tok/s=16018.0]
Inference: 82%|████████▏ | 4833/5920 [54:26<01:50, 9.88it/s, tok/s=16018.9]
Inference: 82%|████████▏ | 4834/5920 [54:27<01:58, 9.13it/s, tok/s=16018.9]
Inference: 82%|████████▏ | 4834/5920 [54:27<01:58, 9.13it/s, tok/s=16020.8]
Inference: 82%|████████▏ | 4835/5920 [54:27<02:10, 8.35it/s, tok/s=16020.8]
Inference: 82%|████████▏ | 4835/5920 [54:27<02:10, 8.35it/s, tok/s=16024.9]
Inference: 82%|████████▏ | 4836/5920 [54:27<03:29, 5.16it/s, tok/s=16024.9]
Inference: 82%|████████▏ | 4836/5920 [54:27<03:29, 5.16it/s, tok/s=16028.4]
Inference: 82%|████████▏ | 4837/5920 [54:28<07:09, 2.52it/s, tok/s=16028.4]
Inference: 82%|████████▏ | 4837/5920 [54:28<07:09, 2.52it/s, tok/s=16026.8]
Inference: 82%|████████▏ | 4838/5920 [54:28<06:01, 2.99it/s, tok/s=16026.8]
Inference: 82%|████████▏ | 4838/5920 [54:28<06:01, 2.99it/s, tok/s=16026.7]
Inference: 82%|████████▏ | 4839/5920 [54:29<06:15, 2.88it/s, tok/s=16026.7]
Inference: 82%|████████▏ | 4839/5920 [54:29<06:15, 2.88it/s, tok/s=16025.2]
Inference: 82%|████████▏ | 4840/5920 [54:29<05:04, 3.55it/s, tok/s=16025.2]
Inference: 82%|████████▏ | 4840/5920 [54:29<05:04, 3.55it/s, tok/s=16025.3]
Inference: 82%|████████▏ | 4841/5920 [54:29<05:04, 3.55it/s, tok/s=16025.6]
Inference: 82%|████████▏ | 4842/5920 [54:29<03:27, 5.19it/s, tok/s=16025.6]
Inference: 82%|████████▏ | 4842/5920 [54:29<03:27, 5.19it/s, tok/s=16026.1]
Inference: 82%|████████▏ | 4843/5920 [54:29<03:13, 5.58it/s, tok/s=16026.1]
Inference: 82%|████████▏ | 4843/5920 [54:29<03:13, 5.58it/s, tok/s=16026.0]
Inference: 82%|████████▏ | 4844/5920 [54:30<04:03, 4.42it/s, tok/s=16026.0]
Inference: 82%|████████▏ | 4844/5920 [54:30<04:03, 4.42it/s, tok/s=16024.8]
Inference: 82%|████████▏ | 4845/5920 [54:30<04:03, 4.42it/s, tok/s=16026.1]
Inference: 82%|████████▏ | 4846/5920 [54:30<03:49, 4.68it/s, tok/s=16026.1]
Inference: 82%|████████▏ | 4846/5920 [54:30<03:49, 4.68it/s, tok/s=16026.2]
Inference: 82%|████████▏ | 4847/5920 [54:31<06:09, 2.90it/s, tok/s=16026.2]
Inference: 82%|████████▏ | 4847/5920 [54:31<06:09, 2.90it/s, tok/s=16023.3]
Inference: 82%|████████▏ | 4848/5920 [54:31<06:09, 2.90it/s, tok/s=16023.7]
Inference: 82%|████████▏ | 4849/5920 [54:31<04:27, 4.00it/s, tok/s=16023.7]
Inference: 82%|████████▏ | 4849/5920 [54:31<04:27, 4.00it/s, tok/s=16026.1]
Inference: 82%|████████▏ | 4850/5920 [54:31<04:30, 3.96it/s, tok/s=16026.1]
Inference: 82%|████████▏ | 4850/5920 [54:31<04:30, 3.96it/s, tok/s=16025.6]
Inference: 82%|████████▏ | 4851/5920 [54:32<08:25, 2.11it/s, tok/s=16025.6]
Inference: 82%|████████▏ | 4851/5920 [54:32<08:25, 2.11it/s, tok/s=16020.4]
Inference: 82%|████████▏ | 4852/5920 [54:33<06:59, 2.55it/s, tok/s=16020.4]
Inference: 82%|████████▏ | 4852/5920 [54:33<06:59, 2.55it/s, tok/s=16021.7]
Inference: 82%|████████▏ | 4853/5920 [54:33<06:59, 2.55it/s, tok/s=16022.1]
Inference: 82%|████████▏ | 4854/5920 [54:33<04:28, 3.97it/s, tok/s=16022.1]
Inference: 82%|████████▏ | 4854/5920 [54:33<04:28, 3.97it/s, tok/s=16022.7]
Inference: 82%|████████▏ | 4855/5920 [54:33<04:18, 4.13it/s, tok/s=16022.7]
Inference: 82%|████████▏ | 4855/5920 [54:33<04:18, 4.13it/s, tok/s=16023.6]
Inference: 82%|████████▏ | 4856/5920 [54:33<04:02, 4.39it/s, tok/s=16023.6]
Inference: 82%|████████▏ | 4856/5920 [54:33<04:02, 4.39it/s, tok/s=16023.4]
Inference: 82%|████████▏ | 4857/5920 [54:33<04:24, 4.02it/s, tok/s=16023.4]
Inference: 82%|████████▏ | 4857/5920 [54:33<04:24, 4.02it/s, tok/s=16023.9]
Inference: 82%|████████▏ | 4858/5920 [54:34<04:09, 4.25it/s, tok/s=16023.9]
Inference: 82%|████████▏ | 4858/5920 [54:34<04:09, 4.25it/s, tok/s=16023.9]
Inference: 82%|████████▏ | 4859/5920 [54:34<04:09, 4.25it/s, tok/s=16024.3]
Inference: 82%|████████▏ | 4860/5920 [54:34<04:09, 4.25it/s, tok/s=16026.0]
Inference: 82%|████████▏ | 4861/5920 [54:34<02:50, 6.19it/s, tok/s=16026.0]
Inference: 82%|████████▏ | 4861/5920 [54:34<02:50, 6.19it/s, tok/s=16025.6]
Inference: 82%|████████▏ | 4862/5920 [54:34<02:50, 6.19it/s, tok/s=16025.7]
Inference: 82%|████████▏ | 4863/5920 [54:34<02:16, 7.75it/s, tok/s=16025.7]
Inference: 82%|████████▏ | 4863/5920 [54:34<02:16, 7.75it/s, tok/s=16028.9]
Inference: 82%|████████▏ | 4864/5920 [54:34<02:29, 7.06it/s, tok/s=16028.9]
Inference: 82%|████████▏ | 4864/5920 [54:34<02:29, 7.06it/s, tok/s=16036.3]
Inference: 82%|████████▏ | 4865/5920 [54:34<02:29, 7.06it/s, tok/s=16041.6]
Inference: 82%|████████▏ | 4866/5920 [54:34<01:56, 9.01it/s, tok/s=16041.6]
Inference: 82%|████████▏ | 4866/5920 [54:34<01:56, 9.01it/s, tok/s=16042.6]
Inference: 82%|████████▏ | 4867/5920 [54:35<01:56, 9.01it/s, tok/s=16046.5]
Inference: 82%|████████▏ | 4868/5920 [54:35<03:06, 5.63it/s, tok/s=16046.5]
Inference: 82%|████████▏ | 4868/5920 [54:35<03:06, 5.63it/s, tok/s=16053.7]
Inference: 82%|████████▏ | 4869/5920 [54:35<03:06, 5.63it/s, tok/s=16055.4]
Inference: 82%|████████▏ | 4870/5920 [54:35<02:45, 6.36it/s, tok/s=16055.4]
Inference: 82%|████████▏ | 4870/5920 [54:35<02:45, 6.36it/s, tok/s=16057.1]
Inference: 82%|████████▏ | 4871/5920 [54:35<02:43, 6.41it/s, tok/s=16057.1]
Inference: 82%|████████▏ | 4871/5920 [54:35<02:43, 6.41it/s, tok/s=16057.1]
Inference: 82%|████████▏ | 4872/5920 [54:36<02:51, 6.11it/s, tok/s=16057.1]
Inference: 82%|████████▏ | 4872/5920 [54:36<02:51, 6.11it/s, tok/s=16059.4]
Inference: 82%|████████▏ | 4873/5920 [54:36<03:24, 5.12it/s, tok/s=16059.4]
Inference: 82%|████████▏ | 4873/5920 [54:36<03:24, 5.12it/s, tok/s=16060.1]
Inference: 82%|████████▏ | 4874/5920 [54:36<04:16, 4.08it/s, tok/s=16060.1]
Inference: 82%|████████▏ | 4874/5920 [54:36<04:16, 4.08it/s, tok/s=16062.6]
Inference: 82%|████████▏ | 4875/5920 [54:37<05:05, 3.42it/s, tok/s=16062.6]
Inference: 82%|████████▏ | 4875/5920 [54:37<05:05, 3.42it/s, tok/s=16069.2]
Inference: 82%|████████▏ | 4876/5920 [54:37<05:05, 3.42it/s, tok/s=16070.4]
Inference: 82%|████████▏ | 4877/5920 [54:37<03:45, 4.63it/s, tok/s=16070.4]
Inference: 82%|████████▏ | 4877/5920 [54:37<03:45, 4.63it/s, tok/s=16070.1]
Inference: 82%|████████▏ | 4878/5920 [54:37<03:47, 4.57it/s, tok/s=16070.1]
Inference: 82%|████████▏ | 4878/5920 [54:37<03:47, 4.57it/s, tok/s=16069.6]
Inference: 82%|████████▏ | 4879/5920 [54:37<03:50, 4.52it/s, tok/s=16069.6]
Inference: 82%|████████▏ | 4879/5920 [54:37<03:50, 4.52it/s, tok/s=16069.0]
Inference: 82%|████████▏ | 4880/5920 [54:37<03:26, 5.03it/s, tok/s=16069.0]
Inference: 82%|████████▏ | 4880/5920 [54:37<03:26, 5.03it/s, tok/s=16068.7]
Inference: 82%|████████▏ | 4881/5920 [54:38<04:41, 3.69it/s, tok/s=16068.7]
Inference: 82%|████████▏ | 4881/5920 [54:38<04:41, 3.69it/s, tok/s=16067.2]
Inference: 82%|████████▏ | 4882/5920 [54:38<05:17, 3.27it/s, tok/s=16067.2]
Inference: 82%|████████▏ | 4882/5920 [54:38<05:17, 3.27it/s, tok/s=16066.3]
Inference: 82%|████████▏ | 4883/5920 [54:38<04:28, 3.86it/s, tok/s=16066.3]
Inference: 82%|████████▏ | 4883/5920 [54:38<04:28, 3.86it/s, tok/s=16066.6]
Inference: 82%|████████▎ | 4884/5920 [54:39<03:41, 4.67it/s, tok/s=16066.6]
Inference: 82%|████████▎ | 4884/5920 [54:39<03:41, 4.67it/s, tok/s=16066.7]
Inference: 83%|████████▎ | 4885/5920 [54:39<03:37, 4.76it/s, tok/s=16066.7]
Inference: 83%|████████▎ | 4885/5920 [54:39<03:37, 4.76it/s, tok/s=16069.3]
Inference: 83%|████████▎ | 4886/5920 [54:39<03:37, 4.76it/s, tok/s=16070.1]
Inference: 83%|████████▎ | 4887/5920 [54:39<04:02, 4.26it/s, tok/s=16070.1]
Inference: 83%|████████▎ | 4887/5920 [54:39<04:02, 4.26it/s, tok/s=16068.3]
Inference: 83%|████████▎ | 4888/5920 [54:40<04:29, 3.82it/s, tok/s=16068.3]
Inference: 83%|████████▎ | 4888/5920 [54:40<04:29, 3.82it/s, tok/s=16068.6]
Inference: 83%|████████▎ | 4889/5920 [54:40<04:43, 3.64it/s, tok/s=16068.6]
Inference: 83%|████████▎ | 4889/5920 [54:40<04:43, 3.64it/s, tok/s=16067.5]
Inference: 83%|████████▎ | 4890/5920 [54:40<04:25, 3.88it/s, tok/s=16067.5]
Inference: 83%|████████▎ | 4890/5920 [54:40<04:25, 3.88it/s, tok/s=16067.3]
Inference: 83%|████████▎ | 4891/5920 [54:40<04:15, 4.03it/s, tok/s=16067.3]
Inference: 83%|████████▎ | 4891/5920 [54:40<04:15, 4.03it/s, tok/s=16066.8]
Inference: 83%|████████▎ | 4892/5920 [54:41<04:58, 3.44it/s, tok/s=16066.8]
Inference: 83%|████████▎ | 4892/5920 [54:41<04:58, 3.44it/s, tok/s=16066.0]
Inference: 83%|████████▎ | 4893/5920 [54:41<04:09, 4.12it/s, tok/s=16066.0]
Inference: 83%|████████▎ | 4893/5920 [54:41<04:09, 4.12it/s, tok/s=16065.9]
Inference: 83%|████████▎ | 4894/5920 [54:41<04:08, 4.12it/s, tok/s=16068.2]
Inference: 83%|████████▎ | 4895/5920 [54:41<02:48, 6.09it/s, tok/s=16068.2]
Inference: 83%|████████▎ | 4895/5920 [54:41<02:48, 6.09it/s, tok/s=16070.7]
Inference: 83%|████████▎ | 4896/5920 [54:41<03:00, 5.66it/s, tok/s=16070.7]
Inference: 83%|████████▎ | 4896/5920 [54:41<03:00, 5.66it/s, tok/s=16070.2]
Inference: 83%|████████▎ | 4897/5920 [54:42<03:47, 4.50it/s, tok/s=16070.2]
Inference: 83%|████████▎ | 4897/5920 [54:42<03:47, 4.50it/s, tok/s=16069.0]
Inference: 83%|████████▎ | 4898/5920 [54:42<03:23, 5.02it/s, tok/s=16069.0]
Inference: 83%|████████▎ | 4898/5920 [54:42<03:23, 5.02it/s, tok/s=16069.5]
Inference: 83%|████████▎ | 4899/5920 [54:42<04:04, 4.18it/s, tok/s=16069.5]
Inference: 83%|████████▎ | 4899/5920 [54:42<04:04, 4.18it/s, tok/s=16073.6]
Inference: 83%|████████▎ | 4900/5920 [54:42<03:26, 4.94it/s, tok/s=16073.6]
Inference: 83%|████████▎ | 4900/5920 [54:42<03:26, 4.94it/s, tok/s=16076.2]
Inference: 83%|████████▎ | 4901/5920 [54:42<03:26, 4.94it/s, tok/s=16078.8]
Inference: 83%|████████▎ | 4902/5920 [54:42<02:57, 5.72it/s, tok/s=16078.8]
Inference: 83%|████████▎ | 4902/5920 [54:42<02:57, 5.72it/s, tok/s=16078.3]
Inference: 83%|████████▎ | 4903/5920 [54:43<02:57, 5.72it/s, tok/s=16078.4]
Inference: 83%|████████▎ | 4904/5920 [54:43<03:41, 4.59it/s, tok/s=16078.4]
Inference: 83%|████████▎ | 4904/5920 [54:43<03:41, 4.59it/s, tok/s=16078.5]
Inference: 83%|████████▎ | 4905/5920 [54:44<05:01, 3.37it/s, tok/s=16078.5]
Inference: 83%|████████▎ | 4905/5920 [54:44<05:01, 3.37it/s, tok/s=16076.3]
Inference: 83%|████████▎ | 4906/5920 [54:44<05:12, 3.25it/s, tok/s=16076.3]
Inference: 83%|████████▎ | 4906/5920 [54:44<05:12, 3.25it/s, tok/s=16077.8]
Inference: 83%|████████▎ | 4907/5920 [54:44<05:17, 3.19it/s, tok/s=16077.8]
Inference: 83%|████████▎ | 4907/5920 [54:44<05:17, 3.19it/s, tok/s=16076.8]
Inference: 83%|████████▎ | 4908/5920 [54:46<10:11, 1.66it/s, tok/s=16076.8]
Inference: 83%|████████▎ | 4908/5920 [54:46<10:11, 1.66it/s, tok/s=16075.6]
Inference: 83%|████████▎ | 4909/5920 [54:46<10:10, 1.66it/s, tok/s=16078.8]
Inference: 83%|████████▎ | 4910/5920 [54:46<06:45, 2.49it/s, tok/s=16078.8]
Inference: 83%|████████▎ | 4910/5920 [54:46<06:45, 2.49it/s, tok/s=16078.7]
Inference: 83%|████████▎ | 4911/5920 [54:47<08:41, 1.94it/s, tok/s=16078.7]
Inference: 83%|████████▎ | 4911/5920 [54:47<08:41, 1.94it/s, tok/s=16083.5]
Inference: 83%|████████▎ | 4912/5920 [54:47<08:31, 1.97it/s, tok/s=16083.5]
Inference: 83%|████████▎ | 4912/5920 [54:47<08:31, 1.97it/s, tok/s=16081.8]
Inference: 83%|████████▎ | 4913/5920 [54:47<08:31, 1.97it/s, tok/s=16090.1]
Inference: 83%|████████▎ | 4914/5920 [54:48<05:31, 3.03it/s, tok/s=16090.1]
Inference: 83%|████████▎ | 4914/5920 [54:48<05:31, 3.03it/s, tok/s=16091.6]
Inference: 83%|████████▎ | 4915/5920 [54:48<05:57, 2.81it/s, tok/s=16091.6]
Inference: 83%|████████▎ | 4915/5920 [54:48<05:57, 2.81it/s, tok/s=16089.9]
Inference: 83%|████████▎ | 4916/5920 [54:48<06:05, 2.75it/s, tok/s=16089.9]
Inference: 83%|████████▎ | 4916/5920 [54:48<06:05, 2.75it/s, tok/s=16088.9]
Inference: 83%|████████▎ | 4917/5920 [54:48<04:59, 3.35it/s, tok/s=16088.9]
Inference: 83%|████████▎ | 4917/5920 [54:48<04:59, 3.35it/s, tok/s=16089.4]
Inference: 83%|████████▎ | 4918/5920 [54:49<05:39, 2.95it/s, tok/s=16089.4]
Inference: 83%|████████▎ | 4918/5920 [54:49<05:39, 2.95it/s, tok/s=16087.7]
Inference: 83%|████████▎ | 4919/5920 [54:49<05:05, 3.27it/s, tok/s=16087.7]
Inference: 83%|████████▎ | 4919/5920 [54:49<05:05, 3.27it/s, tok/s=16087.2]
Inference: 83%|████████▎ | 4920/5920 [54:50<07:41, 2.17it/s, tok/s=16087.2]
Inference: 83%|████████▎ | 4920/5920 [54:50<07:41, 2.17it/s, tok/s=16083.8]
Inference: 83%|████████▎ | 4921/5920 [54:50<07:41, 2.17it/s, tok/s=16084.4]
Inference: 83%|████████▎ | 4922/5920 [54:50<05:02, 3.30it/s, tok/s=16084.4]
Inference: 83%|████████▎ | 4922/5920 [54:50<05:02, 3.30it/s, tok/s=16084.2]
Inference: 83%|████████▎ | 4923/5920 [54:51<05:18, 3.13it/s, tok/s=16084.2]
Inference: 83%|████████▎ | 4923/5920 [54:51<05:18, 3.13it/s, tok/s=16083.0]
Inference: 83%|████████▎ | 4924/5920 [54:51<04:56, 3.36it/s, tok/s=16083.0]
Inference: 83%|████████▎ | 4924/5920 [54:51<04:56, 3.36it/s, tok/s=16082.9]
Inference: 83%|████████▎ | 4925/5920 [54:51<05:05, 3.26it/s, tok/s=16082.9]
Inference: 83%|████████▎ | 4925/5920 [54:51<05:05, 3.26it/s, tok/s=16084.0]
Inference: 83%|████████▎ | 4926/5920 [54:51<05:05, 3.26it/s, tok/s=16087.3]
Inference: 83%|████████▎ | 4927/5920 [54:51<03:30, 4.71it/s, tok/s=16087.3]
Inference: 83%|████████▎ | 4927/5920 [54:51<03:30, 4.71it/s, tok/s=16089.3]
Inference: 83%|████████▎ | 4928/5920 [54:51<03:13, 5.13it/s, tok/s=16089.3]
Inference: 83%|████████▎ | 4928/5920 [54:51<03:13, 5.13it/s, tok/s=16089.0]
Inference: 83%|████████▎ | 4929/5920 [54:52<02:49, 5.83it/s, tok/s=16089.0]
Inference: 83%|████████▎ | 4929/5920 [54:52<02:49, 5.83it/s, tok/s=16089.1]
Inference: 83%|████████▎ | 4930/5920 [54:52<02:49, 5.83it/s, tok/s=16089.8]
Inference: 83%|████████▎ | 4931/5920 [54:52<02:49, 5.83it/s, tok/s=16091.0]
Inference: 83%|████████▎ | 4932/5920 [54:52<03:21, 4.91it/s, tok/s=16091.0]
Inference: 83%|████████▎ | 4932/5920 [54:52<03:21, 4.91it/s, tok/s=16088.2]
Inference: 83%|████████▎ | 4933/5920 [54:52<03:21, 4.91it/s, tok/s=16090.9]
Inference: 83%|████████▎ | 4934/5920 [54:52<03:20, 4.91it/s, tok/s=16097.8]
Inference: 83%|████████▎ | 4935/5920 [54:53<03:22, 4.86it/s, tok/s=16097.8]
Inference: 83%|████████▎ | 4935/5920 [54:53<03:22, 4.86it/s, tok/s=16095.9]
Inference: 83%|████████▎ | 4936/5920 [54:54<05:10, 3.17it/s, tok/s=16095.9]
Inference: 83%|████████▎ | 4936/5920 [54:54<05:10, 3.17it/s, tok/s=16092.2]
Inference: 83%|████████▎ | 4937/5920 [54:54<05:34, 2.94it/s, tok/s=16092.2]
Inference: 83%|████████▎ | 4937/5920 [54:54<05:34, 2.94it/s, tok/s=16090.4]
Inference: 83%|████████▎ | 4938/5920 [54:54<05:17, 3.09it/s, tok/s=16090.4]
Inference: 83%|████████▎ | 4938/5920 [54:54<05:17, 3.09it/s, tok/s=16089.7]
Inference: 83%|████████▎ | 4939/5920 [54:55<04:51, 3.36it/s, tok/s=16089.7]
Inference: 83%|████████▎ | 4939/5920 [54:55<04:51, 3.36it/s, tok/s=16089.3]
Inference: 83%|████████▎ | 4940/5920 [54:55<04:51, 3.36it/s, tok/s=16091.7]
Inference: 83%|████████▎ | 4941/5920 [54:55<03:24, 4.80it/s, tok/s=16091.7]
Inference: 83%|████████▎ | 4941/5920 [54:55<03:24, 4.80it/s, tok/s=16092.6]
Inference: 83%|████████▎ | 4942/5920 [54:55<03:43, 4.38it/s, tok/s=16092.6]
Inference: 83%|████████▎ | 4942/5920 [54:55<03:43, 4.38it/s, tok/s=16091.7]
Inference: 83%|████████▎ | 4943/5920 [54:56<04:38, 3.51it/s, tok/s=16091.7]
Inference: 83%|████████▎ | 4943/5920 [54:56<04:38, 3.51it/s, tok/s=16090.0]
Inference: 84%|████████▎ | 4944/5920 [54:56<04:27, 3.65it/s, tok/s=16090.0]
Inference: 84%|████████▎ | 4944/5920 [54:56<04:27, 3.65it/s, tok/s=16090.3]
Inference: 84%|████████▎ | 4945/5920 [54:56<04:19, 3.75it/s, tok/s=16090.3]
Inference: 84%|████████▎ | 4945/5920 [54:56<04:19, 3.75it/s, tok/s=16089.9]
Inference: 84%|████████▎ | 4946/5920 [54:56<04:18, 3.77it/s, tok/s=16089.9]
Inference: 84%|████████▎ | 4946/5920 [54:56<04:18, 3.77it/s, tok/s=16090.5]
Inference: 84%|████████▎ | 4947/5920 [54:57<05:25, 2.99it/s, tok/s=16090.5]
Inference: 84%|████████▎ | 4947/5920 [54:57<05:25, 2.99it/s, tok/s=16088.4]
Inference: 84%|████████▎ | 4948/5920 [54:57<05:11, 3.12it/s, tok/s=16088.4]
Inference: 84%|████████▎ | 4948/5920 [54:57<05:11, 3.12it/s, tok/s=16094.1]
Inference: 84%|████████▎ | 4949/5920 [54:57<05:10, 3.12it/s, tok/s=16095.4]
Inference: 84%|████████▎ | 4950/5920 [54:58<04:50, 3.34it/s, tok/s=16095.4]
Inference: 84%|████████▎ | 4950/5920 [54:58<04:50, 3.34it/s, tok/s=16093.8]
Inference: 84%|████████▎ | 4951/5920 [54:58<04:50, 3.34it/s, tok/s=16094.2]
Inference: 84%|████████▎ | 4952/5920 [54:58<03:15, 4.95it/s, tok/s=16094.2]
Inference: 84%|████████▎ | 4952/5920 [54:58<03:15, 4.95it/s, tok/s=16094.1]
Inference: 84%|████████▎ | 4953/5920 [54:58<03:12, 5.02it/s, tok/s=16094.1]
Inference: 84%|████████▎ | 4953/5920 [54:58<03:12, 5.02it/s, tok/s=16094.6]
Inference: 84%|████████▎ | 4954/5920 [54:58<03:12, 5.02it/s, tok/s=16095.2]
Inference: 84%|████████▎ | 4955/5920 [54:58<02:18, 6.99it/s, tok/s=16095.2]
Inference: 84%|████████▎ | 4955/5920 [54:58<02:18, 6.99it/s, tok/s=16095.4]
Inference: 84%|████████▎ | 4956/5920 [54:58<02:17, 6.99it/s, tok/s=16099.6]
Inference: 84%|████████▎ | 4957/5920 [54:58<02:01, 7.92it/s, tok/s=16099.6]
Inference: 84%|████████▎ | 4957/5920 [54:58<02:01, 7.92it/s, tok/s=16101.8]
Inference: 84%|████████▍ | 4958/5920 [54:59<02:01, 7.92it/s, tok/s=16099.2]
Inference: 84%|████████▍ | 4959/5920 [54:59<03:35, 4.45it/s, tok/s=16099.2]
Inference: 84%|████████▍ | 4959/5920 [54:59<03:35, 4.45it/s, tok/s=16102.6]
Inference: 84%|████████▍ | 4960/5920 [54:59<03:35, 4.45it/s, tok/s=16104.9]
Inference: 84%|████████▍ | 4961/5920 [54:59<03:27, 4.62it/s, tok/s=16104.9]
Inference: 84%|████████▍ | 4961/5920 [54:59<03:27, 4.62it/s, tok/s=16111.3]
Inference: 84%|████████▍ | 4962/5920 [55:00<03:53, 4.10it/s, tok/s=16111.3]
Inference: 84%|████████▍ | 4962/5920 [55:00<03:53, 4.10it/s, tok/s=16115.8]
Inference: 84%|████████▍ | 4963/5920 [55:00<03:53, 4.10it/s, tok/s=16116.3]
Inference: 84%|████████▍ | 4964/5920 [55:00<03:16, 4.85it/s, tok/s=16116.3]
Inference: 84%|████████▍ | 4964/5920 [55:00<03:16, 4.85it/s, tok/s=16118.3]
Inference: 84%|████████▍ | 4965/5920 [55:00<03:17, 4.85it/s, tok/s=16118.3]
Inference: 84%|████████▍ | 4965/5920 [55:00<03:17, 4.85it/s, tok/s=16120.5]
Inference: 84%|████████▍ | 4966/5920 [55:00<03:16, 4.85it/s, tok/s=16121.2]
Inference: 84%|████████▍ | 4967/5920 [55:00<03:16, 4.85it/s, tok/s=16124.6]
Inference: 84%|████████▍ | 4968/5920 [55:01<02:34, 6.14it/s, tok/s=16124.6]
Inference: 84%|████████▍ | 4968/5920 [55:01<02:34, 6.14it/s, tok/s=16123.9]
Inference: 84%|████████▍ | 4969/5920 [55:01<02:27, 6.44it/s, tok/s=16123.9]
Inference: 84%|████████▍ | 4969/5920 [55:01<02:27, 6.44it/s, tok/s=16132.4]
Inference: 84%|████████▍ | 4970/5920 [55:01<02:41, 5.89it/s, tok/s=16132.4]
Inference: 84%|████████▍ | 4970/5920 [55:01<02:41, 5.89it/s, tok/s=16132.0]
Inference: 84%|████████▍ | 4971/5920 [55:01<02:41, 5.89it/s, tok/s=16132.7]
Inference: 84%|████████▍ | 4972/5920 [55:01<02:40, 5.89it/s, tok/s=16135.1]
Inference: 84%|████████▍ | 4973/5920 [55:02<03:12, 4.91it/s, tok/s=16135.1]
Inference: 84%|████████▍ | 4973/5920 [55:02<03:12, 4.91it/s, tok/s=16135.2]
Inference: 84%|████████▍ | 4974/5920 [55:02<03:24, 4.63it/s, tok/s=16135.2]
Inference: 84%|████████▍ | 4974/5920 [55:02<03:24, 4.63it/s, tok/s=16136.5]
Inference: 84%|████████▍ | 4975/5920 [55:02<03:24, 4.63it/s, tok/s=16136.7]
Inference: 84%|████████▍ | 4976/5920 [55:02<03:04, 5.13it/s, tok/s=16136.7]
Inference: 84%|████████▍ | 4976/5920 [55:02<03:04, 5.13it/s, tok/s=16136.3]
Inference: 84%|████████▍ | 4977/5920 [55:02<02:48, 5.59it/s, tok/s=16136.3]
Inference: 84%|████████▍ | 4977/5920 [55:02<02:48, 5.59it/s, tok/s=16142.9]
Inference: 84%|████████▍ | 4978/5920 [55:03<03:58, 3.95it/s, tok/s=16142.9]
Inference: 84%|████████▍ | 4978/5920 [55:03<03:58, 3.95it/s, tok/s=16142.3]
Inference: 84%|████████▍ | 4979/5920 [55:03<03:24, 4.60it/s, tok/s=16142.3]
Inference: 84%|████████▍ | 4979/5920 [55:03<03:24, 4.60it/s, tok/s=16142.7]
Inference: 84%|████████▍ | 4980/5920 [55:03<04:10, 3.75it/s, tok/s=16142.7]
Inference: 84%|████████▍ | 4980/5920 [55:03<04:10, 3.75it/s, tok/s=16141.4]
Inference: 84%|████████▍ | 4981/5920 [55:04<04:38, 3.37it/s, tok/s=16141.4]
Inference: 84%|████████▍ | 4981/5920 [55:04<04:38, 3.37it/s, tok/s=16140.9]
Inference: 84%|████████▍ | 4982/5920 [55:05<06:18, 2.48it/s, tok/s=16140.9]
Inference: 84%|████████▍ | 4982/5920 [55:05<06:18, 2.48it/s, tok/s=16138.7]
Inference: 84%|████████▍ | 4983/5920 [55:05<06:17, 2.48it/s, tok/s=16141.0]
Inference: 84%|████████▍ | 4984/5920 [55:05<04:32, 3.44it/s, tok/s=16141.0]
Inference: 84%|████████▍ | 4984/5920 [55:05<04:32, 3.44it/s, tok/s=16142.3]
Inference: 84%|████████▍ | 4985/5920 [55:05<04:30, 3.46it/s, tok/s=16142.3]
Inference: 84%|████████▍ | 4985/5920 [55:05<04:30, 3.46it/s, tok/s=16146.2]
Inference: 84%|████████▍ | 4986/5920 [55:05<03:55, 3.97it/s, tok/s=16146.2]
Inference: 84%|████████▍ | 4986/5920 [55:05<03:55, 3.97it/s, tok/s=16147.8]
Inference: 84%|████████▍ | 4987/5920 [55:06<06:28, 2.40it/s, tok/s=16147.8]
Inference: 84%|████████▍ | 4987/5920 [55:06<06:28, 2.40it/s, tok/s=16144.0]
Inference: 84%|████████▍ | 4988/5920 [55:08<11:32, 1.35it/s, tok/s=16144.0]
Inference: 84%|████████▍ | 4988/5920 [55:08<11:32, 1.35it/s, tok/s=16136.6]
Inference: 84%|████████▍ | 4989/5920 [55:08<11:31, 1.35it/s, tok/s=16137.4]
Inference: 84%|████████▍ | 4990/5920 [55:08<08:45, 1.77it/s, tok/s=16137.4]
Inference: 84%|████████▍ | 4990/5920 [55:08<08:45, 1.77it/s, tok/s=16136.0]
Inference: 84%|████████▍ | 4991/5920 [55:08<08:44, 1.77it/s, tok/s=16135.8]
Inference: 84%|████████▍ | 4992/5920 [55:09<05:54, 2.62it/s, tok/s=16135.8]
Inference: 84%|████████▍ | 4992/5920 [55:09<05:54, 2.62it/s, tok/s=16135.9]
Inference: 84%|████████▍ | 4993/5920 [55:09<05:35, 2.76it/s, tok/s=16135.9]
Inference: 84%|████████▍ | 4993/5920 [55:09<05:35, 2.76it/s, tok/s=16135.8]
Inference: 84%|████████▍ | 4994/5920 [55:09<05:35, 2.76it/s, tok/s=16136.6]
Inference: 84%|████████▍ | 4995/5920 [55:09<04:16, 3.60it/s, tok/s=16136.6]
Inference: 84%|████████▍ | 4995/5920 [55:09<04:16, 3.60it/s, tok/s=16137.3]
Inference: 84%|████████▍ | 4996/5920 [55:09<04:06, 3.75it/s, tok/s=16137.3]
Inference: 84%|████████▍ | 4996/5920 [55:09<04:06, 3.75it/s, tok/s=16139.4]
Inference: 84%|████████▍ | 4997/5920 [55:10<05:04, 3.03it/s, tok/s=16139.4]
Inference: 84%|████████▍ | 4997/5920 [55:10<05:04, 3.03it/s, tok/s=16137.9]
Inference: 84%|████████▍ | 4998/5920 [55:10<04:40, 3.29it/s, tok/s=16137.9]
Inference: 84%|████████▍ | 4998/5920 [55:10<04:40, 3.29it/s, tok/s=16137.3]
Inference: 84%|████████▍ | 4999/5920 [55:10<04:39, 3.29it/s, tok/s=16137.9]
Inference: 84%|████████▍ | 5000/5920 [55:10<03:08, 4.88it/s, tok/s=16137.9]
Inference: 84%|████████▍ | 5000/5920 [55:10<03:08, 4.88it/s, tok/s=16140.1]
Inference: 84%|████████▍ | 5001/5920 [55:10<03:08, 4.88it/s, tok/s=16142.9]
Inference: 84%|████████▍ | 5002/5920 [55:10<03:07, 4.88it/s, tok/s=16143.6]
Inference: 85%|████████▍ | 5003/5920 [55:11<03:07, 4.89it/s, tok/s=16143.6]
Inference: 85%|████████▍ | 5003/5920 [55:11<03:07, 4.89it/s, tok/s=16142.3]
Inference: 85%|████████▍ | 5004/5920 [55:11<03:08, 4.85it/s, tok/s=16142.3]
Inference: 85%|████████▍ | 5004/5920 [55:11<03:08, 4.85it/s, tok/s=16144.6]
Inference: 85%|████████▍ | 5005/5920 [55:11<03:03, 4.97it/s, tok/s=16144.6]
Inference: 85%|████████▍ | 5005/5920 [55:11<03:03, 4.97it/s, tok/s=16147.5]
Inference: 85%|████████▍ | 5006/5920 [55:11<02:47, 5.45it/s, tok/s=16147.5]
Inference: 85%|████████▍ | 5006/5920 [55:11<02:47, 5.45it/s, tok/s=16148.4]
Inference: 85%|████████▍ | 5007/5920 [55:11<02:47, 5.45it/s, tok/s=16149.0]
Inference: 85%|████████▍ | 5008/5920 [55:12<03:21, 4.52it/s, tok/s=16149.0]
Inference: 85%|████████▍ | 5008/5920 [55:12<03:21, 4.52it/s, tok/s=16147.2]
Inference: 85%|████████▍ | 5009/5920 [55:12<03:04, 4.94it/s, tok/s=16147.2]
Inference: 85%|████████▍ | 5009/5920 [55:12<03:04, 4.94it/s, tok/s=16148.8]
Inference: 85%|████████▍ | 5010/5920 [55:12<03:41, 4.10it/s, tok/s=16148.8]
Inference: 85%|████████▍ | 5010/5920 [55:12<03:41, 4.10it/s, tok/s=16148.0]
Inference: 85%|████████▍ | 5011/5920 [55:13<03:41, 4.10it/s, tok/s=16149.4]
Inference: 85%|████████▍ | 5012/5920 [55:13<03:28, 4.35it/s, tok/s=16149.4]
Inference: 85%|████████▍ | 5012/5920 [55:13<03:28, 4.35it/s, tok/s=16148.2]
Inference: 85%|████████▍ | 5013/5920 [55:13<03:28, 4.35it/s, tok/s=16149.9]
Inference: 85%|████████▍ | 5014/5920 [55:13<03:08, 4.81it/s, tok/s=16149.9]
Inference: 85%|████████▍ | 5014/5920 [55:13<03:08, 4.81it/s, tok/s=16150.2]
Inference: 85%|████████▍ | 5015/5920 [55:13<03:08, 4.81it/s, tok/s=16150.6]
Inference: 85%|████████▍ | 5016/5920 [55:13<02:35, 5.80it/s, tok/s=16150.6]
Inference: 85%|████████▍ | 5016/5920 [55:13<02:35, 5.80it/s, tok/s=16150.6]
Inference: 85%|████████▍ | 5017/5920 [55:14<02:35, 5.80it/s, tok/s=16152.2]
Inference: 85%|████████▍ | 5018/5920 [55:14<02:11, 6.88it/s, tok/s=16152.2]
Inference: 85%|████████▍ | 5018/5920 [55:14<02:11, 6.88it/s, tok/s=16152.3]
Inference: 85%|████████▍ | 5019/5920 [55:14<03:23, 4.42it/s, tok/s=16152.3]
Inference: 85%|████████▍ | 5019/5920 [55:14<03:23, 4.42it/s, tok/s=16151.0]
Inference: 85%|████████▍ | 5020/5920 [55:14<03:18, 4.55it/s, tok/s=16151.0]
Inference: 85%|████████▍ | 5020/5920 [55:14<03:18, 4.55it/s, tok/s=16150.9]
Inference: 85%|████████▍ | 5021/5920 [55:15<02:58, 5.03it/s, tok/s=16150.9]
Inference: 85%|████████▍ | 5021/5920 [55:15<02:58, 5.03it/s, tok/s=16152.4]
Inference: 85%|████████▍ | 5022/5920 [55:15<03:26, 4.34it/s, tok/s=16152.4]
Inference: 85%|████████▍ | 5022/5920 [55:15<03:26, 4.34it/s, tok/s=16151.8]
Inference: 85%|████████▍ | 5023/5920 [55:16<06:30, 2.30it/s, tok/s=16151.8]
Inference: 85%|████████▍ | 5023/5920 [55:16<06:30, 2.30it/s, tok/s=16147.5]
Inference: 85%|████████▍ | 5024/5920 [55:16<05:09, 2.90it/s, tok/s=16147.5]
Inference: 85%|████████▍ | 5024/5920 [55:16<05:09, 2.90it/s, tok/s=16148.5]
Inference: 85%|████████▍ | 5025/5920 [55:16<05:13, 2.86it/s, tok/s=16148.5]
Inference: 85%|████████▍ | 5025/5920 [55:16<05:13, 2.86it/s, tok/s=16149.9]
Inference: 85%|████████▍ | 5026/5920 [55:17<05:55, 2.51it/s, tok/s=16149.9]
Inference: 85%|████████▍ | 5026/5920 [55:17<05:55, 2.51it/s, tok/s=16148.4]
Inference: 85%|████████▍ | 5027/5920 [55:17<05:34, 2.67it/s, tok/s=16148.4]
Inference: 85%|████████▍ | 5027/5920 [55:17<05:34, 2.67it/s, tok/s=16149.7]
Inference: 85%|████████▍ | 5028/5920 [55:17<05:34, 2.67it/s, tok/s=16149.8]
Inference: 85%|████████▍ | 5029/5920 [55:18<06:26, 2.31it/s, tok/s=16149.8]
Inference: 85%|████████▍ | 5029/5920 [55:18<06:26, 2.31it/s, tok/s=16145.7]
Inference: 85%|████████▍ | 5030/5920 [55:18<05:25, 2.73it/s, tok/s=16145.7]
Inference: 85%|████████▍ | 5030/5920 [55:18<05:25, 2.73it/s, tok/s=16145.5]
Inference: 85%|████████▍ | 5031/5920 [55:19<07:38, 1.94it/s, tok/s=16145.5]
Inference: 85%|████████▍ | 5031/5920 [55:19<07:38, 1.94it/s, tok/s=16144.3]
Inference: 85%|████████▌ | 5032/5920 [55:19<07:38, 1.94it/s, tok/s=16145.5]
Inference: 85%|████████▌ | 5033/5920 [55:20<05:44, 2.58it/s, tok/s=16145.5]
Inference: 85%|████████▌ | 5033/5920 [55:20<05:44, 2.58it/s, tok/s=16144.8]
Inference: 85%|████████▌ | 5034/5920 [55:20<05:41, 2.60it/s, tok/s=16144.8]
Inference: 85%|████████▌ | 5034/5920 [55:20<05:41, 2.60it/s, tok/s=16144.1]
Inference: 85%|████████▌ | 5035/5920 [55:20<05:40, 2.60it/s, tok/s=16144.5]
Inference: 85%|████████▌ | 5036/5920 [55:20<04:02, 3.64it/s, tok/s=16144.5]
Inference: 85%|████████▌ | 5036/5920 [55:20<04:02, 3.64it/s, tok/s=16144.2]
Inference: 85%|████████▌ | 5037/5920 [55:20<03:39, 4.03it/s, tok/s=16144.2]
Inference: 85%|████████▌ | 5037/5920 [55:20<03:39, 4.03it/s, tok/s=16143.8]
Inference: 85%|████████▌ | 5038/5920 [55:21<03:57, 3.71it/s, tok/s=16143.8]
Inference: 85%|████████▌ | 5038/5920 [55:21<03:57, 3.71it/s, tok/s=16143.3]
Inference: 85%|████████▌ | 5039/5920 [55:21<03:45, 3.91it/s, tok/s=16143.3]
Inference: 85%|████████▌ | 5039/5920 [55:21<03:45, 3.91it/s, tok/s=16143.4]
Inference: 85%|████████▌ | 5040/5920 [55:21<04:28, 3.28it/s, tok/s=16143.4]
Inference: 85%|████████▌ | 5040/5920 [55:21<04:28, 3.28it/s, tok/s=16142.2]
Inference: 85%|████████▌ | 5041/5920 [55:22<04:27, 3.28it/s, tok/s=16142.4]
Inference: 85%|████████▌ | 5042/5920 [55:22<03:56, 3.71it/s, tok/s=16142.4]
Inference: 85%|████████▌ | 5042/5920 [55:22<03:56, 3.71it/s, tok/s=16141.3]
Inference: 85%|████████▌ | 5043/5920 [55:23<06:32, 2.24it/s, tok/s=16141.3]
Inference: 85%|████████▌ | 5043/5920 [55:23<06:32, 2.24it/s, tok/s=16136.9]
Inference: 85%|████████▌ | 5044/5920 [55:24<08:39, 1.69it/s, tok/s=16136.9]
Inference: 85%|████████▌ | 5044/5920 [55:24<08:39, 1.69it/s, tok/s=16135.2]
Inference: 85%|████████▌ | 5045/5920 [55:24<07:10, 2.03it/s, tok/s=16135.2]
Inference: 85%|████████▌ | 5045/5920 [55:24<07:10, 2.03it/s, tok/s=16138.1]
Inference: 85%|████████▌ | 5046/5920 [55:24<07:09, 2.03it/s, tok/s=16138.5]
Inference: 85%|████████▌ | 5047/5920 [55:24<07:09, 2.03it/s, tok/s=16144.5]
Inference: 85%|████████▌ | 5048/5920 [55:24<07:08, 2.03it/s, tok/s=16146.0]
Inference: 85%|████████▌ | 5049/5920 [55:24<03:16, 4.44it/s, tok/s=16146.0]
Inference: 85%|████████▌ | 5049/5920 [55:24<03:16, 4.44it/s, tok/s=16146.1]
Inference: 85%|████████▌ | 5050/5920 [55:25<04:24, 3.29it/s, tok/s=16146.1]
Inference: 85%|████████▌ | 5050/5920 [55:25<04:24, 3.29it/s, tok/s=16145.9]
Inference: 85%|████████▌ | 5051/5920 [55:25<04:19, 3.35it/s, tok/s=16145.9]
Inference: 85%|████████▌ | 5051/5920 [55:25<04:19, 3.35it/s, tok/s=16145.6]
Inference: 85%|████████▌ | 5052/5920 [55:26<04:50, 2.99it/s, tok/s=16145.6]
Inference: 85%|████████▌ | 5052/5920 [55:26<04:50, 2.99it/s, tok/s=16144.9]
Inference: 85%|████████▌ | 5053/5920 [55:26<04:07, 3.50it/s, tok/s=16144.9]
Inference: 85%|████████▌ | 5053/5920 [55:26<04:07, 3.50it/s, tok/s=16145.8]
Inference: 85%|████████▌ | 5054/5920 [55:26<03:51, 3.74it/s, tok/s=16145.8]
Inference: 85%|████████▌ | 5054/5920 [55:26<03:51, 3.74it/s, tok/s=16145.4]
Inference: 85%|████████▌ | 5055/5920 [55:26<03:36, 3.99it/s, tok/s=16145.4]
Inference: 85%|████████▌ | 5055/5920 [55:26<03:36, 3.99it/s, tok/s=16145.2]
Inference: 85%|████████▌ | 5056/5920 [55:26<03:36, 3.99it/s, tok/s=16145.6]
Inference: 85%|████████▌ | 5057/5920 [55:27<03:28, 4.14it/s, tok/s=16145.6]
Inference: 85%|████████▌ | 5057/5920 [55:27<03:28, 4.14it/s, tok/s=16144.9]
Inference: 85%|████████▌ | 5058/5920 [55:27<03:28, 4.14it/s, tok/s=16145.9]
Inference: 85%|████████▌ | 5059/5920 [55:28<06:30, 2.20it/s, tok/s=16145.9]
Inference: 85%|████████▌ | 5059/5920 [55:28<06:30, 2.20it/s, tok/s=16139.9]
Inference: 85%|████████▌ | 5060/5920 [55:28<05:30, 2.60it/s, tok/s=16139.9]
Inference: 85%|████████▌ | 5060/5920 [55:28<05:30, 2.60it/s, tok/s=16143.4]
Inference: 85%|████████▌ | 5061/5920 [55:29<05:30, 2.60it/s, tok/s=16145.0]
Inference: 86%|████████▌ | 5062/5920 [55:29<03:57, 3.61it/s, tok/s=16145.0]
Inference: 86%|████████▌ | 5062/5920 [55:29<03:57, 3.61it/s, tok/s=16145.8]
Inference: 86%|████████▌ | 5063/5920 [55:29<03:57, 3.61it/s, tok/s=16152.5]
Inference: 86%|████████▌ | 5064/5920 [55:29<04:19, 3.30it/s, tok/s=16152.5]
Inference: 86%|████████▌ | 5064/5920 [55:29<04:19, 3.30it/s, tok/s=16151.2]
Inference: 86%|████████▌ | 5065/5920 [55:29<04:18, 3.30it/s, tok/s=16153.8]
Inference: 86%|████████▌ | 5066/5920 [55:30<04:10, 3.41it/s, tok/s=16153.8]
Inference: 86%|████████▌ | 5066/5920 [55:30<04:10, 3.41it/s, tok/s=16152.4]
Inference: 86%|████████▌ | 5067/5920 [55:30<03:49, 3.72it/s, tok/s=16152.4]
Inference: 86%|████████▌ | 5067/5920 [55:30<03:49, 3.72it/s, tok/s=16155.2]
Inference: 86%|████████▌ | 5068/5920 [55:31<05:56, 2.39it/s, tok/s=16155.2]
Inference: 86%|████████▌ | 5068/5920 [55:31<05:56, 2.39it/s, tok/s=16151.6]
Inference: 86%|████████▌ | 5069/5920 [55:32<06:04, 2.33it/s, tok/s=16151.6]
Inference: 86%|████████▌ | 5069/5920 [55:32<06:04, 2.33it/s, tok/s=16153.2]
Inference: 86%|████████▌ | 5070/5920 [55:32<06:21, 2.23it/s, tok/s=16153.2]
Inference: 86%|████████▌ | 5070/5920 [55:32<06:21, 2.23it/s, tok/s=16154.8]
Inference: 86%|████████▌ | 5071/5920 [55:32<06:21, 2.23it/s, tok/s=16155.7]
Inference: 86%|████████▌ | 5072/5920 [55:32<04:35, 3.07it/s, tok/s=16155.7]
Inference: 86%|████████▌ | 5072/5920 [55:32<04:35, 3.07it/s, tok/s=16156.8]
Inference: 86%|████████▌ | 5073/5920 [55:33<04:02, 3.49it/s, tok/s=16156.8]
Inference: 86%|████████▌ | 5073/5920 [55:33<04:02, 3.49it/s, tok/s=16160.2]
Inference: 86%|████████▌ | 5074/5920 [55:33<04:02, 3.49it/s, tok/s=16162.8]
Inference: 86%|████████▌ | 5075/5920 [55:33<02:51, 4.92it/s, tok/s=16162.8]
Inference: 86%|████████▌ | 5075/5920 [55:33<02:51, 4.92it/s, tok/s=16170.7]
Inference: 86%|████████▌ | 5076/5920 [55:33<02:52, 4.89it/s, tok/s=16170.7]
Inference: 86%|████████▌ | 5076/5920 [55:33<02:52, 4.89it/s, tok/s=16171.0]
Inference: 86%|████████▌ | 5077/5920 [55:33<04:16, 3.28it/s, tok/s=16171.0]
Inference: 86%|████████▌ | 5077/5920 [55:33<04:16, 3.28it/s, tok/s=16170.8]
Inference: 86%|████████▌ | 5078/5920 [55:34<04:24, 3.18it/s, tok/s=16170.8]
Inference: 86%|████████▌ | 5078/5920 [55:34<04:24, 3.18it/s, tok/s=16173.4]
Inference: 86%|████████▌ | 5079/5920 [55:34<04:24, 3.18it/s, tok/s=16173.6]
Inference: 86%|████████▌ | 5080/5920 [55:34<04:24, 3.18it/s, tok/s=16174.5]
Inference: 86%|████████▌ | 5081/5920 [55:34<02:50, 4.92it/s, tok/s=16174.5]
Inference: 86%|████████▌ | 5081/5920 [55:34<02:50, 4.92it/s, tok/s=16174.4]
Inference: 86%|████████▌ | 5082/5920 [55:35<03:33, 3.93it/s, tok/s=16174.4]
Inference: 86%|████████▌ | 5082/5920 [55:35<03:33, 3.93it/s, tok/s=16176.0]
Inference: 86%|████████▌ | 5083/5920 [55:35<03:33, 3.93it/s, tok/s=16176.6]
Inference: 86%|████████▌ | 5084/5920 [55:35<03:32, 3.93it/s, tok/s=16178.7]
Inference: 86%|████████▌ | 5085/5920 [55:35<02:43, 5.11it/s, tok/s=16178.7]
Inference: 86%|████████▌ | 5085/5920 [55:35<02:43, 5.11it/s, tok/s=16181.1]
Inference: 86%|████████▌ | 5086/5920 [55:36<04:11, 3.32it/s, tok/s=16181.1]
Inference: 86%|████████▌ | 5086/5920 [55:36<04:11, 3.32it/s, tok/s=16179.1]
Inference: 86%|████████▌ | 5087/5920 [55:36<04:11, 3.32it/s, tok/s=16179.6]
Inference: 86%|████████▌ | 5088/5920 [55:36<03:04, 4.50it/s, tok/s=16179.6]
Inference: 86%|████████▌ | 5088/5920 [55:36<03:04, 4.50it/s, tok/s=16181.1]
Inference: 86%|████████▌ | 5089/5920 [55:36<03:04, 4.50it/s, tok/s=16187.5]
Inference: 86%|████████▌ | 5090/5920 [55:36<03:04, 4.50it/s, tok/s=16188.7]
Inference: 86%|████████▌ | 5091/5920 [55:36<02:05, 6.59it/s, tok/s=16188.7]
Inference: 86%|████████▌ | 5091/5920 [55:36<02:05, 6.59it/s, tok/s=16189.1]
Inference: 86%|████████▌ | 5092/5920 [55:36<02:05, 6.59it/s, tok/s=16193.0]
Inference: 86%|████████▌ | 5093/5920 [55:36<01:45, 7.82it/s, tok/s=16193.0]
Inference: 86%|████████▌ | 5093/5920 [55:36<01:45, 7.82it/s, tok/s=16193.4]
Inference: 86%|████████▌ | 5094/5920 [55:36<01:45, 7.82it/s, tok/s=16193.3]
Inference: 86%|████████▌ | 5095/5920 [55:37<02:02, 6.74it/s, tok/s=16193.3]
Inference: 86%|████████▌ | 5095/5920 [55:37<02:02, 6.74it/s, tok/s=16193.6]
Inference: 86%|████████▌ | 5096/5920 [55:37<02:02, 6.74it/s, tok/s=16194.5]
Inference: 86%|████████▌ | 5097/5920 [55:37<02:34, 5.33it/s, tok/s=16194.5]
Inference: 86%|████████▌ | 5097/5920 [55:37<02:34, 5.33it/s, tok/s=16192.5]
Inference: 86%|████████▌ | 5098/5920 [55:38<04:51, 2.82it/s, tok/s=16192.5]
Inference: 86%|████████▌ | 5098/5920 [55:38<04:51, 2.82it/s, tok/s=16188.1]
Inference: 86%|████████▌ | 5099/5920 [55:39<05:27, 2.51it/s, tok/s=16188.1]
Inference: 86%|████████▌ | 5099/5920 [55:39<05:27, 2.51it/s, tok/s=16188.3]
Inference: 86%|████████▌ | 5100/5920 [55:39<05:27, 2.51it/s, tok/s=16188.9]
Inference: 86%|████████▌ | 5101/5920 [55:39<05:26, 2.51it/s, tok/s=16191.0]
Inference: 86%|████████▌ | 5102/5920 [55:39<03:44, 3.65it/s, tok/s=16191.0]
Inference: 86%|████████▌ | 5102/5920 [55:39<03:44, 3.65it/s, tok/s=16192.1]
Inference: 86%|████████▌ | 5103/5920 [55:40<04:00, 3.40it/s, tok/s=16192.1]
Inference: 86%|████████▌ | 5103/5920 [55:40<04:00, 3.40it/s, tok/s=16191.7]
Inference: 86%|████████▌ | 5104/5920 [55:40<04:00, 3.40it/s, tok/s=16193.7]
Inference: 86%|████████▌ | 5105/5920 [55:40<02:57, 4.60it/s, tok/s=16193.7]
Inference: 86%|████████▌ | 5105/5920 [55:40<02:57, 4.60it/s, tok/s=16195.5]
Inference: 86%|████████▋ | 5106/5920 [55:41<04:50, 2.80it/s, tok/s=16195.5]
Inference: 86%|████████▋ | 5106/5920 [55:41<04:50, 2.80it/s, tok/s=16192.7]
Inference: 86%|████████▋ | 5107/5920 [55:41<04:58, 2.72it/s, tok/s=16192.7]
Inference: 86%|████████▋ | 5107/5920 [55:41<04:58, 2.72it/s, tok/s=16191.9]
Inference: 86%|████████▋ | 5108/5920 [55:41<04:58, 2.72it/s, tok/s=16192.0]
Inference: 86%|████████▋ | 5109/5920 [55:42<04:04, 3.31it/s, tok/s=16192.0]
Inference: 86%|████████▋ | 5109/5920 [55:42<04:04, 3.31it/s, tok/s=16191.4]
Inference: 86%|████████▋ | 5110/5920 [55:42<04:03, 3.33it/s, tok/s=16191.4]
Inference: 86%|████████▋ | 5110/5920 [55:42<04:03, 3.33it/s, tok/s=16190.7]
Inference: 86%|████████▋ | 5111/5920 [55:42<03:42, 3.63it/s, tok/s=16190.7]
Inference: 86%|████████▋ | 5111/5920 [55:42<03:42, 3.63it/s, tok/s=16190.7]
Inference: 86%|████████▋ | 5112/5920 [55:42<03:59, 3.38it/s, tok/s=16190.7]
Inference: 86%|████████▋ | 5112/5920 [55:42<03:59, 3.38it/s, tok/s=16193.3]
Inference: 86%|████████▋ | 5113/5920 [55:43<05:13, 2.57it/s, tok/s=16193.3]
Inference: 86%|████████▋ | 5113/5920 [55:43<05:13, 2.57it/s, tok/s=16191.7]
Inference: 86%|████████▋ | 5114/5920 [55:43<05:13, 2.57it/s, tok/s=16195.4]
Inference: 86%|████████▋ | 5115/5920 [55:44<04:54, 2.73it/s, tok/s=16195.4]
Inference: 86%|████████▋ | 5115/5920 [55:44<04:54, 2.73it/s, tok/s=16194.7]
Inference: 86%|████████▋ | 5116/5920 [55:44<05:11, 2.58it/s, tok/s=16194.7]
Inference: 86%|████████▋ | 5116/5920 [55:44<05:11, 2.58it/s, tok/s=16193.4]
Inference: 86%|████████▋ | 5117/5920 [55:44<04:19, 3.09it/s, tok/s=16193.4]
Inference: 86%|████████▋ | 5117/5920 [55:44<04:19, 3.09it/s, tok/s=16193.5]
Inference: 86%|████████▋ | 5118/5920 [55:45<04:01, 3.33it/s, tok/s=16193.5]
Inference: 86%|████████▋ | 5118/5920 [55:45<04:01, 3.33it/s, tok/s=16193.0]
Inference: 86%|████████▋ | 5119/5920 [55:45<04:33, 2.93it/s, tok/s=16193.0]
Inference: 86%|████████▋ | 5119/5920 [55:45<04:33, 2.93it/s, tok/s=16191.5]
Inference: 86%|████████▋ | 5120/5920 [55:45<04:32, 2.93it/s, tok/s=16196.2]
Inference: 87%|████████▋ | 5121/5920 [55:45<04:32, 2.93it/s, tok/s=16197.0]
Inference: 87%|████████▋ | 5122/5920 [55:45<02:33, 5.19it/s, tok/s=16197.0]
Inference: 87%|████████▋ | 5122/5920 [55:45<02:33, 5.19it/s, tok/s=16197.4]
Inference: 87%|████████▋ | 5123/5920 [55:45<02:39, 5.01it/s, tok/s=16197.4]
Inference: 87%|████████▋ | 5123/5920 [55:45<02:39, 5.01it/s, tok/s=16197.4]
Inference: 87%|████████▋ | 5124/5920 [55:46<02:39, 5.00it/s, tok/s=16197.4]
Inference: 87%|████████▋ | 5124/5920 [55:46<02:39, 5.00it/s, tok/s=16200.1]
Inference: 87%|████████▋ | 5125/5920 [55:46<03:07, 4.24it/s, tok/s=16200.1]
Inference: 87%|████████▋ | 5125/5920 [55:46<03:07, 4.24it/s, tok/s=16202.8]
Inference: 87%|████████▋ | 5126/5920 [55:46<03:51, 3.43it/s, tok/s=16202.8]
Inference: 87%|████████▋ | 5126/5920 [55:46<03:51, 3.43it/s, tok/s=16202.7]
Inference: 87%|████████▋ | 5127/5920 [55:46<03:51, 3.43it/s, tok/s=16203.6]
Inference: 87%|████████▋ | 5128/5920 [55:47<02:32, 5.20it/s, tok/s=16203.6]
Inference: 87%|████████▋ | 5128/5920 [55:47<02:32, 5.20it/s, tok/s=16204.5]
Inference: 87%|████████▋ | 5129/5920 [55:47<02:32, 5.20it/s, tok/s=16206.2]
Inference: 87%|████████▋ | 5130/5920 [55:47<02:32, 5.20it/s, tok/s=16208.0]
Inference: 87%|████████▋ | 5131/5920 [55:47<01:56, 6.78it/s, tok/s=16208.0]
Inference: 87%|████████▋ | 5131/5920 [55:47<01:56, 6.78it/s, tok/s=16208.6]
Inference: 87%|████████▋ | 5132/5920 [55:47<02:12, 5.95it/s, tok/s=16208.6]
Inference: 87%|████████▋ | 5132/5920 [55:47<02:12, 5.95it/s, tok/s=16208.5]
Inference: 87%|████████▋ | 5133/5920 [55:48<02:59, 4.40it/s, tok/s=16208.5]
Inference: 87%|████████▋ | 5133/5920 [55:48<02:59, 4.40it/s, tok/s=16210.2]
Inference: 87%|████████▋ | 5134/5920 [55:48<02:37, 4.98it/s, tok/s=16210.2]
Inference: 87%|████████▋ | 5134/5920 [55:48<02:37, 4.98it/s, tok/s=16212.2]
Inference: 87%|████████▋ | 5135/5920 [55:48<04:17, 3.05it/s, tok/s=16212.2]
Inference: 87%|████████▋ | 5135/5920 [55:48<04:17, 3.05it/s, tok/s=16209.2]
Inference: 87%|████████▋ | 5136/5920 [55:48<04:17, 3.05it/s, tok/s=16210.0]
Inference: 87%|████████▋ | 5137/5920 [55:49<04:30, 2.89it/s, tok/s=16210.0]
Inference: 87%|████████▋ | 5137/5920 [55:49<04:30, 2.89it/s, tok/s=16210.5]
Inference: 87%|████████▋ | 5138/5920 [55:50<05:08, 2.53it/s, tok/s=16210.5]
Inference: 87%|████████▋ | 5138/5920 [55:50<05:08, 2.53it/s, tok/s=16208.8]
Inference: 87%|████████▋ | 5139/5920 [55:50<05:08, 2.53it/s, tok/s=16209.4]
Inference: 87%|████████▋ | 5140/5920 [55:51<07:17, 1.78it/s, tok/s=16209.4]
Inference: 87%|████████▋ | 5140/5920 [55:51<07:17, 1.78it/s, tok/s=16202.4]
Inference: 87%|████████▋ | 5141/5920 [55:53<09:29, 1.37it/s, tok/s=16202.4]
Inference: 87%|████████▋ | 5141/5920 [55:53<09:29, 1.37it/s, tok/s=16196.6]
Inference: 87%|████████▋ | 5142/5920 [55:53<08:59, 1.44it/s, tok/s=16196.6]
Inference: 87%|████████▋ | 5142/5920 [55:53<08:59, 1.44it/s, tok/s=16198.6]
Inference: 87%|████████▋ | 5143/5920 [55:54<07:46, 1.66it/s, tok/s=16198.6]
Inference: 87%|████████▋ | 5143/5920 [55:54<07:46, 1.66it/s, tok/s=16198.3]
Inference: 87%|████████▋ | 5144/5920 [55:54<09:05, 1.42it/s, tok/s=16198.3]
Inference: 87%|████████▋ | 5144/5920 [55:54<09:05, 1.42it/s, tok/s=16198.0]
Inference: 87%|████████▋ | 5145/5920 [55:55<07:09, 1.80it/s, tok/s=16198.0]
Inference: 87%|████████▋ | 5145/5920 [55:55<07:09, 1.80it/s, tok/s=16198.6]
Inference: 87%|████████▋ | 5146/5920 [55:55<07:09, 1.80it/s, tok/s=16201.3]
Inference: 87%|████████▋ | 5147/5920 [55:55<05:02, 2.56it/s, tok/s=16201.3]
Inference: 87%|████████▋ | 5147/5920 [55:55<05:02, 2.56it/s, tok/s=16202.3]
Inference: 87%|████████▋ | 5148/5920 [55:55<05:02, 2.56it/s, tok/s=16203.8]
Inference: 87%|████████▋ | 5149/5920 [55:55<05:01, 2.56it/s, tok/s=16208.2]
Inference: 87%|████████▋ | 5150/5920 [55:56<04:29, 2.85it/s, tok/s=16208.2]
Inference: 87%|████████▋ | 5150/5920 [55:56<04:29, 2.85it/s, tok/s=16205.1]
Inference: 87%|████████▋ | 5151/5920 [55:56<04:29, 2.85it/s, tok/s=16205.8]
Inference: 87%|████████▋ | 5152/5920 [55:56<03:42, 3.45it/s, tok/s=16205.8]
Inference: 87%|████████▋ | 5152/5920 [55:56<03:42, 3.45it/s, tok/s=16205.4]
Inference: 87%|████████▋ | 5153/5920 [55:57<04:03, 3.15it/s, tok/s=16205.4]
Inference: 87%|████████▋ | 5153/5920 [55:57<04:03, 3.15it/s, tok/s=16204.8]
Inference: 87%|████████▋ | 5154/5920 [55:57<04:02, 3.15it/s, tok/s=16206.3]
Inference: 87%|████████▋ | 5155/5920 [55:57<03:35, 3.55it/s, tok/s=16206.3]
Inference: 87%|████████▋ | 5155/5920 [55:57<03:35, 3.55it/s, tok/s=16205.7]
Inference: 87%|████████▋ | 5156/5920 [55:57<03:44, 3.41it/s, tok/s=16205.7]
Inference: 87%|████████▋ | 5156/5920 [55:57<03:44, 3.41it/s, tok/s=16209.1]
Inference: 87%|████████▋ | 5157/5920 [55:58<04:19, 2.94it/s, tok/s=16209.1]
Inference: 87%|████████▋ | 5157/5920 [55:58<04:19, 2.94it/s, tok/s=16209.5]
Inference: 87%|████████▋ | 5158/5920 [55:58<04:18, 2.94it/s, tok/s=16212.8]
Inference: 87%|████████▋ | 5159/5920 [55:58<03:30, 3.61it/s, tok/s=16212.8]
Inference: 87%|████████▋ | 5159/5920 [55:58<03:30, 3.61it/s, tok/s=16213.3]
Inference: 87%|████████▋ | 5160/5920 [55:59<04:33, 2.78it/s, tok/s=16213.3]
Inference: 87%|████████▋ | 5160/5920 [55:59<04:33, 2.78it/s, tok/s=16215.3]
Inference: 87%|████████▋ | 5161/5920 [56:00<06:02, 2.09it/s, tok/s=16215.3]
Inference: 87%|████████▋ | 5161/5920 [56:00<06:02, 2.09it/s, tok/s=16214.7]
Inference: 87%|████████▋ | 5162/5920 [56:01<07:06, 1.78it/s, tok/s=16214.7]
Inference: 87%|████████▋ | 5162/5920 [56:01<07:06, 1.78it/s, tok/s=16211.9]
Inference: 87%|████████▋ | 5163/5920 [56:01<07:05, 1.78it/s, tok/s=16213.1]
Inference: 87%|████████▋ | 5164/5920 [56:01<06:10, 2.04it/s, tok/s=16213.1]
Inference: 87%|████████▋ | 5164/5920 [56:01<06:10, 2.04it/s, tok/s=16213.4]
Inference: 87%|████████▋ | 5165/5920 [56:02<05:28, 2.30it/s, tok/s=16213.4]
Inference: 87%|████████▋ | 5165/5920 [56:02<05:28, 2.30it/s, tok/s=16213.6]
Inference: 87%|████████▋ | 5166/5920 [56:02<05:36, 2.24it/s, tok/s=16213.6]
Inference: 87%|████████▋ | 5166/5920 [56:02<05:36, 2.24it/s, tok/s=16212.6]
Inference: 87%|████████▋ | 5167/5920 [56:02<04:35, 2.73it/s, tok/s=16212.6]
Inference: 87%|████████▋ | 5167/5920 [56:02<04:35, 2.73it/s, tok/s=16216.3]
Inference: 87%|████████▋ | 5168/5920 [56:03<05:52, 2.13it/s, tok/s=16216.3]
Inference: 87%|████████▋ | 5168/5920 [56:03<05:52, 2.13it/s, tok/s=16214.1]
Inference: 87%|████████▋ | 5169/5920 [56:04<06:47, 1.84it/s, tok/s=16214.1]
Inference: 87%|████████▋ | 5169/5920 [56:04<06:47, 1.84it/s, tok/s=16212.6]
Inference: 87%|████████▋ | 5170/5920 [56:05<07:48, 1.60it/s, tok/s=16212.6]
Inference: 87%|████████▋ | 5170/5920 [56:05<07:48, 1.60it/s, tok/s=16209.9]
Inference: 87%|████████▋ | 5171/5920 [56:05<07:38, 1.63it/s, tok/s=16209.9]
Inference: 87%|████████▋ | 5171/5920 [56:05<07:38, 1.63it/s, tok/s=16211.1]
Inference: 87%|████████▋ | 5172/5920 [56:05<05:47, 2.15it/s, tok/s=16211.1]
Inference: 87%|████████▋ | 5172/5920 [56:05<05:47, 2.15it/s, tok/s=16211.7]
Inference: 87%|████████▋ | 5173/5920 [56:05<05:47, 2.15it/s, tok/s=16214.5]
Inference: 87%|████████▋ | 5174/5920 [56:06<04:53, 2.54it/s, tok/s=16214.5]
Inference: 87%|████████▋ | 5174/5920 [56:06<04:53, 2.54it/s, tok/s=16214.3]
Inference: 87%|████████▋ | 5175/5920 [56:06<04:38, 2.67it/s, tok/s=16214.3]
Inference: 87%|████████▋ | 5175/5920 [56:06<04:38, 2.67it/s, tok/s=16214.2]
Inference: 87%|████████▋ | 5176/5920 [56:07<04:27, 2.78it/s, tok/s=16214.2]
Inference: 87%|████████▋ | 5176/5920 [56:07<04:27, 2.78it/s, tok/s=16214.3]
Inference: 87%|████████▋ | 5177/5920 [56:07<04:30, 2.75it/s, tok/s=16214.3]
Inference: 87%|████████▋ | 5177/5920 [56:07<04:30, 2.75it/s, tok/s=16215.4]
Inference: 87%|████████▋ | 5178/5920 [56:07<04:29, 2.75it/s, tok/s=16224.2]
Inference: 87%|████████▋ | 5179/5920 [56:07<03:26, 3.59it/s, tok/s=16224.2]
Inference: 87%|████████▋ | 5179/5920 [56:07<03:26, 3.59it/s, tok/s=16227.0]
Inference: 88%|████████▊ | 5180/5920 [56:07<02:58, 4.16it/s, tok/s=16227.0]
Inference: 88%|████████▊ | 5180/5920 [56:07<02:58, 4.16it/s, tok/s=16231.2]
Inference: 88%|████████▊ | 5181/5920 [56:08<03:18, 3.71it/s, tok/s=16231.2]
Inference: 88%|████████▊ | 5181/5920 [56:08<03:18, 3.71it/s, tok/s=16230.5]
Inference: 88%|████████▊ | 5182/5920 [56:08<03:18, 3.71it/s, tok/s=16231.7]
Inference: 88%|████████▊ | 5183/5920 [56:08<02:30, 4.91it/s, tok/s=16231.7]
Inference: 88%|████████▊ | 5183/5920 [56:08<02:30, 4.91it/s, tok/s=16238.5]
Inference: 88%|████████▊ | 5184/5920 [56:08<02:44, 4.47it/s, tok/s=16238.5]
Inference: 88%|████████▊ | 5184/5920 [56:08<02:44, 4.47it/s, tok/s=16239.0]
Inference: 88%|████████▊ | 5185/5920 [56:10<06:20, 1.93it/s, tok/s=16239.0]
Inference: 88%|████████▊ | 5185/5920 [56:10<06:20, 1.93it/s, tok/s=16234.9]
Inference: 88%|████████▊ | 5186/5920 [56:10<05:24, 2.26it/s, tok/s=16234.9]
Inference: 88%|████████▊ | 5186/5920 [56:10<05:24, 2.26it/s, tok/s=16235.8]
Inference: 88%|████████▊ | 5187/5920 [56:11<08:06, 1.51it/s, tok/s=16235.8]
Inference: 88%|████████▊ | 5187/5920 [56:11<08:06, 1.51it/s, tok/s=16231.3]
Inference: 88%|████████▊ | 5188/5920 [56:11<08:06, 1.51it/s, tok/s=16232.9]
Inference: 88%|████████▊ | 5189/5920 [56:11<05:12, 2.34it/s, tok/s=16232.9]
Inference: 88%|████████▊ | 5189/5920 [56:11<05:12, 2.34it/s, tok/s=16236.7]
Inference: 88%|████████▊ | 5190/5920 [56:11<05:11, 2.34it/s, tok/s=16241.6]
Inference: 88%|████████▊ | 5191/5920 [56:12<03:47, 3.20it/s, tok/s=16241.6]
Inference: 88%|████████▊ | 5191/5920 [56:12<03:47, 3.20it/s, tok/s=16245.3]
Inference: 88%|████████▊ | 5192/5920 [56:12<03:47, 3.20it/s, tok/s=16254.1]
Inference: 88%|████████▊ | 5193/5920 [56:12<02:49, 4.30it/s, tok/s=16254.1]
Inference: 88%|████████▊ | 5193/5920 [56:12<02:49, 4.30it/s, tok/s=16257.6]
Inference: 88%|████████▊ | 5194/5920 [56:12<03:11, 3.79it/s, tok/s=16257.6]
Inference: 88%|████████▊ | 5194/5920 [56:12<03:11, 3.79it/s, tok/s=16262.6]
Inference: 88%|████████▊ | 5195/5920 [56:13<05:01, 2.40it/s, tok/s=16262.6]
Inference: 88%|████████▊ | 5195/5920 [56:13<05:01, 2.40it/s, tok/s=16259.6]
Inference: 88%|████████▊ | 5196/5920 [56:13<05:01, 2.40it/s, tok/s=16264.1]
Inference: 88%|████████▊ | 5197/5920 [56:14<05:29, 2.19it/s, tok/s=16264.1]
Inference: 88%|████████▊ | 5197/5920 [56:14<05:29, 2.19it/s, tok/s=16260.5]
Inference: 88%|████████▊ | 5198/5920 [56:16<08:12, 1.46it/s, tok/s=16260.5]
Inference: 88%|████████▊ | 5198/5920 [56:16<08:12, 1.46it/s, tok/s=16255.2]
Inference: 88%|████████▊ | 5199/5920 [56:16<06:58, 1.72it/s, tok/s=16255.2]
Inference: 88%|████████▊ | 5199/5920 [56:16<06:58, 1.72it/s, tok/s=16257.8]
Inference: 88%|████████▊ | 5200/5920 [56:16<06:57, 1.72it/s, tok/s=16262.3]
Inference: 88%|████████▊ | 5201/5920 [56:16<05:12, 2.30it/s, tok/s=16262.3]
Inference: 88%|████████▊ | 5201/5920 [56:16<05:12, 2.30it/s, tok/s=16261.3]
Inference: 88%|████████▊ | 5202/5920 [56:16<05:12, 2.30it/s, tok/s=16265.5]
Inference: 88%|████████▊ | 5203/5920 [56:17<03:49, 3.13it/s, tok/s=16265.5]
Inference: 88%|████████▊ | 5203/5920 [56:17<03:49, 3.13it/s, tok/s=16266.4]
Inference: 88%|████████▊ | 5204/5920 [56:17<03:48, 3.13it/s, tok/s=16268.0]
Inference: 88%|████████▊ | 5205/5920 [56:17<03:48, 3.13it/s, tok/s=16276.6]
Inference: 88%|████████▊ | 5206/5920 [56:17<03:48, 3.13it/s, tok/s=16285.6]
Inference: 88%|████████▊ | 5207/5920 [56:17<03:02, 3.92it/s, tok/s=16285.6]
Inference: 88%|████████▊ | 5207/5920 [56:17<03:02, 3.92it/s, tok/s=16283.2]
Inference: 88%|████████▊ | 5208/5920 [56:17<03:01, 3.92it/s, tok/s=16288.4]
Inference: 88%|████████▊ | 5209/5920 [56:18<02:30, 4.72it/s, tok/s=16288.4]
Inference: 88%|████████▊ | 5209/5920 [56:18<02:30, 4.72it/s, tok/s=16289.0]
Inference: 88%|████████▊ | 5210/5920 [56:18<02:56, 4.03it/s, tok/s=16289.0]
Inference: 88%|████████▊ | 5210/5920 [56:18<02:56, 4.03it/s, tok/s=16289.6]
Inference: 88%|████████▊ | 5211/5920 [56:19<03:49, 3.09it/s, tok/s=16289.6]
Inference: 88%|████████▊ | 5211/5920 [56:19<03:49, 3.09it/s, tok/s=16289.4]
Inference: 88%|████████▊ | 5212/5920 [56:19<04:37, 2.55it/s, tok/s=16289.4]
Inference: 88%|████████▊ | 5212/5920 [56:19<04:37, 2.55it/s, tok/s=16287.4]
Inference: 88%|████████▊ | 5213/5920 [56:20<05:56, 1.98it/s, tok/s=16287.4]
Inference: 88%|████████▊ | 5213/5920 [56:20<05:56, 1.98it/s, tok/s=16284.5]
Inference: 88%|████████▊ | 5214/5920 [56:20<05:18, 2.22it/s, tok/s=16284.5]
Inference: 88%|████████▊ | 5214/5920 [56:20<05:18, 2.22it/s, tok/s=16284.5]
Inference: 88%|████████▊ | 5215/5920 [56:21<05:17, 2.22it/s, tok/s=16285.5]
Inference: 88%|████████▊ | 5216/5920 [56:22<06:13, 1.89it/s, tok/s=16285.5]
Inference: 88%|████████▊ | 5216/5920 [56:22<06:13, 1.89it/s, tok/s=16282.3]
Inference: 88%|████████▊ | 5217/5920 [56:22<05:03, 2.31it/s, tok/s=16282.3]
Inference: 88%|████████▊ | 5217/5920 [56:22<05:03, 2.31it/s, tok/s=16283.4]
Inference: 88%|████████▊ | 5218/5920 [56:22<05:03, 2.31it/s, tok/s=16292.2]
Inference: 88%|████████▊ | 5219/5920 [56:22<03:50, 3.04it/s, tok/s=16292.2]
Inference: 88%|████████▊ | 5219/5920 [56:22<03:50, 3.04it/s, tok/s=16292.3]
Inference: 88%|████████▊ | 5220/5920 [56:22<03:41, 3.16it/s, tok/s=16292.3]
Inference: 88%|████████▊ | 5220/5920 [56:22<03:41, 3.16it/s, tok/s=16293.9]
Inference: 88%|████████▊ | 5221/5920 [56:23<03:41, 3.16it/s, tok/s=16295.0]
Inference: 88%|████████▊ | 5222/5920 [56:24<05:03, 2.30it/s, tok/s=16295.0]
Inference: 88%|████████▊ | 5222/5920 [56:24<05:03, 2.30it/s, tok/s=16289.9]
Inference: 88%|████████▊ | 5223/5920 [56:24<04:19, 2.69it/s, tok/s=16289.9]
Inference: 88%|████████▊ | 5223/5920 [56:24<04:19, 2.69it/s, tok/s=16291.0]
Inference: 88%|████████▊ | 5224/5920 [56:24<03:34, 3.24it/s, tok/s=16291.0]
Inference: 88%|████████▊ | 5224/5920 [56:24<03:34, 3.24it/s, tok/s=16293.5]
Inference: 88%|████████▊ | 5225/5920 [56:25<05:43, 2.02it/s, tok/s=16293.5]
Inference: 88%|████████▊ | 5225/5920 [56:25<05:43, 2.02it/s, tok/s=16289.6]
Inference: 88%|████████▊ | 5226/5920 [56:25<04:54, 2.35it/s, tok/s=16289.6]
Inference: 88%|████████▊ | 5226/5920 [56:25<04:54, 2.35it/s, tok/s=16289.6]
Inference: 88%|████████▊ | 5227/5920 [56:25<04:54, 2.35it/s, tok/s=16296.2]
Inference: 88%|████████▊ | 5228/5920 [56:27<06:00, 1.92it/s, tok/s=16296.2]
Inference: 88%|████████▊ | 5228/5920 [56:27<06:00, 1.92it/s, tok/s=16293.7]
Inference: 88%|████████▊ | 5229/5920 [56:27<06:16, 1.84it/s, tok/s=16293.7]
Inference: 88%|████████▊ | 5229/5920 [56:27<06:16, 1.84it/s, tok/s=16293.6]
Inference: 88%|████████▊ | 5230/5920 [56:28<07:13, 1.59it/s, tok/s=16293.6]
Inference: 88%|████████▊ | 5230/5920 [56:28<07:13, 1.59it/s, tok/s=16292.4]
Inference: 88%|████████▊ | 5231/5920 [56:29<09:30, 1.21it/s, tok/s=16292.4]
Inference: 88%|████████▊ | 5231/5920 [56:29<09:30, 1.21it/s, tok/s=16287.4]
Inference: 88%|████████▊ | 5232/5920 [56:29<09:29, 1.21it/s, tok/s=16291.4]
Inference: 88%|████████▊ | 5233/5920 [56:30<06:21, 1.80it/s, tok/s=16291.4]
Inference: 88%|████████▊ | 5233/5920 [56:30<06:21, 1.80it/s, tok/s=16291.1]
Inference: 88%|████████▊ | 5234/5920 [56:31<06:57, 1.64it/s, tok/s=16291.1]
Inference: 88%|████████▊ | 5234/5920 [56:31<06:57, 1.64it/s, tok/s=16288.3]
Inference: 88%|████████▊ | 5235/5920 [56:31<06:29, 1.76it/s, tok/s=16288.3]
Inference: 88%|████████▊ | 5235/5920 [56:31<06:29, 1.76it/s, tok/s=16288.8]
Inference: 88%|████████▊ | 5236/5920 [56:31<06:29, 1.76it/s, tok/s=16289.8]
Inference: 88%|████████▊ | 5237/5920 [56:31<04:34, 2.48it/s, tok/s=16289.8]
Inference: 88%|████████▊ | 5237/5920 [56:31<04:34, 2.48it/s, tok/s=16289.7]
Inference: 88%|████████▊ | 5238/5920 [56:33<07:44, 1.47it/s, tok/s=16289.7]
Inference: 88%|████████▊ | 5238/5920 [56:33<07:44, 1.47it/s, tok/s=16285.2]
Inference: 88%|████████▊ | 5239/5920 [56:34<07:12, 1.57it/s, tok/s=16285.2]
Inference: 88%|████████▊ | 5239/5920 [56:34<07:12, 1.57it/s, tok/s=16285.1]
Inference: 89%|████████▊ | 5240/5920 [56:34<07:12, 1.57it/s, tok/s=16286.8]
Inference: 89%|████████▊ | 5241/5920 [56:34<05:56, 1.90it/s, tok/s=16286.8]
Inference: 89%|████████▊ | 5241/5920 [56:34<05:56, 1.90it/s, tok/s=16283.8]
Inference: 89%|████████▊ | 5242/5920 [56:34<04:52, 2.32it/s, tok/s=16283.8]
Inference: 89%|████████▊ | 5242/5920 [56:34<04:52, 2.32it/s, tok/s=16285.1]
Inference: 89%|████████▊ | 5243/5920 [56:35<04:34, 2.47it/s, tok/s=16285.1]
Inference: 89%|████████▊ | 5243/5920 [56:35<04:34, 2.47it/s, tok/s=16284.8]
Inference: 89%|████████▊ | 5244/5920 [56:35<03:40, 3.06it/s, tok/s=16284.8]
Inference: 89%|████████▊ | 5244/5920 [56:35<03:40, 3.06it/s, tok/s=16286.1]
Inference: 89%|████████▊ | 5245/5920 [56:35<03:30, 3.20it/s, tok/s=16286.1]
Inference: 89%|████████▊ | 5245/5920 [56:35<03:30, 3.20it/s, tok/s=16290.4]
Inference: 89%|████████▊ | 5246/5920 [56:35<03:30, 3.20it/s, tok/s=16291.2]
Inference: 89%|████████▊ | 5247/5920 [56:36<03:19, 3.37it/s, tok/s=16291.2]
Inference: 89%|████████▊ | 5247/5920 [56:36<03:19, 3.37it/s, tok/s=16290.7]
Inference: 89%|████████▊ | 5248/5920 [56:36<03:36, 3.11it/s, tok/s=16290.7]
Inference: 89%|████████▊ | 5248/5920 [56:36<03:36, 3.11it/s, tok/s=16291.5]
Inference: 89%|████████▊ | 5249/5920 [56:37<04:02, 2.77it/s, tok/s=16291.5]
Inference: 89%|████████▊ | 5249/5920 [56:37<04:02, 2.77it/s, tok/s=16290.0]
Inference: 89%|████████▊ | 5250/5920 [56:37<04:33, 2.45it/s, tok/s=16290.0]
Inference: 89%|████████▊ | 5250/5920 [56:37<04:33, 2.45it/s, tok/s=16288.6]
Inference: 89%|████████▊ | 5251/5920 [56:37<04:22, 2.55it/s, tok/s=16288.6]
Inference: 89%|████████▊ | 5251/5920 [56:37<04:22, 2.55it/s, tok/s=16288.3]
Inference: 89%|████████▊ | 5252/5920 [56:37<04:22, 2.55it/s, tok/s=16289.2]
Inference: 89%|████████▊ | 5253/5920 [56:38<02:54, 3.83it/s, tok/s=16289.2]
Inference: 89%|████████▊ | 5253/5920 [56:38<02:54, 3.83it/s, tok/s=16290.9]
Inference: 89%|████████▉ | 5254/5920 [56:38<03:49, 2.90it/s, tok/s=16290.9]
Inference: 89%|████████▉ | 5254/5920 [56:38<03:49, 2.90it/s, tok/s=16289.0]
Inference: 89%|████████▉ | 5255/5920 [56:39<05:57, 1.86it/s, tok/s=16289.0]
Inference: 89%|████████▉ | 5255/5920 [56:39<05:57, 1.86it/s, tok/s=16285.6]
Inference: 89%|████████▉ | 5256/5920 [56:41<08:40, 1.27it/s, tok/s=16285.6]
Inference: 89%|████████▉ | 5256/5920 [56:41<08:40, 1.27it/s, tok/s=16283.7]
Inference: 89%|████████▉ | 5257/5920 [56:41<07:02, 1.57it/s, tok/s=16283.7]
Inference: 89%|████████▉ | 5257/5920 [56:41<07:02, 1.57it/s, tok/s=16284.3]
Inference: 89%|████████▉ | 5258/5920 [56:41<06:20, 1.74it/s, tok/s=16284.3]
Inference: 89%|████████▉ | 5258/5920 [56:41<06:20, 1.74it/s, tok/s=16284.0]
Inference: 89%|████████▉ | 5259/5920 [56:43<09:07, 1.21it/s, tok/s=16284.0]
Inference: 89%|████████▉ | 5259/5920 [56:43<09:07, 1.21it/s, tok/s=16280.0]
Inference: 89%|████████▉ | 5260/5920 [56:45<13:43, 1.25s/it, tok/s=16280.0]
Inference: 89%|████████▉ | 5260/5920 [56:45<13:43, 1.25s/it, tok/s=16271.0]
Inference: 89%|████████▉ | 5261/5920 [56:45<10:30, 1.05it/s, tok/s=16271.0]
Inference: 89%|████████▉ | 5261/5920 [56:45<10:30, 1.05it/s, tok/s=16273.3]
Inference: 89%|████████▉ | 5262/5920 [56:46<08:24, 1.30it/s, tok/s=16273.3]
Inference: 89%|████████▉ | 5262/5920 [56:46<08:24, 1.30it/s, tok/s=16274.7]
Inference: 89%|████████▉ | 5263/5920 [56:46<06:26, 1.70it/s, tok/s=16274.7]
Inference: 89%|████████▉ | 5263/5920 [56:46<06:26, 1.70it/s, tok/s=16275.2]
Inference: 89%|████████▉ | 5264/5920 [56:46<06:25, 1.70it/s, tok/s=16276.6]
Inference: 89%|████████▉ | 5265/5920 [56:46<04:33, 2.40it/s, tok/s=16276.6]
Inference: 89%|████████▉ | 5265/5920 [56:46<04:33, 2.40it/s, tok/s=16279.0]
Inference: 89%|████████▉ | 5266/5920 [56:47<03:57, 2.75it/s, tok/s=16279.0]
Inference: 89%|████████▉ | 5266/5920 [56:47<03:57, 2.75it/s, tok/s=16281.0]
Inference: 89%|████████▉ | 5267/5920 [56:48<06:43, 1.62it/s, tok/s=16281.0]
Inference: 89%|████████▉ | 5267/5920 [56:48<06:43, 1.62it/s, tok/s=16276.9]
Inference: 89%|████████▉ | 5268/5920 [56:48<05:18, 2.04it/s, tok/s=16276.9]
Inference: 89%|████████▉ | 5268/5920 [56:48<05:18, 2.04it/s, tok/s=16279.0]
Inference: 89%|████████▉ | 5269/5920 [56:49<06:15, 1.73it/s, tok/s=16279.0]
Inference: 89%|████████▉ | 5269/5920 [56:49<06:15, 1.73it/s, tok/s=16280.6]
Inference: 89%|████████▉ | 5270/5920 [56:50<07:10, 1.51it/s, tok/s=16280.6]
Inference: 89%|████████▉ | 5270/5920 [56:50<07:10, 1.51it/s, tok/s=16279.2]
Inference: 89%|████████▉ | 5271/5920 [56:50<07:29, 1.44it/s, tok/s=16279.2]
Inference: 89%|████████▉ | 5271/5920 [56:50<07:29, 1.44it/s, tok/s=16279.4]
Inference: 89%|████████▉ | 5272/5920 [56:52<11:39, 1.08s/it, tok/s=16279.4]
Inference: 89%|████████▉ | 5272/5920 [56:52<11:39, 1.08s/it, tok/s=16272.4]
Inference: 89%|████████▉ | 5273/5920 [56:53<10:07, 1.06it/s, tok/s=16272.4]
Inference: 89%|████████▉ | 5273/5920 [56:53<10:07, 1.06it/s, tok/s=16270.9]
Inference: 89%|████████▉ | 5274/5920 [56:53<10:06, 1.06it/s, tok/s=16272.4]
Inference: 89%|████████▉ | 5275/5920 [56:54<06:51, 1.57it/s, tok/s=16272.4]
Inference: 89%|████████▉ | 5275/5920 [56:54<06:51, 1.57it/s, tok/s=16273.9]
Inference: 89%|████████▉ | 5276/5920 [56:54<06:49, 1.57it/s, tok/s=16273.9]
Inference: 89%|████████▉ | 5276/5920 [56:54<06:49, 1.57it/s, tok/s=16272.8]
Inference: 89%|████████▉ | 5277/5920 [56:54<05:41, 1.88it/s, tok/s=16272.8]
Inference: 89%|████████▉ | 5277/5920 [56:54<05:41, 1.88it/s, tok/s=16273.7]
Inference: 89%|████████▉ | 5278/5920 [56:55<05:52, 1.82it/s, tok/s=16273.7]
Inference: 89%|████████▉ | 5278/5920 [56:55<05:52, 1.82it/s, tok/s=16273.6]
Inference: 89%|████████▉ | 5279/5920 [56:55<04:36, 2.32it/s, tok/s=16273.6]
Inference: 89%|████████▉ | 5279/5920 [56:55<04:36, 2.32it/s, tok/s=16276.0]
Inference: 89%|████████▉ | 5280/5920 [56:56<04:18, 2.47it/s, tok/s=16276.0]
Inference: 89%|████████▉ | 5280/5920 [56:56<04:18, 2.47it/s, tok/s=16277.1]
Inference: 89%|████████▉ | 5281/5920 [56:56<04:18, 2.47it/s, tok/s=16278.7]
Inference: 89%|████████▉ | 5282/5920 [56:56<03:12, 3.32it/s, tok/s=16278.7]
Inference: 89%|████████▉ | 5282/5920 [56:56<03:12, 3.32it/s, tok/s=16280.3]
Inference: 89%|████████▉ | 5283/5920 [56:56<03:11, 3.32it/s, tok/s=16283.0]
Inference: 89%|████████▉ | 5284/5920 [56:56<02:56, 3.61it/s, tok/s=16283.0]
Inference: 89%|████████▉ | 5284/5920 [56:56<02:56, 3.61it/s, tok/s=16284.1]
Inference: 89%|████████▉ | 5285/5920 [56:57<03:34, 2.96it/s, tok/s=16284.1]
Inference: 89%|████████▉ | 5285/5920 [56:57<03:34, 2.96it/s, tok/s=16282.1]
Inference: 89%|████████▉ | 5286/5920 [56:58<05:53, 1.80it/s, tok/s=16282.1]
Inference: 89%|████████▉ | 5286/5920 [56:58<05:53, 1.80it/s, tok/s=16278.2]
Inference: 89%|████████▉ | 5287/5920 [56:59<05:12, 2.02it/s, tok/s=16278.2]
Inference: 89%|████████▉ | 5287/5920 [56:59<05:12, 2.02it/s, tok/s=16281.0]
Inference: 89%|████████▉ | 5288/5920 [57:00<07:29, 1.41it/s, tok/s=16281.0]
Inference: 89%|████████▉ | 5288/5920 [57:00<07:29, 1.41it/s, tok/s=16281.9]
Inference: 89%|████████▉ | 5289/5920 [57:00<06:00, 1.75it/s, tok/s=16281.9]
Inference: 89%|████████▉ | 5289/5920 [57:00<06:00, 1.75it/s, tok/s=16281.5]
Inference: 89%|████████▉ | 5290/5920 [57:00<05:13, 2.01it/s, tok/s=16281.5]
Inference: 89%|████████▉ | 5290/5920 [57:00<05:13, 2.01it/s, tok/s=16282.6]
Inference: 89%|████████▉ | 5291/5920 [57:01<04:16, 2.45it/s, tok/s=16282.6]
Inference: 89%|████████▉ | 5291/5920 [57:01<04:16, 2.45it/s, tok/s=16284.5]
Inference: 89%|████████▉ | 5292/5920 [57:01<03:49, 2.74it/s, tok/s=16284.5]
Inference: 89%|████████▉ | 5292/5920 [57:01<03:49, 2.74it/s, tok/s=16284.4]
Inference: 89%|████████▉ | 5293/5920 [57:02<05:15, 1.99it/s, tok/s=16284.4]
Inference: 89%|████████▉ | 5293/5920 [57:02<05:15, 1.99it/s, tok/s=16283.3]
Inference: 89%|████████▉ | 5294/5920 [57:02<05:39, 1.84it/s, tok/s=16283.3]
Inference: 89%|████████▉ | 5294/5920 [57:02<05:39, 1.84it/s, tok/s=16283.2]
Inference: 89%|████████▉ | 5295/5920 [57:02<05:39, 1.84it/s, tok/s=16286.5]
Inference: 89%|████████▉ | 5296/5920 [57:03<03:39, 2.84it/s, tok/s=16286.5]
Inference: 89%|████████▉ | 5296/5920 [57:03<03:39, 2.84it/s, tok/s=16288.0]
Inference: 89%|████████▉ | 5297/5920 [57:03<03:16, 3.18it/s, tok/s=16288.0]
Inference: 89%|████████▉ | 5297/5920 [57:03<03:16, 3.18it/s, tok/s=16287.8]
Inference: 89%|████████▉ | 5298/5920 [57:04<05:48, 1.78it/s, tok/s=16287.8]
Inference: 89%|████████▉ | 5298/5920 [57:04<05:48, 1.78it/s, tok/s=16283.4]
Inference: 90%|████████▉ | 5299/5920 [57:04<05:06, 2.03it/s, tok/s=16283.4]
Inference: 90%|████████▉ | 5299/5920 [57:04<05:06, 2.03it/s, tok/s=16284.0]
Inference: 90%|████████▉ | 5300/5920 [57:04<05:05, 2.03it/s, tok/s=16284.5]
Inference: 90%|████████▉ | 5301/5920 [57:04<03:13, 3.20it/s, tok/s=16284.5]
Inference: 90%|████████▉ | 5301/5920 [57:04<03:13, 3.20it/s, tok/s=16289.1]
Inference: 90%|████████▉ | 5302/5920 [57:05<02:49, 3.64it/s, tok/s=16289.1]
Inference: 90%|████████▉ | 5302/5920 [57:05<02:49, 3.64it/s, tok/s=16289.8]
Inference: 90%|████████▉ | 5303/5920 [57:05<02:52, 3.57it/s, tok/s=16289.8]
Inference: 90%|████████▉ | 5303/5920 [57:05<02:52, 3.57it/s, tok/s=16289.6]
Inference: 90%|████████▉ | 5304/5920 [57:06<05:34, 1.84it/s, tok/s=16289.6]
Inference: 90%|████████▉ | 5304/5920 [57:06<05:34, 1.84it/s, tok/s=16284.7]
Inference: 90%|████████▉ | 5305/5920 [57:06<05:34, 1.84it/s, tok/s=16287.9]
Inference: 90%|████████▉ | 5306/5920 [57:06<03:34, 2.86it/s, tok/s=16287.9]
Inference: 90%|████████▉ | 5306/5920 [57:06<03:34, 2.86it/s, tok/s=16288.5]
Inference: 90%|████████▉ | 5307/5920 [57:09<08:16, 1.23it/s, tok/s=16288.5]
Inference: 90%|████████▉ | 5307/5920 [57:09<08:16, 1.23it/s, tok/s=16279.0]
Inference: 90%|████████▉ | 5308/5920 [57:09<06:53, 1.48it/s, tok/s=16279.0]
Inference: 90%|████████▉ | 5308/5920 [57:09<06:53, 1.48it/s, tok/s=16279.0]
Inference: 90%|████████▉ | 5309/5920 [57:09<05:47, 1.76it/s, tok/s=16279.0]
Inference: 90%|████████▉ | 5309/5920 [57:09<05:47, 1.76it/s, tok/s=16282.9]
Inference: 90%|████████▉ | 5310/5920 [57:10<05:03, 2.01it/s, tok/s=16282.9]
Inference: 90%|████████▉ | 5310/5920 [57:10<05:03, 2.01it/s, tok/s=16282.5]
Inference: 90%|████████▉ | 5311/5920 [57:10<04:58, 2.04it/s, tok/s=16282.5]
Inference: 90%|████████▉ | 5311/5920 [57:10<04:58, 2.04it/s, tok/s=16281.5]
Inference: 90%|████████▉ | 5312/5920 [57:11<05:58, 1.70it/s, tok/s=16281.5]
Inference: 90%|████████▉ | 5312/5920 [57:11<05:58, 1.70it/s, tok/s=16278.9]
Inference: 90%|████████▉ | 5313/5920 [57:11<05:51, 1.73it/s, tok/s=16278.9]
Inference: 90%|████████▉ | 5313/5920 [57:11<05:51, 1.73it/s, tok/s=16279.1]
Inference: 90%|████████▉ | 5314/5920 [57:12<05:37, 1.80it/s, tok/s=16279.1]
Inference: 90%|████████▉ | 5314/5920 [57:12<05:37, 1.80it/s, tok/s=16280.3]
Inference: 90%|████████▉ | 5315/5920 [57:13<08:27, 1.19it/s, tok/s=16280.3]
Inference: 90%|████████▉ | 5315/5920 [57:13<08:27, 1.19it/s, tok/s=16275.1]
Inference: 90%|████████▉ | 5316/5920 [57:14<07:59, 1.26it/s, tok/s=16275.1]
Inference: 90%|████████▉ | 5316/5920 [57:14<07:59, 1.26it/s, tok/s=16274.1]
Inference: 90%|████████▉ | 5317/5920 [57:14<05:55, 1.70it/s, tok/s=16274.1]
Inference: 90%|████████▉ | 5317/5920 [57:14<05:55, 1.70it/s, tok/s=16276.3]
Inference: 90%|████████▉ | 5318/5920 [57:14<04:39, 2.15it/s, tok/s=16276.3]
Inference: 90%|████████▉ | 5318/5920 [57:14<04:39, 2.15it/s, tok/s=16276.0]
Inference: 90%|████████▉ | 5319/5920 [57:15<03:45, 2.66it/s, tok/s=16276.0]
Inference: 90%|████████▉ | 5319/5920 [57:15<03:45, 2.66it/s, tok/s=16275.8]
Inference: 90%|████████▉ | 5320/5920 [57:15<03:12, 3.12it/s, tok/s=16275.8]
Inference: 90%|████████▉ | 5320/5920 [57:15<03:12, 3.12it/s, tok/s=16278.6]
Inference: 90%|████████▉ | 5321/5920 [57:15<03:11, 3.12it/s, tok/s=16280.0]
Inference: 90%|████████▉ | 5322/5920 [57:15<03:09, 3.16it/s, tok/s=16280.0]
Inference: 90%|████████▉ | 5322/5920 [57:15<03:09, 3.16it/s, tok/s=16280.7]
Inference: 90%|████████▉ | 5323/5920 [57:15<03:08, 3.16it/s, tok/s=16282.8]
Inference: 90%|████████▉ | 5324/5920 [57:16<02:20, 4.24it/s, tok/s=16282.8]
Inference: 90%|████████▉ | 5324/5920 [57:16<02:20, 4.24it/s, tok/s=16284.5]
Inference: 90%|████████▉ | 5325/5920 [57:16<02:16, 4.35it/s, tok/s=16284.5]
Inference: 90%|████████▉ | 5325/5920 [57:16<02:16, 4.35it/s, tok/s=16290.6]
Inference: 90%|████████▉ | 5326/5920 [57:16<02:12, 4.48it/s, tok/s=16290.6]
Inference: 90%|████████▉ | 5326/5920 [57:16<02:12, 4.48it/s, tok/s=16290.7]
Inference: 90%|████████▉ | 5327/5920 [57:17<03:00, 3.28it/s, tok/s=16290.7]
Inference: 90%|████████▉ | 5327/5920 [57:17<03:00, 3.28it/s, tok/s=16290.3]
Inference: 90%|█████████ | 5328/5920 [57:17<04:47, 2.06it/s, tok/s=16290.3]
Inference: 90%|█████████ | 5328/5920 [57:17<04:47, 2.06it/s, tok/s=16286.3]
Inference: 90%|█████████ | 5329/5920 [57:18<05:14, 1.88it/s, tok/s=16286.3]
Inference: 90%|█████████ | 5329/5920 [57:18<05:14, 1.88it/s, tok/s=16284.6]
Inference: 90%|█████████ | 5330/5920 [57:18<04:36, 2.14it/s, tok/s=16284.6]
Inference: 90%|█████████ | 5330/5920 [57:18<04:36, 2.14it/s, tok/s=16284.0]
Inference: 90%|█████████ | 5331/5920 [57:20<06:44, 1.46it/s, tok/s=16284.0]
Inference: 90%|█████████ | 5331/5920 [57:20<06:44, 1.46it/s, tok/s=16279.3]
Inference: 90%|█████████ | 5332/5920 [57:21<07:55, 1.24it/s, tok/s=16279.3]
Inference: 90%|█████████ | 5332/5920 [57:21<07:55, 1.24it/s, tok/s=16277.3]
Inference: 90%|█████████ | 5333/5920 [57:21<06:16, 1.56it/s, tok/s=16277.3]
Inference: 90%|█████████ | 5333/5920 [57:21<06:16, 1.56it/s, tok/s=16279.3]
Inference: 90%|█████████ | 5334/5920 [57:22<07:51, 1.24it/s, tok/s=16279.3]
Inference: 90%|█████████ | 5334/5920 [57:22<07:51, 1.24it/s, tok/s=16275.6]
Inference: 90%|█████████ | 5335/5920 [57:22<05:57, 1.64it/s, tok/s=16275.6]
Inference: 90%|█████████ | 5335/5920 [57:22<05:57, 1.64it/s, tok/s=16276.7]
Inference: 90%|█████████ | 5336/5920 [57:23<05:36, 1.74it/s, tok/s=16276.7]
Inference: 90%|█████████ | 5336/5920 [57:23<05:36, 1.74it/s, tok/s=16275.9]
Inference: 90%|█████████ | 5337/5920 [57:23<05:35, 1.74it/s, tok/s=16279.0]
Inference: 90%|█████████ | 5338/5920 [57:23<03:16, 2.96it/s, tok/s=16279.0]
Inference: 90%|█████████ | 5338/5920 [57:23<03:16, 2.96it/s, tok/s=16279.6]
Inference: 90%|█████████ | 5339/5920 [57:23<03:16, 2.96it/s, tok/s=16282.3]
Inference: 90%|█████████ | 5340/5920 [57:23<02:57, 3.26it/s, tok/s=16282.3]
Inference: 90%|█████████ | 5340/5920 [57:23<02:57, 3.26it/s, tok/s=16281.4]
Inference: 90%|█████████ | 5341/5920 [57:24<03:05, 3.11it/s, tok/s=16281.4]
Inference: 90%|█████████ | 5341/5920 [57:24<03:05, 3.11it/s, tok/s=16282.2]
Inference: 90%|█████████ | 5342/5920 [57:24<03:05, 3.11it/s, tok/s=16282.9]
Inference: 90%|█████████ | 5343/5920 [57:24<03:05, 3.11it/s, tok/s=16287.6]
Inference: 90%|█████████ | 5344/5920 [57:24<02:05, 4.59it/s, tok/s=16287.6]
Inference: 90%|█████████ | 5344/5920 [57:24<02:05, 4.59it/s, tok/s=16286.6]
Inference: 90%|█████████ | 5345/5920 [57:24<02:05, 4.59it/s, tok/s=16289.1]
Inference: 90%|█████████ | 5346/5920 [57:24<01:53, 5.06it/s, tok/s=16289.1]
Inference: 90%|█████████ | 5346/5920 [57:24<01:53, 5.06it/s, tok/s=16288.3]
Inference: 90%|█████████ | 5347/5920 [57:25<01:45, 5.44it/s, tok/s=16288.3]
Inference: 90%|█████████ | 5347/5920 [57:25<01:45, 5.44it/s, tok/s=16288.7]
Inference: 90%|█████████ | 5348/5920 [57:26<04:26, 2.15it/s, tok/s=16288.7]
Inference: 90%|█████████ | 5348/5920 [57:26<04:26, 2.15it/s, tok/s=16283.9]
Inference: 90%|█████████ | 5349/5920 [57:26<03:47, 2.51it/s, tok/s=16283.9]
Inference: 90%|█████████ | 5349/5920 [57:26<03:47, 2.51it/s, tok/s=16286.2]
Inference: 90%|█████████ | 5350/5920 [57:27<03:35, 2.65it/s, tok/s=16286.2]
Inference: 90%|█████████ | 5350/5920 [57:27<03:35, 2.65it/s, tok/s=16287.0]
Inference: 90%|█████████ | 5351/5920 [57:27<03:06, 3.06it/s, tok/s=16287.0]
Inference: 90%|█████████ | 5351/5920 [57:27<03:06, 3.06it/s, tok/s=16288.1]
Inference: 90%|█████████ | 5352/5920 [57:27<03:05, 3.06it/s, tok/s=16291.5]
Inference: 90%|█████████ | 5353/5920 [57:27<02:12, 4.28it/s, tok/s=16291.5]
Inference: 90%|█████████ | 5353/5920 [57:27<02:12, 4.28it/s, tok/s=16292.5]
Inference: 90%|█████████ | 5354/5920 [57:27<02:31, 3.73it/s, tok/s=16292.5]
Inference: 90%|█████████ | 5354/5920 [57:27<02:31, 3.73it/s, tok/s=16291.5]
Inference: 90%|█████████ | 5355/5920 [57:27<02:31, 3.73it/s, tok/s=16295.2]
Inference: 90%|█████████ | 5356/5920 [57:28<02:20, 4.01it/s, tok/s=16295.2]
Inference: 90%|█████████ | 5356/5920 [57:28<02:20, 4.01it/s, tok/s=16294.1]
Inference: 90%|█████████ | 5357/5920 [57:28<02:05, 4.50it/s, tok/s=16294.1]
Inference: 90%|█████████ | 5357/5920 [57:28<02:05, 4.50it/s, tok/s=16293.9]
Inference: 91%|█████████ | 5358/5920 [57:28<01:57, 4.78it/s, tok/s=16293.9]
Inference: 91%|█████████ | 5358/5920 [57:28<01:57, 4.78it/s, tok/s=16295.6]
Inference: 91%|█████████ | 5359/5920 [57:28<01:54, 4.89it/s, tok/s=16295.6]
Inference: 91%|█████████ | 5359/5920 [57:28<01:54, 4.89it/s, tok/s=16295.4]
Inference: 91%|█████████ | 5360/5920 [57:29<02:46, 3.36it/s, tok/s=16295.4]
Inference: 91%|█████████ | 5360/5920 [57:29<02:46, 3.36it/s, tok/s=16293.4]
Inference: 91%|█████████ | 5361/5920 [57:30<03:38, 2.56it/s, tok/s=16293.4]
Inference: 91%|█████████ | 5361/5920 [57:30<03:38, 2.56it/s, tok/s=16292.2]
Inference: 91%|█████████ | 5362/5920 [57:30<03:51, 2.41it/s, tok/s=16292.2]
Inference: 91%|█████████ | 5362/5920 [57:30<03:51, 2.41it/s, tok/s=16292.0]
Inference: 91%|█████████ | 5363/5920 [57:30<03:04, 3.03it/s, tok/s=16292.0]
Inference: 91%|█████████ | 5363/5920 [57:30<03:04, 3.03it/s, tok/s=16294.6]
Inference: 91%|█████████ | 5364/5920 [57:30<03:03, 3.03it/s, tok/s=16297.2]
Inference: 91%|█████████ | 5365/5920 [57:30<03:03, 3.03it/s, tok/s=16298.3]
Inference: 91%|█████████ | 5366/5920 [57:32<04:35, 2.01it/s, tok/s=16298.3]
Inference: 91%|█████████ | 5366/5920 [57:32<04:35, 2.01it/s, tok/s=16291.4]
Inference: 91%|█████████ | 5367/5920 [57:32<04:13, 2.18it/s, tok/s=16291.4]
Inference: 91%|█████████ | 5367/5920 [57:32<04:13, 2.18it/s, tok/s=16290.4]
Inference: 91%|█████████ | 5368/5920 [57:32<04:12, 2.18it/s, tok/s=16293.8]
Inference: 91%|█████████ | 5369/5920 [57:32<04:12, 2.18it/s, tok/s=16295.7]
Inference: 91%|█████████ | 5370/5920 [57:33<03:06, 2.95it/s, tok/s=16295.7]
Inference: 91%|█████████ | 5370/5920 [57:33<03:06, 2.95it/s, tok/s=16293.6]
Inference: 91%|█████████ | 5371/5920 [57:34<04:27, 2.05it/s, tok/s=16293.6]
Inference: 91%|█████████ | 5371/5920 [57:34<04:27, 2.05it/s, tok/s=16292.8]
Inference: 91%|█████████ | 5372/5920 [57:35<05:56, 1.54it/s, tok/s=16292.8]
Inference: 91%|█████████ | 5372/5920 [57:35<05:56, 1.54it/s, tok/s=16287.3]
Inference: 91%|█████████ | 5373/5920 [57:36<05:17, 1.72it/s, tok/s=16287.3]
Inference: 91%|█████████ | 5373/5920 [57:36<05:17, 1.72it/s, tok/s=16288.5]
Inference: 91%|█████████ | 5374/5920 [57:37<07:40, 1.18it/s, tok/s=16288.5]
Inference: 91%|█████████ | 5374/5920 [57:37<07:40, 1.18it/s, tok/s=16282.1]
Inference: 91%|█████████ | 5375/5920 [57:38<07:28, 1.22it/s, tok/s=16282.1]
Inference: 91%|█████████ | 5375/5920 [57:38<07:28, 1.22it/s, tok/s=16280.8]
Inference: 91%|█████████ | 5376/5920 [57:38<07:27, 1.22it/s, tok/s=16283.2]
Inference: 91%|█████████ | 5377/5920 [57:38<04:34, 1.98it/s, tok/s=16283.2]
Inference: 91%|█████████ | 5377/5920 [57:38<04:34, 1.98it/s, tok/s=16283.9]
Inference: 91%|█████████ | 5378/5920 [57:39<03:58, 2.27it/s, tok/s=16283.9]
Inference: 91%|█████████ | 5378/5920 [57:39<03:58, 2.27it/s, tok/s=16285.2]
Inference: 91%|█████████ | 5379/5920 [57:40<05:47, 1.55it/s, tok/s=16285.2]
Inference: 91%|█████████ | 5379/5920 [57:40<05:47, 1.55it/s, tok/s=16285.1]
Inference: 91%|█████████ | 5380/5920 [57:40<04:57, 1.82it/s, tok/s=16285.1]
Inference: 91%|█████████ | 5380/5920 [57:40<04:57, 1.82it/s, tok/s=16285.7]
Inference: 91%|█████████ | 5381/5920 [57:41<05:41, 1.58it/s, tok/s=16285.7]
Inference: 91%|█████████ | 5381/5920 [57:41<05:41, 1.58it/s, tok/s=16282.5]
Inference: 91%|█████████ | 5382/5920 [57:41<05:40, 1.58it/s, tok/s=16285.2]
Inference: 91%|█████████ | 5383/5920 [57:41<04:06, 2.18it/s, tok/s=16285.2]
Inference: 91%|█████████ | 5383/5920 [57:41<04:06, 2.18it/s, tok/s=16283.7]
Inference: 91%|█████████ | 5384/5920 [57:43<05:34, 1.60it/s, tok/s=16283.7]
Inference: 91%|█████████ | 5384/5920 [57:43<05:34, 1.60it/s, tok/s=16279.2]
Inference: 91%|█████████ | 5385/5920 [57:43<04:29, 1.99it/s, tok/s=16279.2]
Inference: 91%|█████████ | 5385/5920 [57:43<04:29, 1.99it/s, tok/s=16282.5]
Inference: 91%|█████████ | 5386/5920 [57:44<05:37, 1.58it/s, tok/s=16282.5]
Inference: 91%|█████████ | 5386/5920 [57:44<05:37, 1.58it/s, tok/s=16280.9]
Inference: 91%|█████████ | 5387/5920 [57:44<04:39, 1.91it/s, tok/s=16280.9]
Inference: 91%|█████████ | 5387/5920 [57:44<04:39, 1.91it/s, tok/s=16280.9]
Inference: 91%|█████████ | 5388/5920 [57:44<03:38, 2.43it/s, tok/s=16280.9]
Inference: 91%|█████████ | 5388/5920 [57:44<03:38, 2.43it/s, tok/s=16282.7]
Inference: 91%|█████████ | 5389/5920 [57:44<03:31, 2.51it/s, tok/s=16282.7]
Inference: 91%|█████████ | 5389/5920 [57:44<03:31, 2.51it/s, tok/s=16281.4]
Inference: 91%|█████████ | 5390/5920 [57:45<04:44, 1.86it/s, tok/s=16281.4]
Inference: 91%|█████████ | 5390/5920 [57:45<04:44, 1.86it/s, tok/s=16279.4]
Inference: 91%|█████████ | 5391/5920 [57:45<03:48, 2.31it/s, tok/s=16279.4]
Inference: 91%|█████████ | 5391/5920 [57:45<03:48, 2.31it/s, tok/s=16282.0]
Inference: 91%|█████████ | 5392/5920 [57:46<05:08, 1.71it/s, tok/s=16282.0]
Inference: 91%|█████████ | 5392/5920 [57:46<05:08, 1.71it/s, tok/s=16281.9]
Inference: 91%|█████████ | 5393/5920 [57:47<05:35, 1.57it/s, tok/s=16281.9]
Inference: 91%|█████████ | 5393/5920 [57:47<05:35, 1.57it/s, tok/s=16283.9]
Inference: 91%|█████████ | 5394/5920 [57:47<04:15, 2.06it/s, tok/s=16283.9]
Inference: 91%|█████████ | 5394/5920 [57:47<04:15, 2.06it/s, tok/s=16287.6]
Inference: 91%|█████████ | 5395/5920 [57:47<04:15, 2.06it/s, tok/s=16288.3]
Inference: 91%|█████████ | 5396/5920 [57:47<02:38, 3.32it/s, tok/s=16288.3]
Inference: 91%|█████████ | 5396/5920 [57:47<02:38, 3.32it/s, tok/s=16290.5]
Inference: 91%|█████████ | 5397/5920 [57:48<03:51, 2.26it/s, tok/s=16290.5]
Inference: 91%|█████████ | 5397/5920 [57:48<03:51, 2.26it/s, tok/s=16294.1]
Inference: 91%|█████████ | 5398/5920 [57:48<03:12, 2.71it/s, tok/s=16294.1]
Inference: 91%|█████████ | 5398/5920 [57:48<03:12, 2.71it/s, tok/s=16294.0]
Inference: 91%|█████████ | 5399/5920 [57:49<04:03, 2.14it/s, tok/s=16294.0]
Inference: 91%|█████████ | 5399/5920 [57:49<04:03, 2.14it/s, tok/s=16292.2]
Inference: 91%|█████████ | 5400/5920 [57:50<03:49, 2.26it/s, tok/s=16292.2]
Inference: 91%|█████████ | 5400/5920 [57:50<03:49, 2.26it/s, tok/s=16292.8]
Inference: 91%|█████████ | 5401/5920 [57:50<03:08, 2.76it/s, tok/s=16292.8]
Inference: 91%|█████████ | 5401/5920 [57:50<03:08, 2.76it/s, tok/s=16293.8]
Inference: 91%|█████████▏| 5402/5920 [57:50<03:34, 2.42it/s, tok/s=16293.8]
Inference: 91%|█████████▏| 5402/5920 [57:50<03:34, 2.42it/s, tok/s=16291.7]
Inference: 91%|█████████▏| 5403/5920 [57:50<02:59, 2.88it/s, tok/s=16291.7]
Inference: 91%|█████████▏| 5403/5920 [57:50<02:59, 2.88it/s, tok/s=16291.3]
Inference: 91%|█████████▏| 5404/5920 [57:51<02:25, 3.54it/s, tok/s=16291.3]
Inference: 91%|█████████▏| 5404/5920 [57:51<02:25, 3.54it/s, tok/s=16295.2]
Inference: 91%|█████████▏| 5405/5920 [57:51<02:25, 3.54it/s, tok/s=16295.6]
Inference: 91%|█████████▏| 5406/5920 [57:51<01:46, 4.84it/s, tok/s=16295.6]
Inference: 91%|█████████▏| 5406/5920 [57:51<01:46, 4.84it/s, tok/s=16295.3]
Inference: 91%|█████████▏| 5407/5920 [57:52<02:53, 2.96it/s, tok/s=16295.3]
Inference: 91%|█████████▏| 5407/5920 [57:52<02:53, 2.96it/s, tok/s=16292.3]
Inference: 91%|█████████▏| 5408/5920 [57:52<02:24, 3.55it/s, tok/s=16292.3]
Inference: 91%|█████████▏| 5408/5920 [57:52<02:24, 3.55it/s, tok/s=16295.0]
Inference: 91%|█████████▏| 5409/5920 [57:52<02:23, 3.57it/s, tok/s=16295.0]
Inference: 91%|█████████▏| 5409/5920 [57:52<02:23, 3.57it/s, tok/s=16296.0]
Inference: 91%|█████████▏| 5410/5920 [57:52<02:02, 4.17it/s, tok/s=16296.0]
Inference: 91%|█████████▏| 5410/5920 [57:52<02:02, 4.17it/s, tok/s=16299.6]
Inference: 91%|█████████▏| 5411/5920 [57:52<01:49, 4.66it/s, tok/s=16299.6]
Inference: 91%|█████████▏| 5411/5920 [57:52<01:49, 4.66it/s, tok/s=16301.3]
Inference: 91%|█████████▏| 5412/5920 [57:53<03:49, 2.22it/s, tok/s=16301.3]
Inference: 91%|█████████▏| 5412/5920 [57:53<03:49, 2.22it/s, tok/s=16299.8]
Inference: 91%|█████████▏| 5413/5920 [57:53<03:48, 2.22it/s, tok/s=16300.1]
Inference: 91%|█████████▏| 5414/5920 [57:54<02:36, 3.23it/s, tok/s=16300.1]
Inference: 91%|█████████▏| 5414/5920 [57:54<02:36, 3.23it/s, tok/s=16301.9]
Inference: 91%|█████████▏| 5415/5920 [57:54<02:34, 3.27it/s, tok/s=16301.9]
Inference: 91%|█████████▏| 5415/5920 [57:54<02:34, 3.27it/s, tok/s=16301.0]
Inference: 91%|█████████▏| 5416/5920 [57:54<02:34, 3.27it/s, tok/s=16301.0]
Inference: 92%|█████████▏| 5417/5920 [57:54<01:50, 4.54it/s, tok/s=16301.0]
Inference: 92%|█████████▏| 5417/5920 [57:54<01:50, 4.54it/s, tok/s=16301.3]
Inference: 92%|█████████▏| 5418/5920 [57:54<01:50, 4.54it/s, tok/s=16303.1]
Inference: 92%|█████████▏| 5419/5920 [57:54<01:19, 6.29it/s, tok/s=16303.1]
Inference: 92%|█████████▏| 5419/5920 [57:54<01:19, 6.29it/s, tok/s=16304.2]
Inference: 92%|█████████▏| 5420/5920 [57:55<02:15, 3.68it/s, tok/s=16304.2]
Inference: 92%|█████████▏| 5420/5920 [57:55<02:15, 3.68it/s, tok/s=16307.1]
Inference: 92%|█████████▏| 5421/5920 [57:55<02:15, 3.68it/s, tok/s=16308.8]
Inference: 92%|█████████▏| 5422/5920 [57:55<01:56, 4.28it/s, tok/s=16308.8]
Inference: 92%|█████████▏| 5422/5920 [57:55<01:56, 4.28it/s, tok/s=16309.5]
Inference: 92%|█████████▏| 5423/5920 [57:56<02:30, 3.29it/s, tok/s=16309.5]
Inference: 92%|█████████▏| 5423/5920 [57:56<02:30, 3.29it/s, tok/s=16308.2]
Inference: 92%|█████████▏| 5424/5920 [57:56<02:30, 3.29it/s, tok/s=16310.9]
Inference: 92%|█████████▏| 5425/5920 [57:57<02:44, 3.01it/s, tok/s=16310.9]
Inference: 92%|█████████▏| 5425/5920 [57:57<02:44, 3.01it/s, tok/s=16308.3]
Inference: 92%|█████████▏| 5426/5920 [57:57<02:44, 3.01it/s, tok/s=16312.5]
Inference: 92%|█████████▏| 5427/5920 [57:57<01:55, 4.27it/s, tok/s=16312.5]
Inference: 92%|█████████▏| 5427/5920 [57:57<01:55, 4.27it/s, tok/s=16316.5]
Inference: 92%|█████████▏| 5428/5920 [57:57<02:45, 2.97it/s, tok/s=16316.5]
Inference: 92%|█████████▏| 5428/5920 [57:57<02:45, 2.97it/s, tok/s=16313.5]
Inference: 92%|█████████▏| 5429/5920 [57:58<02:21, 3.47it/s, tok/s=16313.5]
Inference: 92%|█████████▏| 5429/5920 [57:58<02:21, 3.47it/s, tok/s=16313.6]
Inference: 92%|█████████▏| 5430/5920 [57:58<02:17, 3.57it/s, tok/s=16313.6]
Inference: 92%|█████████▏| 5430/5920 [57:58<02:17, 3.57it/s, tok/s=16316.5]
Inference: 92%|█████████▏| 5431/5920 [57:58<02:53, 2.81it/s, tok/s=16316.5]
Inference: 92%|█████████▏| 5431/5920 [57:58<02:53, 2.81it/s, tok/s=16315.5]
Inference: 92%|█████████▏| 5432/5920 [57:59<03:46, 2.15it/s, tok/s=16315.5]
Inference: 92%|█████████▏| 5432/5920 [57:59<03:46, 2.15it/s, tok/s=16313.5]
Inference: 92%|█████████▏| 5433/5920 [57:59<03:32, 2.29it/s, tok/s=16313.5]
Inference: 92%|█████████▏| 5433/5920 [57:59<03:32, 2.29it/s, tok/s=16312.3]
Inference: 92%|█████████▏| 5434/5920 [58:00<03:17, 2.46it/s, tok/s=16312.3]
Inference: 92%|█████████▏| 5434/5920 [58:00<03:17, 2.46it/s, tok/s=16313.0]
Inference: 92%|█████████▏| 5435/5920 [58:00<03:16, 2.47it/s, tok/s=16313.0]
Inference: 92%|█████████▏| 5435/5920 [58:00<03:16, 2.47it/s, tok/s=16311.7]
Inference: 92%|█████████▏| 5436/5920 [58:01<03:13, 2.50it/s, tok/s=16311.7]
Inference: 92%|█████████▏| 5436/5920 [58:01<03:13, 2.50it/s, tok/s=16313.5]
Inference: 92%|█████████▏| 5437/5920 [58:01<02:34, 3.12it/s, tok/s=16313.5]
Inference: 92%|█████████▏| 5437/5920 [58:01<02:34, 3.12it/s, tok/s=16314.7]
Inference: 92%|█████████▏| 5438/5920 [58:01<02:50, 2.83it/s, tok/s=16314.7]
Inference: 92%|█████████▏| 5438/5920 [58:01<02:50, 2.83it/s, tok/s=16313.3]
Inference: 92%|█████████▏| 5439/5920 [58:01<02:44, 2.92it/s, tok/s=16313.3]
Inference: 92%|█████████▏| 5439/5920 [58:01<02:44, 2.92it/s, tok/s=16315.1]
Inference: 92%|█████████▏| 5440/5920 [58:01<02:44, 2.92it/s, tok/s=16316.5]
Inference: 92%|█████████▏| 5441/5920 [58:02<01:54, 4.18it/s, tok/s=16316.5]
Inference: 92%|█████████▏| 5441/5920 [58:02<01:54, 4.18it/s, tok/s=16317.5]
Inference: 92%|█████████▏| 5442/5920 [58:02<02:02, 3.90it/s, tok/s=16317.5]
Inference: 92%|█████████▏| 5442/5920 [58:02<02:02, 3.90it/s, tok/s=16319.5]
Inference: 92%|█████████▏| 5443/5920 [58:03<02:59, 2.66it/s, tok/s=16319.5]
Inference: 92%|█████████▏| 5443/5920 [58:03<02:59, 2.66it/s, tok/s=16318.0]
Inference: 92%|█████████▏| 5444/5920 [58:03<02:58, 2.66it/s, tok/s=16318.1]
Inference: 92%|█████████▏| 5445/5920 [58:03<02:45, 2.87it/s, tok/s=16318.1]
Inference: 92%|█████████▏| 5445/5920 [58:03<02:45, 2.87it/s, tok/s=16320.6]
Inference: 92%|█████████▏| 5446/5920 [58:04<02:33, 3.08it/s, tok/s=16320.6]
Inference: 92%|█████████▏| 5446/5920 [58:04<02:33, 3.08it/s, tok/s=16325.7]
Inference: 92%|█████████▏| 5447/5920 [58:04<02:37, 3.00it/s, tok/s=16325.7]
Inference: 92%|█████████▏| 5447/5920 [58:04<02:37, 3.00it/s, tok/s=16327.2]
Inference: 92%|█████████▏| 5448/5920 [58:05<03:02, 2.59it/s, tok/s=16327.2]
Inference: 92%|█████████▏| 5448/5920 [58:05<03:02, 2.59it/s, tok/s=16325.1]
Inference: 92%|█████████▏| 5449/5920 [58:05<03:03, 2.57it/s, tok/s=16325.1]
Inference: 92%|█████████▏| 5449/5920 [58:05<03:03, 2.57it/s, tok/s=16325.7]
Inference: 92%|█████████▏| 5450/5920 [58:05<02:30, 3.13it/s, tok/s=16325.7]
Inference: 92%|█████████▏| 5450/5920 [58:05<02:30, 3.13it/s, tok/s=16327.4]
Inference: 92%|█████████▏| 5451/5920 [58:06<02:54, 2.68it/s, tok/s=16327.4]
Inference: 92%|█████████▏| 5451/5920 [58:06<02:54, 2.68it/s, tok/s=16325.6]
Inference: 92%|█████████▏| 5452/5920 [58:06<02:19, 3.36it/s, tok/s=16325.6]
Inference: 92%|█████████▏| 5452/5920 [58:06<02:19, 3.36it/s, tok/s=16329.2]
Inference: 92%|█████████▏| 5453/5920 [58:06<01:57, 3.96it/s, tok/s=16329.2]
Inference: 92%|█████████▏| 5453/5920 [58:06<01:57, 3.96it/s, tok/s=16331.6]
Inference: 92%|█████████▏| 5454/5920 [58:06<01:57, 3.96it/s, tok/s=16332.0]
Inference: 92%|█████████▏| 5455/5920 [58:06<01:17, 6.01it/s, tok/s=16332.0]
Inference: 92%|█████████▏| 5455/5920 [58:06<01:17, 6.01it/s, tok/s=16332.3]
Inference: 92%|█████████▏| 5456/5920 [58:06<01:17, 6.01it/s, tok/s=16335.7]
Inference: 92%|█████████▏| 5457/5920 [58:06<01:19, 5.79it/s, tok/s=16335.7]
Inference: 92%|█████████▏| 5457/5920 [58:06<01:19, 5.79it/s, tok/s=16337.5]
Inference: 92%|█████████▏| 5458/5920 [58:06<01:22, 5.63it/s, tok/s=16337.5]
Inference: 92%|█████████▏| 5458/5920 [58:06<01:22, 5.63it/s, tok/s=16339.9]
Inference: 92%|█████████▏| 5459/5920 [58:07<01:15, 6.14it/s, tok/s=16339.9]
Inference: 92%|█████████▏| 5459/5920 [58:07<01:15, 6.14it/s, tok/s=16339.9]
Inference: 92%|█████████▏| 5460/5920 [58:07<01:48, 4.25it/s, tok/s=16339.9]
Inference: 92%|█████████▏| 5460/5920 [58:07<01:48, 4.25it/s, tok/s=16341.0]
Inference: 92%|█████████▏| 5461/5920 [58:07<01:32, 4.99it/s, tok/s=16341.0]
Inference: 92%|█████████▏| 5461/5920 [58:07<01:32, 4.99it/s, tok/s=16343.5]
Inference: 92%|█████████▏| 5462/5920 [58:07<01:36, 4.75it/s, tok/s=16343.5]
Inference: 92%|█████████▏| 5462/5920 [58:07<01:36, 4.75it/s, tok/s=16344.4]
Inference: 92%|█████████▏| 5463/5920 [58:08<01:29, 5.11it/s, tok/s=16344.4]
Inference: 92%|█████████▏| 5463/5920 [58:08<01:29, 5.11it/s, tok/s=16344.7]
Inference: 92%|█████████▏| 5464/5920 [58:08<01:29, 5.11it/s, tok/s=16344.9]
Inference: 92%|█████████▏| 5465/5920 [58:08<01:26, 5.28it/s, tok/s=16344.9]
Inference: 92%|█████████▏| 5465/5920 [58:08<01:26, 5.28it/s, tok/s=16344.3]
Inference: 92%|█████████▏| 5466/5920 [58:09<03:12, 2.36it/s, tok/s=16344.3]
Inference: 92%|█████████▏| 5466/5920 [58:09<03:12, 2.36it/s, tok/s=16339.8]
Inference: 92%|█████████▏| 5467/5920 [58:09<03:11, 2.36it/s, tok/s=16340.2]
Inference: 92%|█████████▏| 5468/5920 [58:09<02:12, 3.41it/s, tok/s=16340.2]
Inference: 92%|█████████▏| 5468/5920 [58:09<02:12, 3.41it/s, tok/s=16340.6]
Inference: 92%|█████████▏| 5469/5920 [58:10<02:09, 3.48it/s, tok/s=16340.6]
Inference: 92%|█████████▏| 5469/5920 [58:10<02:09, 3.48it/s, tok/s=16348.0]
Inference: 92%|█████████▏| 5470/5920 [58:10<03:03, 2.45it/s, tok/s=16348.0]
Inference: 92%|█████████▏| 5470/5920 [58:10<03:03, 2.45it/s, tok/s=16345.2]
Inference: 92%|█████████▏| 5471/5920 [58:11<02:46, 2.69it/s, tok/s=16345.2]
Inference: 92%|█████████▏| 5471/5920 [58:11<02:46, 2.69it/s, tok/s=16347.5]
Inference: 92%|█████████▏| 5472/5920 [58:11<02:46, 2.69it/s, tok/s=16348.3]
Inference: 92%|█████████▏| 5473/5920 [58:11<01:59, 3.73it/s, tok/s=16348.3]
Inference: 92%|█████████▏| 5473/5920 [58:11<01:59, 3.73it/s, tok/s=16350.2]
Inference: 92%|█████████▏| 5474/5920 [58:12<04:10, 1.78it/s, tok/s=16350.2]
Inference: 92%|█████████▏| 5474/5920 [58:12<04:10, 1.78it/s, tok/s=16345.8]
Inference: 92%|█████████▏| 5475/5920 [58:13<04:39, 1.59it/s, tok/s=16345.8]
Inference: 92%|█████████▏| 5475/5920 [58:13<04:39, 1.59it/s, tok/s=16342.9]
Inference: 92%|█████████▎| 5476/5920 [58:13<04:38, 1.59it/s, tok/s=16348.9]
Inference: 93%|█████████▎| 5477/5920 [58:13<02:59, 2.47it/s, tok/s=16348.9]
Inference: 93%|█████████▎| 5477/5920 [58:13<02:59, 2.47it/s, tok/s=16350.5]
Inference: 93%|█████████▎| 5478/5920 [58:14<02:47, 2.64it/s, tok/s=16350.5]
Inference: 93%|█████████▎| 5478/5920 [58:14<02:47, 2.64it/s, tok/s=16355.7]
Inference: 93%|█████████▎| 5479/5920 [58:14<02:46, 2.64it/s, tok/s=16358.1]
Inference: 93%|█████████▎| 5480/5920 [58:15<02:55, 2.51it/s, tok/s=16358.1]
Inference: 93%|█████████▎| 5480/5920 [58:15<02:55, 2.51it/s, tok/s=16355.0]
Inference: 93%|█████████▎| 5481/5920 [58:15<02:54, 2.51it/s, tok/s=16355.0]
Inference: 93%|█████████▎| 5481/5920 [58:15<02:54, 2.51it/s, tok/s=16354.8]
Inference: 93%|█████████▎| 5482/5920 [58:15<02:54, 2.51it/s, tok/s=16357.5]
Inference: 93%|█████████▎| 5483/5920 [58:16<02:39, 2.74it/s, tok/s=16357.5]
Inference: 93%|█████████▎| 5483/5920 [58:16<02:39, 2.74it/s, tok/s=16356.9]
Inference: 93%|█████████▎| 5484/5920 [58:16<03:10, 2.29it/s, tok/s=16356.9]
Inference: 93%|█████████▎| 5484/5920 [58:16<03:10, 2.29it/s, tok/s=16358.2]
Inference: 93%|█████████▎| 5485/5920 [58:16<02:45, 2.64it/s, tok/s=16358.2]
Inference: 93%|█████████▎| 5485/5920 [58:16<02:45, 2.64it/s, tok/s=16358.5]
Inference: 93%|█████████▎| 5486/5920 [58:17<02:16, 3.17it/s, tok/s=16358.5]
Inference: 93%|█████████▎| 5486/5920 [58:17<02:16, 3.17it/s, tok/s=16360.2]
Inference: 93%|█████████▎| 5487/5920 [58:17<02:45, 2.62it/s, tok/s=16360.2]
Inference: 93%|█████████▎| 5487/5920 [58:17<02:45, 2.62it/s, tok/s=16358.7]
Inference: 93%|█████████▎| 5488/5920 [58:18<02:42, 2.67it/s, tok/s=16358.7]
Inference: 93%|█████████▎| 5488/5920 [58:18<02:42, 2.67it/s, tok/s=16362.0]
Inference: 93%|█████████▎| 5489/5920 [58:18<03:37, 1.98it/s, tok/s=16362.0]
Inference: 93%|█████████▎| 5489/5920 [58:18<03:37, 1.98it/s, tok/s=16362.0]
Inference: 93%|█████████▎| 5490/5920 [58:19<03:05, 2.32it/s, tok/s=16362.0]
Inference: 93%|█████████▎| 5490/5920 [58:19<03:05, 2.32it/s, tok/s=16362.8]
Inference: 93%|█████████▎| 5491/5920 [58:19<02:31, 2.84it/s, tok/s=16362.8]
Inference: 93%|█████████▎| 5491/5920 [58:19<02:31, 2.84it/s, tok/s=16368.5]
Inference: 93%|█████████▎| 5492/5920 [58:19<02:30, 2.84it/s, tok/s=16368.8]
Inference: 93%|█████████▎| 5493/5920 [58:20<04:00, 1.78it/s, tok/s=16368.8]
Inference: 93%|█████████▎| 5493/5920 [58:20<04:00, 1.78it/s, tok/s=16363.8]
Inference: 93%|█████████▎| 5494/5920 [58:21<03:41, 1.92it/s, tok/s=16363.8]
Inference: 93%|█████████▎| 5494/5920 [58:21<03:41, 1.92it/s, tok/s=16362.7]
Inference: 93%|█████████▎| 5495/5920 [58:22<04:30, 1.57it/s, tok/s=16362.7]
Inference: 93%|█████████▎| 5495/5920 [58:22<04:30, 1.57it/s, tok/s=16359.3]
Inference: 93%|█████████▎| 5496/5920 [58:22<03:40, 1.93it/s, tok/s=16359.3]
Inference: 93%|█████████▎| 5496/5920 [58:22<03:40, 1.93it/s, tok/s=16358.9]
Inference: 93%|█████████▎| 5497/5920 [58:22<03:39, 1.93it/s, tok/s=16359.3]
Inference: 93%|█████████▎| 5498/5920 [58:22<02:14, 3.14it/s, tok/s=16359.3]
Inference: 93%|█████████▎| 5498/5920 [58:22<02:14, 3.14it/s, tok/s=16360.7]
Inference: 93%|█████████▎| 5499/5920 [58:23<03:27, 2.03it/s, tok/s=16360.7]
Inference: 93%|█████████▎| 5499/5920 [58:23<03:27, 2.03it/s, tok/s=16356.5]
Inference: 93%|█████████▎| 5500/5920 [58:23<03:04, 2.28it/s, tok/s=16356.5]
Inference: 93%|█████████▎| 5500/5920 [58:23<03:04, 2.28it/s, tok/s=16356.5]
Inference: 93%|█████████▎| 5501/5920 [58:24<03:25, 2.04it/s, tok/s=16356.5]
Inference: 93%|█████████▎| 5501/5920 [58:24<03:25, 2.04it/s, tok/s=16355.4]
Inference: 93%|█████████▎| 5502/5920 [58:24<03:24, 2.04it/s, tok/s=16357.0]
Inference: 93%|█████████▎| 5503/5920 [58:25<02:51, 2.43it/s, tok/s=16357.0]
Inference: 93%|█████████▎| 5503/5920 [58:25<02:51, 2.43it/s, tok/s=16355.1]
Inference: 93%|█████████▎| 5504/5920 [58:26<04:21, 1.59it/s, tok/s=16355.1]
Inference: 93%|█████████▎| 5504/5920 [58:26<04:21, 1.59it/s, tok/s=16353.5]
Inference: 93%|█████████▎| 5505/5920 [58:26<03:43, 1.85it/s, tok/s=16353.5]
Inference: 93%|█████████▎| 5505/5920 [58:26<03:43, 1.85it/s, tok/s=16354.2]
Inference: 93%|█████████▎| 5506/5920 [58:28<05:07, 1.35it/s, tok/s=16354.2]
Inference: 93%|█████████▎| 5506/5920 [58:28<05:07, 1.35it/s, tok/s=16349.0]
Inference: 93%|█████████▎| 5507/5920 [58:28<04:36, 1.50it/s, tok/s=16349.0]
Inference: 93%|█████████▎| 5507/5920 [58:28<04:36, 1.50it/s, tok/s=16350.6]
Inference: 93%|█████████▎| 5508/5920 [58:28<03:40, 1.87it/s, tok/s=16350.6]
Inference: 93%|█████████▎| 5508/5920 [58:28<03:40, 1.87it/s, tok/s=16352.7]
Inference: 93%|█████████▎| 5509/5920 [58:29<03:39, 1.87it/s, tok/s=16352.7]
Inference: 93%|█████████▎| 5509/5920 [58:29<03:39, 1.87it/s, tok/s=16350.9]
Inference: 93%|█████████▎| 5510/5920 [58:29<03:58, 1.72it/s, tok/s=16350.9]
Inference: 93%|█████████▎| 5510/5920 [58:29<03:58, 1.72it/s, tok/s=16350.3]
Inference: 93%|█████████▎| 5511/5920 [58:30<03:58, 1.72it/s, tok/s=16352.2]
Inference: 93%|█████████▎| 5512/5920 [58:30<02:51, 2.38it/s, tok/s=16352.2]
Inference: 93%|█████████▎| 5512/5920 [58:30<02:51, 2.38it/s, tok/s=16351.7]
Inference: 93%|█████████▎| 5513/5920 [58:31<03:36, 1.88it/s, tok/s=16351.7]
Inference: 93%|█████████▎| 5513/5920 [58:31<03:36, 1.88it/s, tok/s=16351.0]
Inference: 93%|█████████▎| 5514/5920 [58:31<03:30, 1.93it/s, tok/s=16351.0]
Inference: 93%|█████████▎| 5514/5920 [58:31<03:30, 1.93it/s, tok/s=16349.4]
Inference: 93%|█████████▎| 5515/5920 [58:31<02:47, 2.42it/s, tok/s=16349.4]
Inference: 93%|█████████▎| 5515/5920 [58:31<02:47, 2.42it/s, tok/s=16353.1]
Inference: 93%|█████████▎| 5516/5920 [58:32<02:42, 2.49it/s, tok/s=16353.1]
Inference: 93%|█████████▎| 5516/5920 [58:32<02:42, 2.49it/s, tok/s=16354.5]
Inference: 93%|█████████▎| 5517/5920 [58:32<02:28, 2.71it/s, tok/s=16354.5]
Inference: 93%|█████████▎| 5517/5920 [58:32<02:28, 2.71it/s, tok/s=16356.6]
Inference: 93%|█████████▎| 5518/5920 [58:32<02:28, 2.72it/s, tok/s=16356.6]
Inference: 93%|█████████▎| 5518/5920 [58:32<02:28, 2.72it/s, tok/s=16355.7]
Inference: 93%|█████████▎| 5519/5920 [58:33<01:59, 3.34it/s, tok/s=16355.7]
Inference: 93%|█████████▎| 5519/5920 [58:33<01:59, 3.34it/s, tok/s=16358.3]
Inference: 93%|█████████▎| 5520/5920 [58:33<01:42, 3.91it/s, tok/s=16358.3]
Inference: 93%|█████████▎| 5520/5920 [58:33<01:42, 3.91it/s, tok/s=16360.4]
Inference: 93%|█████████▎| 5521/5920 [58:33<01:25, 4.69it/s, tok/s=16360.4]
Inference: 93%|█████████▎| 5521/5920 [58:33<01:25, 4.69it/s, tok/s=16363.7]
Inference: 93%|█████████▎| 5522/5920 [58:33<01:33, 4.24it/s, tok/s=16363.7]
Inference: 93%|█████████▎| 5522/5920 [58:33<01:33, 4.24it/s, tok/s=16364.4]
Inference: 93%|█████████▎| 5523/5920 [58:33<01:19, 4.97it/s, tok/s=16364.4]
Inference: 93%|█████████▎| 5523/5920 [58:33<01:19, 4.97it/s, tok/s=16366.2]
Inference: 93%|█████████▎| 5524/5920 [58:34<01:43, 3.84it/s, tok/s=16366.2]
Inference: 93%|█████████▎| 5524/5920 [58:34<01:43, 3.84it/s, tok/s=16367.9]
Inference: 93%|█████████▎| 5525/5920 [58:34<01:42, 3.84it/s, tok/s=16370.2]
Inference: 93%|█████████▎| 5526/5920 [58:34<01:54, 3.45it/s, tok/s=16370.2]
Inference: 93%|█████████▎| 5526/5920 [58:34<01:54, 3.45it/s, tok/s=16371.3]
Inference: 93%|█████████▎| 5527/5920 [58:35<02:26, 2.68it/s, tok/s=16371.3]
Inference: 93%|█████████▎| 5527/5920 [58:35<02:26, 2.68it/s, tok/s=16375.1]
Inference: 93%|█████████▎| 5528/5920 [58:36<03:26, 1.90it/s, tok/s=16375.1]
Inference: 93%|█████████▎| 5528/5920 [58:36<03:26, 1.90it/s, tok/s=16373.1]
Inference: 93%|█████████▎| 5529/5920 [58:36<03:37, 1.80it/s, tok/s=16373.1]
Inference: 93%|█████████▎| 5529/5920 [58:36<03:37, 1.80it/s, tok/s=16373.5]
Inference: 93%|█████████▎| 5530/5920 [58:38<04:47, 1.35it/s, tok/s=16373.5]
Inference: 93%|█████████▎| 5530/5920 [58:38<04:47, 1.35it/s, tok/s=16370.8]
Inference: 93%|█████████▎| 5531/5920 [58:38<04:47, 1.35it/s, tok/s=16371.7]
Inference: 93%|█████████▎| 5532/5920 [58:38<03:23, 1.91it/s, tok/s=16371.7]
Inference: 93%|█████████▎| 5532/5920 [58:38<03:23, 1.91it/s, tok/s=16373.3]
Inference: 93%|█████████▎| 5533/5920 [58:38<02:55, 2.20it/s, tok/s=16373.3]
Inference: 93%|█████████▎| 5533/5920 [58:38<02:55, 2.20it/s, tok/s=16375.8]
Inference: 93%|█████████▎| 5534/5920 [58:39<02:55, 2.20it/s, tok/s=16377.6]
Inference: 93%|█████████▎| 5535/5920 [58:39<02:07, 3.02it/s, tok/s=16377.6]
Inference: 93%|█████████▎| 5535/5920 [58:39<02:07, 3.02it/s, tok/s=16379.5]
Inference: 94%|█████████▎| 5536/5920 [58:39<02:07, 3.02it/s, tok/s=16381.6]
Inference: 94%|█████████▎| 5537/5920 [58:39<01:46, 3.59it/s, tok/s=16381.6]
Inference: 94%|█████████▎| 5537/5920 [58:39<01:46, 3.59it/s, tok/s=16382.0]
Inference: 94%|█████████▎| 5538/5920 [58:39<01:46, 3.59it/s, tok/s=16382.3]
Inference: 94%|█████████▎| 5539/5920 [58:40<01:51, 3.41it/s, tok/s=16382.3]
Inference: 94%|█████████▎| 5539/5920 [58:40<01:51, 3.41it/s, tok/s=16381.0]
Inference: 94%|█████████▎| 5540/5920 [58:41<02:41, 2.35it/s, tok/s=16381.0]
Inference: 94%|█████████▎| 5540/5920 [58:41<02:41, 2.35it/s, tok/s=16377.2]
Inference: 94%|█████████▎| 5541/5920 [58:41<02:18, 2.74it/s, tok/s=16377.2]
Inference: 94%|█████████▎| 5541/5920 [58:41<02:18, 2.74it/s, tok/s=16377.8]
Inference: 94%|█████████▎| 5542/5920 [58:41<02:29, 2.53it/s, tok/s=16377.8]
Inference: 94%|█████████▎| 5542/5920 [58:41<02:29, 2.53it/s, tok/s=16376.8]
Inference: 94%|█████████▎| 5543/5920 [58:43<03:41, 1.70it/s, tok/s=16376.8]
Inference: 94%|█████████▎| 5543/5920 [58:43<03:41, 1.70it/s, tok/s=16375.3]
Inference: 94%|█████████▎| 5544/5920 [58:43<03:41, 1.70it/s, tok/s=16378.6]
Inference: 94%|█████████▎| 5545/5920 [58:43<02:47, 2.24it/s, tok/s=16378.6]
Inference: 94%|█████████▎| 5545/5920 [58:43<02:47, 2.24it/s, tok/s=16379.6]
Inference: 94%|█████████▎| 5546/5920 [58:43<02:32, 2.45it/s, tok/s=16379.6]
Inference: 94%|█████████▎| 5546/5920 [58:43<02:32, 2.45it/s, tok/s=16379.1]
Inference: 94%|█████████▎| 5547/5920 [58:43<02:32, 2.45it/s, tok/s=16382.2]
Inference: 94%|█████████▎| 5548/5920 [58:43<01:45, 3.52it/s, tok/s=16382.2]
Inference: 94%|█████████▎| 5548/5920 [58:43<01:45, 3.52it/s, tok/s=16382.6]
Inference: 94%|█████████▎| 5549/5920 [58:44<01:32, 4.00it/s, tok/s=16382.6]
Inference: 94%|█████████▎| 5549/5920 [58:44<01:32, 4.00it/s, tok/s=16384.2]
Inference: 94%|█████████▍| 5550/5920 [58:44<01:40, 3.68it/s, tok/s=16384.2]
Inference: 94%|█████████▍| 5550/5920 [58:44<01:40, 3.68it/s, tok/s=16384.4]
Inference: 94%|█████████▍| 5551/5920 [58:44<01:28, 4.16it/s, tok/s=16384.4]
Inference: 94%|█████████▍| 5551/5920 [58:44<01:28, 4.16it/s, tok/s=16386.5]
Inference: 94%|█████████▍| 5552/5920 [58:44<01:25, 4.29it/s, tok/s=16386.5]
Inference: 94%|█████████▍| 5552/5920 [58:44<01:25, 4.29it/s, tok/s=16387.0]
Inference: 94%|█████████▍| 5553/5920 [58:45<01:59, 3.07it/s, tok/s=16387.0]
Inference: 94%|█████████▍| 5553/5920 [58:45<01:59, 3.07it/s, tok/s=16385.3]
Inference: 94%|█████████▍| 5554/5920 [58:46<02:32, 2.40it/s, tok/s=16385.3]
Inference: 94%|█████████▍| 5554/5920 [58:46<02:32, 2.40it/s, tok/s=16383.2]
Inference: 94%|█████████▍| 5555/5920 [58:46<02:32, 2.40it/s, tok/s=16384.0]
Inference: 94%|█████████▍| 5556/5920 [58:46<01:42, 3.54it/s, tok/s=16384.0]
Inference: 94%|█████████▍| 5556/5920 [58:46<01:42, 3.54it/s, tok/s=16385.3]
Inference: 94%|█████████▍| 5557/5920 [58:47<03:00, 2.01it/s, tok/s=16385.3]
Inference: 94%|█████████▍| 5557/5920 [58:47<03:00, 2.01it/s, tok/s=16381.3]
Inference: 94%|█████████▍| 5558/5920 [58:47<02:24, 2.51it/s, tok/s=16381.3]
Inference: 94%|█████████▍| 5558/5920 [58:47<02:24, 2.51it/s, tok/s=16382.8]
Inference: 94%|█████████▍| 5559/5920 [58:47<02:07, 2.83it/s, tok/s=16382.8]
Inference: 94%|█████████▍| 5559/5920 [58:47<02:07, 2.83it/s, tok/s=16384.5]
Inference: 94%|█████████▍| 5560/5920 [58:48<02:32, 2.35it/s, tok/s=16384.5]
Inference: 94%|█████████▍| 5560/5920 [58:48<02:32, 2.35it/s, tok/s=16384.1]
Inference: 94%|█████████▍| 5561/5920 [58:49<03:38, 1.64it/s, tok/s=16384.1]
Inference: 94%|█████████▍| 5561/5920 [58:49<03:38, 1.64it/s, tok/s=16382.5]
Inference: 94%|█████████▍| 5562/5920 [58:49<03:37, 1.64it/s, tok/s=16386.2]
Inference: 94%|█████████▍| 5563/5920 [58:49<02:41, 2.21it/s, tok/s=16386.2]
Inference: 94%|█████████▍| 5563/5920 [58:49<02:41, 2.21it/s, tok/s=16384.9]
Inference: 94%|█████████▍| 5564/5920 [58:50<02:17, 2.59it/s, tok/s=16384.9]
Inference: 94%|█████████▍| 5564/5920 [58:50<02:17, 2.59it/s, tok/s=16391.1]
Inference: 94%|█████████▍| 5565/5920 [58:50<01:51, 3.19it/s, tok/s=16391.1]
Inference: 94%|█████████▍| 5565/5920 [58:50<01:51, 3.19it/s, tok/s=16391.7]
Inference: 94%|█████████▍| 5566/5920 [58:50<01:53, 3.12it/s, tok/s=16391.7]
Inference: 94%|█████████▍| 5566/5920 [58:50<01:53, 3.12it/s, tok/s=16392.9]
Inference: 94%|█████████▍| 5567/5920 [58:50<01:53, 3.12it/s, tok/s=16396.5]
Inference: 94%|█████████▍| 5568/5920 [58:50<01:13, 4.82it/s, tok/s=16396.5]
Inference: 94%|█████████▍| 5568/5920 [58:50<01:13, 4.82it/s, tok/s=16396.6]
Inference: 94%|█████████▍| 5569/5920 [58:51<01:23, 4.23it/s, tok/s=16396.6]
Inference: 94%|█████████▍| 5569/5920 [58:51<01:23, 4.23it/s, tok/s=16399.1]
Inference: 94%|█████████▍| 5570/5920 [58:51<01:41, 3.46it/s, tok/s=16399.1]
Inference: 94%|█████████▍| 5570/5920 [58:51<01:41, 3.46it/s, tok/s=16399.3]
Inference: 94%|█████████▍| 5571/5920 [58:51<01:53, 3.09it/s, tok/s=16399.3]
Inference: 94%|█████████▍| 5571/5920 [58:51<01:53, 3.09it/s, tok/s=16399.4]
Inference: 94%|█████████▍| 5572/5920 [58:52<01:32, 3.74it/s, tok/s=16399.4]
Inference: 94%|█████████▍| 5572/5920 [58:52<01:32, 3.74it/s, tok/s=16403.2]
Inference: 94%|█████████▍| 5573/5920 [58:52<01:32, 3.74it/s, tok/s=16404.9]
Inference: 94%|█████████▍| 5574/5920 [58:52<01:18, 4.38it/s, tok/s=16404.9]
Inference: 94%|█████████▍| 5574/5920 [58:52<01:18, 4.38it/s, tok/s=16405.8]
Inference: 94%|█████████▍| 5575/5920 [58:52<01:18, 4.38it/s, tok/s=16409.5]
Inference: 94%|█████████▍| 5576/5920 [58:52<01:20, 4.25it/s, tok/s=16409.5]
Inference: 94%|█████████▍| 5576/5920 [58:52<01:20, 4.25it/s, tok/s=16409.3]
Inference: 94%|█████████▍| 5577/5920 [58:53<02:15, 2.54it/s, tok/s=16409.3]
Inference: 94%|█████████▍| 5577/5920 [58:53<02:15, 2.54it/s, tok/s=16406.1]
Inference: 94%|█████████▍| 5578/5920 [58:54<02:10, 2.61it/s, tok/s=16406.1]
Inference: 94%|█████████▍| 5578/5920 [58:54<02:10, 2.61it/s, tok/s=16405.8]
Inference: 94%|█████████▍| 5579/5920 [58:55<03:15, 1.74it/s, tok/s=16405.8]
Inference: 94%|█████████▍| 5579/5920 [58:55<03:15, 1.74it/s, tok/s=16402.8]
Inference: 94%|█████████▍| 5580/5920 [58:55<03:00, 1.89it/s, tok/s=16402.8]
Inference: 94%|█████████▍| 5580/5920 [58:55<03:00, 1.89it/s, tok/s=16403.4]
Inference: 94%|█████████▍| 5581/5920 [58:55<02:25, 2.32it/s, tok/s=16403.4]
Inference: 94%|█████████▍| 5581/5920 [58:55<02:25, 2.32it/s, tok/s=16406.2]
Inference: 94%|█████████▍| 5582/5920 [58:56<02:55, 1.92it/s, tok/s=16406.2]
Inference: 94%|█████████▍| 5582/5920 [58:56<02:55, 1.92it/s, tok/s=16405.5]
Inference: 94%|█████████▍| 5583/5920 [58:56<02:23, 2.35it/s, tok/s=16405.5]
Inference: 94%|█████████▍| 5583/5920 [58:56<02:23, 2.35it/s, tok/s=16405.8]
Inference: 94%|█████████▍| 5584/5920 [58:57<02:33, 2.18it/s, tok/s=16405.8]
Inference: 94%|█████████▍| 5584/5920 [58:57<02:33, 2.18it/s, tok/s=16407.0]
Inference: 94%|█████████▍| 5585/5920 [58:58<04:09, 1.34it/s, tok/s=16407.0]
Inference: 94%|█████████▍| 5585/5920 [58:58<04:09, 1.34it/s, tok/s=16403.4]
Inference: 94%|█████████▍| 5586/5920 [58:58<03:09, 1.76it/s, tok/s=16403.4]
Inference: 94%|█████████▍| 5586/5920 [58:58<03:09, 1.76it/s, tok/s=16406.5]
Inference: 94%|█████████▍| 5587/5920 [58:59<02:48, 1.98it/s, tok/s=16406.5]
Inference: 94%|█████████▍| 5587/5920 [58:59<02:48, 1.98it/s, tok/s=16409.8]
Inference: 94%|█████████▍| 5588/5920 [58:59<02:55, 1.89it/s, tok/s=16409.8]
Inference: 94%|█████████▍| 5588/5920 [58:59<02:55, 1.89it/s, tok/s=16410.8]
Inference: 94%|█████████▍| 5589/5920 [59:00<03:35, 1.54it/s, tok/s=16410.8]
Inference: 94%|█████████▍| 5589/5920 [59:00<03:35, 1.54it/s, tok/s=16409.6]
Inference: 94%|█████████▍| 5590/5920 [59:01<03:11, 1.73it/s, tok/s=16409.6]
Inference: 94%|█████████▍| 5590/5920 [59:01<03:11, 1.73it/s, tok/s=16409.3]
Inference: 94%|█████████▍| 5591/5920 [59:02<04:47, 1.14it/s, tok/s=16409.3]
Inference: 94%|█████████▍| 5591/5920 [59:02<04:47, 1.14it/s, tok/s=16403.5]
Inference: 94%|█████████▍| 5592/5920 [59:03<04:18, 1.27it/s, tok/s=16403.5]
Inference: 94%|█████████▍| 5592/5920 [59:03<04:18, 1.27it/s, tok/s=16403.0]
Inference: 94%|█████████▍| 5593/5920 [59:04<03:58, 1.37it/s, tok/s=16403.0]
Inference: 94%|█████████▍| 5593/5920 [59:04<03:58, 1.37it/s, tok/s=16402.8]
Inference: 94%|█████████▍| 5594/5920 [59:04<03:57, 1.37it/s, tok/s=16404.6]
Inference: 95%|█████████▍| 5595/5920 [59:05<03:40, 1.47it/s, tok/s=16404.6]
Inference: 95%|█████████▍| 5595/5920 [59:05<03:40, 1.47it/s, tok/s=16402.9]
Inference: 95%|█████████▍| 5596/5920 [59:05<03:00, 1.80it/s, tok/s=16402.9]
Inference: 95%|█████████▍| 5596/5920 [59:05<03:00, 1.80it/s, tok/s=16406.0]
Inference: 95%|█████████▍| 5597/5920 [59:05<02:33, 2.11it/s, tok/s=16406.0]
Inference: 95%|█████████▍| 5597/5920 [59:05<02:33, 2.11it/s, tok/s=16407.1]
Inference: 95%|█████████▍| 5598/5920 [59:06<03:20, 1.61it/s, tok/s=16407.1]
Inference: 95%|█████████▍| 5598/5920 [59:06<03:20, 1.61it/s, tok/s=16406.8]
Inference: 95%|█████████▍| 5599/5920 [59:07<02:58, 1.80it/s, tok/s=16406.8]
Inference: 95%|█████████▍| 5599/5920 [59:07<02:58, 1.80it/s, tok/s=16407.8]
Inference: 95%|█████████▍| 5600/5920 [59:07<02:28, 2.15it/s, tok/s=16407.8]
Inference: 95%|█████████▍| 5600/5920 [59:07<02:28, 2.15it/s, tok/s=16408.6]
Inference: 95%|█████████▍| 5601/5920 [59:07<02:01, 2.62it/s, tok/s=16408.6]
Inference: 95%|█████████▍| 5601/5920 [59:07<02:01, 2.62it/s, tok/s=16413.0]
Inference: 95%|█████████▍| 5602/5920 [59:07<01:44, 3.05it/s, tok/s=16413.0]
Inference: 95%|█████████▍| 5602/5920 [59:07<01:44, 3.05it/s, tok/s=16413.1]
Inference: 95%|█████████▍| 5603/5920 [59:08<02:56, 1.80it/s, tok/s=16413.1]
Inference: 95%|█████████▍| 5603/5920 [59:08<02:56, 1.80it/s, tok/s=16410.3]
Inference: 95%|█████████▍| 5604/5920 [59:09<02:57, 1.78it/s, tok/s=16410.3]
Inference: 95%|█████████▍| 5604/5920 [59:09<02:57, 1.78it/s, tok/s=16410.8]
Inference: 95%|█████████▍| 5605/5920 [59:09<02:18, 2.28it/s, tok/s=16410.8]
Inference: 95%|█████████▍| 5605/5920 [59:09<02:18, 2.28it/s, tok/s=16415.6]
Inference: 95%|█████████▍| 5606/5920 [59:09<01:50, 2.85it/s, tok/s=16415.6]
Inference: 95%|█████████▍| 5606/5920 [59:09<01:50, 2.85it/s, tok/s=16419.0]
Inference: 95%|█████████▍| 5607/5920 [59:10<02:47, 1.86it/s, tok/s=16419.0]
Inference: 95%|█████████▍| 5607/5920 [59:10<02:47, 1.86it/s, tok/s=16416.9]
Inference: 95%|█████████▍| 5608/5920 [59:10<02:09, 2.41it/s, tok/s=16416.9]
Inference: 95%|█████████▍| 5608/5920 [59:10<02:09, 2.41it/s, tok/s=16418.1]
Inference: 95%|█████████▍| 5609/5920 [59:10<02:09, 2.41it/s, tok/s=16420.3]
Inference: 95%|█████████▍| 5610/5920 [59:11<01:36, 3.20it/s, tok/s=16420.3]
Inference: 95%|█████████▍| 5610/5920 [59:11<01:36, 3.20it/s, tok/s=16419.9]
Inference: 95%|█████████▍| 5611/5920 [59:12<02:59, 1.72it/s, tok/s=16419.9]
Inference: 95%|█████████▍| 5611/5920 [59:12<02:59, 1.72it/s, tok/s=16417.0]
Inference: 95%|█████████▍| 5612/5920 [59:12<02:35, 1.98it/s, tok/s=16417.0]
Inference: 95%|█████████▍| 5612/5920 [59:12<02:35, 1.98it/s, tok/s=16417.0]
Inference: 95%|█████████▍| 5613/5920 [59:13<02:22, 2.15it/s, tok/s=16417.0]
Inference: 95%|█████████▍| 5613/5920 [59:13<02:22, 2.15it/s, tok/s=16416.9]
Inference: 95%|█████████▍| 5614/5920 [59:13<01:51, 2.74it/s, tok/s=16416.9]
Inference: 95%|█████████▍| 5614/5920 [59:13<01:51, 2.74it/s, tok/s=16418.5]
Inference: 95%|█████████▍| 5615/5920 [59:13<01:59, 2.55it/s, tok/s=16418.5]
Inference: 95%|█████████▍| 5615/5920 [59:13<01:59, 2.55it/s, tok/s=16417.9]
Inference: 95%|█████████▍| 5616/5920 [59:14<02:35, 1.95it/s, tok/s=16417.9]
Inference: 95%|█████████▍| 5616/5920 [59:14<02:35, 1.95it/s, tok/s=16417.3]
Inference: 95%|█████████▍| 5617/5920 [59:14<02:35, 1.95it/s, tok/s=16424.6]
Inference: 95%|█████████▍| 5618/5920 [59:15<02:13, 2.26it/s, tok/s=16424.6]
Inference: 95%|█████████▍| 5618/5920 [59:15<02:13, 2.26it/s, tok/s=16427.0]
Inference: 95%|█████████▍| 5619/5920 [59:15<02:16, 2.21it/s, tok/s=16427.0]
Inference: 95%|█████████▍| 5619/5920 [59:15<02:16, 2.21it/s, tok/s=16428.5]
Inference: 95%|█████████▍| 5620/5920 [59:16<02:20, 2.14it/s, tok/s=16428.5]
Inference: 95%|█████████▍| 5620/5920 [59:16<02:20, 2.14it/s, tok/s=16429.7]
Inference: 95%|█████████▍| 5621/5920 [59:17<03:34, 1.40it/s, tok/s=16429.7]
Inference: 95%|█████████▍| 5621/5920 [59:17<03:34, 1.40it/s, tok/s=16425.4]
Inference: 95%|█████████▍| 5622/5920 [59:18<03:30, 1.42it/s, tok/s=16425.4]
Inference: 95%|█████████▍| 5622/5920 [59:18<03:30, 1.42it/s, tok/s=16424.9]
Inference: 95%|█████████▍| 5623/5920 [59:18<03:23, 1.46it/s, tok/s=16424.9]
Inference: 95%|█████████▍| 5623/5920 [59:18<03:23, 1.46it/s, tok/s=16423.1]
Inference: 95%|█████████▌| 5624/5920 [59:19<02:56, 1.68it/s, tok/s=16423.1]
Inference: 95%|█████████▌| 5624/5920 [59:19<02:56, 1.68it/s, tok/s=16424.3]
Inference: 95%|█████████▌| 5625/5920 [59:20<03:32, 1.39it/s, tok/s=16424.3]
Inference: 95%|█████████▌| 5625/5920 [59:20<03:32, 1.39it/s, tok/s=16426.0]
Inference: 95%|█████████▌| 5626/5920 [59:21<03:38, 1.35it/s, tok/s=16426.0]
Inference: 95%|█████████▌| 5626/5920 [59:21<03:38, 1.35it/s, tok/s=16426.6]
Inference: 95%|█████████▌| 5627/5920 [59:21<02:58, 1.64it/s, tok/s=16426.6]
Inference: 95%|█████████▌| 5627/5920 [59:21<02:58, 1.64it/s, tok/s=16427.4]
Inference: 95%|█████████▌| 5628/5920 [59:21<02:23, 2.04it/s, tok/s=16427.4]
Inference: 95%|█████████▌| 5628/5920 [59:21<02:23, 2.04it/s, tok/s=16429.3]
Inference: 95%|█████████▌| 5629/5920 [59:21<02:22, 2.04it/s, tok/s=16430.9]
Inference: 95%|█████████▌| 5630/5920 [59:22<02:18, 2.09it/s, tok/s=16430.9]
Inference: 95%|█████████▌| 5630/5920 [59:22<02:18, 2.09it/s, tok/s=16432.1]
Inference: 95%|█████████▌| 5631/5920 [59:23<03:02, 1.58it/s, tok/s=16432.1]
Inference: 95%|█████████▌| 5631/5920 [59:23<03:02, 1.58it/s, tok/s=16429.2]
Inference: 95%|█████████▌| 5632/5920 [59:23<03:02, 1.58it/s, tok/s=16432.3]
Inference: 95%|█████████▌| 5633/5920 [59:23<03:01, 1.58it/s, tok/s=16434.8]
Inference: 95%|█████████▌| 5634/5920 [59:23<01:41, 2.83it/s, tok/s=16434.8]
Inference: 95%|█████████▌| 5634/5920 [59:23<01:41, 2.83it/s, tok/s=16435.3]
Inference: 95%|█████████▌| 5635/5920 [59:24<01:40, 2.83it/s, tok/s=16440.3]
Inference: 95%|█████████▌| 5636/5920 [59:24<01:37, 2.92it/s, tok/s=16440.3]
Inference: 95%|█████████▌| 5636/5920 [59:24<01:37, 2.92it/s, tok/s=16438.6]
Inference: 95%|█████████▌| 5637/5920 [59:25<02:09, 2.19it/s, tok/s=16438.6]
Inference: 95%|█████████▌| 5637/5920 [59:25<02:09, 2.19it/s, tok/s=16436.6]
Inference: 95%|█████████▌| 5638/5920 [59:25<02:08, 2.19it/s, tok/s=16439.3]
Inference: 95%|█████████▌| 5639/5920 [59:27<03:20, 1.40it/s, tok/s=16439.3]
Inference: 95%|█████████▌| 5639/5920 [59:27<03:20, 1.40it/s, tok/s=16431.6]
Inference: 95%|█████████▌| 5640/5920 [59:28<02:59, 1.56it/s, tok/s=16431.6]
Inference: 95%|█████████▌| 5640/5920 [59:28<02:59, 1.56it/s, tok/s=16432.8]
Inference: 95%|█████████▌| 5641/5920 [59:28<02:58, 1.56it/s, tok/s=16434.0]
Inference: 95%|█████████▌| 5642/5920 [59:28<02:25, 1.92it/s, tok/s=16434.0]
Inference: 95%|█████████▌| 5642/5920 [59:28<02:25, 1.92it/s, tok/s=16432.7]
Inference: 95%|█████████▌| 5643/5920 [59:29<02:20, 1.98it/s, tok/s=16432.7]
Inference: 95%|█████████▌| 5643/5920 [59:29<02:20, 1.98it/s, tok/s=16433.7]
Inference: 95%|█████████▌| 5644/5920 [59:32<04:34, 1.01it/s, tok/s=16433.7]
Inference: 95%|█████████▌| 5644/5920 [59:32<04:34, 1.01it/s, tok/s=16424.0]
Inference: 95%|█████████▌| 5645/5920 [59:32<04:33, 1.01it/s, tok/s=16425.8]
Inference: 95%|█████████▌| 5646/5920 [59:32<04:32, 1.01it/s, tok/s=16428.9]
Inference: 95%|█████████▌| 5647/5920 [59:32<02:22, 1.91it/s, tok/s=16428.9]
Inference: 95%|█████████▌| 5647/5920 [59:32<02:22, 1.91it/s, tok/s=16431.2]
Inference: 95%|█████████▌| 5648/5920 [59:34<04:03, 1.12it/s, tok/s=16431.2]
Inference: 95%|█████████▌| 5648/5920 [59:34<04:03, 1.12it/s, tok/s=16421.5]
Inference: 95%|█████████▌| 5649/5920 [59:35<03:40, 1.23it/s, tok/s=16421.5]
Inference: 95%|█████████▌| 5649/5920 [59:35<03:40, 1.23it/s, tok/s=16422.8]
Inference: 95%|█████████▌| 5650/5920 [59:35<02:57, 1.52it/s, tok/s=16422.8]
Inference: 95%|█████████▌| 5650/5920 [59:35<02:57, 1.52it/s, tok/s=16424.4]
Inference: 95%|█████████▌| 5651/5920 [59:35<02:50, 1.57it/s, tok/s=16424.4]
Inference: 95%|█████████▌| 5651/5920 [59:35<02:50, 1.57it/s, tok/s=16426.2]
Inference: 95%|█████████▌| 5652/5920 [59:35<02:50, 1.57it/s, tok/s=16428.6]
Inference: 95%|█████████▌| 5653/5920 [59:35<01:46, 2.50it/s, tok/s=16428.6]
Inference: 95%|█████████▌| 5653/5920 [59:35<01:46, 2.50it/s, tok/s=16430.3]
Inference: 96%|█████████▌| 5654/5920 [59:36<02:12, 2.01it/s, tok/s=16430.3]
Inference: 96%|█████████▌| 5654/5920 [59:36<02:12, 2.01it/s, tok/s=16428.2]
Inference: 96%|█████████▌| 5655/5920 [59:37<02:27, 1.79it/s, tok/s=16428.2]
Inference: 96%|█████████▌| 5655/5920 [59:37<02:27, 1.79it/s, tok/s=16428.9]
Inference: 96%|█████████▌| 5656/5920 [59:37<02:12, 2.00it/s, tok/s=16428.9]
Inference: 96%|█████████▌| 5656/5920 [59:37<02:12, 2.00it/s, tok/s=16430.3]
Inference: 96%|█████████▌| 5657/5920 [59:38<01:46, 2.46it/s, tok/s=16430.3]
Inference: 96%|█████████▌| 5657/5920 [59:38<01:46, 2.46it/s, tok/s=16433.0]
Inference: 96%|█████████▌| 5658/5920 [59:39<02:32, 1.72it/s, tok/s=16433.0]
Inference: 96%|█████████▌| 5658/5920 [59:39<02:32, 1.72it/s, tok/s=16430.4]
Inference: 96%|█████████▌| 5659/5920 [59:39<02:32, 1.72it/s, tok/s=16431.5]
Inference: 96%|█████████▌| 5660/5920 [59:39<01:45, 2.46it/s, tok/s=16431.5]
Inference: 96%|█████████▌| 5660/5920 [59:39<01:45, 2.46it/s, tok/s=16433.8]
Inference: 96%|█████████▌| 5661/5920 [59:41<02:57, 1.46it/s, tok/s=16433.8]
Inference: 96%|█████████▌| 5661/5920 [59:41<02:57, 1.46it/s, tok/s=16428.7]
Inference: 96%|█████████▌| 5662/5920 [59:41<02:56, 1.46it/s, tok/s=16430.8]
Inference: 96%|█████████▌| 5663/5920 [59:41<01:57, 2.19it/s, tok/s=16430.8]
Inference: 96%|█████████▌| 5663/5920 [59:41<01:57, 2.19it/s, tok/s=16433.0]
Inference: 96%|█████████▌| 5664/5920 [59:41<01:52, 2.28it/s, tok/s=16433.0]
Inference: 96%|█████████▌| 5664/5920 [59:41<01:52, 2.28it/s, tok/s=16434.9]
Inference: 96%|█████████▌| 5665/5920 [59:42<01:58, 2.14it/s, tok/s=16434.9]
Inference: 96%|█████████▌| 5665/5920 [59:42<01:58, 2.14it/s, tok/s=16436.5]
Inference: 96%|█████████▌| 5666/5920 [59:43<02:21, 1.80it/s, tok/s=16436.5]
Inference: 96%|█████████▌| 5666/5920 [59:43<02:21, 1.80it/s, tok/s=16435.1]
Inference: 96%|█████████▌| 5667/5920 [59:43<02:09, 1.95it/s, tok/s=16435.1]
Inference: 96%|█████████▌| 5667/5920 [59:43<02:09, 1.95it/s, tok/s=16438.1]
Inference: 96%|█████████▌| 5668/5920 [59:44<02:14, 1.87it/s, tok/s=16438.1]
Inference: 96%|█████████▌| 5668/5920 [59:44<02:14, 1.87it/s, tok/s=16437.3]
Inference: 96%|█████████▌| 5669/5920 [59:44<02:14, 1.87it/s, tok/s=16437.3]
Inference: 96%|█████████▌| 5669/5920 [59:44<02:14, 1.87it/s, tok/s=16436.9]
Inference: 96%|█████████▌| 5670/5920 [59:45<02:18, 1.80it/s, tok/s=16436.9]
Inference: 96%|█████████▌| 5670/5920 [59:45<02:18, 1.80it/s, tok/s=16436.0]
Inference: 96%|█████████▌| 5671/5920 [59:45<02:04, 2.00it/s, tok/s=16436.0]
Inference: 96%|█████████▌| 5671/5920 [59:45<02:04, 2.00it/s, tok/s=16438.1]
Inference: 96%|█████████▌| 5672/5920 [59:45<01:45, 2.34it/s, tok/s=16438.1]
Inference: 96%|█████████▌| 5672/5920 [59:45<01:45, 2.34it/s, tok/s=16440.8]
Inference: 96%|█████████▌| 5673/5920 [59:46<02:44, 1.50it/s, tok/s=16440.8]
Inference: 96%|█████████▌| 5673/5920 [59:46<02:44, 1.50it/s, tok/s=16437.9]
Inference: 96%|█████████▌| 5674/5920 [59:47<02:06, 1.95it/s, tok/s=16437.9]
Inference: 96%|█████████▌| 5674/5920 [59:47<02:06, 1.95it/s, tok/s=16439.5]
Inference: 96%|█████████▌| 5675/5920 [59:47<02:05, 1.95it/s, tok/s=16441.9]
Inference: 96%|█████████▌| 5676/5920 [59:47<01:15, 3.25it/s, tok/s=16441.9]
Inference: 96%|█████████▌| 5676/5920 [59:47<01:15, 3.25it/s, tok/s=16443.7]
Inference: 96%|█████████▌| 5677/5920 [59:48<01:45, 2.30it/s, tok/s=16443.7]
Inference: 96%|█████████▌| 5677/5920 [59:48<01:45, 2.30it/s, tok/s=16442.9]
Inference: 96%|█████████▌| 5678/5920 [59:48<02:10, 1.86it/s, tok/s=16442.9]
Inference: 96%|█████████▌| 5678/5920 [59:48<02:10, 1.86it/s, tok/s=16445.8]
Inference: 96%|█████████▌| 5679/5920 [59:49<01:45, 2.29it/s, tok/s=16445.8]
Inference: 96%|█████████▌| 5679/5920 [59:49<01:45, 2.29it/s, tok/s=16447.2]
Inference: 96%|█████████▌| 5680/5920 [59:49<01:50, 2.18it/s, tok/s=16447.2]
Inference: 96%|█████████▌| 5680/5920 [59:49<01:50, 2.18it/s, tok/s=16448.9]
Inference: 96%|█████████▌| 5681/5920 [59:49<01:26, 2.78it/s, tok/s=16448.9]
Inference: 96%|█████████▌| 5681/5920 [59:49<01:26, 2.78it/s, tok/s=16451.0]
Inference: 96%|█████████▌| 5682/5920 [59:50<02:17, 1.73it/s, tok/s=16451.0]
Inference: 96%|█████████▌| 5682/5920 [59:50<02:17, 1.73it/s, tok/s=16448.4]
Inference: 96%|█████████▌| 5683/5920 [59:51<02:00, 1.97it/s, tok/s=16448.4]
Inference: 96%|█████████▌| 5683/5920 [59:51<02:00, 1.97it/s, tok/s=16449.8]
Inference: 96%|█████████▌| 5684/5920 [59:52<02:49, 1.39it/s, tok/s=16449.8]
Inference: 96%|█████████▌| 5684/5920 [59:52<02:49, 1.39it/s, tok/s=16447.8]
Inference: 96%|█████████▌| 5685/5920 [59:52<02:34, 1.52it/s, tok/s=16447.8]
Inference: 96%|█████████▌| 5685/5920 [59:52<02:34, 1.52it/s, tok/s=16449.6]
Inference: 96%|█████████▌| 5686/5920 [59:53<02:22, 1.65it/s, tok/s=16449.6]
Inference: 96%|█████████▌| 5686/5920 [59:53<02:22, 1.65it/s, tok/s=16450.2]
Inference: 96%|█████████▌| 5687/5920 [59:53<01:53, 2.05it/s, tok/s=16450.2]
Inference: 96%|█████████▌| 5687/5920 [59:53<01:53, 2.05it/s, tok/s=16452.0]
Inference: 96%|█████████▌| 5688/5920 [59:55<03:42, 1.04it/s, tok/s=16452.0]
Inference: 96%|█████████▌| 5688/5920 [59:55<03:42, 1.04it/s, tok/s=16444.4]
Inference: 96%|█████████▌| 5689/5920 [59:56<03:42, 1.04it/s, tok/s=16444.4]
Inference: 96%|█████████▌| 5689/5920 [59:56<03:42, 1.04it/s, tok/s=16445.6]
Inference: 96%|█████████▌| 5690/5920 [59:56<02:46, 1.38it/s, tok/s=16445.6]
Inference: 96%|█████████▌| 5690/5920 [59:56<02:46, 1.38it/s, tok/s=16449.3]
Inference: 96%|█████████▌| 5691/5920 [59:57<02:39, 1.44it/s, tok/s=16449.3]
Inference: 96%|█████████▌| 5691/5920 [59:57<02:39, 1.44it/s, tok/s=16450.7]
Inference: 96%|█████████▌| 5692/5920 [59:57<02:01, 1.87it/s, tok/s=16450.7]
Inference: 96%|█████████▌| 5692/5920 [59:57<02:01, 1.87it/s, tok/s=16454.3]
Inference: 96%|█████████▌| 5693/5920 [59:57<01:44, 2.17it/s, tok/s=16454.3]
Inference: 96%|█████████▌| 5693/5920 [59:57<01:44, 2.17it/s, tok/s=16455.1]
Inference: 96%|█████████▌| 5694/5920 [59:58<01:59, 1.89it/s, tok/s=16455.1]
Inference: 96%|█████████▌| 5694/5920 [59:58<01:59, 1.89it/s, tok/s=16454.9]
Inference: 96%|█████████▌| 5695/5920 [59:58<01:45, 2.13it/s, tok/s=16454.9]
Inference: 96%|█████████▌| 5695/5920 [59:58<01:45, 2.13it/s, tok/s=16458.8]
Inference: 96%|█████████▌| 5696/5920 [59:59<01:55, 1.94it/s, tok/s=16458.8]
Inference: 96%|█████████▌| 5696/5920 [59:59<01:55, 1.94it/s, tok/s=16460.4]
Inference: 96%|█████████▌| 5697/5920 [1:00:01<03:31, 1.05it/s, tok/s=16460.4]
Inference: 96%|█████████▌| 5697/5920 [1:00:01<03:31, 1.05it/s, tok/s=16455.8]
Inference: 96%|█████████▋| 5698/5920 [1:00:01<02:46, 1.34it/s, tok/s=16455.8]
Inference: 96%|█████████▋| 5698/5920 [1:00:01<02:46, 1.34it/s, tok/s=16456.7]
Inference: 96%|█████████▋| 5699/5920 [1:00:02<02:31, 1.45it/s, tok/s=16456.7]
Inference: 96%|█████████▋| 5699/5920 [1:00:02<02:31, 1.45it/s, tok/s=16456.7]
Inference: 96%|█████████▋| 5700/5920 [1:00:03<03:13, 1.14it/s, tok/s=16456.7]
Inference: 96%|█████████▋| 5700/5920 [1:00:03<03:13, 1.14it/s, tok/s=16453.5]
Inference: 96%|█████████▋| 5701/5920 [1:00:04<03:10, 1.15it/s, tok/s=16453.5]
Inference: 96%|█████████▋| 5701/5920 [1:00:04<03:10, 1.15it/s, tok/s=16452.4]
Inference: 96%|█████████▋| 5702/5920 [1:00:04<02:19, 1.57it/s, tok/s=16452.4]
Inference: 96%|█████████▋| 5702/5920 [1:00:04<02:19, 1.57it/s, tok/s=16455.5]
Inference: 96%|█████████▋| 5703/5920 [1:00:04<02:18, 1.57it/s, tok/s=16458.8]
Inference: 96%|█████████▋| 5704/5920 [1:00:04<01:30, 2.38it/s, tok/s=16458.8]
Inference: 96%|█████████▋| 5704/5920 [1:00:04<01:30, 2.38it/s, tok/s=16462.9]
Inference: 96%|█████████▋| 5705/5920 [1:00:05<01:53, 1.89it/s, tok/s=16462.9]
Inference: 96%|█████████▋| 5705/5920 [1:00:05<01:53, 1.89it/s, tok/s=16461.5]
Inference: 96%|█████████▋| 5706/5920 [1:00:07<02:52, 1.24it/s, tok/s=16461.5]
Inference: 96%|█████████▋| 5706/5920 [1:00:07<02:52, 1.24it/s, tok/s=16457.6]
Inference: 96%|█████████▋| 5707/5920 [1:00:07<02:51, 1.24it/s, tok/s=16461.6]
Inference: 96%|█████████▋| 5708/5920 [1:00:07<02:50, 1.24it/s, tok/s=16466.0]
Inference: 96%|█████████▋| 5709/5920 [1:00:09<02:21, 1.49it/s, tok/s=16466.0]
Inference: 96%|█████████▋| 5709/5920 [1:00:09<02:21, 1.49it/s, tok/s=16462.4]
Inference: 96%|█████████▋| 5710/5920 [1:00:11<03:18, 1.06it/s, tok/s=16462.4]
Inference: 96%|█████████▋| 5710/5920 [1:00:11<03:18, 1.06it/s, tok/s=16457.3]
Inference: 96%|█████████▋| 5711/5920 [1:00:11<03:00, 1.16it/s, tok/s=16457.3]
Inference: 96%|█████████▋| 5711/5920 [1:00:11<03:00, 1.16it/s, tok/s=16458.8]
Inference: 96%|█████████▋| 5712/5920 [1:00:12<02:53, 1.20it/s, tok/s=16458.8]
Inference: 96%|█████████▋| 5712/5920 [1:00:12<02:53, 1.20it/s, tok/s=16457.6]
Inference: 97%|█████████▋| 5713/5920 [1:00:12<02:13, 1.55it/s, tok/s=16457.6]
Inference: 97%|█████████▋| 5713/5920 [1:00:12<02:13, 1.55it/s, tok/s=16465.4]
Inference: 97%|█████████▋| 5714/5920 [1:00:14<03:23, 1.01it/s, tok/s=16465.4]
Inference: 97%|█████████▋| 5714/5920 [1:00:14<03:23, 1.01it/s, tok/s=16459.3]
Inference: 97%|█████████▋| 5715/5920 [1:00:14<03:22, 1.01it/s, tok/s=16466.6]
Inference: 97%|█████████▋| 5716/5920 [1:00:15<02:44, 1.24it/s, tok/s=16466.6]
Inference: 97%|█████████▋| 5716/5920 [1:00:15<02:44, 1.24it/s, tok/s=16464.8]
Inference: 97%|█████████▋| 5717/5920 [1:00:19<05:35, 1.65s/it, tok/s=16464.8]
Inference: 97%|█████████▋| 5717/5920 [1:00:19<05:35, 1.65s/it, tok/s=16453.2]
Inference: 97%|█████████▋| 5718/5920 [1:00:20<05:01, 1.49s/it, tok/s=16453.2]
Inference: 97%|█████████▋| 5718/5920 [1:00:20<05:01, 1.49s/it, tok/s=16451.8]
Inference: 97%|█████████▋| 5719/5920 [1:00:21<03:58, 1.19s/it, tok/s=16451.8]
Inference: 97%|█████████▋| 5719/5920 [1:00:21<03:58, 1.19s/it, tok/s=16452.3]
Inference: 97%|█████████▋| 5720/5920 [1:00:21<03:57, 1.19s/it, tok/s=16455.0]
Inference: 97%|█████████▋| 5721/5920 [1:00:22<02:53, 1.15it/s, tok/s=16455.0]
Inference: 97%|█████████▋| 5721/5920 [1:00:22<02:53, 1.15it/s, tok/s=16454.9]
Inference: 97%|█████████▋| 5722/5920 [1:00:22<02:28, 1.34it/s, tok/s=16454.9]
Inference: 97%|█████████▋| 5722/5920 [1:00:22<02:28, 1.34it/s, tok/s=16456.0]
Inference: 97%|█████████▋| 5723/5920 [1:00:22<02:27, 1.34it/s, tok/s=16464.3]
Inference: 97%|█████████▋| 5724/5920 [1:00:23<01:55, 1.69it/s, tok/s=16464.3]
Inference: 97%|█████████▋| 5724/5920 [1:00:23<01:55, 1.69it/s, tok/s=16464.9]
Inference: 97%|█████████▋| 5725/5920 [1:00:23<01:47, 1.81it/s, tok/s=16464.9]
Inference: 97%|█████████▋| 5725/5920 [1:00:23<01:47, 1.81it/s, tok/s=16465.8]
Inference: 97%|█████████▋| 5726/5920 [1:00:25<02:35, 1.25it/s, tok/s=16465.8]
Inference: 97%|█████████▋| 5726/5920 [1:00:25<02:35, 1.25it/s, tok/s=16463.0]
Inference: 97%|█████████▋| 5727/5920 [1:00:26<02:53, 1.11it/s, tok/s=16463.0]
Inference: 97%|█████████▋| 5727/5920 [1:00:26<02:53, 1.11it/s, tok/s=16461.8]
Inference: 97%|█████████▋| 5728/5920 [1:00:26<02:20, 1.36it/s, tok/s=16461.8]
Inference: 97%|█████████▋| 5728/5920 [1:00:26<02:20, 1.36it/s, tok/s=16466.9]
Inference: 97%|█████████▋| 5729/5920 [1:00:28<03:08, 1.02it/s, tok/s=16466.9]
Inference: 97%|█████████▋| 5729/5920 [1:00:28<03:08, 1.02it/s, tok/s=16461.8]
Inference: 97%|█████████▋| 5730/5920 [1:00:30<04:00, 1.27s/it, tok/s=16461.8]
Inference: 97%|█████████▋| 5730/5920 [1:00:30<04:00, 1.27s/it, tok/s=16455.9]
Inference: 97%|█████████▋| 5731/5920 [1:00:33<05:24, 1.72s/it, tok/s=16455.9]
Inference: 97%|█████████▋| 5731/5920 [1:00:33<05:24, 1.72s/it, tok/s=16446.6]
Inference: 97%|█████████▋| 5732/5920 [1:00:33<04:16, 1.36s/it, tok/s=16446.6]
Inference: 97%|█████████▋| 5732/5920 [1:00:33<04:16, 1.36s/it, tok/s=16447.0]
Inference: 97%|█████████▋| 5733/5920 [1:00:34<03:57, 1.27s/it, tok/s=16447.0]
Inference: 97%|█████████▋| 5733/5920 [1:00:34<03:57, 1.27s/it, tok/s=16448.5]
Inference: 97%|█████████▋| 5734/5920 [1:00:34<02:52, 1.08it/s, tok/s=16448.5]
Inference: 97%|█████████▋| 5734/5920 [1:00:34<02:52, 1.08it/s, tok/s=16453.2]
Inference: 97%|█████████▋| 5735/5920 [1:00:34<02:51, 1.08it/s, tok/s=16454.9]
Inference: 97%|█████████▋| 5736/5920 [1:00:36<02:49, 1.08it/s, tok/s=16454.9]
Inference: 97%|█████████▋| 5736/5920 [1:00:36<02:49, 1.08it/s, tok/s=16450.5]
Inference: 97%|█████████▋| 5737/5920 [1:00:40<04:41, 1.54s/it, tok/s=16450.5]
Inference: 97%|█████████▋| 5737/5920 [1:00:40<04:41, 1.54s/it, tok/s=16438.7]
Inference: 97%|█████████▋| 5738/5920 [1:00:41<04:27, 1.47s/it, tok/s=16438.7]
Inference: 97%|█████████▋| 5738/5920 [1:00:41<04:27, 1.47s/it, tok/s=16439.2]
Inference: 97%|█████████▋| 5739/5920 [1:00:41<03:23, 1.13s/it, tok/s=16439.2]
Inference: 97%|█████████▋| 5739/5920 [1:00:41<03:23, 1.13s/it, tok/s=16441.1]
Inference: 97%|█████████▋| 5740/5920 [1:00:42<02:52, 1.04it/s, tok/s=16441.1]
Inference: 97%|█████████▋| 5740/5920 [1:00:42<02:52, 1.04it/s, tok/s=16443.5]
Inference: 97%|█████████▋| 5741/5920 [1:00:42<02:40, 1.11it/s, tok/s=16443.5]
Inference: 97%|█████████▋| 5741/5920 [1:00:42<02:40, 1.11it/s, tok/s=16443.4]
Inference: 97%|█████████▋| 5742/5920 [1:00:43<02:31, 1.18it/s, tok/s=16443.4]
Inference: 97%|█████████▋| 5742/5920 [1:00:43<02:31, 1.18it/s, tok/s=16443.9]
Inference: 97%|█████████▋| 5743/5920 [1:00:43<02:30, 1.18it/s, tok/s=16447.4]
Inference: 97%|█████████▋| 5744/5920 [1:00:43<01:26, 2.05it/s, tok/s=16447.4]
Inference: 97%|█████████▋| 5744/5920 [1:00:43<01:26, 2.05it/s, tok/s=16452.6]
Inference: 97%|█████████▋| 5745/5920 [1:00:44<01:25, 2.06it/s, tok/s=16452.6]
Inference: 97%|█████████▋| 5745/5920 [1:00:44<01:25, 2.06it/s, tok/s=16458.6]
Inference: 97%|█████████▋| 5746/5920 [1:00:45<02:13, 1.31it/s, tok/s=16458.6]
Inference: 97%|█████████▋| 5746/5920 [1:00:45<02:13, 1.31it/s, tok/s=16455.9]
Inference: 97%|█████████▋| 5747/5920 [1:00:50<05:33, 1.93s/it, tok/s=16455.9]
Inference: 97%|█████████▋| 5747/5920 [1:00:50<05:33, 1.93s/it, tok/s=16436.7]
Inference: 97%|█████████▋| 5748/5920 [1:00:50<05:31, 1.93s/it, tok/s=16438.7]
Inference: 97%|█████████▋| 5749/5920 [1:00:51<03:16, 1.15s/it, tok/s=16438.7]
Inference: 97%|█████████▋| 5749/5920 [1:00:51<03:16, 1.15s/it, tok/s=16443.9]
Inference: 97%|█████████▋| 5750/5920 [1:00:51<02:59, 1.06s/it, tok/s=16443.9]
Inference: 97%|█████████▋| 5750/5920 [1:00:51<02:59, 1.06s/it, tok/s=16442.2]
Inference: 97%|█████████▋| 5751/5920 [1:00:51<02:20, 1.21it/s, tok/s=16442.2]
Inference: 97%|█████████▋| 5751/5920 [1:00:51<02:20, 1.21it/s, tok/s=16444.3]
Inference: 97%|█████████▋| 5752/5920 [1:00:54<03:19, 1.19s/it, tok/s=16444.3]
Inference: 97%|█████████▋| 5752/5920 [1:00:54<03:19, 1.19s/it, tok/s=16437.5]
Inference: 97%|█████████▋| 5753/5920 [1:00:55<03:18, 1.19s/it, tok/s=16437.5]
Inference: 97%|█████████▋| 5753/5920 [1:00:55<03:18, 1.19s/it, tok/s=16436.0]
Inference: 97%|█████████▋| 5754/5920 [1:00:56<02:52, 1.04s/it, tok/s=16436.0]
Inference: 97%|█████████▋| 5754/5920 [1:00:56<02:52, 1.04s/it, tok/s=16434.9]
Inference: 97%|█████████▋| 5755/5920 [1:00:56<02:25, 1.14it/s, tok/s=16434.9]
Inference: 97%|█████████▋| 5755/5920 [1:00:56<02:25, 1.14it/s, tok/s=16435.9]
Inference: 97%|█████████▋| 5756/5920 [1:00:56<01:51, 1.47it/s, tok/s=16435.9]
Inference: 97%|█████████▋| 5756/5920 [1:00:56<01:51, 1.47it/s, tok/s=16439.0]
Inference: 97%|█████████▋| 5757/5920 [1:00:59<03:39, 1.35s/it, tok/s=16439.0]
Inference: 97%|█████████▋| 5757/5920 [1:00:59<03:39, 1.35s/it, tok/s=16430.7]
Inference: 97%|█████████▋| 5758/5920 [1:00:59<03:38, 1.35s/it, tok/s=16434.3]
Inference: 97%|█████████▋| 5759/5920 [1:01:00<02:13, 1.21it/s, tok/s=16434.3]
Inference: 97%|█████████▋| 5759/5920 [1:01:00<02:13, 1.21it/s, tok/s=16437.7]
Inference: 97%|█████████▋| 5760/5920 [1:01:00<02:02, 1.30it/s, tok/s=16437.7]
Inference: 97%|█████████▋| 5760/5920 [1:01:00<02:02, 1.30it/s, tok/s=16443.0]
Inference: 97%|█████████▋| 5761/5920 [1:01:00<02:01, 1.30it/s, tok/s=16446.0]
Inference: 97%|█████████▋| 5762/5920 [1:01:02<02:06, 1.25it/s, tok/s=16446.0]
Inference: 97%|█████████▋| 5762/5920 [1:01:02<02:06, 1.25it/s, tok/s=16442.7]
Inference: 97%|█████████▋| 5763/5920 [1:01:03<02:16, 1.15it/s, tok/s=16442.7]
Inference: 97%|█████████▋| 5763/5920 [1:01:03<02:16, 1.15it/s, tok/s=16440.9]
Inference: 97%|█████████▋| 5764/5920 [1:01:04<02:04, 1.25it/s, tok/s=16440.9]
Inference: 97%|█████████▋| 5764/5920 [1:01:04<02:04, 1.25it/s, tok/s=16440.1]
Inference: 97%|█████████▋| 5765/5920 [1:01:04<01:36, 1.60it/s, tok/s=16440.1]
Inference: 97%|█████████▋| 5765/5920 [1:01:04<01:36, 1.60it/s, tok/s=16445.9]
Inference: 97%|█████████▋| 5766/5920 [1:01:04<01:36, 1.60it/s, tok/s=16452.0]
Inference: 97%|█████████▋| 5767/5920 [1:01:05<01:27, 1.75it/s, tok/s=16452.0]
Inference: 97%|█████████▋| 5767/5920 [1:01:05<01:27, 1.75it/s, tok/s=16450.4]
Inference: 97%|█████████▋| 5768/5920 [1:01:05<01:15, 2.02it/s, tok/s=16450.4]
Inference: 97%|█████████▋| 5768/5920 [1:01:05<01:15, 2.02it/s, tok/s=16455.4]
Inference: 97%|█████████▋| 5769/5920 [1:01:06<01:27, 1.73it/s, tok/s=16455.4]
Inference: 97%|█████████▋| 5769/5920 [1:01:06<01:27, 1.73it/s, tok/s=16457.4]
Inference: 97%|█████████▋| 5770/5920 [1:01:07<01:35, 1.58it/s, tok/s=16457.4]
Inference: 97%|█████████▋| 5770/5920 [1:01:07<01:35, 1.58it/s, tok/s=16458.9]
Inference: 97%|█████████▋| 5771/5920 [1:01:07<01:25, 1.75it/s, tok/s=16458.9]
Inference: 97%|█████████▋| 5771/5920 [1:01:07<01:25, 1.75it/s, tok/s=16460.8]
Inference: 98%|█████████▊| 5772/5920 [1:01:08<01:42, 1.45it/s, tok/s=16460.8]
Inference: 98%|█████████▊| 5772/5920 [1:01:08<01:42, 1.45it/s, tok/s=16460.7]
Inference: 98%|█████████▊| 5773/5920 [1:01:09<01:53, 1.29it/s, tok/s=16460.7]
Inference: 98%|█████████▊| 5773/5920 [1:01:09<01:53, 1.29it/s, tok/s=16461.3]
Inference: 98%|█████████▊| 5774/5920 [1:01:09<01:41, 1.44it/s, tok/s=16461.3]
Inference: 98%|█████████▊| 5774/5920 [1:01:09<01:41, 1.44it/s, tok/s=16461.9]
Inference: 98%|█████████▊| 5775/5920 [1:01:10<01:29, 1.62it/s, tok/s=16461.9]
Inference: 98%|█████████▊| 5775/5920 [1:01:10<01:29, 1.62it/s, tok/s=16463.4]
Inference: 98%|█████████▊| 5776/5920 [1:01:10<01:28, 1.62it/s, tok/s=16468.0]
Inference: 98%|█████████▊| 5777/5920 [1:01:10<00:55, 2.59it/s, tok/s=16468.0]
Inference: 98%|█████████▊| 5777/5920 [1:01:10<00:55, 2.59it/s, tok/s=16474.4]
Inference: 98%|█████████▊| 5778/5920 [1:01:10<00:54, 2.59it/s, tok/s=16477.6]
Inference: 98%|█████████▊| 5779/5920 [1:01:13<01:43, 1.36it/s, tok/s=16477.6]
Inference: 98%|█████████▊| 5779/5920 [1:01:13<01:43, 1.36it/s, tok/s=16472.6]
Inference: 98%|█████████▊| 5780/5920 [1:01:14<02:16, 1.03it/s, tok/s=16472.6]
Inference: 98%|█████████▊| 5780/5920 [1:01:14<02:16, 1.03it/s, tok/s=16467.1]
Inference: 98%|█████████▊| 5781/5920 [1:01:17<03:03, 1.32s/it, tok/s=16467.1]
Inference: 98%|█████████▊| 5781/5920 [1:01:17<03:03, 1.32s/it, tok/s=16462.3]
Inference: 98%|█████████▊| 5782/5920 [1:01:17<02:20, 1.02s/it, tok/s=16462.3]
Inference: 98%|█████████▊| 5782/5920 [1:01:17<02:20, 1.02s/it, tok/s=16466.8]
Inference: 98%|█████████▊| 5783/5920 [1:01:18<02:16, 1.00it/s, tok/s=16466.8]
Inference: 98%|█████████▊| 5783/5920 [1:01:18<02:16, 1.00it/s, tok/s=16470.8]
Inference: 98%|█████████▊| 5784/5920 [1:01:19<02:27, 1.08s/it, tok/s=16470.8]
Inference: 98%|█████████▊| 5784/5920 [1:01:19<02:27, 1.08s/it, tok/s=16468.0]
Inference: 98%|█████████▊| 5785/5920 [1:01:19<01:54, 1.17it/s, tok/s=16468.0]
Inference: 98%|█████████▊| 5785/5920 [1:01:19<01:54, 1.17it/s, tok/s=16469.7]
Inference: 98%|█████████▊| 5786/5920 [1:01:20<01:30, 1.49it/s, tok/s=16469.7]
Inference: 98%|█████████▊| 5786/5920 [1:01:20<01:30, 1.49it/s, tok/s=16474.2]
Inference: 98%|█████████▊| 5787/5920 [1:01:21<01:57, 1.13it/s, tok/s=16474.2]
Inference: 98%|█████████▊| 5787/5920 [1:01:21<01:57, 1.13it/s, tok/s=16473.4]
Inference: 98%|█████████▊| 5788/5920 [1:01:21<01:27, 1.51it/s, tok/s=16473.4]
Inference: 98%|█████████▊| 5788/5920 [1:01:21<01:27, 1.51it/s, tok/s=16479.3]
Inference: 98%|█████████▊| 5789/5920 [1:01:25<03:25, 1.57s/it, tok/s=16479.3]
Inference: 98%|█████████▊| 5789/5920 [1:01:25<03:25, 1.57s/it, tok/s=16465.7]
Inference: 98%|█████████▊| 5790/5920 [1:01:25<02:31, 1.16s/it, tok/s=16465.7]
Inference: 98%|█████████▊| 5790/5920 [1:01:25<02:31, 1.16s/it, tok/s=16470.9]
Inference: 98%|█████████▊| 5791/5920 [1:01:26<02:06, 1.02it/s, tok/s=16470.9]
Inference: 98%|█████████▊| 5791/5920 [1:01:26<02:06, 1.02it/s, tok/s=16471.7]
Inference: 98%|█████████▊| 5792/5920 [1:01:26<01:35, 1.35it/s, tok/s=16471.7]
Inference: 98%|█████████▊| 5792/5920 [1:01:26<01:35, 1.35it/s, tok/s=16474.9]
Inference: 98%|█████████▊| 5793/5920 [1:01:27<01:53, 1.12it/s, tok/s=16474.9]
Inference: 98%|█████████▊| 5793/5920 [1:01:27<01:53, 1.12it/s, tok/s=16475.4]
Inference: 98%|█████████▊| 5794/5920 [1:01:28<01:36, 1.30it/s, tok/s=16475.4]
Inference: 98%|█████████▊| 5794/5920 [1:01:28<01:36, 1.30it/s, tok/s=16478.1]
Inference: 98%|█████████▊| 5795/5920 [1:01:28<01:18, 1.58it/s, tok/s=16478.1]
Inference: 98%|█████████▊| 5795/5920 [1:01:28<01:18, 1.58it/s, tok/s=16482.8]
Inference: 98%|█████████▊| 5796/5920 [1:01:29<01:18, 1.59it/s, tok/s=16482.8]
Inference: 98%|█████████▊| 5796/5920 [1:01:29<01:18, 1.59it/s, tok/s=16485.1]
Inference: 98%|█████████▊| 5797/5920 [1:01:29<01:25, 1.45it/s, tok/s=16485.1]
Inference: 98%|█████████▊| 5797/5920 [1:01:29<01:25, 1.45it/s, tok/s=16487.7]
Inference: 98%|█████████▊| 5798/5920 [1:01:30<01:38, 1.24it/s, tok/s=16487.7]
Inference: 98%|█████████▊| 5798/5920 [1:01:30<01:38, 1.24it/s, tok/s=16485.5]
Inference: 98%|█████████▊| 5799/5920 [1:01:35<03:49, 1.90s/it, tok/s=16485.5]
Inference: 98%|█████████▊| 5799/5920 [1:01:35<03:49, 1.90s/it, tok/s=16469.7]
Inference: 98%|█████████▊| 5800/5920 [1:01:35<02:58, 1.48s/it, tok/s=16469.7]
Inference: 98%|█████████▊| 5800/5920 [1:01:35<02:58, 1.48s/it, tok/s=16474.8]
Inference: 98%|█████████▊| 5801/5920 [1:01:36<02:18, 1.16s/it, tok/s=16474.8]
Inference: 98%|█████████▊| 5801/5920 [1:01:36<02:18, 1.16s/it, tok/s=16475.1]
Inference: 98%|█████████▊| 5802/5920 [1:01:37<02:22, 1.21s/it, tok/s=16475.1]
Inference: 98%|█████████▊| 5802/5920 [1:01:37<02:22, 1.21s/it, tok/s=16472.0]
Inference: 98%|█████████▊| 5803/5920 [1:01:38<02:10, 1.12s/it, tok/s=16472.0]
Inference: 98%|█████████▊| 5803/5920 [1:01:38<02:10, 1.12s/it, tok/s=16475.5]
Inference: 98%|█████████▊| 5804/5920 [1:01:41<02:58, 1.54s/it, tok/s=16475.5]
Inference: 98%|█████████▊| 5804/5920 [1:01:41<02:58, 1.54s/it, tok/s=16468.7]
Inference: 98%|█████████▊| 5805/5920 [1:01:42<02:52, 1.50s/it, tok/s=16468.7]
Inference: 98%|█████████▊| 5805/5920 [1:01:42<02:52, 1.50s/it, tok/s=16470.0]
Inference: 98%|█████████▊| 5806/5920 [1:01:44<03:00, 1.58s/it, tok/s=16470.0]
Inference: 98%|█████████▊| 5806/5920 [1:01:44<03:00, 1.58s/it, tok/s=16464.8]
Inference: 98%|█████████▊| 5807/5920 [1:01:46<03:06, 1.65s/it, tok/s=16464.8]
Inference: 98%|█████████▊| 5807/5920 [1:01:46<03:06, 1.65s/it, tok/s=16460.3]
Inference: 98%|█████████▊| 5808/5920 [1:01:46<03:04, 1.65s/it, tok/s=16463.3]
Inference: 98%|█████████▊| 5809/5920 [1:01:46<01:42, 1.08it/s, tok/s=16463.3]
Inference: 98%|█████████▊| 5809/5920 [1:01:46<01:42, 1.08it/s, tok/s=16469.2]
Inference: 98%|█████████▊| 5810/5920 [1:01:48<02:13, 1.22s/it, tok/s=16469.2]
Inference: 98%|█████████▊| 5810/5920 [1:01:48<02:13, 1.22s/it, tok/s=16463.5]
Inference: 98%|█████████▊| 5811/5920 [1:01:48<01:54, 1.05s/it, tok/s=16463.5]
Inference: 98%|█████████▊| 5811/5920 [1:01:48<01:54, 1.05s/it, tok/s=16465.0]
Inference: 98%|█████████▊| 5812/5920 [1:01:50<02:01, 1.12s/it, tok/s=16465.0]
Inference: 98%|█████████▊| 5812/5920 [1:01:50<02:01, 1.12s/it, tok/s=16465.3]
Inference: 98%|█████████▊| 5813/5920 [1:01:50<01:41, 1.06it/s, tok/s=16465.3]
Inference: 98%|█████████▊| 5813/5920 [1:01:50<01:41, 1.06it/s, tok/s=16468.6]
Inference: 98%|█████████▊| 5814/5920 [1:01:51<01:49, 1.03s/it, tok/s=16468.6]
Inference: 98%|█████████▊| 5814/5920 [1:01:51<01:49, 1.03s/it, tok/s=16466.9]
Inference: 98%|█████████▊| 5815/5920 [1:01:54<02:23, 1.37s/it, tok/s=16466.9]
Inference: 98%|█████████▊| 5815/5920 [1:01:54<02:23, 1.37s/it, tok/s=16460.2]
Inference: 98%|█████████▊| 5816/5920 [1:01:54<02:03, 1.19s/it, tok/s=16460.2]
Inference: 98%|█████████▊| 5816/5920 [1:01:54<02:03, 1.19s/it, tok/s=16461.6]
Inference: 98%|█████████▊| 5817/5920 [1:01:56<02:25, 1.41s/it, tok/s=16461.6]
Inference: 98%|█████████▊| 5817/5920 [1:01:56<02:25, 1.41s/it, tok/s=16459.0]
Inference: 98%|█████████▊| 5818/5920 [1:02:00<03:37, 2.13s/it, tok/s=16459.0]
Inference: 98%|█████████▊| 5818/5920 [1:02:00<03:37, 2.13s/it, tok/s=16445.2]
Inference: 98%|█████████▊| 5819/5920 [1:02:02<03:17, 1.95s/it, tok/s=16445.2]
Inference: 98%|█████████▊| 5819/5920 [1:02:02<03:17, 1.95s/it, tok/s=16441.3]
Inference: 98%|█████████▊| 5820/5920 [1:02:03<02:42, 1.62s/it, tok/s=16441.3]
Inference: 98%|█████████▊| 5820/5920 [1:02:03<02:42, 1.62s/it, tok/s=16443.8]
Inference: 98%|█████████▊| 5821/5920 [1:02:03<02:08, 1.30s/it, tok/s=16443.8]
Inference: 98%|█████████▊| 5821/5920 [1:02:03<02:08, 1.30s/it, tok/s=16446.1]
Inference: 98%|█████████▊| 5822/5920 [1:02:06<02:46, 1.69s/it, tok/s=16446.1]
Inference: 98%|█████████▊| 5822/5920 [1:02:06<02:46, 1.69s/it, tok/s=16438.6]
Inference: 98%|█████████▊| 5823/5920 [1:02:06<02:14, 1.39s/it, tok/s=16438.6]
Inference: 98%|█████████▊| 5823/5920 [1:02:06<02:14, 1.39s/it, tok/s=16440.0]
Inference: 98%|█████████▊| 5824/5920 [1:02:08<02:16, 1.42s/it, tok/s=16440.0]
Inference: 98%|█████████▊| 5824/5920 [1:02:08<02:16, 1.42s/it, tok/s=16439.1]
Inference: 98%|█████████▊| 5825/5920 [1:02:09<02:09, 1.36s/it, tok/s=16439.1]
Inference: 98%|█████████▊| 5825/5920 [1:02:09<02:09, 1.36s/it, tok/s=16441.8]
Inference: 98%|█████████▊| 5826/5920 [1:02:10<02:04, 1.32s/it, tok/s=16441.8]
Inference: 98%|█████████▊| 5826/5920 [1:02:10<02:04, 1.32s/it, tok/s=16441.8]
Inference: 98%|█████████▊| 5827/5920 [1:02:11<01:52, 1.21s/it, tok/s=16441.8]
Inference: 98%|█████████▊| 5827/5920 [1:02:11<01:52, 1.21s/it, tok/s=16442.8]
Inference: 98%|█████████▊| 5828/5920 [1:02:12<01:39, 1.09s/it, tok/s=16442.8]
Inference: 98%|█████████▊| 5828/5920 [1:02:12<01:39, 1.09s/it, tok/s=16445.4]
Inference: 98%|█████████▊| 5829/5920 [1:02:13<01:33, 1.02s/it, tok/s=16445.4]
Inference: 98%|█████████▊| 5829/5920 [1:02:13<01:33, 1.02s/it, tok/s=16445.4]
Inference: 98%|█████████▊| 5830/5920 [1:02:13<01:32, 1.02s/it, tok/s=16450.7]
Inference: 98%|█████████▊| 5831/5920 [1:02:14<01:16, 1.17it/s, tok/s=16450.7]
Inference: 98%|█████████▊| 5831/5920 [1:02:14<01:16, 1.17it/s, tok/s=16449.5]
Inference: 99%|█████████▊| 5832/5920 [1:02:15<01:17, 1.13it/s, tok/s=16449.5]
Inference: 99%|█████████▊| 5832/5920 [1:02:15<01:17, 1.13it/s, tok/s=16452.7]
Inference: 99%|█████████▊| 5833/5920 [1:02:15<01:00, 1.43it/s, tok/s=16452.7]
Inference: 99%|█████████▊| 5833/5920 [1:02:15<01:00, 1.43it/s, tok/s=16455.4]
Inference: 99%|█████████▊| 5834/5920 [1:02:19<02:05, 1.45s/it, tok/s=16455.4]
Inference: 99%|█████████▊| 5834/5920 [1:02:19<02:05, 1.45s/it, tok/s=16443.8]
Inference: 99%|█████████▊| 5835/5920 [1:02:20<02:03, 1.45s/it, tok/s=16443.8]
Inference: 99%|█████████▊| 5835/5920 [1:02:20<02:03, 1.45s/it, tok/s=16444.3]
Inference: 99%|█████████▊| 5836/5920 [1:02:20<02:01, 1.45s/it, tok/s=16447.8]
Inference: 99%|█████████▊| 5837/5920 [1:02:21<01:27, 1.06s/it, tok/s=16447.8]
Inference: 99%|█████████▊| 5837/5920 [1:02:21<01:27, 1.06s/it, tok/s=16450.6]
Inference: 99%|█████████▊| 5838/5920 [1:02:23<01:33, 1.14s/it, tok/s=16450.6]
Inference: 99%|█████████▊| 5838/5920 [1:02:23<01:33, 1.14s/it, tok/s=16449.0]
Inference: 99%|█████████▊| 5839/5920 [1:02:23<01:11, 1.13it/s, tok/s=16449.0]
Inference: 99%|█████████▊| 5839/5920 [1:02:23<01:11, 1.13it/s, tok/s=16456.5]
Inference: 99%|█████████▊| 5840/5920 [1:02:24<01:15, 1.06it/s, tok/s=16456.5]
Inference: 99%|█████████▊| 5840/5920 [1:02:24<01:15, 1.06it/s, tok/s=16456.8]
Inference: 99%|█████████▊| 5841/5920 [1:02:25<01:02, 1.26it/s, tok/s=16456.8]
Inference: 99%|█████████▊| 5841/5920 [1:02:25<01:02, 1.26it/s, tok/s=16458.4]
Inference: 99%|█████████▊| 5842/5920 [1:02:29<02:21, 1.82s/it, tok/s=16458.4]
Inference: 99%|█████████▊| 5842/5920 [1:02:29<02:21, 1.82s/it, tok/s=16443.8]
Inference: 99%|█████████▊| 5843/5920 [1:02:30<02:12, 1.73s/it, tok/s=16443.8]
Inference: 99%|█████████▊| 5843/5920 [1:02:30<02:12, 1.73s/it, tok/s=16444.3]
Inference: 99%|█████████▊| 5844/5920 [1:02:31<01:49, 1.44s/it, tok/s=16444.3]
Inference: 99%|█████████▊| 5844/5920 [1:02:31<01:49, 1.44s/it, tok/s=16445.2]
Inference: 99%|█████████▊| 5845/5920 [1:02:32<01:29, 1.19s/it, tok/s=16445.2]
Inference: 99%|█████████▊| 5845/5920 [1:02:32<01:29, 1.19s/it, tok/s=16448.5]
Inference: 99%|█████████▉| 5846/5920 [1:02:32<01:04, 1.14it/s, tok/s=16448.5]
Inference: 99%|█████████▉| 5846/5920 [1:02:32<01:04, 1.14it/s, tok/s=16451.6]
Inference: 99%|█████████▉| 5847/5920 [1:02:35<01:51, 1.53s/it, tok/s=16451.6]
Inference: 99%|█████████▉| 5847/5920 [1:02:35<01:51, 1.53s/it, tok/s=16442.5]
Inference: 99%|█████████▉| 5848/5920 [1:02:37<02:04, 1.73s/it, tok/s=16442.5]
Inference: 99%|█████████▉| 5848/5920 [1:02:37<02:04, 1.73s/it, tok/s=16437.6]
Inference: 99%|█████████▉| 5849/5920 [1:02:37<01:31, 1.29s/it, tok/s=16437.6]
Inference: 99%|█████████▉| 5849/5920 [1:02:37<01:31, 1.29s/it, tok/s=16444.1]
Inference: 99%|█████████▉| 5850/5920 [1:02:39<01:27, 1.26s/it, tok/s=16444.1]
Inference: 99%|█████████▉| 5850/5920 [1:02:39<01:27, 1.26s/it, tok/s=16444.7]
Inference: 99%|█████████▉| 5851/5920 [1:02:39<01:03, 1.09it/s, tok/s=16444.7]
Inference: 99%|█████████▉| 5851/5920 [1:02:39<01:03, 1.09it/s, tok/s=16449.3]
Inference: 99%|█████████▉| 5852/5920 [1:02:40<01:05, 1.03it/s, tok/s=16449.3]
Inference: 99%|█████████▉| 5852/5920 [1:02:40<01:05, 1.03it/s, tok/s=16451.3]
Inference: 99%|█████████▉| 5853/5920 [1:02:41<01:08, 1.02s/it, tok/s=16451.3]
Inference: 99%|█████████▉| 5853/5920 [1:02:41<01:08, 1.02s/it, tok/s=16450.2]
Inference: 99%|█████████▉| 5854/5920 [1:02:42<01:00, 1.10it/s, tok/s=16450.2]
Inference: 99%|█████████▉| 5854/5920 [1:02:42<01:00, 1.10it/s, tok/s=16451.3]
Inference: 99%|█████████▉| 5855/5920 [1:02:42<00:43, 1.49it/s, tok/s=16451.3]
Inference: 99%|█████████▉| 5855/5920 [1:02:42<00:43, 1.49it/s, tok/s=16454.5]
Inference: 99%|█████████▉| 5856/5920 [1:02:46<01:48, 1.69s/it, tok/s=16454.5]
Inference: 99%|█████████▉| 5856/5920 [1:02:46<01:48, 1.69s/it, tok/s=16440.1]
Inference: 99%|█████████▉| 5857/5920 [1:02:47<01:35, 1.52s/it, tok/s=16440.1]
Inference: 99%|█████████▉| 5857/5920 [1:02:47<01:35, 1.52s/it, tok/s=16440.9]
Inference: 99%|█████████▉| 5858/5920 [1:02:48<01:17, 1.25s/it, tok/s=16440.9]
Inference: 99%|█████████▉| 5858/5920 [1:02:48<01:17, 1.25s/it, tok/s=16442.3]
Inference: 99%|█████████▉| 5859/5920 [1:02:48<01:16, 1.25s/it, tok/s=16446.5]
Inference: 99%|█████████▉| 5860/5920 [1:02:48<00:42, 1.42it/s, tok/s=16446.5]
Inference: 99%|█████████▉| 5860/5920 [1:02:48<00:42, 1.42it/s, tok/s=16449.5]
Inference: 99%|█████████▉| 5861/5920 [1:02:51<01:25, 1.46s/it, tok/s=16449.5]
Inference: 99%|█████████▉| 5861/5920 [1:02:51<01:25, 1.46s/it, tok/s=16439.8]
Inference: 99%|█████████▉| 5862/5920 [1:02:52<01:05, 1.13s/it, tok/s=16439.8]
Inference: 99%|█████████▉| 5862/5920 [1:02:52<01:05, 1.13s/it, tok/s=16443.1]
Inference: 99%|█████████▉| 5863/5920 [1:02:53<01:13, 1.29s/it, tok/s=16443.1]
Inference: 99%|█████████▉| 5863/5920 [1:02:53<01:13, 1.29s/it, tok/s=16438.6]
Inference: 99%|█████████▉| 5864/5920 [1:02:57<01:47, 1.91s/it, tok/s=16438.6]
Inference: 99%|█████████▉| 5864/5920 [1:02:57<01:47, 1.91s/it, tok/s=16430.2]
Inference: 99%|█████████▉| 5865/5920 [1:02:58<01:28, 1.62s/it, tok/s=16430.2]
Inference: 99%|█████████▉| 5865/5920 [1:02:58<01:28, 1.62s/it, tok/s=16429.8]
Inference: 99%|█████████▉| 5866/5920 [1:02:58<01:12, 1.34s/it, tok/s=16429.8]
Inference: 99%|█████████▉| 5866/5920 [1:02:58<01:12, 1.34s/it, tok/s=16431.8]
Inference: 99%|█████████▉| 5867/5920 [1:02:59<00:55, 1.04s/it, tok/s=16431.8]
Inference: 99%|█████████▉| 5867/5920 [1:02:59<00:55, 1.04s/it, tok/s=16434.5]
Inference: 99%|█████████▉| 5868/5920 [1:02:59<00:43, 1.21it/s, tok/s=16434.5]
Inference: 99%|█████████▉| 5868/5920 [1:02:59<00:43, 1.21it/s, tok/s=16440.2]
Inference: 99%|█████████▉| 5869/5920 [1:03:00<00:51, 1.01s/it, tok/s=16440.2]
Inference: 99%|█████████▉| 5869/5920 [1:03:00<00:51, 1.01s/it, tok/s=16439.8]
Inference: 99%|█████████▉| 5870/5920 [1:03:07<02:12, 2.64s/it, tok/s=16439.8]
Inference: 99%|█████████▉| 5870/5920 [1:03:07<02:12, 2.64s/it, tok/s=16415.1]
Inference: 99%|█████████▉| 5871/5920 [1:03:08<01:43, 2.11s/it, tok/s=16415.1]
Inference: 99%|█████████▉| 5871/5920 [1:03:08<01:43, 2.11s/it, tok/s=16416.3]
Inference: 99%|█████████▉| 5872/5920 [1:03:08<01:18, 1.63s/it, tok/s=16416.3]
Inference: 99%|█████████▉| 5872/5920 [1:03:08<01:18, 1.63s/it, tok/s=16418.3]
Inference: 99%|█████████▉| 5873/5920 [1:03:10<01:10, 1.50s/it, tok/s=16418.3]
Inference: 99%|█████████▉| 5873/5920 [1:03:10<01:10, 1.50s/it, tok/s=16421.1]
Inference: 99%|█████████▉| 5874/5920 [1:03:13<01:34, 2.05s/it, tok/s=16421.1]
Inference: 99%|█████████▉| 5874/5920 [1:03:13<01:34, 2.05s/it, tok/s=16413.7]
Inference: 99%|█████████▉| 5875/5920 [1:03:13<01:06, 1.48s/it, tok/s=16413.7]
Inference: 99%|█████████▉| 5875/5920 [1:03:13<01:06, 1.48s/it, tok/s=16417.0]
Inference: 99%|█████████▉| 5876/5920 [1:03:13<00:48, 1.09s/it, tok/s=16417.0]
Inference: 99%|█████████▉| 5876/5920 [1:03:13<00:48, 1.09s/it, tok/s=16420.1]
Inference: 99%|█████████▉| 5877/5920 [1:03:14<00:45, 1.05s/it, tok/s=16420.1]
Inference: 99%|█████████▉| 5877/5920 [1:03:14<00:45, 1.05s/it, tok/s=16420.6]
Inference: 99%|█████████▉| 5878/5920 [1:03:16<00:56, 1.35s/it, tok/s=16420.6]
Inference: 99%|█████████▉| 5878/5920 [1:03:16<00:56, 1.35s/it, tok/s=16416.6]
Inference: 99%|█████████▉| 5879/5920 [1:03:17<00:53, 1.31s/it, tok/s=16416.6]
Inference: 99%|█████████▉| 5879/5920 [1:03:17<00:53, 1.31s/it, tok/s=16418.1]
Inference: 99%|█████████▉| 5880/5920 [1:03:18<00:47, 1.18s/it, tok/s=16418.1]
Inference: 99%|█████████▉| 5880/5920 [1:03:18<00:47, 1.18s/it, tok/s=16421.0]
Inference: 99%|█████████▉| 5881/5920 [1:03:22<01:13, 1.89s/it, tok/s=16421.0]
Inference: 99%|█████████▉| 5881/5920 [1:03:22<01:13, 1.89s/it, tok/s=16412.7]
Inference: 99%|█████████▉| 5882/5920 [1:03:23<01:02, 1.65s/it, tok/s=16412.7]
Inference: 99%|█████████▉| 5882/5920 [1:03:23<01:02, 1.65s/it, tok/s=16414.1]
Inference: 99%|█████████▉| 5883/5920 [1:03:26<01:14, 2.02s/it, tok/s=16414.1]
Inference: 99%|█████████▉| 5883/5920 [1:03:26<01:14, 2.02s/it, tok/s=16407.7]
Inference: 99%|█████████▉| 5884/5920 [1:03:27<00:59, 1.66s/it, tok/s=16407.7]
Inference: 99%|█████████▉| 5884/5920 [1:03:27<00:59, 1.66s/it, tok/s=16410.4]
Inference: 99%|█████████▉| 5885/5920 [1:03:27<00:43, 1.24s/it, tok/s=16410.4]
Inference: 99%|█████████▉| 5885/5920 [1:03:27<00:43, 1.24s/it, tok/s=16414.0]
Inference: 99%|█████████▉| 5886/5920 [1:03:33<01:34, 2.78s/it, tok/s=16414.0]
Inference: 99%|█████████▉| 5886/5920 [1:03:33<01:34, 2.78s/it, tok/s=16392.3]
Inference: 99%|█████████▉| 5887/5920 [1:03:36<01:28, 2.67s/it, tok/s=16392.3]
Inference: 99%|█████████▉| 5887/5920 [1:03:36<01:28, 2.67s/it, tok/s=16386.0]
Inference: 99%|█████████▉| 5888/5920 [1:03:36<01:01, 1.91s/it, tok/s=16386.0]
Inference: 99%|█████████▉| 5888/5920 [1:03:36<01:01, 1.91s/it, tok/s=16391.0]
Inference: 99%|█████████▉| 5889/5920 [1:03:43<01:44, 3.38s/it, tok/s=16391.0]
Inference: 99%|█████████▉| 5889/5920 [1:03:43<01:44, 3.38s/it, tok/s=16366.6]
Inference: 99%|█████████▉| 5890/5920 [1:03:43<01:16, 2.56s/it, tok/s=16366.6]
Inference: 99%|█████████▉| 5890/5920 [1:03:43<01:16, 2.56s/it, tok/s=16368.1]
Inference: 100%|█████████▉| 5891/5920 [1:03:44<00:58, 2.00s/it, tok/s=16368.1]
Inference: 100%|█████████▉| 5891/5920 [1:03:44<00:58, 2.00s/it, tok/s=16368.8]
Inference: 100%|█████████▉| 5892/5920 [1:03:49<01:18, 2.81s/it, tok/s=16368.8]
Inference: 100%|█████████▉| 5892/5920 [1:03:49<01:18, 2.81s/it, tok/s=16355.1]
Inference: 100%|█████████▉| 5893/5920 [1:03:54<01:37, 3.61s/it, tok/s=16355.1]
Inference: 100%|█████████▉| 5893/5920 [1:03:54<01:37, 3.61s/it, tok/s=16337.0]
Inference: 100%|█████████▉| 5894/5920 [1:03:57<01:28, 3.40s/it, tok/s=16337.0]
Inference: 100%|█████████▉| 5894/5920 [1:03:57<01:28, 3.40s/it, tok/s=16329.6]
Inference: 100%|█████████▉| 5895/5920 [1:03:59<01:16, 3.05s/it, tok/s=16329.6]
Inference: 100%|█████████▉| 5895/5920 [1:03:59<01:16, 3.05s/it, tok/s=16325.3]
Inference: 100%|█████████▉| 5896/5920 [1:04:02<01:10, 2.92s/it, tok/s=16325.3]
Inference: 100%|█████████▉| 5896/5920 [1:04:02<01:10, 2.92s/it, tok/s=16318.4]
Inference: 100%|█████████▉| 5897/5920 [1:04:03<00:54, 2.38s/it, tok/s=16318.4]
Inference: 100%|█████████▉| 5897/5920 [1:04:03<00:54, 2.38s/it, tok/s=16320.1]
Inference: 100%|█████████▉| 5898/5920 [1:04:07<01:02, 2.82s/it, tok/s=16320.1]
Inference: 100%|█████████▉| 5898/5920 [1:04:07<01:02, 2.82s/it, tok/s=16311.6]
Inference: 100%|█████████▉| 5899/5920 [1:04:07<00:44, 2.12s/it, tok/s=16311.6]
Inference: 100%|█████████▉| 5899/5920 [1:04:07<00:44, 2.12s/it, tok/s=16314.6]
Inference: 100%|█████████▉| 5900/5920 [1:04:08<00:35, 1.80s/it, tok/s=16314.6]
Inference: 100%|█████████▉| 5900/5920 [1:04:08<00:35, 1.80s/it, tok/s=16315.1]
Inference: 100%|█████████▉| 5901/5920 [1:04:12<00:44, 2.35s/it, tok/s=16315.1]
Inference: 100%|█████████▉| 5901/5920 [1:04:12<00:44, 2.35s/it, tok/s=16304.9]
Inference: 100%|█████████▉| 5902/5920 [1:04:13<00:33, 1.88s/it, tok/s=16304.9]
Inference: 100%|█████████▉| 5902/5920 [1:04:13<00:33, 1.88s/it, tok/s=16309.4]
Inference: 100%|█████████▉| 5903/5920 [1:04:18<00:46, 2.75s/it, tok/s=16309.4]
Inference: 100%|█████████▉| 5903/5920 [1:04:18<00:46, 2.75s/it, tok/s=16294.6]
Inference: 100%|█████████▉| 5904/5920 [1:04:18<00:33, 2.07s/it, tok/s=16294.6]
Inference: 100%|█████████▉| 5904/5920 [1:04:18<00:33, 2.07s/it, tok/s=16300.3]
Inference: 100%|█████████▉| 5905/5920 [1:04:20<00:30, 2.05s/it, tok/s=16300.3]
Inference: 100%|█████████▉| 5905/5920 [1:04:20<00:30, 2.05s/it, tok/s=16296.6]
Inference: 100%|█████████▉| 5906/5920 [1:04:22<00:28, 2.07s/it, tok/s=16296.6]
Inference: 100%|█████████▉| 5906/5920 [1:04:22<00:28, 2.07s/it, tok/s=16294.5]
Inference: 100%|█████████▉| 5907/5920 [1:04:28<00:41, 3.23s/it, tok/s=16294.5]
Inference: 100%|█████████▉| 5907/5920 [1:04:28<00:41, 3.23s/it, tok/s=16276.1]
Inference: 100%|█████████▉| 5908/5920 [1:04:32<00:39, 3.27s/it, tok/s=16276.1]
Inference: 100%|█████████▉| 5908/5920 [1:04:32<00:39, 3.27s/it, tok/s=16266.8]
Inference: 100%|█████████▉| 5909/5920 [1:04:38<00:45, 4.12s/it, tok/s=16266.8]
Inference: 100%|█████████▉| 5909/5920 [1:04:38<00:45, 4.12s/it, tok/s=16248.9]
Inference: 100%|█████████▉| 5910/5920 [1:04:38<00:30, 3.10s/it, tok/s=16248.9]
Inference: 100%|█████████▉| 5910/5920 [1:04:38<00:30, 3.10s/it, tok/s=16250.6]
Inference: 100%|█████████▉| 5911/5920 [1:04:43<00:32, 3.56s/it, tok/s=16250.6]
Inference: 100%|█████████▉| 5911/5920 [1:04:43<00:32, 3.56s/it, tok/s=16239.0]
Inference: 100%|█████████▉| 5912/5920 [1:04:46<00:26, 3.31s/it, tok/s=16239.0]
Inference: 100%|█████████▉| 5912/5920 [1:04:46<00:26, 3.31s/it, tok/s=16235.4]
Inference: 100%|█████████▉| 5913/5920 [1:04:54<00:33, 4.77s/it, tok/s=16235.4]
Inference: 100%|█████████▉| 5913/5920 [1:04:54<00:33, 4.77s/it, tok/s=16207.1]
Inference: 100%|█████████▉| 5914/5920 [1:05:15<00:57, 9.59s/it, tok/s=16207.1]
Inference: 100%|█████████▉| 5914/5920 [1:05:15<00:57, 9.59s/it, tok/s=16127.3]
Inference: 100%|█████████▉| 5915/5920 [1:05:35<01:03, 12.78s/it, tok/s=16127.3]
Inference: 100%|█████████▉| 5915/5920 [1:05:35<01:03, 12.78s/it, tok/s=16052.0]
Inference: 100%|█████████▉| 5916/5920 [1:05:46<00:48, 12.16s/it, tok/s=16052.0]
Inference: 100%|█████████▉| 5916/5920 [1:05:46<00:48, 12.16s/it, tok/s=16016.0]
Inference: 100%|█████████▉| 5916/5920 [1:05:46<00:02, 1.50it/s, tok/s=16016.0]
Timeout during comparison
Timeout during comparison
Timeout during comparison
Timeout during comparison
{"stage": "all", "step01": {"task_count": 6, "instance_count": 5920, "pass_k_by_task": {"gpqa_diamond": 4, "hmmt2025": 4, "aime2024": 32, "aime2025": 32, "math500": 4, "minerva": 4}, "task_sizes": {"gpqa_diamond": 792, "hmmt2025": 120, "aime2024": 960, "aime2025": 960, "math500": 2000, "minerva": 1088}, "output_path": "/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-20260403_091551/iter_0001023-hf/eval_20260409163705/step01_prepared.jsonl"}, "step02": {"backend": "offline", "input_path": "/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-20260403_091551/iter_0001023-hf/eval_20260409163705/step01_prepared.jsonl", "output_path": "/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-20260403_091551/iter_0001023-hf/eval_20260409163705/step02_inference.jsonl"}, "step03": {"input_path": "/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-20260403_091551/iter_0001023-hf/eval_20260409163705/step02_inference.jsonl", "score_output_path": "/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-20260403_091551/iter_0001023-hf/eval_20260409163705/step03_score.jsonl", "final_eval_output_path": "/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-20260403_091551/iter_0001023-hf/eval_20260409163705/step03_final_eval.jsonl", "metrics": {"gpqa_diamond": {"avg_k": 0.5812182741116751, "pass_k": 0.7614213197969543, "avg_total_tokens": 10732.243654822336, "avg_thinking_tokens": 0.0, "max_thinking_tokens": 0.0, "min_thinking_tokens": 0.0}, "hmmt2025": {"avg_k": 0.375, "pass_k": 0.5333333333333333, "avg_total_tokens": 18450.008333333335, "avg_thinking_tokens": 0.0, "max_thinking_tokens": 0.0, "min_thinking_tokens": 0.0}, "aime2024": {"avg_k": 0.7020833333333333, "pass_k": 0.9333333333333333, "avg_total_tokens": 13840.913541666667, "avg_thinking_tokens": 0.0, "max_thinking_tokens": 0.0, "min_thinking_tokens": 0.0}, "aime2025": {"avg_k": 0.6010416666666667, "pass_k": 0.8666666666666667, "avg_total_tokens": 15299.15, "avg_thinking_tokens": 0.0, "max_thinking_tokens": 0.0, "min_thinking_tokens": 0.0}, "math500": {"avg_k": 0.952, "pass_k": 0.98, "avg_total_tokens": 4403.267, "avg_thinking_tokens": 0.0, "max_thinking_tokens": 0.0, "min_thinking_tokens": 0.0}, "minerva": {"avg_k": 0.484375, "pass_k": 0.5661764705882353, "avg_total_tokens": 6378.409007352941, "avg_thinking_tokens": 0.0, "max_thinking_tokens": 0.0, "min_thinking_tokens": 0.0}, "overall": {"avg_k": 0.7074036511156186, "pass_k": 0.8158640226628895, "avg_total_tokens": 9194.001521298174, "avg_thinking_tokens": 0.0, "max_thinking_tokens": 0.0, "min_thinking_tokens": 0.0}}}}
Traceback (most recent call last):
File "/usr/lib/python3.12/multiprocessing/resource_tracker.py", line 239, in main
cache[rtype].remove(name)
KeyError: '/mp-e49dr1x4'
Reference in New Issue View Git Blame Copy Permalink