CCCCCyx/Qwen3-8B-onpolicy-profiling-muon-20260413_090005
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
8383ebb77f2a5242b2fc594ee26a2dd2df6d863e
Qwen3-8B-onpolicy-profiling…/eval_20260417051759/run.log
ModelHub XC 8383ebb77f 初始化项目，由ModelHub XC社区提供模型 
Model: CCCCCyx/Qwen3-8B-onpolicy-profiling-muon-20260413_090005
Source: Original Platform
2026-05-15 15:19:53 +08:00

127 lines
998 KiB
Plaintext
Raw Blame History

[2026-04-17 05:17:59 nanoeval] run.py:42 - Step 0/1 system_prompt: None
[2026-04-17 05:17:59 nanoeval] task.py:107 - prepare_eval_input: system_prompt=None
[2026-04-17 05:18:02 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-muon-20260413_090005/iter_0001023-hf/eval_20260417051759/step01_prepared.jsonl'}
[2026-04-17 05:18:03 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.
[2026-04-17 05:18:09 nanoeval] server_args.py:1837 - Attention backend not specified. Use fa3 backend by default.
[2026-04-17 05:18:09 nanoeval] engine.py:157 - server_args=ServerArgs(model_path='/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-muon-20260413_090005/iter_0001023-hf', tokenizer_path='/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-muon-20260413_090005/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=298022117, 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-muon-20260413_090005/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.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.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 [00:03<00:10, 3.48s/it]
Loading safetensors checkpoint shards: 25% Completed | 1/4 [00:03<00:10, 3.37s/it]
Loading safetensors checkpoint shards: 25% Completed | 1/4 [00:03<00:10, 3.61s/it]
Loading safetensors checkpoint shards: 25% Completed | 1/4 [00:03<00:10, 3.48s/it]
Loading safetensors checkpoint shards: 25% Completed | 1/4 [00:03<00:10, 3.47s/it]
Loading safetensors checkpoint shards: 25% Completed | 1/4 [00:03<00:10, 3.45s/it]
Loading safetensors checkpoint shards: 25% Completed | 1/4 [00:03<00:10, 3.49s/it]
Loading safetensors checkpoint shards: 25% Completed | 1/4 [00:03<00:10, 3.45s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:06<00:06, 3.24s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:06<00:06, 3.18s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:06<00:06, 3.17s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:06<00:06, 3.19s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:06<00:06, 3.14s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:06<00:06, 3.18s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:06<00:06, 3.19s/it]
Loading safetensors checkpoint shards: 50% Completed | 2/4 [00:06<00:06, 3.17s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:09<00:03, 3.10s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:09<00:03, 3.09s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:09<00:03, 3.10s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:09<00:03, 3.07s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:09<00:03, 3.10s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:09<00:03, 3.09s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:09<00:03, 3.09s/it]
Loading safetensors checkpoint shards: 75% Completed | 3/4 [00:09<00:03, 3.12s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.15s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.56s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.17s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.55s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.19s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.18s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.19s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.56s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.55s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.17s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.56s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.53s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.18s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.55s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.19s/it]
Loading safetensors checkpoint shards: 100% Completed | 4/4 [00:10<00:00, 2.56s/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, 4.98it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:07, 4.98it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:07, 4.98it/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, 4.98it/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): 11%|█ | 4/36 [00:00<00:02, 13.52it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:02, 13.52it/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=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.52it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.52it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 3%|▎ | 1/36 [00:00<00:07, 4.98it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:07, 4.98it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 3%|▎ | 1/36 [00:00<00:07, 4.99it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:07, 4.99it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.89it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.89it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:07, 4.98it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:07, 4.99it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.89it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 3%|▎ | 1/36 [00:00<00:06, 5.20it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:06, 5.20it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:07, 4.98it/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=256 avail_mem=13.09 GB): 3%|▎ | 1/36 [00:00<00:06, 5.38it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:06, 5.38it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:07, 4.99it/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=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.89it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.20it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 11%|█ | 4/36 [00:00<00:02, 13.48it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:02, 13.48it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.38it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 11%|█ | 4/36 [00:00<00:02, 13.54it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:02, 13.54it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.90it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.90it/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<00:06, 5.20it/s]
Capturing batches (bs=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.48it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.38it/s]
Capturing batches (bs=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.54it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.90it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 11%|█ | 4/36 [00:00<00:02, 13.82it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:02, 13.82it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 11%|█ | 4/36 [00:00<00:02, 14.10it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:02, 14.10it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.48it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.54it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.90it/s]
Capturing batches (bs=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.82it/s]
Capturing batches (bs=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 14.10it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.70it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.70it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 3%|▎ | 1/36 [00:00<00:06, 5.15it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:06, 5.15it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.17it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.17it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.80it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.80it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 3%|▎ | 1/36 [00:00<00:06, 5.31it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:06, 5.31it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.82it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 14.10it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.70it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.15it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.17it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.80it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.31it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:01, 17.01it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:01, 17.01it/s]
Capturing batches (bs=256 avail_mem=13.09 GB): 3%|▎ | 1/36 [00:00<00:06, 5.25it/s]
Capturing batches (bs=248 avail_mem=12.84 GB): 3%|▎ | 1/36 [00:00<00:06, 5.25it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:01, 17.27it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:01, 17.27it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.70it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.15it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.17it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.80it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.31it/s]
Capturing batches (bs=240 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.25it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 17.01it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 17.27it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.55it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.55it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 11%|█ | 4/36 [00:00<00:02, 13.75it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:02, 13.75it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.72it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.72it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.66it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.66it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 11%|█ | 4/36 [00:00<00:02, 13.96it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:02, 13.96it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 3%|▎ | 1/36 [00:00<00:06, 5.25it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 17.01it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 17.27it/s]
Capturing batches (bs=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.75it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.55it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.66it/s]
Capturing batches (bs=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.96it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.72it/s]
Capturing batches (bs=232 avail_mem=12.83 GB): 11%|█ | 4/36 [00:00<00:02, 13.78it/s]
Capturing batches (bs=224 avail_mem=12.82 GB): 11%|█ | 4/36 [00:00<00:02, 13.78it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.76it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.76it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 19.07it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 19.07it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.55it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.75it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.66it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.72it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.96it/s]
Capturing batches (bs=216 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.78it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.76it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 19.07it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.65it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.65it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.86it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.86it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.79it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.79it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 21.04it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 21.04it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:01, 17.09it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:01, 17.09it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.76it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 11%|█ | 4/36 [00:00<00:02, 13.78it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 19.07it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.65it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.86it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.79it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 21.04it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 17.09it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.98it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.98it/s]
Capturing batches (bs=208 avail_mem=12.81 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.98it/s]
Capturing batches (bs=200 avail_mem=12.80 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.98it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.33it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.33it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.65it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.86it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.79it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 17.09it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 21.04it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.98it/s]
Capturing batches (bs=192 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.98it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.33it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.21it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.21it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.79it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.79it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.31it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.31it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.93it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.93it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 20.74it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.74it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.98it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 19%|█▉ | 7/36 [00:00<00:01, 16.98it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.33it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.79it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.21it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.93it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.31it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.51it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.51it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.74it/s]
Capturing batches (bs=184 avail_mem=12.79 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.82it/s]
Capturing batches (bs=176 avail_mem=12.78 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.82it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.84it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.84it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.79it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.21it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.93it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.31it/s]
Capturing batches (bs=168 avail_mem=12.77 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.82it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.51it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 61%|██████ | 22/36 [00:01<00:00, 20.74it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.84it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.07it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.07it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.56it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.56it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.07it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.07it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.69it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.69it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 28%|██▊ | 10/36 [00:00<00:01, 18.82it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.51it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.75it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.75it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.84it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.07it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.07it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.56it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.69it/s]
Capturing batches (bs=160 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.95it/s]
Capturing batches (bs=152 avail_mem=12.76 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.95it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.72it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.72it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.75it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 21.13it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 21.13it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.07it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 20.07it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.56it/s]
Capturing batches (bs=144 avail_mem=12.75 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.95it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.69it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.72it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.75it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 21.13it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.49it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.49it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.56it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.56it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 20.26it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.26it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 36%|███▌ | 13/36 [00:00<00:01, 19.95it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 20.36it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.36it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.72it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.69it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.69it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 21.13it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.49it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.56it/s]
Capturing batches (bs=136 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.50it/s]
Capturing batches (bs=128 avail_mem=12.74 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.50it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.26it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.36it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 20.44it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.69it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.44it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 20.82it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.82it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.49it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.56it/s]
Capturing batches (bs=120 avail_mem=12.72 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.50it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 61%|██████ | 22/36 [00:01<00:00, 20.26it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 61%|██████ | 22/36 [00:01<00:00, 20.36it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.69it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.44it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.88it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.88it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.82it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.88it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.88it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 44%|████▍ | 16/36 [00:00<00:00, 20.50it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.29it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.29it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.35it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.35it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 61%|██████ | 22/36 [00:01<00:00, 20.44it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 61%|██████ | 22/36 [00:01<00:00, 20.82it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.88it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.88it/s]
Capturing batches (bs=112 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.84it/s]
Capturing batches (bs=104 avail_mem=12.71 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.84it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 19.13it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 19.13it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.29it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.35it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.43it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.43it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.75it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.75it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.88it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.88it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 86%|████████▌ | 31/36 [00:01<00:00, 19.13it/s]
Capturing batches (bs=96 avail_mem=12.70 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.84it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.29it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.35it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.43it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.75it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 20.60it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.60it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 20.54it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.54it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 86%|████████▌ | 31/36 [00:01<00:00, 19.13it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 53%|█████▎ | 19/36 [00:01<00:00, 20.84it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.22it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.22it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.28it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.28it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.43it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.75it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.60it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 94%|█████████▍| 34/36 [00:01<00:00, 20.35it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 20.35it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.54it/s]
Capturing batches (bs=88 avail_mem=12.69 GB): 61%|██████ | 22/36 [00:01<00:00, 20.51it/s]
Capturing batches (bs=80 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.51it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.22it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.28it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.35it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.35it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 20.35it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.69it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.69it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 61%|██████ | 22/36 [00:01<00:00, 20.60it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 61%|██████ | 22/36 [00:01<00:00, 20.54it/s]
Capturing batches (bs=72 avail_mem=12.68 GB): 61%|██████ | 22/36 [00:01<00:00, 20.51it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:01<00:00, 19.59it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.22it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.28it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.35it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.69it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.60it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.60it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.51it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.51it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 61%|██████ | 22/36 [00:01<00:00, 20.51it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.35it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.69it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.60it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.51it/s]
Capturing batches (bs=64 avail_mem=12.67 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.47it/s]
Capturing batches (bs=56 avail_mem=12.66 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.47it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.70it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.70it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.75it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.75it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.60it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.70it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.51it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.75it/s]
Capturing batches (bs=48 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.47it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.82it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.82it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 19.11it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 19.11it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.52it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.52it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.70it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.40it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.40it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.75it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.82it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 69%|██████▉ | 25/36 [00:01<00:00, 20.47it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 86%|████████▌ | 31/36 [00:01<00:00, 19.11it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.52it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 94%|█████████▍| 34/36 [00:01<00:00, 19.84it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 19.84it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 94%|█████████▍| 34/36 [00:01<00:00, 19.91it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 19.91it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.40it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.82it/s]
Capturing batches (bs=40 avail_mem=12.65 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.33it/s]
Capturing batches (bs=32 avail_mem=12.64 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.33it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 86%|████████▌ | 31/36 [00:01<00:00, 19.11it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 19.84it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 19.91it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.52it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 94%|█████████▍| 34/36 [00:01<00:00, 19.98it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 19.98it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.40it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 94%|█████████▍| 34/36 [00:01<00:00, 20.28it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 20.28it/s]
Capturing batches (bs=24 avail_mem=12.63 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.33it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:01<00:00, 19.17it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:01<00:00, 19.26it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 19.98it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 20.28it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 78%|███████▊ | 28/36 [00:01<00:00, 20.33it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:01<00:00, 19.38it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 19.00it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 19.00it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:01<00:00, 19.71it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.85it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.85it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 86%|████████▌ | 31/36 [00:01<00:00, 19.00it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.85it/s]
Capturing batches (bs=16 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.84it/s]
Capturing batches (bs=12 avail_mem=12.62 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.84it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 86%|████████▌ | 31/36 [00:01<00:00, 19.00it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.85it/s]
Capturing batches (bs=8 avail_mem=12.61 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.84it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 94%|█████████▍| 34/36 [00:01<00:00, 20.19it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 20.19it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 94%|█████████▍| 34/36 [00:01<00:00, 20.02it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 20.02it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 86%|████████▌ | 31/36 [00:01<00:00, 18.84it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 20.19it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 20.02it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:01<00:00, 19.49it/s]
Capturing batches (bs=4 avail_mem=12.60 GB): 94%|█████████▍| 34/36 [00:01<00:00, 20.00it/s]
Capturing batches (bs=2 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 20.00it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:01<00:00, 19.47it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 94%|█████████▍| 34/36 [00:01<00:00, 20.00it/s]
Capturing batches (bs=1 avail_mem=12.59 GB): 100%|██████████| 36/36 [00:01<00:00, 19.40it/s]
Inference: 0%| | 0/5920 [00:00<?, ?it/s][2026-04-17 05:18:43] 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.
[2026-04-17 05:18:43] 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.
[2026-04-17 05:18:43] 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.
[2026-04-17 05:18:43] 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 [00:08<13:48:02, 8.39s/it]
Inference: 0%| | 1/5920 [00:08<13:48:02, 8.39s/it, tok/s=125.1]
Inference: 0%| | 2/5920 [00:08<13:47:53, 8.39s/it, tok/s=250.0]
Inference: 0%| | 3/5920 [00:08<13:47:45, 8.39s/it, tok/s=374.6]
Inference: 0%| | 4/5920 [00:08<13:47:36, 8.39s/it, tok/s=495.7]
Inference: 0%| | 5/5920 [00:10<2:45:47, 1.68s/it, tok/s=495.7]
Inference: 0%| | 5/5920 [00:10<2:45:47, 1.68s/it, tok/s=527.7]
Inference: 0%| | 6/5920 [00:12<2:51:26, 1.74s/it, tok/s=527.7]
Inference: 0%| | 6/5920 [00:12<2:51:26, 1.74s/it, tok/s=575.3]
Inference: 0%| | 7/5920 [00:13<2:25:31, 1.48s/it, tok/s=575.3]
Inference: 0%| | 7/5920 [00:13<2:25:31, 1.48s/it, tok/s=672.9]
Inference: 0%| | 8/5920 [00:13<1:59:14, 1.21s/it, tok/s=672.9]
Inference: 0%| | 8/5920 [00:13<1:59:14, 1.21s/it, tok/s=777.3]
Inference: 0%| | 9/5920 [00:14<2:02:26, 1.24s/it, tok/s=777.3]
Inference: 0%| | 9/5920 [00:14<2:02:26, 1.24s/it, tok/s=825.0]
Inference: 0%| | 10/5920 [00:15<1:42:16, 1.04s/it, tok/s=825.0]
Inference: 0%| | 10/5920 [00:15<1:42:16, 1.04s/it, tok/s=914.1]
Inference: 0%| | 11/5920 [00:16<1:32:07, 1.07it/s, tok/s=914.1]
Inference: 0%| | 11/5920 [00:16<1:32:07, 1.07it/s, tok/s=985.8]
Inference: 0%| | 12/5920 [00:16<1:32:06, 1.07it/s, tok/s=1101.2]
Inference: 0%| | 13/5920 [00:16<1:06:36, 1.48it/s, tok/s=1101.2]
Inference: 0%| | 13/5920 [00:16<1:06:36, 1.48it/s, tok/s=1163.6]
Inference: 0%| | 14/5920 [00:17<56:33, 1.74it/s, tok/s=1163.6]
Inference: 0%| | 14/5920 [00:17<56:33, 1.74it/s, tok/s=1254.9]
Inference: 0%| | 15/5920 [00:19<1:40:37, 1.02s/it, tok/s=1254.9]
Inference: 0%| | 15/5920 [00:19<1:40:37, 1.02s/it, tok/s=1213.1]
Inference: 0%| | 16/5920 [00:19<1:18:40, 1.25it/s, tok/s=1213.1]
Inference: 0%| | 16/5920 [00:19<1:18:40, 1.25it/s, tok/s=1314.5]
Inference: 0%| | 17/5920 [00:19<1:00:19, 1.63it/s, tok/s=1314.5]
Inference: 0%| | 17/5920 [00:19<1:00:19, 1.63it/s, tok/s=1414.9]
Inference: 0%| | 18/5920 [00:19<48:46, 2.02it/s, tok/s=1414.9]
Inference: 0%| | 18/5920 [00:19<48:46, 2.02it/s, tok/s=1519.6]
Inference: 0%| | 19/5920 [00:20<46:56, 2.10it/s, tok/s=1519.6]
Inference: 0%| | 19/5920 [00:20<46:56, 2.10it/s, tok/s=1600.2]
Inference: 0%| | 20/5920 [00:20<40:48, 2.41it/s, tok/s=1600.2]
Inference: 0%| | 20/5920 [00:20<40:48, 2.41it/s, tok/s=1691.9]
Inference: 0%| | 21/5920 [00:20<35:06, 2.80it/s, tok/s=1691.9]
Inference: 0%| | 21/5920 [00:20<35:06, 2.80it/s, tok/s=1790.6]
Inference: 0%| | 22/5920 [00:21<38:31, 2.55it/s, tok/s=1790.6]
Inference: 0%| | 22/5920 [00:21<38:31, 2.55it/s, tok/s=1867.3]
Inference: 0%| | 23/5920 [00:21<31:37, 3.11it/s, tok/s=1867.3]
Inference: 0%| | 23/5920 [00:21<31:37, 3.11it/s, tok/s=1958.8]
Inference: 0%| | 24/5920 [00:22<46:35, 2.11it/s, tok/s=1958.8]
Inference: 0%| | 24/5920 [00:22<46:35, 2.11it/s, tok/s=1990.2]
Inference: 0%| | 25/5920 [00:22<40:22, 2.43it/s, tok/s=1990.2]
Inference: 0%| | 25/5920 [00:22<40:22, 2.43it/s, tok/s=2072.8]
Inference: 0%| | 26/5920 [00:25<1:52:15, 1.14s/it, tok/s=2072.8]
Inference: 0%| | 26/5920 [00:25<1:52:15, 1.14s/it, tok/s=1946.8]
Inference: 0%| | 27/5920 [00:27<2:16:27, 1.39s/it, tok/s=1946.8]
Inference: 0%| | 27/5920 [00:27<2:16:27, 1.39s/it, tok/s=1903.8]
Inference: 0%| | 28/5920 [00:27<1:45:39, 1.08s/it, tok/s=1903.8]
Inference: 0%| | 28/5920 [00:27<1:45:39, 1.08s/it, tok/s=1984.0]
Inference: 0%| | 29/5920 [00:27<1:45:37, 1.08s/it, tok/s=2085.2]
Inference: 1%| | 30/5920 [00:28<1:15:58, 1.29it/s, tok/s=2085.2]
Inference: 1%| | 30/5920 [00:28<1:15:58, 1.29it/s, tok/s=2129.9]
Inference: 1%| | 31/5920 [00:30<1:50:58, 1.13s/it, tok/s=2129.9]
Inference: 1%| | 31/5920 [00:30<1:50:58, 1.13s/it, tok/s=2072.0]
Inference: 1%| | 32/5920 [00:31<1:53:17, 1.15s/it, tok/s=2072.0]
Inference: 1%| | 32/5920 [00:31<1:53:17, 1.15s/it, tok/s=2088.6]
Inference: 1%| | 33/5920 [00:32<1:41:24, 1.03s/it, tok/s=2088.6]
Inference: 1%| | 33/5920 [00:32<1:41:24, 1.03s/it, tok/s=2138.2]
Inference: 1%| | 34/5920 [00:33<1:43:52, 1.06s/it, tok/s=2138.2]
Inference: 1%| | 34/5920 [00:33<1:43:52, 1.06s/it, tok/s=2161.3]
Inference: 1%| | 35/5920 [00:35<1:51:33, 1.14s/it, tok/s=2161.3]
Inference: 1%| | 35/5920 [00:35<1:51:33, 1.14s/it, tok/s=2172.7]
Inference: 1%| | 36/5920 [00:35<1:24:00, 1.17it/s, tok/s=2172.7]
Inference: 1%| | 36/5920 [00:35<1:24:00, 1.17it/s, tok/s=2255.5]
Inference: 1%| | 37/5920 [00:40<3:24:34, 2.09s/it, tok/s=2255.5]
Inference: 1%| | 37/5920 [00:40<3:24:34, 2.09s/it, tok/s=2060.4]
Inference: 1%| | 38/5920 [00:40<2:43:15, 1.67s/it, tok/s=2060.4]
Inference: 1%| | 38/5920 [00:40<2:43:15, 1.67s/it, tok/s=2116.8]
Inference: 1%| | 39/5920 [00:42<2:31:12, 1.54s/it, tok/s=2116.8]
Inference: 1%| | 39/5920 [00:42<2:31:12, 1.54s/it, tok/s=2150.7]
Inference: 1%| | 40/5920 [00:42<2:00:53, 1.23s/it, tok/s=2150.7]
Inference: 1%| | 40/5920 [00:42<2:00:53, 1.23s/it, tok/s=2215.4]
Inference: 1%| | 41/5920 [00:44<2:30:00, 1.53s/it, tok/s=2215.4]
Inference: 1%| | 41/5920 [00:44<2:30:00, 1.53s/it, tok/s=2193.2]
Inference: 1%| | 42/5920 [00:45<2:29:59, 1.53s/it, tok/s=2274.6]
Inference: 1%| | 43/5920 [00:46<1:57:59, 1.20s/it, tok/s=2274.6]
Inference: 1%| | 43/5920 [00:46<1:57:59, 1.20s/it, tok/s=2284.7]
Inference: 1%| | 44/5920 [00:46<1:57:58, 1.20s/it, tok/s=2366.1]
Inference: 1%| | 45/5920 [00:49<2:04:56, 1.28s/it, tok/s=2366.1]
Inference: 1%| | 45/5920 [00:49<2:04:56, 1.28s/it, tok/s=2321.7]
Inference: 1%| | 46/5920 [00:49<1:42:33, 1.05s/it, tok/s=2321.7]
Inference: 1%| | 46/5920 [00:49<1:42:33, 1.05s/it, tok/s=2396.5]
Inference: 1%| | 47/5920 [00:51<2:02:06, 1.25s/it, tok/s=2396.5]
Inference: 1%| | 47/5920 [00:51<2:02:06, 1.25s/it, tok/s=2393.0]
Inference: 1%| | 48/5920 [00:53<2:27:16, 1.50s/it, tok/s=2393.0]
Inference: 1%| | 48/5920 [00:53<2:27:16, 1.50s/it, tok/s=2373.7]
Inference: 1%| | 49/5920 [00:54<2:08:25, 1.31s/it, tok/s=2373.7]
Inference: 1%| | 49/5920 [00:54<2:08:25, 1.31s/it, tok/s=2421.5]
Inference: 1%| | 50/5920 [00:55<2:02:54, 1.26s/it, tok/s=2421.5]
Inference: 1%| | 50/5920 [00:55<2:02:54, 1.26s/it, tok/s=2460.6]
Inference: 1%| | 51/5920 [00:56<2:02:57, 1.26s/it, tok/s=2460.6]
Inference: 1%| | 51/5920 [00:56<2:02:57, 1.26s/it, tok/s=2487.3]
Inference: 1%| | 52/5920 [00:57<1:45:59, 1.08s/it, tok/s=2487.3]
Inference: 1%| | 52/5920 [00:57<1:45:59, 1.08s/it, tok/s=2539.2]
Inference: 1%| | 53/5920 [00:58<1:52:27, 1.15s/it, tok/s=2539.2]
Inference: 1%| | 53/5920 [00:58<1:52:27, 1.15s/it, tok/s=2564.2]
Inference: 1%| | 54/5920 [00:59<1:26:15, 1.13it/s, tok/s=2564.2]
Inference: 1%| | 54/5920 [00:59<1:26:15, 1.13it/s, tok/s=2639.6]
Inference: 1%| | 55/5920 [01:00<1:37:04, 1.01it/s, tok/s=2639.6]
Inference: 1%| | 55/5920 [01:00<1:37:04, 1.01it/s, tok/s=2662.3]
Inference: 1%| | 56/5920 [01:00<1:37:03, 1.01it/s, tok/s=2737.7]
Inference: 1%| | 57/5920 [01:00<59:21, 1.65it/s, tok/s=2737.7]
Inference: 1%| | 57/5920 [01:00<59:21, 1.65it/s, tok/s=2807.0]
Inference: 1%| | 58/5920 [01:00<50:29, 1.94it/s, tok/s=2807.0]
Inference: 1%| | 58/5920 [01:00<50:29, 1.94it/s, tok/s=2874.0]
Inference: 1%| | 59/5920 [01:01<43:56, 2.22it/s, tok/s=2874.0]
Inference: 1%| | 59/5920 [01:01<43:56, 2.22it/s, tok/s=2942.4]
Inference: 1%| | 60/5920 [01:02<1:05:17, 1.50it/s, tok/s=2942.4]
Inference: 1%| | 60/5920 [01:02<1:05:17, 1.50it/s, tok/s=2960.8]
Inference: 1%| | 61/5920 [01:02<1:05:16, 1.50it/s, tok/s=3036.9]
Inference: 1%| | 62/5920 [01:03<1:08:15, 1.43it/s, tok/s=3036.9]
Inference: 1%| | 62/5920 [01:03<1:08:15, 1.43it/s, tok/s=3050.4]
Inference: 1%| | 63/5920 [01:04<1:06:30, 1.47it/s, tok/s=3050.4]
Inference: 1%| | 63/5920 [01:04<1:06:30, 1.47it/s, tok/s=3106.0]
Inference: 1%| | 64/5920 [01:05<1:18:05, 1.25it/s, tok/s=3106.0]
Inference: 1%| | 64/5920 [01:05<1:18:05, 1.25it/s, tok/s=3129.5]
Inference: 1%| | 65/5920 [01:06<1:08:56, 1.42it/s, tok/s=3129.5]
Inference: 1%| | 65/5920 [01:06<1:08:56, 1.42it/s, tok/s=3184.5]
Inference: 1%| | 66/5920 [01:06<59:06, 1.65it/s, tok/s=3184.5]
Inference: 1%| | 66/5920 [01:06<59:06, 1.65it/s, tok/s=3246.7]
Inference: 1%| | 67/5920 [01:06<59:05, 1.65it/s, tok/s=3319.1]
Inference: 1%| | 68/5920 [01:07<56:06, 1.74it/s, tok/s=3319.1]
Inference: 1%| | 68/5920 [01:07<56:06, 1.74it/s, tok/s=3348.8]
Inference: 1%| | 69/5920 [01:07<45:09, 2.16it/s, tok/s=3348.8]
Inference: 1%| | 69/5920 [01:07<45:09, 2.16it/s, tok/s=3421.1]
Inference: 1%| | 70/5920 [01:07<37:36, 2.59it/s, tok/s=3421.1]
Inference: 1%| | 70/5920 [01:07<37:36, 2.59it/s, tok/s=3490.8]
Inference: 1%| | 71/5920 [01:07<31:41, 3.08it/s, tok/s=3490.8]
Inference: 1%| | 71/5920 [01:07<31:41, 3.08it/s, tok/s=3560.3]
Inference: 1%| | 72/5920 [01:08<32:09, 3.03it/s, tok/s=3560.3]
Inference: 1%| | 72/5920 [01:08<32:09, 3.03it/s, tok/s=3618.8]
Inference: 1%| | 73/5920 [01:12<2:18:30, 1.42s/it, tok/s=3618.8]
Inference: 1%| | 73/5920 [01:12<2:18:30, 1.42s/it, tok/s=3484.7]
Inference: 1%|▏ | 74/5920 [01:13<2:11:29, 1.35s/it, tok/s=3484.7]
Inference: 1%|▏ | 74/5920 [01:13<2:11:29, 1.35s/it, tok/s=3503.7]
Inference: 1%|▏ | 75/5920 [01:14<2:01:08, 1.24s/it, tok/s=3503.7]
Inference: 1%|▏ | 75/5920 [01:14<2:01:08, 1.24s/it, tok/s=3531.8]
Inference: 1%|▏ | 76/5920 [01:15<1:54:08, 1.17s/it, tok/s=3531.8]
Inference: 1%|▏ | 76/5920 [01:15<1:54:08, 1.17s/it, tok/s=3558.0]
Inference: 1%|▏ | 77/5920 [01:15<1:27:52, 1.11it/s, tok/s=3558.0]
Inference: 1%|▏ | 77/5920 [01:15<1:27:52, 1.11it/s, tok/s=3620.4]
Inference: 1%|▏ | 78/5920 [01:17<1:52:28, 1.16s/it, tok/s=3620.4]
Inference: 1%|▏ | 78/5920 [01:17<1:52:28, 1.16s/it, tok/s=3612.8]
Inference: 1%|▏ | 79/5920 [01:17<1:25:36, 1.14it/s, tok/s=3612.8]
Inference: 1%|▏ | 79/5920 [01:17<1:25:36, 1.14it/s, tok/s=3676.0]
Inference: 1%|▏ | 80/5920 [01:18<1:15:03, 1.30it/s, tok/s=3676.0]
Inference: 1%|▏ | 80/5920 [01:18<1:15:03, 1.30it/s, tok/s=3724.4]
Inference: 1%|▏ | 81/5920 [01:18<1:15:02, 1.30it/s, tok/s=3794.2]
Inference: 1%|▏ | 82/5920 [01:18<49:11, 1.98it/s, tok/s=3794.2]
Inference: 1%|▏ | 82/5920 [01:18<49:11, 1.98it/s, tok/s=3851.6]
Inference: 1%|▏ | 83/5920 [01:19<46:51, 2.08it/s, tok/s=3851.6]
Inference: 1%|▏ | 83/5920 [01:19<46:51, 2.08it/s, tok/s=3902.7]
Inference: 1%|▏ | 84/5920 [01:21<1:20:49, 1.20it/s, tok/s=3902.7]
Inference: 1%|▏ | 84/5920 [01:21<1:20:49, 1.20it/s, tok/s=3890.1]
Inference: 1%|▏ | 85/5920 [01:23<1:57:37, 1.21s/it, tok/s=3890.1]
Inference: 1%|▏ | 85/5920 [01:23<1:57:37, 1.21s/it, tok/s=3857.8]
Inference: 1%|▏ | 86/5920 [01:23<1:40:42, 1.04s/it, tok/s=3857.8]
Inference: 1%|▏ | 86/5920 [01:23<1:40:42, 1.04s/it, tok/s=3901.4]
Inference: 1%|▏ | 87/5920 [01:24<1:21:25, 1.19it/s, tok/s=3901.4]
Inference: 1%|▏ | 87/5920 [01:24<1:21:25, 1.19it/s, tok/s=3955.4]
Inference: 1%|▏ | 88/5920 [01:25<1:39:05, 1.02s/it, tok/s=3955.4]
Inference: 1%|▏ | 88/5920 [01:25<1:39:05, 1.02s/it, tok/s=3958.9]
Inference: 2%|▏ | 89/5920 [01:26<1:32:40, 1.05it/s, tok/s=3958.9]
Inference: 2%|▏ | 89/5920 [01:26<1:32:40, 1.05it/s, tok/s=3993.0]
Inference: 2%|▏ | 90/5920 [01:26<1:18:10, 1.24it/s, tok/s=3993.0]
Inference: 2%|▏ | 90/5920 [01:26<1:18:10, 1.24it/s, tok/s=4042.5]
Inference: 2%|▏ | 91/5920 [01:26<1:18:09, 1.24it/s, tok/s=4108.5]
Inference: 2%|▏ | 92/5920 [01:27<54:51, 1.77it/s, tok/s=4108.5]
Inference: 2%|▏ | 92/5920 [01:27<54:51, 1.77it/s, tok/s=4155.0]
Inference: 2%|▏ | 93/5920 [01:28<1:03:09, 1.54it/s, tok/s=4155.0]
Inference: 2%|▏ | 93/5920 [01:28<1:03:09, 1.54it/s, tok/s=4181.8]
Inference: 2%|▏ | 94/5920 [01:29<1:07:33, 1.44it/s, tok/s=4181.8]
Inference: 2%|▏ | 94/5920 [01:29<1:07:33, 1.44it/s, tok/s=4211.9]
Inference: 2%|▏ | 95/5920 [01:36<4:05:20, 2.53s/it, tok/s=4211.9]
Inference: 2%|▏ | 95/5920 [01:36<4:05:20, 2.53s/it, tok/s=3955.2]
Inference: 2%|▏ | 96/5920 [01:38<4:00:19, 2.48s/it, tok/s=3955.2]
Inference: 2%|▏ | 96/5920 [01:38<4:00:19, 2.48s/it, tok/s=3929.0]
Inference: 2%|▏ | 97/5920 [01:39<2:55:51, 1.81s/it, tok/s=3929.0]
Inference: 2%|▏ | 97/5920 [01:39<2:55:51, 1.81s/it, tok/s=3990.3]
Inference: 2%|▏ | 98/5920 [01:39<2:24:26, 1.49s/it, tok/s=3990.3]
Inference: 2%|▏ | 98/5920 [01:39<2:24:26, 1.49s/it, tok/s=4029.1]
Inference: 2%|▏ | 99/5920 [01:40<1:49:24, 1.13s/it, tok/s=4029.1]
Inference: 2%|▏ | 99/5920 [01:40<1:49:24, 1.13s/it, tok/s=4086.1]
Inference: 2%|▏ | 100/5920 [01:42<2:33:25, 1.58s/it, tok/s=4086.1]
Inference: 2%|▏ | 100/5920 [01:42<2:33:25, 1.58s/it, tok/s=4046.3]
Inference: 2%|▏ | 101/5920 [01:42<1:55:08, 1.19s/it, tok/s=4046.3]
Inference: 2%|▏ | 101/5920 [01:42<1:55:08, 1.19s/it, tok/s=4103.9]
Inference: 2%|▏ | 102/5920 [01:44<1:50:58, 1.14s/it, tok/s=4103.9]
Inference: 2%|▏ | 102/5920 [01:44<1:50:58, 1.14s/it, tok/s=4128.0]
Inference: 2%|▏ | 103/5920 [01:48<3:34:37, 2.21s/it, tok/s=4128.0]
Inference: 2%|▏ | 103/5920 [01:48<3:34:37, 2.21s/it, tok/s=4013.2]
Inference: 2%|▏ | 104/5920 [01:50<3:13:40, 2.00s/it, tok/s=4013.2]
Inference: 2%|▏ | 104/5920 [01:50<3:13:40, 2.00s/it, tok/s=4023.4]
Inference: 2%|▏ | 105/5920 [01:55<4:56:22, 3.06s/it, tok/s=4023.4]
Inference: 2%|▏ | 105/5920 [01:55<4:56:22, 3.06s/it, tok/s=3896.0]
Inference: 2%|▏ | 106/5920 [01:56<3:47:50, 2.35s/it, tok/s=3896.0]
Inference: 2%|▏ | 106/5920 [01:56<3:47:50, 2.35s/it, tok/s=3935.4]
Inference: 2%|▏ | 107/5920 [01:57<3:16:56, 2.03s/it, tok/s=3935.4]
Inference: 2%|▏ | 107/5920 [01:57<3:16:56, 2.03s/it, tok/s=3955.9]
Inference: 2%|▏ | 108/5920 [02:02<4:37:40, 2.87s/it, tok/s=3955.9]
Inference: 2%|▏ | 108/5920 [02:02<4:37:40, 2.87s/it, tok/s=3862.2]
Inference: 2%|▏ | 109/5920 [02:03<3:27:41, 2.14s/it, tok/s=3862.2]
Inference: 2%|▏ | 109/5920 [02:03<3:27:41, 2.14s/it, tok/s=3909.8]
Inference: 2%|▏ | 110/5920 [02:03<2:34:14, 1.59s/it, tok/s=3909.8]
Inference: 2%|▏ | 110/5920 [02:03<2:34:14, 1.59s/it, tok/s=3961.4]
Inference: 2%|▏ | 111/5920 [02:03<1:55:24, 1.19s/it, tok/s=3961.4]
Inference: 2%|▏ | 111/5920 [02:03<1:55:24, 1.19s/it, tok/s=4015.5]
Inference: 2%|▏ | 112/5920 [02:03<1:55:23, 1.19s/it, tok/s=4077.1]
Inference: 2%|▏ | 113/5920 [02:04<1:16:32, 1.26it/s, tok/s=4077.1]
Inference: 2%|▏ | 113/5920 [02:04<1:16:32, 1.26it/s, tok/s=4119.7]
Inference: 2%|▏ | 114/5920 [02:04<1:00:41, 1.59it/s, tok/s=4119.7]
Inference: 2%|▏ | 114/5920 [02:04<1:00:41, 1.59it/s, tok/s=4178.0]
Inference: 2%|▏ | 115/5920 [02:04<50:52, 1.90it/s, tok/s=4178.0]
Inference: 2%|▏ | 115/5920 [02:04<50:52, 1.90it/s, tok/s=4231.9]
Inference: 2%|▏ | 116/5920 [02:04<50:52, 1.90it/s, tok/s=4290.3]
Inference: 2%|▏ | 117/5920 [02:05<54:03, 1.79it/s, tok/s=4290.3]
Inference: 2%|▏ | 117/5920 [02:05<54:03, 1.79it/s, tok/s=4313.7]
Inference: 2%|▏ | 118/5920 [02:06<1:03:23, 1.53it/s, tok/s=4313.7]
Inference: 2%|▏ | 118/5920 [02:06<1:03:23, 1.53it/s, tok/s=4341.3]
Inference: 2%|▏ | 119/5920 [02:06<51:02, 1.89it/s, tok/s=4341.3]
Inference: 2%|▏ | 119/5920 [02:06<51:02, 1.89it/s, tok/s=4396.8]
Inference: 2%|▏ | 120/5920 [02:07<41:08, 2.35it/s, tok/s=4396.8]
Inference: 2%|▏ | 120/5920 [02:07<41:08, 2.35it/s, tok/s=4401.8]
Inference: 2%|▏ | 121/5920 [02:08<1:07:31, 1.43it/s, tok/s=4401.8]
Inference: 2%|▏ | 121/5920 [02:08<1:07:31, 1.43it/s, tok/s=4413.7]
Inference: 2%|▏ | 122/5920 [02:11<2:01:25, 1.26s/it, tok/s=4413.7]
Inference: 2%|▏ | 122/5920 [02:11<2:01:25, 1.26s/it, tok/s=4383.6]
Inference: 2%|▏ | 123/5920 [02:14<3:06:25, 1.93s/it, tok/s=4383.6]
Inference: 2%|▏ | 123/5920 [02:14<3:06:25, 1.93s/it, tok/s=4326.7]
Inference: 2%|▏ | 124/5920 [02:15<2:35:19, 1.61s/it, tok/s=4326.7]
Inference: 2%|▏ | 124/5920 [02:15<2:35:19, 1.61s/it, tok/s=4361.5]
Inference: 2%|▏ | 125/5920 [02:16<2:07:21, 1.32s/it, tok/s=4361.5]
Inference: 2%|▏ | 125/5920 [02:16<2:07:21, 1.32s/it, tok/s=4405.5]
Inference: 2%|▏ | 126/5920 [02:17<2:04:40, 1.29s/it, tok/s=4405.5]
Inference: 2%|▏ | 126/5920 [02:17<2:04:40, 1.29s/it, tok/s=4425.2]
Inference: 2%|▏ | 127/5920 [02:17<2:04:39, 1.29s/it, tok/s=4484.5]
Inference: 2%|▏ | 128/5920 [02:17<1:14:25, 1.30it/s, tok/s=4484.5]
Inference: 2%|▏ | 128/5920 [02:17<1:14:25, 1.30it/s, tok/s=4535.7]
Inference: 2%|▏ | 129/5920 [02:19<1:43:28, 1.07s/it, tok/s=4535.7]
Inference: 2%|▏ | 129/5920 [02:19<1:43:28, 1.07s/it, tok/s=4531.1]
Inference: 2%|▏ | 130/5920 [02:19<1:20:18, 1.20it/s, tok/s=4531.1]
Inference: 2%|▏ | 130/5920 [02:19<1:20:18, 1.20it/s, tok/s=4585.0]
Inference: 2%|▏ | 131/5920 [02:20<1:19:09, 1.22it/s, tok/s=4585.0]
Inference: 2%|▏ | 131/5920 [02:20<1:19:09, 1.22it/s, tok/s=4617.9]
Inference: 2%|▏ | 132/5920 [02:21<1:10:00, 1.38it/s, tok/s=4617.9]
Inference: 2%|▏ | 132/5920 [02:21<1:10:00, 1.38it/s, tok/s=4660.9]
Inference: 2%|▏ | 133/5920 [02:21<59:38, 1.62it/s, tok/s=4660.9]
Inference: 2%|▏ | 133/5920 [02:21<59:38, 1.62it/s, tok/s=4709.8]
Inference: 2%|▏ | 134/5920 [02:22<1:08:14, 1.41it/s, tok/s=4709.8]
Inference: 2%|▏ | 134/5920 [02:22<1:08:14, 1.41it/s, tok/s=4739.3]
Inference: 2%|▏ | 135/5920 [02:22<52:58, 1.82it/s, tok/s=4739.3]
Inference: 2%|▏ | 135/5920 [02:22<52:58, 1.82it/s, tok/s=4746.4]
Inference: 2%|▏ | 136/5920 [02:22<52:57, 1.82it/s, tok/s=4803.1]
Inference: 2%|▏ | 137/5920 [02:23<58:22, 1.65it/s, tok/s=4803.1]
Inference: 2%|▏ | 137/5920 [02:23<58:22, 1.65it/s, tok/s=4819.4]
Inference: 2%|▏ | 138/5920 [02:25<1:31:07, 1.06it/s, tok/s=4819.4]
Inference: 2%|▏ | 138/5920 [02:25<1:31:07, 1.06it/s, tok/s=4812.9]
Inference: 2%|▏ | 139/5920 [02:26<1:20:11, 1.20it/s, tok/s=4812.9]
Inference: 2%|▏ | 139/5920 [02:26<1:20:11, 1.20it/s, tok/s=4808.4]
Inference: 2%|▏ | 140/5920 [02:28<1:48:31, 1.13s/it, tok/s=4808.4]
Inference: 2%|▏ | 140/5920 [02:28<1:48:31, 1.13s/it, tok/s=4803.4]
Inference: 2%|▏ | 141/5920 [02:28<1:48:30, 1.13s/it, tok/s=4859.7]
Inference: 2%|▏ | 142/5920 [02:29<1:16:57, 1.25it/s, tok/s=4859.7]
Inference: 2%|▏ | 142/5920 [02:29<1:16:57, 1.25it/s, tok/s=4895.7]
Inference: 2%|▏ | 143/5920 [02:29<1:07:04, 1.44it/s, tok/s=4895.7]
Inference: 2%|▏ | 143/5920 [02:29<1:07:04, 1.44it/s, tok/s=4942.2]
Inference: 2%|▏ | 144/5920 [02:29<59:27, 1.62it/s, tok/s=4942.2]
Inference: 2%|▏ | 144/5920 [02:29<59:27, 1.62it/s, tok/s=4987.4]
Inference: 2%|▏ | 145/5920 [02:30<50:52, 1.89it/s, tok/s=4987.4]
Inference: 2%|▏ | 145/5920 [02:30<50:52, 1.89it/s, tok/s=5037.2]
Inference: 2%|▏ | 146/5920 [02:33<1:59:25, 1.24s/it, tok/s=5037.2]
Inference: 2%|▏ | 146/5920 [02:33<1:59:25, 1.24s/it, tok/s=4992.6]
Inference: 2%|▏ | 147/5920 [02:34<1:56:26, 1.21s/it, tok/s=4992.6]
Inference: 2%|▏ | 147/5920 [02:34<1:56:26, 1.21s/it, tok/s=4970.5]
Inference: 2%|▎ | 148/5920 [02:34<1:35:41, 1.01it/s, tok/s=4970.5]
Inference: 2%|▎ | 148/5920 [02:34<1:35:41, 1.01it/s, tok/s=5013.2]
Inference: 3%|▎ | 149/5920 [02:37<2:18:56, 1.44s/it, tok/s=5013.2]
Inference: 3%|▎ | 149/5920 [02:37<2:18:56, 1.44s/it, tok/s=4989.7]
Inference: 3%|▎ | 150/5920 [02:38<2:12:12, 1.37s/it, tok/s=4989.7]
Inference: 3%|▎ | 150/5920 [02:38<2:12:12, 1.37s/it, tok/s=5008.7]
Inference: 3%|▎ | 151/5920 [02:38<2:12:11, 1.37s/it, tok/s=5067.5]
Inference: 3%|▎ | 152/5920 [02:39<1:22:15, 1.17it/s, tok/s=5067.5]
Inference: 3%|▎ | 152/5920 [02:39<1:22:15, 1.17it/s, tok/s=5111.0]
Inference: 3%|▎ | 153/5920 [02:39<1:12:38, 1.32it/s, tok/s=5111.0]
Inference: 3%|▎ | 153/5920 [02:39<1:12:38, 1.32it/s, tok/s=5152.9]
Inference: 3%|▎ | 154/5920 [02:41<1:33:25, 1.03it/s, tok/s=5152.9]
Inference: 3%|▎ | 154/5920 [02:41<1:33:25, 1.03it/s, tok/s=5159.0]
Inference: 3%|▎ | 155/5920 [02:42<1:33:09, 1.03it/s, tok/s=5159.0]
Inference: 3%|▎ | 155/5920 [02:42<1:33:09, 1.03it/s, tok/s=5187.1]
Inference: 3%|▎ | 156/5920 [02:44<2:03:54, 1.29s/it, tok/s=5187.1]
Inference: 3%|▎ | 156/5920 [02:44<2:03:54, 1.29s/it, tok/s=5175.9]
Inference: 3%|▎ | 157/5920 [02:44<1:40:48, 1.05s/it, tok/s=5175.9]
Inference: 3%|▎ | 157/5920 [02:44<1:40:48, 1.05s/it, tok/s=5218.1]
Inference: 3%|▎ | 158/5920 [02:45<1:42:43, 1.07s/it, tok/s=5218.1]
Inference: 3%|▎ | 158/5920 [02:45<1:42:43, 1.07s/it, tok/s=5239.5]
Inference: 3%|▎ | 159/5920 [02:46<1:44:58, 1.09s/it, tok/s=5239.5]
Inference: 3%|▎ | 159/5920 [02:46<1:44:58, 1.09s/it, tok/s=5257.5]
Inference: 3%|▎ | 160/5920 [02:47<1:17:24, 1.24it/s, tok/s=5257.5]
Inference: 3%|▎ | 160/5920 [02:47<1:17:24, 1.24it/s, tok/s=5308.2]
Inference: 3%|▎ | 161/5920 [02:47<1:04:35, 1.49it/s, tok/s=5308.2]
Inference: 3%|▎ | 161/5920 [02:47<1:04:35, 1.49it/s, tok/s=5351.8]
Inference: 3%|▎ | 162/5920 [02:48<1:03:23, 1.51it/s, tok/s=5351.8]
Inference: 3%|▎ | 162/5920 [02:48<1:03:23, 1.51it/s, tok/s=5389.4]
Inference: 3%|▎ | 163/5920 [02:51<2:22:13, 1.48s/it, tok/s=5389.4]
Inference: 3%|▎ | 163/5920 [02:51<2:22:13, 1.48s/it, tok/s=5337.8]
Inference: 3%|▎ | 164/5920 [02:51<1:43:03, 1.07s/it, tok/s=5337.8]
Inference: 3%|▎ | 164/5920 [02:51<1:43:03, 1.07s/it, tok/s=5388.7]
Inference: 3%|▎ | 165/5920 [02:52<1:37:45, 1.02s/it, tok/s=5388.7]
Inference: 3%|▎ | 165/5920 [02:52<1:37:45, 1.02s/it, tok/s=5416.4]
Inference: 3%|▎ | 166/5920 [02:52<1:13:10, 1.31it/s, tok/s=5416.4]
Inference: 3%|▎ | 166/5920 [02:52<1:13:10, 1.31it/s, tok/s=5467.3]
Inference: 3%|▎ | 167/5920 [02:54<1:45:18, 1.10s/it, tok/s=5467.3]
Inference: 3%|▎ | 167/5920 [02:54<1:45:18, 1.10s/it, tok/s=5464.8]
Inference: 3%|▎ | 168/5920 [02:54<1:45:16, 1.10s/it, tok/s=5520.8]
Inference: 3%|▎ | 169/5920 [02:56<1:44:20, 1.09s/it, tok/s=5520.8]
Inference: 3%|▎ | 169/5920 [02:56<1:44:20, 1.09s/it, tok/s=5510.3]
Inference: 3%|▎ | 170/5920 [02:57<1:26:54, 1.10it/s, tok/s=5510.3]
Inference: 3%|▎ | 170/5920 [02:57<1:26:54, 1.10it/s, tok/s=5554.5]
Inference: 3%|▎ | 171/5920 [02:57<1:26:53, 1.10it/s, tok/s=5606.4]
Inference: 3%|▎ | 172/5920 [02:57<56:47, 1.69it/s, tok/s=5606.4]
Inference: 3%|▎ | 172/5920 [02:57<56:47, 1.69it/s, tok/s=5653.7]
Inference: 3%|▎ | 173/5920 [02:57<56:46, 1.69it/s, tok/s=5705.8]
Inference: 3%|▎ | 174/5920 [02:57<45:32, 2.10it/s, tok/s=5705.8]
Inference: 3%|▎ | 174/5920 [02:57<45:32, 2.10it/s, tok/s=5743.8]
Inference: 3%|▎ | 175/5920 [02:58<56:41, 1.69it/s, tok/s=5743.8]
Inference: 3%|▎ | 175/5920 [02:58<56:41, 1.69it/s, tok/s=5764.1]
Inference: 3%|▎ | 176/5920 [03:00<1:24:14, 1.14it/s, tok/s=5764.1]
Inference: 3%|▎ | 176/5920 [03:00<1:24:14, 1.14it/s, tok/s=5761.9]
Inference: 3%|▎ | 177/5920 [03:02<1:54:09, 1.19s/it, tok/s=5761.9]
Inference: 3%|▎ | 177/5920 [03:02<1:54:09, 1.19s/it, tok/s=5748.8]
Inference: 3%|▎ | 178/5920 [03:02<1:54:08, 1.19s/it, tok/s=5768.3]
Inference: 3%|▎ | 179/5920 [03:03<1:25:15, 1.12it/s, tok/s=5768.3]
Inference: 3%|▎ | 179/5920 [03:03<1:25:15, 1.12it/s, tok/s=5795.2]
Inference: 3%|▎ | 180/5920 [03:06<1:57:06, 1.22s/it, tok/s=5795.2]
Inference: 3%|▎ | 180/5920 [03:06<1:57:06, 1.22s/it, tok/s=5776.0]
Inference: 3%|▎ | 181/5920 [03:07<1:53:54, 1.19s/it, tok/s=5776.0]
Inference: 3%|▎ | 181/5920 [03:07<1:53:54, 1.19s/it, tok/s=5797.4]
Inference: 3%|▎ | 182/5920 [03:07<1:41:05, 1.06s/it, tok/s=5797.4]
Inference: 3%|▎ | 182/5920 [03:07<1:41:05, 1.06s/it, tok/s=5835.0]
Inference: 3%|▎ | 183/5920 [03:09<2:05:11, 1.31s/it, tok/s=5835.0]
Inference: 3%|▎ | 183/5920 [03:09<2:05:11, 1.31s/it, tok/s=5826.3]
Inference: 3%|▎ | 184/5920 [03:11<2:13:46, 1.40s/it, tok/s=5826.3]
Inference: 3%|▎ | 184/5920 [03:11<2:13:46, 1.40s/it, tok/s=5831.7]
Inference: 3%|▎ | 185/5920 [03:12<2:10:47, 1.37s/it, tok/s=5831.7]
Inference: 3%|▎ | 185/5920 [03:12<2:10:47, 1.37s/it, tok/s=5845.9]
Inference: 3%|▎ | 186/5920 [03:14<2:22:06, 1.49s/it, tok/s=5845.9]
Inference: 3%|▎ | 186/5920 [03:14<2:22:06, 1.49s/it, tok/s=5818.0]
Inference: 3%|▎ | 187/5920 [03:14<1:49:43, 1.15s/it, tok/s=5818.0]
Inference: 3%|▎ | 187/5920 [03:14<1:49:43, 1.15s/it, tok/s=5862.2]
Inference: 3%|▎ | 188/5920 [03:15<1:40:29, 1.05s/it, tok/s=5862.2]
Inference: 3%|▎ | 188/5920 [03:15<1:40:29, 1.05s/it, tok/s=5862.1]
Inference: 3%|▎ | 189/5920 [03:15<1:14:48, 1.28it/s, tok/s=5862.1]
Inference: 3%|▎ | 189/5920 [03:15<1:14:48, 1.28it/s, tok/s=5910.3]
Inference: 3%|▎ | 190/5920 [03:17<1:29:42, 1.06it/s, tok/s=5910.3]
Inference: 3%|▎ | 190/5920 [03:17<1:29:42, 1.06it/s, tok/s=5924.3]
Inference: 3%|▎ | 191/5920 [03:17<1:09:49, 1.37it/s, tok/s=5924.3]
Inference: 3%|▎ | 191/5920 [03:17<1:09:49, 1.37it/s, tok/s=5969.6]
Inference: 3%|▎ | 192/5920 [03:18<1:05:26, 1.46it/s, tok/s=5969.6]
Inference: 3%|▎ | 192/5920 [03:18<1:05:26, 1.46it/s, tok/s=6005.2]
Inference: 3%|▎ | 193/5920 [03:18<50:44, 1.88it/s, tok/s=6005.2]
Inference: 3%|▎ | 193/5920 [03:18<50:44, 1.88it/s, tok/s=6052.4]
Inference: 3%|▎ | 194/5920 [03:18<50:43, 1.88it/s, tok/s=6103.9]
Inference: 3%|▎ | 195/5920 [03:18<50:43, 1.88it/s, tok/s=6157.4]
Inference: 3%|▎ | 196/5920 [03:19<42:32, 2.24it/s, tok/s=6157.4]
Inference: 3%|▎ | 196/5920 [03:19<42:32, 2.24it/s, tok/s=6176.6]
Inference: 3%|▎ | 197/5920 [03:20<47:34, 2.01it/s, tok/s=6176.6]
Inference: 3%|▎ | 197/5920 [03:20<47:34, 2.01it/s, tok/s=6208.8]
Inference: 3%|▎ | 198/5920 [03:22<1:24:18, 1.13it/s, tok/s=6208.8]
Inference: 3%|▎ | 198/5920 [03:22<1:24:18, 1.13it/s, tok/s=6194.2]
Inference: 3%|▎ | 199/5920 [03:23<1:23:57, 1.14it/s, tok/s=6194.2]
Inference: 3%|▎ | 199/5920 [03:23<1:23:57, 1.14it/s, tok/s=6220.1]
Inference: 3%|▎ | 200/5920 [03:23<1:09:21, 1.37it/s, tok/s=6220.1]
Inference: 3%|▎ | 200/5920 [03:23<1:09:21, 1.37it/s, tok/s=6264.4]
Inference: 3%|▎ | 201/5920 [03:25<1:33:01, 1.02it/s, tok/s=6264.4]
Inference: 3%|▎ | 201/5920 [03:25<1:33:01, 1.02it/s, tok/s=6267.9]
Inference: 3%|▎ | 202/5920 [03:26<1:51:52, 1.17s/it, tok/s=6267.9]
Inference: 3%|▎ | 202/5920 [03:26<1:51:52, 1.17s/it, tok/s=6269.0]
Inference: 3%|▎ | 203/5920 [03:27<1:27:21, 1.09it/s, tok/s=6269.0]
Inference: 3%|▎ | 203/5920 [03:27<1:27:21, 1.09it/s, tok/s=6311.2]
Inference: 3%|▎ | 204/5920 [03:27<1:24:23, 1.13it/s, tok/s=6311.2]
Inference: 3%|▎ | 204/5920 [03:27<1:24:23, 1.13it/s, tok/s=6340.5]
Inference: 3%|▎ | 205/5920 [03:28<1:23:53, 1.14it/s, tok/s=6340.5]
Inference: 3%|▎ | 205/5920 [03:28<1:23:53, 1.14it/s, tok/s=6367.4]
Inference: 3%|▎ | 206/5920 [03:29<1:21:35, 1.17it/s, tok/s=6367.4]
Inference: 3%|▎ | 206/5920 [03:29<1:21:35, 1.17it/s, tok/s=6395.3]
Inference: 3%|▎ | 207/5920 [03:30<1:13:41, 1.29it/s, tok/s=6395.3]
Inference: 3%|▎ | 207/5920 [03:30<1:13:41, 1.29it/s, tok/s=6430.1]
Inference: 4%|▎ | 208/5920 [03:30<1:13:41, 1.29it/s, tok/s=6484.5]
Inference: 4%|▎ | 209/5920 [03:30<59:35, 1.60it/s, tok/s=6484.5]
Inference: 4%|▎ | 209/5920 [03:30<59:35, 1.60it/s, tok/s=6510.1]
Inference: 4%|▎ | 210/5920 [03:34<1:57:05, 1.23s/it, tok/s=6510.1]
Inference: 4%|▎ | 210/5920 [03:34<1:57:05, 1.23s/it, tok/s=6469.7]
Inference: 4%|▎ | 211/5920 [03:34<1:38:16, 1.03s/it, tok/s=6469.7]
Inference: 4%|▎ | 211/5920 [03:34<1:38:16, 1.03s/it, tok/s=6507.8]
Inference: 4%|▎ | 212/5920 [03:34<1:38:15, 1.03s/it, tok/s=6557.7]
Inference: 4%|▎ | 213/5920 [03:35<1:07:34, 1.41it/s, tok/s=6557.7]
Inference: 4%|▎ | 213/5920 [03:35<1:07:34, 1.41it/s, tok/s=6595.6]
Inference: 4%|▎ | 214/5920 [03:35<1:03:46, 1.49it/s, tok/s=6595.6]
Inference: 4%|▎ | 214/5920 [03:35<1:03:46, 1.49it/s, tok/s=6631.4]
Inference: 4%|▎ | 215/5920 [03:36<1:15:35, 1.26it/s, tok/s=6631.4]
Inference: 4%|▎ | 215/5920 [03:36<1:15:35, 1.26it/s, tok/s=6646.6]
Inference: 4%|▎ | 216/5920 [03:37<1:12:40, 1.31it/s, tok/s=6646.6]
Inference: 4%|▎ | 216/5920 [03:37<1:12:40, 1.31it/s, tok/s=6678.4]
Inference: 4%|▎ | 217/5920 [03:38<1:15:27, 1.26it/s, tok/s=6678.4]
Inference: 4%|▎ | 217/5920 [03:38<1:15:27, 1.26it/s, tok/s=6703.2]
Inference: 4%|▎ | 218/5920 [03:38<1:07:31, 1.41it/s, tok/s=6703.2]
Inference: 4%|▎ | 218/5920 [03:38<1:07:31, 1.41it/s, tok/s=6739.2]
Inference: 4%|▎ | 219/5920 [03:39<57:52, 1.64it/s, tok/s=6739.2]
Inference: 4%|▎ | 219/5920 [03:39<57:52, 1.64it/s, tok/s=6779.4]
Inference: 4%|▎ | 220/5920 [03:39<54:46, 1.73it/s, tok/s=6779.4]
Inference: 4%|▎ | 220/5920 [03:39<54:46, 1.73it/s, tok/s=6814.5]
Inference: 4%|▎ | 221/5920 [03:40<1:04:53, 1.46it/s, tok/s=6814.5]
Inference: 4%|▎ | 221/5920 [03:40<1:04:53, 1.46it/s, tok/s=6838.5]
Inference: 4%|▍ | 222/5920 [03:41<1:02:15, 1.53it/s, tok/s=6838.5]
Inference: 4%|▍ | 222/5920 [03:41<1:02:15, 1.53it/s, tok/s=6856.2]
Inference: 4%|▍ | 223/5920 [03:41<57:35, 1.65it/s, tok/s=6856.2]
Inference: 4%|▍ | 223/5920 [03:41<57:35, 1.65it/s, tok/s=6895.6]
Inference: 4%|▍ | 224/5920 [03:42<54:09, 1.75it/s, tok/s=6895.6]
Inference: 4%|▍ | 224/5920 [03:42<54:09, 1.75it/s, tok/s=6910.5]
Inference: 4%|▍ | 225/5920 [03:42<53:15, 1.78it/s, tok/s=6910.5]
Inference: 4%|▍ | 225/5920 [03:42<53:15, 1.78it/s, tok/s=6946.2]
Inference: 4%|▍ | 226/5920 [03:43<1:13:26, 1.29it/s, tok/s=6946.2]
Inference: 4%|▍ | 226/5920 [03:43<1:13:26, 1.29it/s, tok/s=6958.2]
Inference: 4%|▍ | 227/5920 [03:44<55:57, 1.70it/s, tok/s=6958.2]
Inference: 4%|▍ | 227/5920 [03:44<55:57, 1.70it/s, tok/s=7004.3]
Inference: 4%|▍ | 228/5920 [03:44<58:53, 1.61it/s, tok/s=7004.3]
Inference: 4%|▍ | 228/5920 [03:44<58:53, 1.61it/s, tok/s=7034.8]
Inference: 4%|▍ | 229/5920 [03:45<1:06:39, 1.42it/s, tok/s=7034.8]
Inference: 4%|▍ | 229/5920 [03:45<1:06:39, 1.42it/s, tok/s=7057.7]
Inference: 4%|▍ | 230/5920 [03:45<1:06:38, 1.42it/s, tok/s=7106.0]
Inference: 4%|▍ | 231/5920 [03:46<1:00:39, 1.56it/s, tok/s=7106.0]
Inference: 4%|▍ | 231/5920 [03:46<1:00:39, 1.56it/s, tok/s=7122.1]
Inference: 4%|▍ | 232/5920 [03:47<1:04:12, 1.48it/s, tok/s=7122.1]
Inference: 4%|▍ | 232/5920 [03:47<1:04:12, 1.48it/s, tok/s=7149.1]
Inference: 4%|▍ | 233/5920 [03:48<57:15, 1.66it/s, tok/s=7149.1]
Inference: 4%|▍ | 233/5920 [03:48<57:15, 1.66it/s, tok/s=7186.7]
Inference: 4%|▍ | 234/5920 [03:48<45:38, 2.08it/s, tok/s=7186.7]
Inference: 4%|▍ | 234/5920 [03:48<45:38, 2.08it/s, tok/s=7230.7]
Inference: 4%|▍ | 235/5920 [03:48<52:38, 1.80it/s, tok/s=7230.7]
Inference: 4%|▍ | 235/5920 [03:48<52:38, 1.80it/s, tok/s=7258.8]
Inference: 4%|▍ | 236/5920 [03:49<48:38, 1.95it/s, tok/s=7258.8]
Inference: 4%|▍ | 236/5920 [03:49<48:38, 1.95it/s, tok/s=7297.2]
Inference: 4%|▍ | 237/5920 [03:49<45:48, 2.07it/s, tok/s=7297.2]
Inference: 4%|▍ | 237/5920 [03:49<45:48, 2.07it/s, tok/s=7310.0]
Inference: 4%|▍ | 238/5920 [03:51<1:11:01, 1.33it/s, tok/s=7310.0]
Inference: 4%|▍ | 238/5920 [03:51<1:11:01, 1.33it/s, tok/s=7315.9]
Inference: 4%|▍ | 239/5920 [03:52<1:17:27, 1.22it/s, tok/s=7315.9]
Inference: 4%|▍ | 239/5920 [03:52<1:17:27, 1.22it/s, tok/s=7336.7]
Inference: 4%|▍ | 240/5920 [03:53<1:42:40, 1.08s/it, tok/s=7336.7]
Inference: 4%|▍ | 240/5920 [03:53<1:42:40, 1.08s/it, tok/s=7300.4]
Inference: 4%|▍ | 241/5920 [03:54<1:20:36, 1.17it/s, tok/s=7300.4]
Inference: 4%|▍ | 241/5920 [03:54<1:20:36, 1.17it/s, tok/s=7342.2]
Inference: 4%|▍ | 242/5920 [03:54<1:17:47, 1.22it/s, tok/s=7342.2]
Inference: 4%|▍ | 242/5920 [03:54<1:17:47, 1.22it/s, tok/s=7368.7]
Inference: 4%|▍ | 243/5920 [03:55<1:12:38, 1.30it/s, tok/s=7368.7]
Inference: 4%|▍ | 243/5920 [03:55<1:12:38, 1.30it/s, tok/s=7399.1]
Inference: 4%|▍ | 244/5920 [03:55<55:50, 1.69it/s, tok/s=7399.1]
Inference: 4%|▍ | 244/5920 [03:55<55:50, 1.69it/s, tok/s=7444.3]
Inference: 4%|▍ | 245/5920 [03:56<50:27, 1.87it/s, tok/s=7444.3]
Inference: 4%|▍ | 245/5920 [03:56<50:27, 1.87it/s, tok/s=7482.4]
Inference: 4%|▍ | 246/5920 [03:56<45:21, 2.08it/s, tok/s=7482.4]
Inference: 4%|▍ | 246/5920 [03:56<45:21, 2.08it/s, tok/s=7522.0]
Inference: 4%|▍ | 247/5920 [03:56<42:42, 2.21it/s, tok/s=7522.0]
Inference: 4%|▍ | 247/5920 [03:56<42:42, 2.21it/s, tok/s=7526.4]
Inference: 4%|▍ | 248/5920 [03:57<38:05, 2.48it/s, tok/s=7526.4]
Inference: 4%|▍ | 248/5920 [03:57<38:05, 2.48it/s, tok/s=7565.2]
Inference: 4%|▍ | 249/5920 [03:58<53:25, 1.77it/s, tok/s=7565.2]
Inference: 4%|▍ | 249/5920 [03:58<53:25, 1.77it/s, tok/s=7564.2]
Inference: 4%|▍ | 250/5920 [03:59<1:17:32, 1.22it/s, tok/s=7564.2]
Inference: 4%|▍ | 250/5920 [03:59<1:17:32, 1.22it/s, tok/s=7569.5]
Inference: 4%|▍ | 251/5920 [03:59<58:56, 1.60it/s, tok/s=7569.5]
Inference: 4%|▍ | 251/5920 [03:59<58:56, 1.60it/s, tok/s=7614.0]
Inference: 4%|▍ | 252/5920 [03:59<48:02, 1.97it/s, tok/s=7614.0]
Inference: 4%|▍ | 252/5920 [03:59<48:02, 1.97it/s, tok/s=7660.0]
Inference: 4%|▍ | 253/5920 [04:00<56:21, 1.68it/s, tok/s=7660.0]
Inference: 4%|▍ | 253/5920 [04:00<56:21, 1.68it/s, tok/s=7683.9]
Inference: 4%|▍ | 254/5920 [04:00<45:57, 2.05it/s, tok/s=7683.9]
Inference: 4%|▍ | 254/5920 [04:00<45:57, 2.05it/s, tok/s=7725.9]
Inference: 4%|▍ | 255/5920 [04:01<41:47, 2.26it/s, tok/s=7725.9]
Inference: 4%|▍ | 255/5920 [04:01<41:47, 2.26it/s, tok/s=7765.3]
Inference: 4%|▍ | 256/5920 [04:02<56:04, 1.68it/s, tok/s=7765.3]
Inference: 4%|▍ | 256/5920 [04:02<56:04, 1.68it/s, tok/s=7783.4]
Inference: 4%|▍ | 257/5920 [04:02<53:33, 1.76it/s, tok/s=7783.4]
Inference: 4%|▍ | 257/5920 [04:02<53:33, 1.76it/s, tok/s=7792.5]
Inference: 4%|▍ | 258/5920 [04:03<44:27, 2.12it/s, tok/s=7792.5]
Inference: 4%|▍ | 258/5920 [04:03<44:27, 2.12it/s, tok/s=7833.8]
Inference: 4%|▍ | 259/5920 [04:03<44:26, 2.12it/s, tok/s=7880.6]
Inference: 4%|▍ | 260/5920 [04:03<44:26, 2.12it/s, tok/s=7925.1]
Inference: 4%|▍ | 261/5920 [04:03<24:24, 3.86it/s, tok/s=7925.1]
Inference: 4%|▍ | 261/5920 [04:03<24:24, 3.86it/s, tok/s=7968.3]
Inference: 4%|▍ | 262/5920 [04:04<34:43, 2.72it/s, tok/s=7968.3]
Inference: 4%|▍ | 262/5920 [04:04<34:43, 2.72it/s, tok/s=7990.7]
Inference: 4%|▍ | 263/5920 [04:04<34:42, 2.72it/s, tok/s=8040.5]
Inference: 4%|▍ | 264/5920 [04:04<26:56, 3.50it/s, tok/s=8040.5]
Inference: 4%|▍ | 264/5920 [04:04<26:56, 3.50it/s, tok/s=8080.7]
Inference: 4%|▍ | 265/5920 [04:05<36:15, 2.60it/s, tok/s=8080.7]
Inference: 4%|▍ | 265/5920 [04:05<36:15, 2.60it/s, tok/s=8105.9]
Inference: 4%|▍ | 266/5920 [04:05<37:27, 2.52it/s, tok/s=8105.9]
Inference: 4%|▍ | 266/5920 [04:05<37:27, 2.52it/s, tok/s=8109.6]
Inference: 5%|▍ | 267/5920 [04:05<30:52, 3.05it/s, tok/s=8109.6]
Inference: 5%|▍ | 267/5920 [04:05<30:52, 3.05it/s, tok/s=8133.0]
Inference: 5%|▍ | 268/5920 [04:06<35:59, 2.62it/s, tok/s=8133.0]
Inference: 5%|▍ | 268/5920 [04:06<35:59, 2.62it/s, tok/s=8155.4]
Inference: 5%|▍ | 269/5920 [04:06<36:01, 2.61it/s, tok/s=8155.4]
Inference: 5%|▍ | 269/5920 [04:06<36:01, 2.61it/s, tok/s=8190.1]
Inference: 5%|▍ | 270/5920 [04:06<36:00, 2.61it/s, tok/s=8236.0]
Inference: 5%|▍ | 271/5920 [04:06<25:54, 3.63it/s, tok/s=8236.0]
Inference: 5%|▍ | 271/5920 [04:06<25:54, 3.63it/s, tok/s=8253.3]
Inference: 5%|▍ | 272/5920 [04:07<25:23, 3.71it/s, tok/s=8253.3]
Inference: 5%|▍ | 272/5920 [04:07<25:23, 3.71it/s, tok/s=8294.8]
Inference: 5%|▍ | 273/5920 [04:07<25:22, 3.71it/s, tok/s=8342.6]
Inference: 5%|▍ | 274/5920 [04:07<19:52, 4.74it/s, tok/s=8342.6]
Inference: 5%|▍ | 274/5920 [04:07<19:52, 4.74it/s, tok/s=8386.8]
Inference: 5%|▍ | 275/5920 [04:07<23:59, 3.92it/s, tok/s=8386.8]
Inference: 5%|▍ | 275/5920 [04:07<23:59, 3.92it/s, tok/s=8385.4]
Inference: 5%|▍ | 276/5920 [04:08<30:34, 3.08it/s, tok/s=8385.4]
Inference: 5%|▍ | 276/5920 [04:08<30:34, 3.08it/s, tok/s=8416.5]
Inference: 5%|▍ | 277/5920 [04:08<29:46, 3.16it/s, tok/s=8416.5]
Inference: 5%|▍ | 277/5920 [04:08<29:46, 3.16it/s, tok/s=8456.9]
Inference: 5%|▍ | 278/5920 [04:08<25:51, 3.64it/s, tok/s=8456.9]
Inference: 5%|▍ | 278/5920 [04:08<25:51, 3.64it/s, tok/s=8499.7]
Inference: 5%|▍ | 279/5920 [04:08<25:51, 3.64it/s, tok/s=8547.1]
Inference: 5%|▍ | 280/5920 [04:10<46:47, 2.01it/s, tok/s=8547.1]
Inference: 5%|▍ | 280/5920 [04:10<46:47, 2.01it/s, tok/s=8540.0]
Inference: 5%|▍ | 281/5920 [04:11<57:43, 1.63it/s, tok/s=8540.0]
Inference: 5%|▍ | 281/5920 [04:11<57:43, 1.63it/s, tok/s=8555.8]
Inference: 5%|▍ | 282/5920 [04:11<57:42, 1.63it/s, tok/s=8603.3]
Inference: 5%|▍ | 283/5920 [04:12<57:02, 1.65it/s, tok/s=8603.3]
Inference: 5%|▍ | 283/5920 [04:12<57:02, 1.65it/s, tok/s=8614.6]
Inference: 5%|▍ | 284/5920 [04:12<46:45, 2.01it/s, tok/s=8614.6]
Inference: 5%|▍ | 284/5920 [04:12<46:45, 2.01it/s, tok/s=8662.2]
Inference: 5%|▍ | 285/5920 [04:13<59:57, 1.57it/s, tok/s=8662.2]
Inference: 5%|▍ | 285/5920 [04:13<59:57, 1.57it/s, tok/s=8674.7]
Inference: 5%|▍ | 286/5920 [04:14<59:17, 1.58it/s, tok/s=8674.7]
Inference: 5%|▍ | 286/5920 [04:14<59:17, 1.58it/s, tok/s=8701.7]
Inference: 5%|▍ | 287/5920 [04:14<59:16, 1.58it/s, tok/s=8748.7]
Inference: 5%|▍ | 288/5920 [04:14<42:44, 2.20it/s, tok/s=8748.7]
Inference: 5%|▍ | 288/5920 [04:14<42:44, 2.20it/s, tok/s=8784.0]
Inference: 5%|▍ | 289/5920 [04:16<1:12:36, 1.29it/s, tok/s=8784.0]
Inference: 5%|▍ | 289/5920 [04:16<1:12:36, 1.29it/s, tok/s=8770.7]
Inference: 5%|▍ | 290/5920 [04:16<59:34, 1.58it/s, tok/s=8770.7]
Inference: 5%|▍ | 290/5920 [04:16<59:34, 1.58it/s, tok/s=8813.8]
Inference: 5%|▍ | 291/5920 [04:16<47:50, 1.96it/s, tok/s=8813.8]
Inference: 5%|▍ | 291/5920 [04:16<47:50, 1.96it/s, tok/s=8856.9]
Inference: 5%|▍ | 292/5920 [04:17<59:42, 1.57it/s, tok/s=8856.9]
Inference: 5%|▍ | 292/5920 [04:17<59:42, 1.57it/s, tok/s=8870.7]
Inference: 5%|▍ | 293/5920 [04:19<1:17:01, 1.22it/s, tok/s=8870.7]
Inference: 5%|▍ | 293/5920 [04:19<1:17:01, 1.22it/s, tok/s=8873.2]
Inference: 5%|▍ | 294/5920 [04:20<1:17:26, 1.21it/s, tok/s=8873.2]
Inference: 5%|▍ | 294/5920 [04:20<1:17:26, 1.21it/s, tok/s=8891.1]
Inference: 5%|▍ | 295/5920 [04:20<1:17:25, 1.21it/s, tok/s=8940.3]
Inference: 5%|▌ | 296/5920 [04:20<46:23, 2.02it/s, tok/s=8940.3]
Inference: 5%|▌ | 296/5920 [04:20<46:23, 2.02it/s, tok/s=8984.8]
Inference: 5%|▌ | 297/5920 [04:20<46:23, 2.02it/s, tok/s=9031.8]
Inference: 5%|▌ | 298/5920 [04:20<33:21, 2.81it/s, tok/s=9031.8]
Inference: 5%|▌ | 298/5920 [04:20<33:21, 2.81it/s, tok/s=9073.0]
Inference: 5%|▌ | 299/5920 [04:20<33:21, 2.81it/s, tok/s=9119.7]
Inference: 5%|▌ | 300/5920 [04:20<33:20, 2.81it/s, tok/s=9136.2]
Inference: 5%|▌ | 301/5920 [04:20<22:00, 4.25it/s, tok/s=9136.2]
Inference: 5%|▌ | 301/5920 [04:20<22:00, 4.25it/s, tok/s=9175.4]
Inference: 5%|▌ | 302/5920 [04:21<33:12, 2.82it/s, tok/s=9175.4]
Inference: 5%|▌ | 302/5920 [04:21<33:12, 2.82it/s, tok/s=9183.3]
Inference: 5%|▌ | 303/5920 [04:21<33:11, 2.82it/s, tok/s=9211.2]
Inference: 5%|▌ | 304/5920 [04:22<36:53, 2.54it/s, tok/s=9211.2]
Inference: 5%|▌ | 304/5920 [04:22<36:53, 2.54it/s, tok/s=9209.6]
Inference: 5%|▌ | 305/5920 [04:22<32:09, 2.91it/s, tok/s=9209.6]
Inference: 5%|▌ | 305/5920 [04:22<32:09, 2.91it/s, tok/s=9252.1]
Inference: 5%|▌ | 306/5920 [04:23<34:14, 2.73it/s, tok/s=9252.1]
Inference: 5%|▌ | 306/5920 [04:23<34:14, 2.73it/s, tok/s=9285.4]
Inference: 5%|▌ | 307/5920 [04:23<29:34, 3.16it/s, tok/s=9285.4]
Inference: 5%|▌ | 307/5920 [04:23<29:34, 3.16it/s, tok/s=9328.7]
Inference: 5%|▌ | 308/5920 [04:23<36:20, 2.57it/s, tok/s=9328.7]
Inference: 5%|▌ | 308/5920 [04:23<36:20, 2.57it/s, tok/s=9357.1]
Inference: 5%|▌ | 309/5920 [04:25<1:00:26, 1.55it/s, tok/s=9357.1]
Inference: 5%|▌ | 309/5920 [04:25<1:00:26, 1.55it/s, tok/s=9357.3]
Inference: 5%|▌ | 310/5920 [04:25<49:27, 1.89it/s, tok/s=9357.3]
Inference: 5%|▌ | 310/5920 [04:25<49:27, 1.89it/s, tok/s=9399.2]
Inference: 5%|▌ | 311/5920 [04:26<56:10, 1.66it/s, tok/s=9399.2]
Inference: 5%|▌ | 311/5920 [04:26<56:10, 1.66it/s, tok/s=9383.7]
Inference: 5%|▌ | 312/5920 [04:26<43:21, 2.16it/s, tok/s=9383.7]
Inference: 5%|▌ | 312/5920 [04:26<43:21, 2.16it/s, tok/s=9425.6]
Inference: 5%|▌ | 313/5920 [04:26<33:32, 2.79it/s, tok/s=9425.6]
Inference: 5%|▌ | 313/5920 [04:26<33:32, 2.79it/s, tok/s=9468.7]
Inference: 5%|▌ | 314/5920 [04:27<35:04, 2.66it/s, tok/s=9468.7]
Inference: 5%|▌ | 314/5920 [04:27<35:04, 2.66it/s, tok/s=9500.5]
Inference: 5%|▌ | 315/5920 [04:27<41:20, 2.26it/s, tok/s=9500.5]
Inference: 5%|▌ | 315/5920 [04:27<41:20, 2.26it/s, tok/s=9527.7]
Inference: 5%|▌ | 316/5920 [04:28<48:10, 1.94it/s, tok/s=9527.7]
Inference: 5%|▌ | 316/5920 [04:28<48:10, 1.94it/s, tok/s=9516.5]
Inference: 5%|▌ | 317/5920 [04:29<58:10, 1.61it/s, tok/s=9516.5]
Inference: 5%|▌ | 317/5920 [04:29<58:10, 1.61it/s, tok/s=9533.6]
Inference: 5%|▌ | 318/5920 [04:29<58:09, 1.61it/s, tok/s=9577.3]
Inference: 5%|▌ | 319/5920 [04:29<34:30, 2.71it/s, tok/s=9577.3]
Inference: 5%|▌ | 319/5920 [04:29<34:30, 2.71it/s, tok/s=9617.4]
Inference: 5%|▌ | 320/5920 [04:29<39:01, 2.39it/s, tok/s=9617.4]
Inference: 5%|▌ | 320/5920 [04:29<39:01, 2.39it/s, tok/s=9625.9]
Inference: 5%|▌ | 321/5920 [04:30<35:07, 2.66it/s, tok/s=9625.9]
Inference: 5%|▌ | 321/5920 [04:30<35:07, 2.66it/s, tok/s=9663.4]
Inference: 5%|▌ | 322/5920 [04:30<35:06, 2.66it/s, tok/s=9682.3]
Inference: 5%|▌ | 323/5920 [04:30<23:57, 3.89it/s, tok/s=9682.3]
Inference: 5%|▌ | 323/5920 [04:30<23:57, 3.89it/s, tok/s=9723.1]
Inference: 5%|▌ | 324/5920 [04:30<23:57, 3.89it/s, tok/s=9771.5]
Inference: 5%|▌ | 325/5920 [04:30<18:06, 5.15it/s, tok/s=9771.5]
Inference: 5%|▌ | 325/5920 [04:30<18:06, 5.15it/s, tok/s=9813.3]
Inference: 6%|▌ | 326/5920 [04:30<19:07, 4.88it/s, tok/s=9813.3]
Inference: 6%|▌ | 326/5920 [04:30<19:07, 4.88it/s, tok/s=9853.4]
Inference: 6%|▌ | 327/5920 [04:32<46:12, 2.02it/s, tok/s=9853.4]
Inference: 6%|▌ | 327/5920 [04:32<46:12, 2.02it/s, tok/s=9847.2]
Inference: 6%|▌ | 328/5920 [04:32<46:11, 2.02it/s, tok/s=9886.6]
Inference: 6%|▌ | 329/5920 [04:32<31:19, 2.97it/s, tok/s=9886.6]
Inference: 6%|▌ | 329/5920 [04:32<31:19, 2.97it/s, tok/s=9926.2]
Inference: 6%|▌ | 330/5920 [04:32<33:26, 2.79it/s, tok/s=9926.2]
Inference: 6%|▌ | 330/5920 [04:32<33:26, 2.79it/s, tok/s=9951.3]
Inference: 6%|▌ | 331/5920 [04:32<33:25, 2.79it/s, tok/s=9997.5]
Inference: 6%|▌ | 332/5920 [04:33<27:10, 3.43it/s, tok/s=9997.5]
Inference: 6%|▌ | 332/5920 [04:33<27:10, 3.43it/s, tok/s=10011.3]
Inference: 6%|▌ | 333/5920 [04:33<29:12, 3.19it/s, tok/s=10011.3]
Inference: 6%|▌ | 333/5920 [04:33<29:12, 3.19it/s, tok/s=10045.0]
Inference: 6%|▌ | 334/5920 [04:34<32:19, 2.88it/s, tok/s=10045.0]
Inference: 6%|▌ | 334/5920 [04:34<32:19, 2.88it/s, tok/s=10047.0]
Inference: 6%|▌ | 335/5920 [04:34<32:19, 2.88it/s, tok/s=10090.7]
Inference: 6%|▌ | 336/5920 [04:34<32:18, 2.88it/s, tok/s=10138.4]
Inference: 6%|▌ | 337/5920 [04:34<21:42, 4.29it/s, tok/s=10138.4]
Inference: 6%|▌ | 337/5920 [04:34<21:42, 4.29it/s, tok/s=10154.9]
Inference: 6%|▌ | 338/5920 [04:35<33:40, 2.76it/s, tok/s=10154.9]
Inference: 6%|▌ | 338/5920 [04:35<33:40, 2.76it/s, tok/s=10170.3]
Inference: 6%|▌ | 339/5920 [04:36<44:45, 2.08it/s, tok/s=10170.3]
Inference: 6%|▌ | 339/5920 [04:36<44:45, 2.08it/s, tok/s=10185.2]
Inference: 6%|▌ | 340/5920 [04:36<44:44, 2.08it/s, tok/s=10231.5]
Inference: 6%|▌ | 341/5920 [04:37<57:03, 1.63it/s, tok/s=10231.5]
Inference: 6%|▌ | 341/5920 [04:37<57:03, 1.63it/s, tok/s=10221.3]
Inference: 6%|▌ | 342/5920 [04:38<50:42, 1.83it/s, tok/s=10221.3]
Inference: 6%|▌ | 342/5920 [04:38<50:42, 1.83it/s, tok/s=10256.5]
Inference: 6%|▌ | 343/5920 [04:38<50:41, 1.83it/s, tok/s=10275.3]
Inference: 6%|▌ | 344/5920 [04:38<33:12, 2.80it/s, tok/s=10275.3]
Inference: 6%|▌ | 344/5920 [04:38<33:12, 2.80it/s, tok/s=10280.6]
Inference: 6%|▌ | 345/5920 [04:38<32:42, 2.84it/s, tok/s=10280.6]
Inference: 6%|▌ | 345/5920 [04:38<32:42, 2.84it/s, tok/s=10285.6]
Inference: 6%|▌ | 346/5920 [04:38<31:03, 2.99it/s, tok/s=10285.6]
Inference: 6%|▌ | 346/5920 [04:38<31:03, 2.99it/s, tok/s=10293.0]
Inference: 6%|▌ | 347/5920 [04:42<1:38:17, 1.06s/it, tok/s=10293.0]
Inference: 6%|▌ | 347/5920 [04:42<1:38:17, 1.06s/it, tok/s=10220.8]
Inference: 6%|▌ | 348/5920 [04:42<1:20:14, 1.16it/s, tok/s=10220.8]
Inference: 6%|▌ | 348/5920 [04:42<1:20:14, 1.16it/s, tok/s=10255.3]
Inference: 6%|▌ | 349/5920 [04:42<1:08:28, 1.36it/s, tok/s=10255.3]
Inference: 6%|▌ | 349/5920 [04:42<1:08:28, 1.36it/s, tok/s=10250.8]
Inference: 6%|▌ | 350/5920 [04:43<1:10:24, 1.32it/s, tok/s=10250.8]
Inference: 6%|▌ | 350/5920 [04:43<1:10:24, 1.32it/s, tok/s=10269.9]
Inference: 6%|▌ | 351/5920 [04:45<1:32:41, 1.00it/s, tok/s=10269.9]
Inference: 6%|▌ | 351/5920 [04:45<1:32:41, 1.00it/s, tok/s=10262.3]
Inference: 6%|▌ | 352/5920 [04:45<1:19:03, 1.17it/s, tok/s=10262.3]
Inference: 6%|▌ | 352/5920 [04:45<1:19:03, 1.17it/s, tok/s=10292.8]
Inference: 6%|▌ | 353/5920 [04:46<1:03:00, 1.47it/s, tok/s=10292.8]
Inference: 6%|▌ | 353/5920 [04:46<1:03:00, 1.47it/s, tok/s=10331.5]
Inference: 6%|▌ | 354/5920 [04:46<48:05, 1.93it/s, tok/s=10331.5]
Inference: 6%|▌ | 354/5920 [04:46<48:05, 1.93it/s, tok/s=10341.7]
Inference: 6%|▌ | 355/5920 [04:46<50:37, 1.83it/s, tok/s=10341.7]
Inference: 6%|▌ | 355/5920 [04:46<50:37, 1.83it/s, tok/s=10365.3]
Inference: 6%|▌ | 356/5920 [04:47<46:58, 1.97it/s, tok/s=10365.3]
Inference: 6%|▌ | 356/5920 [04:47<46:58, 1.97it/s, tok/s=10397.9]
Inference: 6%|▌ | 357/5920 [04:47<46:58, 1.97it/s, tok/s=10441.5]
Inference: 6%|▌ | 358/5920 [04:48<43:19, 2.14it/s, tok/s=10441.5]
Inference: 6%|▌ | 358/5920 [04:48<43:19, 2.14it/s, tok/s=10433.7]
Inference: 6%|▌ | 359/5920 [04:48<36:20, 2.55it/s, tok/s=10433.7]
Inference: 6%|▌ | 359/5920 [04:48<36:20, 2.55it/s, tok/s=10474.3]
Inference: 6%|▌ | 360/5920 [04:48<30:52, 3.00it/s, tok/s=10474.3]
Inference: 6%|▌ | 360/5920 [04:48<30:52, 3.00it/s, tok/s=10517.1]
Inference: 6%|▌ | 361/5920 [04:48<26:09, 3.54it/s, tok/s=10517.1]
Inference: 6%|▌ | 361/5920 [04:48<26:09, 3.54it/s, tok/s=10558.1]
Inference: 6%|▌ | 362/5920 [04:49<40:39, 2.28it/s, tok/s=10558.1]
Inference: 6%|▌ | 362/5920 [04:49<40:39, 2.28it/s, tok/s=10574.8]
Inference: 6%|▌ | 363/5920 [04:49<44:22, 2.09it/s, tok/s=10574.8]
Inference: 6%|▌ | 363/5920 [04:49<44:22, 2.09it/s, tok/s=10602.0]
Inference: 6%|▌ | 364/5920 [04:51<1:19:01, 1.17it/s, tok/s=10602.0]
Inference: 6%|▌ | 364/5920 [04:51<1:19:01, 1.17it/s, tok/s=10583.2]
Inference: 6%|▌ | 365/5920 [04:52<1:27:16, 1.06it/s, tok/s=10583.2]
Inference: 6%|▌ | 365/5920 [04:52<1:27:16, 1.06it/s, tok/s=10588.7]
Inference: 6%|▌ | 366/5920 [04:52<1:27:15, 1.06it/s, tok/s=10624.2]
Inference: 6%|▌ | 367/5920 [04:53<1:02:24, 1.48it/s, tok/s=10624.2]
Inference: 6%|▌ | 367/5920 [04:53<1:02:24, 1.48it/s, tok/s=10625.6]
Inference: 6%|▌ | 368/5920 [04:53<1:02:24, 1.48it/s, tok/s=10668.0]
Inference: 6%|▌ | 369/5920 [04:54<55:41, 1.66it/s, tok/s=10668.0]
Inference: 6%|▌ | 369/5920 [04:54<55:41, 1.66it/s, tok/s=10659.0]
Inference: 6%|▋ | 370/5920 [04:54<46:07, 2.01it/s, tok/s=10659.0]
Inference: 6%|▋ | 370/5920 [04:54<46:07, 2.01it/s, tok/s=10701.4]
Inference: 6%|▋ | 371/5920 [04:55<45:17, 2.04it/s, tok/s=10701.4]
Inference: 6%|▋ | 371/5920 [04:55<45:17, 2.04it/s, tok/s=10731.5]
Inference: 6%|▋ | 372/5920 [04:55<47:37, 1.94it/s, tok/s=10731.5]
Inference: 6%|▋ | 372/5920 [04:55<47:37, 1.94it/s, tok/s=10757.6]
Inference: 6%|▋ | 373/5920 [04:55<47:36, 1.94it/s, tok/s=10784.9]
Inference: 6%|▋ | 374/5920 [04:56<43:20, 2.13it/s, tok/s=10784.9]
Inference: 6%|▋ | 374/5920 [04:56<43:20, 2.13it/s, tok/s=10804.0]
Inference: 6%|▋ | 375/5920 [04:56<38:55, 2.37it/s, tok/s=10804.0]
Inference: 6%|▋ | 375/5920 [04:56<38:55, 2.37it/s, tok/s=10822.3]
Inference: 6%|▋ | 376/5920 [04:56<38:54, 2.37it/s, tok/s=10837.5]
Inference: 6%|▋ | 377/5920 [04:56<26:20, 3.51it/s, tok/s=10837.5]
Inference: 6%|▋ | 377/5920 [04:56<26:20, 3.51it/s, tok/s=10877.7]
Inference: 6%|▋ | 378/5920 [04:57<27:30, 3.36it/s, tok/s=10877.7]
Inference: 6%|▋ | 378/5920 [04:57<27:30, 3.36it/s, tok/s=10876.5]
Inference: 6%|▋ | 379/5920 [04:58<56:28, 1.64it/s, tok/s=10876.5]
Inference: 6%|▋ | 379/5920 [04:58<56:28, 1.64it/s, tok/s=10865.0]
Inference: 6%|▋ | 380/5920 [05:00<1:10:50, 1.30it/s, tok/s=10865.0]
Inference: 6%|▋ | 380/5920 [05:00<1:10:50, 1.30it/s, tok/s=10866.2]
Inference: 6%|▋ | 381/5920 [05:00<59:29, 1.55it/s, tok/s=10866.2]
Inference: 6%|▋ | 381/5920 [05:00<59:29, 1.55it/s, tok/s=10902.5]
Inference: 6%|▋ | 382/5920 [05:00<51:12, 1.80it/s, tok/s=10902.5]
Inference: 6%|▋ | 382/5920 [05:00<51:12, 1.80it/s, tok/s=10906.2]
Inference: 6%|▋ | 383/5920 [05:00<39:42, 2.32it/s, tok/s=10906.2]
Inference: 6%|▋ | 383/5920 [05:00<39:42, 2.32it/s, tok/s=10919.0]
Inference: 6%|▋ | 384/5920 [05:01<34:24, 2.68it/s, tok/s=10919.0]
Inference: 6%|▋ | 384/5920 [05:01<34:24, 2.68it/s, tok/s=10953.6]
Inference: 7%|▋ | 385/5920 [05:01<34:24, 2.68it/s, tok/s=10968.3]
Inference: 7%|▋ | 386/5920 [05:01<31:21, 2.94it/s, tok/s=10968.3]
Inference: 7%|▋ | 386/5920 [05:01<31:21, 2.94it/s, tok/s=10962.2]
Inference: 7%|▋ | 387/5920 [05:02<31:11, 2.96it/s, tok/s=10962.2]
Inference: 7%|▋ | 387/5920 [05:02<31:11, 2.96it/s, tok/s=10997.1]
Inference: 7%|▋ | 388/5920 [05:03<50:53, 1.81it/s, tok/s=10997.1]
Inference: 7%|▋ | 388/5920 [05:03<50:53, 1.81it/s, tok/s=10976.3]
Inference: 7%|▋ | 389/5920 [05:04<1:11:27, 1.29it/s, tok/s=10976.3]
Inference: 7%|▋ | 389/5920 [05:04<1:11:27, 1.29it/s, tok/s=10974.9]
Inference: 7%|▋ | 390/5920 [05:04<1:11:26, 1.29it/s, tok/s=11019.8]
Inference: 7%|▋ | 391/5920 [05:04<48:57, 1.88it/s, tok/s=11019.8]
Inference: 7%|▋ | 391/5920 [05:04<48:57, 1.88it/s, tok/s=11018.0]
Inference: 7%|▋ | 392/5920 [05:05<52:29, 1.76it/s, tok/s=11018.0]
Inference: 7%|▋ | 392/5920 [05:05<52:29, 1.76it/s, tok/s=11004.1]
Inference: 7%|▋ | 393/5920 [05:06<54:13, 1.70it/s, tok/s=11004.1]
Inference: 7%|▋ | 393/5920 [05:06<54:13, 1.70it/s, tok/s=11026.2]
Inference: 7%|▋ | 394/5920 [05:06<54:15, 1.70it/s, tok/s=11026.2]
Inference: 7%|▋ | 394/5920 [05:06<54:15, 1.70it/s, tok/s=11022.6]
Inference: 7%|▋ | 395/5920 [05:07<47:45, 1.93it/s, tok/s=11022.6]
Inference: 7%|▋ | 395/5920 [05:07<47:45, 1.93it/s, tok/s=11033.5]
Inference: 7%|▋ | 396/5920 [05:07<37:19, 2.47it/s, tok/s=11033.5]
Inference: 7%|▋ | 396/5920 [05:07<37:19, 2.47it/s, tok/s=11077.4]
Inference: 7%|▋ | 397/5920 [05:08<43:35, 2.11it/s, tok/s=11077.4]
Inference: 7%|▋ | 397/5920 [05:08<43:35, 2.11it/s, tok/s=11065.2]
Inference: 7%|▋ | 398/5920 [05:08<55:06, 1.67it/s, tok/s=11065.2]
Inference: 7%|▋ | 398/5920 [05:08<55:06, 1.67it/s, tok/s=11079.9]
Inference: 7%|▋ | 399/5920 [05:09<57:42, 1.59it/s, tok/s=11079.9]
Inference: 7%|▋ | 399/5920 [05:09<57:42, 1.59it/s, tok/s=11070.1]
Inference: 7%|▋ | 400/5920 [05:10<1:05:26, 1.41it/s, tok/s=11070.1]
Inference: 7%|▋ | 400/5920 [05:10<1:05:26, 1.41it/s, tok/s=11060.1]
Inference: 7%|▋ | 401/5920 [05:10<52:20, 1.76it/s, tok/s=11060.1]
Inference: 7%|▋ | 401/5920 [05:10<52:20, 1.76it/s, tok/s=11099.3]
Inference: 7%|▋ | 402/5920 [05:11<55:23, 1.66it/s, tok/s=11099.3]
Inference: 7%|▋ | 402/5920 [05:11<55:23, 1.66it/s, tok/s=11120.5]
Inference: 7%|▋ | 403/5920 [05:12<58:23, 1.57it/s, tok/s=11120.5]
Inference: 7%|▋ | 403/5920 [05:12<58:23, 1.57it/s, tok/s=11142.3]
Inference: 7%|▋ | 404/5920 [05:13<1:05:00, 1.41it/s, tok/s=11142.3]
Inference: 7%|▋ | 404/5920 [05:13<1:05:00, 1.41it/s, tok/s=11157.8]
Inference: 7%|▋ | 405/5920 [05:13<1:10:16, 1.31it/s, tok/s=11157.8]
Inference: 7%|▋ | 405/5920 [05:13<1:10:16, 1.31it/s, tok/s=11170.7]
Inference: 7%|▋ | 406/5920 [05:14<1:02:07, 1.48it/s, tok/s=11170.7]
Inference: 7%|▋ | 406/5920 [05:14<1:02:07, 1.48it/s, tok/s=11201.4]
Inference: 7%|▋ | 407/5920 [05:15<1:00:24, 1.52it/s, tok/s=11201.4]
Inference: 7%|▋ | 407/5920 [05:15<1:00:24, 1.52it/s, tok/s=11226.2]
Inference: 7%|▋ | 408/5920 [05:15<50:51, 1.81it/s, tok/s=11226.2]
Inference: 7%|▋ | 408/5920 [05:15<50:51, 1.81it/s, tok/s=11226.6]
Inference: 7%|▋ | 409/5920 [05:15<42:06, 2.18it/s, tok/s=11226.6]
Inference: 7%|▋ | 409/5920 [05:15<42:06, 2.18it/s, tok/s=11239.2]
Inference: 7%|▋ | 410/5920 [05:16<42:32, 2.16it/s, tok/s=11239.2]
Inference: 7%|▋ | 410/5920 [05:16<42:32, 2.16it/s, tok/s=11270.4]
Inference: 7%|▋ | 411/5920 [05:17<56:50, 1.62it/s, tok/s=11270.4]
Inference: 7%|▋ | 411/5920 [05:17<56:50, 1.62it/s, tok/s=11258.2]
Inference: 7%|▋ | 412/5920 [05:17<56:49, 1.62it/s, tok/s=11293.1]
Inference: 7%|▋ | 413/5920 [05:18<54:53, 1.67it/s, tok/s=11293.1]
Inference: 7%|▋ | 413/5920 [05:18<54:53, 1.67it/s, tok/s=11300.5]
Inference: 7%|▋ | 414/5920 [05:18<54:52, 1.67it/s, tok/s=11344.2]
Inference: 7%|▋ | 415/5920 [05:18<47:16, 1.94it/s, tok/s=11344.2]
Inference: 7%|▋ | 415/5920 [05:18<47:16, 1.94it/s, tok/s=11345.7]
Inference: 7%|▋ | 416/5920 [05:18<47:15, 1.94it/s, tok/s=11391.6]
Inference: 7%|▋ | 417/5920 [05:19<35:45, 2.57it/s, tok/s=11391.6]
Inference: 7%|▋ | 417/5920 [05:19<35:45, 2.57it/s, tok/s=11428.1]
Inference: 7%|▋ | 418/5920 [05:19<32:52, 2.79it/s, tok/s=11428.1]
Inference: 7%|▋ | 418/5920 [05:19<32:52, 2.79it/s, tok/s=11466.2]
Inference: 7%|▋ | 419/5920 [05:20<52:05, 1.76it/s, tok/s=11466.2]
Inference: 7%|▋ | 419/5920 [05:20<52:05, 1.76it/s, tok/s=11459.7]
Inference: 7%|▋ | 420/5920 [05:21<1:03:33, 1.44it/s, tok/s=11459.7]
Inference: 7%|▋ | 420/5920 [05:21<1:03:33, 1.44it/s, tok/s=11468.4]
Inference: 7%|▋ | 421/5920 [05:22<1:08:58, 1.33it/s, tok/s=11468.4]
Inference: 7%|▋ | 421/5920 [05:22<1:08:58, 1.33it/s, tok/s=11480.2]
Inference: 7%|▋ | 422/5920 [05:22<1:08:58, 1.33it/s, tok/s=11495.6]
Inference: 7%|▋ | 423/5920 [05:23<48:26, 1.89it/s, tok/s=11495.6]
Inference: 7%|▋ | 423/5920 [05:23<48:26, 1.89it/s, tok/s=11527.6]
Inference: 7%|▋ | 424/5920 [05:23<39:40, 2.31it/s, tok/s=11527.6]
Inference: 7%|▋ | 424/5920 [05:23<39:40, 2.31it/s, tok/s=11570.0]
Inference: 7%|▋ | 425/5920 [05:23<35:08, 2.61it/s, tok/s=11570.0]
Inference: 7%|▋ | 425/5920 [05:23<35:08, 2.61it/s, tok/s=11607.8]
Inference: 7%|▋ | 426/5920 [05:24<51:23, 1.78it/s, tok/s=11607.8]
Inference: 7%|▋ | 426/5920 [05:24<51:23, 1.78it/s, tok/s=11616.4]
Inference: 7%|▋ | 427/5920 [05:24<43:27, 2.11it/s, tok/s=11616.4]
Inference: 7%|▋ | 427/5920 [05:24<43:27, 2.11it/s, tok/s=11637.1]
Inference: 7%|▋ | 428/5920 [05:25<35:51, 2.55it/s, tok/s=11637.1]
Inference: 7%|▋ | 428/5920 [05:25<35:51, 2.55it/s, tok/s=11671.2]
Inference: 7%|▋ | 429/5920 [05:26<52:42, 1.74it/s, tok/s=11671.2]
Inference: 7%|▋ | 429/5920 [05:26<52:42, 1.74it/s, tok/s=11676.8]
Inference: 7%|▋ | 430/5920 [05:26<52:42, 1.74it/s, tok/s=11720.7]
Inference: 7%|▋ | 431/5920 [05:26<45:28, 2.01it/s, tok/s=11720.7]
Inference: 7%|▋ | 431/5920 [05:26<45:28, 2.01it/s, tok/s=11740.2]
Inference: 7%|▋ | 432/5920 [05:27<41:15, 2.22it/s, tok/s=11740.2]
Inference: 7%|▋ | 432/5920 [05:27<41:15, 2.22it/s, tok/s=11739.6]
Inference: 7%|▋ | 433/5920 [05:27<46:19, 1.97it/s, tok/s=11739.6]
Inference: 7%|▋ | 433/5920 [05:27<46:19, 1.97it/s, tok/s=11762.0]
Inference: 7%|▋ | 434/5920 [05:28<47:29, 1.93it/s, tok/s=11762.0]
Inference: 7%|▋ | 434/5920 [05:28<47:29, 1.93it/s, tok/s=11786.7]
Inference: 7%|▋ | 435/5920 [05:28<41:30, 2.20it/s, tok/s=11786.7]
Inference: 7%|▋ | 435/5920 [05:28<41:30, 2.20it/s, tok/s=11822.3]
Inference: 7%|▋ | 436/5920 [05:28<35:32, 2.57it/s, tok/s=11822.3]
Inference: 7%|▋ | 436/5920 [05:28<35:32, 2.57it/s, tok/s=11860.6]
Inference: 7%|▋ | 437/5920 [05:29<33:26, 2.73it/s, tok/s=11860.6]
Inference: 7%|▋ | 437/5920 [05:29<33:26, 2.73it/s, tok/s=11889.6]
Inference: 7%|▋ | 438/5920 [05:30<47:32, 1.92it/s, tok/s=11889.6]
Inference: 7%|▋ | 438/5920 [05:30<47:32, 1.92it/s, tok/s=11885.3]
Inference: 7%|▋ | 439/5920 [05:30<53:20, 1.71it/s, tok/s=11885.3]
Inference: 7%|▋ | 439/5920 [05:30<53:20, 1.71it/s, tok/s=11905.7]
Inference: 7%|▋ | 440/5920 [05:31<41:31, 2.20it/s, tok/s=11905.7]
Inference: 7%|▋ | 440/5920 [05:31<41:31, 2.20it/s, tok/s=11945.7]
Inference: 7%|▋ | 441/5920 [05:31<49:01, 1.86it/s, tok/s=11945.7]
Inference: 7%|▋ | 441/5920 [05:31<49:01, 1.86it/s, tok/s=11931.6]
Inference: 7%|▋ | 442/5920 [05:32<1:06:26, 1.37it/s, tok/s=11931.6]
Inference: 7%|▋ | 442/5920 [05:32<1:06:26, 1.37it/s, tok/s=11898.7]
Inference: 7%|▋ | 443/5920 [05:33<1:05:23, 1.40it/s, tok/s=11898.7]
Inference: 7%|▋ | 443/5920 [05:33<1:05:23, 1.40it/s, tok/s=11905.1]
Inference: 8%|▊ | 444/5920 [05:35<1:23:00, 1.10it/s, tok/s=11905.1]
Inference: 8%|▊ | 444/5920 [05:35<1:23:00, 1.10it/s, tok/s=11901.7]
Inference: 8%|▊ | 445/5920 [05:35<1:03:18, 1.44it/s, tok/s=11901.7]
Inference: 8%|▊ | 445/5920 [05:35<1:03:18, 1.44it/s, tok/s=11939.5]
Inference: 8%|▊ | 446/5920 [05:35<1:03:17, 1.44it/s, tok/s=11950.2]
Inference: 8%|▊ | 447/5920 [05:36<53:25, 1.71it/s, tok/s=11950.2]
Inference: 8%|▊ | 447/5920 [05:36<53:25, 1.71it/s, tok/s=11939.2]
Inference: 8%|▊ | 448/5920 [05:37<1:01:27, 1.48it/s, tok/s=11939.2]
Inference: 8%|▊ | 448/5920 [05:37<1:01:27, 1.48it/s, tok/s=11953.6]
Inference: 8%|▊ | 449/5920 [05:37<1:01:04, 1.49it/s, tok/s=11953.6]
Inference: 8%|▊ | 449/5920 [05:37<1:01:04, 1.49it/s, tok/s=11974.4]
Inference: 8%|▊ | 450/5920 [05:37<48:09, 1.89it/s, tok/s=11974.4]
Inference: 8%|▊ | 450/5920 [05:37<48:09, 1.89it/s, tok/s=12017.3]
Inference: 8%|▊ | 451/5920 [05:37<37:56, 2.40it/s, tok/s=12017.3]
Inference: 8%|▊ | 451/5920 [05:37<37:56, 2.40it/s, tok/s=12036.0]
Inference: 8%|▊ | 452/5920 [05:38<33:26, 2.72it/s, tok/s=12036.0]
Inference: 8%|▊ | 452/5920 [05:38<33:26, 2.72it/s, tok/s=12041.6]
Inference: 8%|▊ | 453/5920 [05:38<32:34, 2.80it/s, tok/s=12041.6]
Inference: 8%|▊ | 453/5920 [05:38<32:34, 2.80it/s, tok/s=12043.5]
Inference: 8%|▊ | 454/5920 [05:39<39:55, 2.28it/s, tok/s=12043.5]
Inference: 8%|▊ | 454/5920 [05:39<39:55, 2.28it/s, tok/s=12045.8]
Inference: 8%|▊ | 455/5920 [05:41<1:35:03, 1.04s/it, tok/s=12045.8]
Inference: 8%|▊ | 455/5920 [05:41<1:35:03, 1.04s/it, tok/s=12003.9]
Inference: 8%|▊ | 456/5920 [05:41<1:11:52, 1.27it/s, tok/s=12003.9]
Inference: 8%|▊ | 456/5920 [05:41<1:11:52, 1.27it/s, tok/s=12042.3]
Inference: 8%|▊ | 457/5920 [05:43<1:25:03, 1.07it/s, tok/s=12042.3]
Inference: 8%|▊ | 457/5920 [05:43<1:25:03, 1.07it/s, tok/s=12043.6]
Inference: 8%|▊ | 458/5920 [05:43<1:04:18, 1.42it/s, tok/s=12043.6]
Inference: 8%|▊ | 458/5920 [05:43<1:04:18, 1.42it/s, tok/s=12083.5]
Inference: 8%|▊ | 459/5920 [05:43<56:03, 1.62it/s, tok/s=12083.5]
Inference: 8%|▊ | 459/5920 [05:43<56:03, 1.62it/s, tok/s=12113.5]
Inference: 8%|▊ | 460/5920 [05:43<43:32, 2.09it/s, tok/s=12113.5]
Inference: 8%|▊ | 460/5920 [05:43<43:32, 2.09it/s, tok/s=12126.5]
Inference: 8%|▊ | 461/5920 [05:44<38:01, 2.39it/s, tok/s=12126.5]
Inference: 8%|▊ | 461/5920 [05:44<38:01, 2.39it/s, tok/s=12147.1]
Inference: 8%|▊ | 462/5920 [05:44<38:01, 2.39it/s, tok/s=12186.5]
Inference: 8%|▊ | 463/5920 [05:44<33:39, 2.70it/s, tok/s=12186.5]
Inference: 8%|▊ | 463/5920 [05:44<33:39, 2.70it/s, tok/s=12186.3]
Inference: 8%|▊ | 464/5920 [05:44<27:38, 3.29it/s, tok/s=12186.3]
Inference: 8%|▊ | 464/5920 [05:44<27:38, 3.29it/s, tok/s=12198.1]
Inference: 8%|▊ | 465/5920 [05:45<46:22, 1.96it/s, tok/s=12198.1]
Inference: 8%|▊ | 465/5920 [05:45<46:22, 1.96it/s, tok/s=12207.2]
Inference: 8%|▊ | 466/5920 [05:46<48:47, 1.86it/s, tok/s=12207.2]
Inference: 8%|▊ | 466/5920 [05:46<48:47, 1.86it/s, tok/s=12230.7]
Inference: 8%|▊ | 467/5920 [05:46<48:46, 1.86it/s, tok/s=12257.8]
Inference: 8%|▊ | 468/5920 [05:47<37:27, 2.43it/s, tok/s=12257.8]
Inference: 8%|▊ | 468/5920 [05:47<37:27, 2.43it/s, tok/s=12261.8]
Inference: 8%|▊ | 469/5920 [05:47<38:26, 2.36it/s, tok/s=12261.8]
Inference: 8%|▊ | 469/5920 [05:47<38:26, 2.36it/s, tok/s=12257.1]
Inference: 8%|▊ | 470/5920 [05:47<37:29, 2.42it/s, tok/s=12257.1]
Inference: 8%|▊ | 470/5920 [05:47<37:29, 2.42it/s, tok/s=12263.6]
Inference: 8%|▊ | 471/5920 [05:49<1:07:39, 1.34it/s, tok/s=12263.6]
Inference: 8%|▊ | 471/5920 [05:49<1:07:39, 1.34it/s, tok/s=12217.5]
Inference: 8%|▊ | 472/5920 [05:51<1:30:39, 1.00it/s, tok/s=12217.5]
Inference: 8%|▊ | 472/5920 [05:51<1:30:39, 1.00it/s, tok/s=12205.2]
Inference: 8%|▊ | 473/5920 [05:51<1:13:18, 1.24it/s, tok/s=12205.2]
Inference: 8%|▊ | 473/5920 [05:51<1:13:18, 1.24it/s, tok/s=12239.5]
Inference: 8%|▊ | 474/5920 [05:51<1:00:59, 1.49it/s, tok/s=12239.5]
Inference: 8%|▊ | 474/5920 [05:51<1:00:59, 1.49it/s, tok/s=12242.1]
Inference: 8%|▊ | 475/5920 [05:52<1:02:32, 1.45it/s, tok/s=12242.1]
Inference: 8%|▊ | 475/5920 [05:52<1:02:32, 1.45it/s, tok/s=12228.3]
Inference: 8%|▊ | 476/5920 [05:54<1:20:27, 1.13it/s, tok/s=12228.3]
Inference: 8%|▊ | 476/5920 [05:54<1:20:27, 1.13it/s, tok/s=12202.9]
Inference: 8%|▊ | 477/5920 [05:54<1:06:50, 1.36it/s, tok/s=12202.9]
Inference: 8%|▊ | 477/5920 [05:54<1:06:50, 1.36it/s, tok/s=12234.5]
Inference: 8%|▊ | 478/5920 [05:54<1:06:49, 1.36it/s, tok/s=12257.0]
Inference: 8%|▊ | 479/5920 [05:54<45:44, 1.98it/s, tok/s=12257.0]
Inference: 8%|▊ | 479/5920 [05:54<45:44, 1.98it/s, tok/s=12286.9]
Inference: 8%|▊ | 480/5920 [05:55<39:51, 2.27it/s, tok/s=12286.9]
Inference: 8%|▊ | 480/5920 [05:55<39:51, 2.27it/s, tok/s=12322.5]
Inference: 8%|▊ | 481/5920 [05:55<39:51, 2.27it/s, tok/s=12369.3]
Inference: 8%|▊ | 482/5920 [05:55<34:37, 2.62it/s, tok/s=12369.3]
Inference: 8%|▊ | 482/5920 [05:55<34:37, 2.62it/s, tok/s=12363.8]
Inference: 8%|▊ | 483/5920 [05:56<36:21, 2.49it/s, tok/s=12363.8]
Inference: 8%|▊ | 483/5920 [05:56<36:21, 2.49it/s, tok/s=12395.2]
Inference: 8%|▊ | 484/5920 [05:56<31:27, 2.88it/s, tok/s=12395.2]
Inference: 8%|▊ | 484/5920 [05:56<31:27, 2.88it/s, tok/s=12433.3]
Inference: 8%|▊ | 485/5920 [05:56<26:21, 3.44it/s, tok/s=12433.3]
Inference: 8%|▊ | 485/5920 [05:56<26:21, 3.44it/s, tok/s=12473.2]
Inference: 8%|▊ | 486/5920 [05:56<28:04, 3.23it/s, tok/s=12473.2]
Inference: 8%|▊ | 486/5920 [05:56<28:04, 3.23it/s, tok/s=12483.7]
Inference: 8%|▊ | 487/5920 [05:57<37:22, 2.42it/s, tok/s=12483.7]
Inference: 8%|▊ | 487/5920 [05:57<37:22, 2.42it/s, tok/s=12505.5]
Inference: 8%|▊ | 488/5920 [05:58<40:13, 2.25it/s, tok/s=12505.5]
Inference: 8%|▊ | 488/5920 [05:58<40:13, 2.25it/s, tok/s=12520.7]
Inference: 8%|▊ | 489/5920 [05:58<40:13, 2.25it/s, tok/s=12532.7]
Inference: 8%|▊ | 490/5920 [05:58<31:51, 2.84it/s, tok/s=12532.7]
Inference: 8%|▊ | 490/5920 [05:58<31:51, 2.84it/s, tok/s=12561.8]
Inference: 8%|▊ | 491/5920 [05:58<28:19, 3.19it/s, tok/s=12561.8]
Inference: 8%|▊ | 491/5920 [05:58<28:19, 3.19it/s, tok/s=12600.0]
Inference: 8%|▊ | 492/5920 [05:59<49:34, 1.82it/s, tok/s=12600.0]
Inference: 8%|▊ | 492/5920 [05:59<49:34, 1.82it/s, tok/s=12602.7]
Inference: 8%|▊ | 493/5920 [06:00<40:30, 2.23it/s, tok/s=12602.7]
Inference: 8%|▊ | 493/5920 [06:00<40:30, 2.23it/s, tok/s=12618.0]
Inference: 8%|▊ | 494/5920 [06:01<56:57, 1.59it/s, tok/s=12618.0]
Inference: 8%|▊ | 494/5920 [06:01<56:57, 1.59it/s, tok/s=12616.9]
Inference: 8%|▊ | 495/5920 [06:01<43:43, 2.07it/s, tok/s=12616.9]
Inference: 8%|▊ | 495/5920 [06:01<43:43, 2.07it/s, tok/s=12626.5]
Inference: 8%|▊ | 496/5920 [06:01<41:26, 2.18it/s, tok/s=12626.5]
Inference: 8%|▊ | 496/5920 [06:01<41:26, 2.18it/s, tok/s=12657.6]
Inference: 8%|▊ | 497/5920 [06:04<1:31:39, 1.01s/it, tok/s=12657.6]
Inference: 8%|▊ | 497/5920 [06:04<1:31:39, 1.01s/it, tok/s=12621.0]
Inference: 8%|▊ | 498/5920 [06:04<1:09:10, 1.31it/s, tok/s=12621.0]
Inference: 8%|▊ | 498/5920 [06:04<1:09:10, 1.31it/s, tok/s=12661.0]
Inference: 8%|▊ | 499/5920 [06:04<57:18, 1.58it/s, tok/s=12661.0]
Inference: 8%|▊ | 499/5920 [06:04<57:18, 1.58it/s, tok/s=12662.0]
Inference: 8%|▊ | 500/5920 [06:05<52:48, 1.71it/s, tok/s=12662.0]
Inference: 8%|▊ | 500/5920 [06:05<52:48, 1.71it/s, tok/s=12691.2]
Inference: 8%|▊ | 501/5920 [06:05<52:48, 1.71it/s, tok/s=12702.4]
Inference: 8%|▊ | 502/5920 [06:05<37:19, 2.42it/s, tok/s=12702.4]
Inference: 8%|▊ | 502/5920 [06:05<37:19, 2.42it/s, tok/s=12705.3]
Inference: 8%|▊ | 503/5920 [06:05<39:13, 2.30it/s, tok/s=12705.3]
Inference: 8%|▊ | 503/5920 [06:05<39:13, 2.30it/s, tok/s=12705.2]
Inference: 9%|▊ | 504/5920 [06:06<51:01, 1.77it/s, tok/s=12705.2]
Inference: 9%|▊ | 504/5920 [06:06<51:01, 1.77it/s, tok/s=12720.6]
Inference: 9%|▊ | 505/5920 [06:07<47:10, 1.91it/s, tok/s=12720.6]
Inference: 9%|▊ | 505/5920 [06:07<47:10, 1.91it/s, tok/s=12751.0]
Inference: 9%|▊ | 506/5920 [06:08<1:01:08, 1.48it/s, tok/s=12751.0]
Inference: 9%|▊ | 506/5920 [06:08<1:01:08, 1.48it/s, tok/s=12734.7]
Inference: 9%|▊ | 507/5920 [06:08<1:01:08, 1.48it/s, tok/s=12755.0]
Inference: 9%|▊ | 508/5920 [06:08<42:42, 2.11it/s, tok/s=12755.0]
Inference: 9%|▊ | 508/5920 [06:08<42:42, 2.11it/s, tok/s=12752.3]
Inference: 9%|▊ | 509/5920 [06:08<34:54, 2.58it/s, tok/s=12752.3]
Inference: 9%|▊ | 509/5920 [06:08<34:54, 2.58it/s, tok/s=12779.1]
Inference: 9%|▊ | 510/5920 [06:09<33:19, 2.71it/s, tok/s=12779.1]
Inference: 9%|▊ | 510/5920 [06:09<33:19, 2.71it/s, tok/s=12812.4]
Inference: 9%|▊ | 511/5920 [06:10<1:05:07, 1.38it/s, tok/s=12812.4]
Inference: 9%|▊ | 511/5920 [06:10<1:05:07, 1.38it/s, tok/s=12798.2]
Inference: 9%|▊ | 512/5920 [06:11<50:42, 1.78it/s, tok/s=12798.2]
Inference: 9%|▊ | 512/5920 [06:11<50:42, 1.78it/s, tok/s=12837.7]
Inference: 9%|▊ | 513/5920 [06:11<42:18, 2.13it/s, tok/s=12837.7]
Inference: 9%|▊ | 513/5920 [06:11<42:18, 2.13it/s, tok/s=12843.1]
Inference: 9%|▊ | 514/5920 [06:11<33:17, 2.71it/s, tok/s=12843.1]
Inference: 9%|▊ | 514/5920 [06:11<33:17, 2.71it/s, tok/s=12883.8]
Inference: 9%|▊ | 515/5920 [06:13<1:30:32, 1.01s/it, tok/s=12883.8]
Inference: 9%|▊ | 515/5920 [06:13<1:30:32, 1.01s/it, tok/s=12840.5]
Inference: 9%|▊ | 516/5920 [06:14<1:09:39, 1.29it/s, tok/s=12840.5]
Inference: 9%|▊ | 516/5920 [06:14<1:09:39, 1.29it/s, tok/s=12878.1]
Inference: 9%|▊ | 517/5920 [06:15<1:11:09, 1.27it/s, tok/s=12878.1]
Inference: 9%|▊ | 517/5920 [06:15<1:11:09, 1.27it/s, tok/s=12894.8]
Inference: 9%|▉ | 518/5920 [06:15<1:01:10, 1.47it/s, tok/s=12894.8]
Inference: 9%|▉ | 518/5920 [06:15<1:01:10, 1.47it/s, tok/s=12928.3]
Inference: 9%|▉ | 519/5920 [06:15<1:01:09, 1.47it/s, tok/s=12971.0]
Inference: 9%|▉ | 520/5920 [06:16<51:46, 1.74it/s, tok/s=12971.0]
Inference: 9%|▉ | 520/5920 [06:16<51:46, 1.74it/s, tok/s=12985.8]
Inference: 9%|▉ | 521/5920 [06:16<49:29, 1.82it/s, tok/s=12985.8]
Inference: 9%|▉ | 521/5920 [06:16<49:29, 1.82it/s, tok/s=13014.5]
Inference: 9%|▉ | 522/5920 [06:16<39:03, 2.30it/s, tok/s=13014.5]
Inference: 9%|▉ | 522/5920 [06:16<39:03, 2.30it/s, tok/s=13048.7]
Inference: 9%|▉ | 523/5920 [06:17<46:46, 1.92it/s, tok/s=13048.7]
Inference: 9%|▉ | 523/5920 [06:17<46:46, 1.92it/s, tok/s=13066.9]
Inference: 9%|▉ | 524/5920 [06:17<36:28, 2.47it/s, tok/s=13066.9]
Inference: 9%|▉ | 524/5920 [06:17<36:28, 2.47it/s, tok/s=13108.2]
Inference: 9%|▉ | 525/5920 [06:18<34:34, 2.60it/s, tok/s=13108.2]
Inference: 9%|▉ | 525/5920 [06:18<34:34, 2.60it/s, tok/s=13140.0]
Inference: 9%|▉ | 526/5920 [06:19<55:51, 1.61it/s, tok/s=13140.0]
Inference: 9%|▉ | 526/5920 [06:19<55:51, 1.61it/s, tok/s=13143.6]
Inference: 9%|▉ | 527/5920 [06:19<48:31, 1.85it/s, tok/s=13143.6]
Inference: 9%|▉ | 527/5920 [06:19<48:31, 1.85it/s, tok/s=13154.1]
Inference: 9%|▉ | 528/5920 [06:21<1:14:34, 1.20it/s, tok/s=13154.1]
Inference: 9%|▉ | 528/5920 [06:21<1:14:34, 1.20it/s, tok/s=13122.0]
Inference: 9%|▉ | 529/5920 [06:22<1:16:13, 1.18it/s, tok/s=13122.0]
Inference: 9%|▉ | 529/5920 [06:22<1:16:13, 1.18it/s, tok/s=13135.5]
Inference: 9%|▉ | 530/5920 [06:22<58:47, 1.53it/s, tok/s=13135.5]
Inference: 9%|▉ | 530/5920 [06:22<58:47, 1.53it/s, tok/s=13173.8]
Inference: 9%|▉ | 531/5920 [06:23<1:09:33, 1.29it/s, tok/s=13173.8]
Inference: 9%|▉ | 531/5920 [06:23<1:09:33, 1.29it/s, tok/s=13180.8]
Inference: 9%|▉ | 532/5920 [06:24<1:16:57, 1.17it/s, tok/s=13180.8]
Inference: 9%|▉ | 532/5920 [06:24<1:16:57, 1.17it/s, tok/s=13182.8]
Inference: 9%|▉ | 533/5920 [06:24<57:25, 1.56it/s, tok/s=13182.8]
Inference: 9%|▉ | 533/5920 [06:24<57:25, 1.56it/s, tok/s=13222.6]
Inference: 9%|▉ | 534/5920 [06:25<1:02:52, 1.43it/s, tok/s=13222.6]
Inference: 9%|▉ | 534/5920 [06:25<1:02:52, 1.43it/s, tok/s=13237.0]
Inference: 9%|▉ | 535/5920 [06:25<51:54, 1.73it/s, tok/s=13237.0]
Inference: 9%|▉ | 535/5920 [06:25<51:54, 1.73it/s, tok/s=13234.6]
Inference: 9%|▉ | 536/5920 [06:25<51:54, 1.73it/s, tok/s=13258.5]
Inference: 9%|▉ | 537/5920 [06:25<51:53, 1.73it/s, tok/s=13301.2]
Inference: 9%|▉ | 538/5920 [06:25<24:29, 3.66it/s, tok/s=13301.2]
Inference: 9%|▉ | 538/5920 [06:25<24:29, 3.66it/s, tok/s=13344.8]
Inference: 9%|▉ | 539/5920 [06:26<24:29, 3.66it/s, tok/s=13348.1]
Inference: 9%|▉ | 540/5920 [06:27<38:06, 2.35it/s, tok/s=13348.1]
Inference: 9%|▉ | 540/5920 [06:27<38:06, 2.35it/s, tok/s=13324.1]
Inference: 9%|▉ | 541/5920 [06:28<45:13, 1.98it/s, tok/s=13324.1]
Inference: 9%|▉ | 541/5920 [06:28<45:13, 1.98it/s, tok/s=13340.7]
Inference: 9%|▉ | 542/5920 [06:29<56:58, 1.57it/s, tok/s=13340.7]
Inference: 9%|▉ | 542/5920 [06:29<56:58, 1.57it/s, tok/s=13322.8]
Inference: 9%|▉ | 543/5920 [06:29<58:38, 1.53it/s, tok/s=13322.8]
Inference: 9%|▉ | 543/5920 [06:29<58:38, 1.53it/s, tok/s=13343.4]
Inference: 9%|▉ | 544/5920 [06:30<48:35, 1.84it/s, tok/s=13343.4]
Inference: 9%|▉ | 544/5920 [06:30<48:35, 1.84it/s, tok/s=13381.0]
Inference: 9%|▉ | 545/5920 [06:30<57:54, 1.55it/s, tok/s=13381.0]
Inference: 9%|▉ | 545/5920 [06:30<57:54, 1.55it/s, tok/s=13367.5]
Inference: 9%|▉ | 546/5920 [06:31<47:41, 1.88it/s, tok/s=13367.5]
Inference: 9%|▉ | 546/5920 [06:31<47:41, 1.88it/s, tok/s=13367.1]
Inference: 9%|▉ | 547/5920 [06:31<37:41, 2.38it/s, tok/s=13367.1]
Inference: 9%|▉ | 547/5920 [06:31<37:41, 2.38it/s, tok/s=13407.2]
Inference: 9%|▉ | 548/5920 [06:31<35:27, 2.53it/s, tok/s=13407.2]
Inference: 9%|▉ | 548/5920 [06:31<35:27, 2.53it/s, tok/s=13434.6]
Inference: 9%|▉ | 549/5920 [06:31<28:10, 3.18it/s, tok/s=13434.6]
Inference: 9%|▉ | 549/5920 [06:31<28:10, 3.18it/s, tok/s=13438.0]
Inference: 9%|▉ | 550/5920 [06:32<30:37, 2.92it/s, tok/s=13438.0]
Inference: 9%|▉ | 550/5920 [06:32<30:37, 2.92it/s, tok/s=13466.2]
Inference: 9%|▉ | 551/5920 [06:32<26:18, 3.40it/s, tok/s=13466.2]
Inference: 9%|▉ | 551/5920 [06:32<26:18, 3.40it/s, tok/s=13468.5]
Inference: 9%|▉ | 552/5920 [06:32<29:45, 3.01it/s, tok/s=13468.5]
Inference: 9%|▉ | 552/5920 [06:32<29:45, 3.01it/s, tok/s=13495.0]
Inference: 9%|▉ | 553/5920 [06:34<1:06:09, 1.35it/s, tok/s=13495.0]
Inference: 9%|▉ | 553/5920 [06:34<1:06:09, 1.35it/s, tok/s=13445.2]
Inference: 9%|▉ | 554/5920 [06:34<50:46, 1.76it/s, tok/s=13445.2]
Inference: 9%|▉ | 554/5920 [06:34<50:46, 1.76it/s, tok/s=13447.2]
Inference: 9%|▉ | 555/5920 [06:36<1:37:15, 1.09s/it, tok/s=13447.2]
Inference: 9%|▉ | 555/5920 [06:36<1:37:15, 1.09s/it, tok/s=13413.7]
Inference: 9%|▉ | 556/5920 [06:37<1:13:51, 1.21it/s, tok/s=13413.7]
Inference: 9%|▉ | 556/5920 [06:37<1:13:51, 1.21it/s, tok/s=13413.9]
Inference: 9%|▉ | 557/5920 [06:37<59:23, 1.50it/s, tok/s=13413.9]
Inference: 9%|▉ | 557/5920 [06:37<59:23, 1.50it/s, tok/s=13449.4]
Inference: 9%|▉ | 558/5920 [06:38<58:18, 1.53it/s, tok/s=13449.4]
Inference: 9%|▉ | 558/5920 [06:38<58:18, 1.53it/s, tok/s=13436.6]
Inference: 9%|▉ | 559/5920 [06:38<58:17, 1.53it/s, tok/s=13450.2]
Inference: 9%|▉ | 560/5920 [06:38<33:43, 2.65it/s, tok/s=13450.2]
Inference: 9%|▉ | 560/5920 [06:38<33:43, 2.65it/s, tok/s=13457.1]
Inference: 9%|▉ | 561/5920 [06:38<36:03, 2.48it/s, tok/s=13457.1]
Inference: 9%|▉ | 561/5920 [06:38<36:03, 2.48it/s, tok/s=13460.0]
Inference: 9%|▉ | 562/5920 [06:38<31:36, 2.82it/s, tok/s=13460.0]
Inference: 9%|▉ | 562/5920 [06:38<31:36, 2.82it/s, tok/s=13497.7]
Inference: 10%|▉ | 563/5920 [06:39<38:34, 2.31it/s, tok/s=13497.7]
Inference: 10%|▉ | 563/5920 [06:39<38:34, 2.31it/s, tok/s=13516.4]
Inference: 10%|▉ | 564/5920 [06:40<45:00, 1.98it/s, tok/s=13516.4]
Inference: 10%|▉ | 564/5920 [06:40<45:00, 1.98it/s, tok/s=13503.1]
Inference: 10%|▉ | 565/5920 [06:40<43:30, 2.05it/s, tok/s=13503.1]
Inference: 10%|▉ | 565/5920 [06:40<43:30, 2.05it/s, tok/s=13501.6]
Inference: 10%|▉ | 566/5920 [06:41<1:01:39, 1.45it/s, tok/s=13501.6]
Inference: 10%|▉ | 566/5920 [06:41<1:01:39, 1.45it/s, tok/s=13506.5]
Inference: 10%|▉ | 567/5920 [06:42<48:01, 1.86it/s, tok/s=13506.5]
Inference: 10%|▉ | 567/5920 [06:42<48:01, 1.86it/s, tok/s=13516.1]
Inference: 10%|▉ | 568/5920 [06:42<37:18, 2.39it/s, tok/s=13516.1]
Inference: 10%|▉ | 568/5920 [06:42<37:18, 2.39it/s, tok/s=13532.5]
Inference: 10%|▉ | 569/5920 [06:42<37:18, 2.39it/s, tok/s=13575.0]
Inference: 10%|▉ | 570/5920 [06:42<26:24, 3.38it/s, tok/s=13575.0]
Inference: 10%|▉ | 570/5920 [06:42<26:24, 3.38it/s, tok/s=13610.3]
Inference: 10%|▉ | 571/5920 [06:42<22:31, 3.96it/s, tok/s=13610.3]
Inference: 10%|▉ | 571/5920 [06:42<22:31, 3.96it/s, tok/s=13650.8]
Inference: 10%|▉ | 572/5920 [06:42<20:28, 4.35it/s, tok/s=13650.8]
Inference: 10%|▉ | 572/5920 [06:42<20:28, 4.35it/s, tok/s=13653.5]
Inference: 10%|▉ | 573/5920 [06:43<42:26, 2.10it/s, tok/s=13653.5]
Inference: 10%|▉ | 573/5920 [06:43<42:26, 2.10it/s, tok/s=13659.2]
Inference: 10%|▉ | 574/5920 [06:44<43:24, 2.05it/s, tok/s=13659.2]
Inference: 10%|▉ | 574/5920 [06:44<43:24, 2.05it/s, tok/s=13650.2]
Inference: 10%|▉ | 575/5920 [06:45<46:39, 1.91it/s, tok/s=13650.2]
Inference: 10%|▉ | 575/5920 [06:45<46:39, 1.91it/s, tok/s=13640.3]
Inference: 10%|▉ | 576/5920 [06:45<40:49, 2.18it/s, tok/s=13640.3]
Inference: 10%|▉ | 576/5920 [06:45<40:49, 2.18it/s, tok/s=13644.6]
Inference: 10%|▉ | 577/5920 [06:45<40:33, 2.20it/s, tok/s=13644.6]
Inference: 10%|▉ | 577/5920 [06:45<40:33, 2.20it/s, tok/s=13674.6]
Inference: 10%|▉ | 578/5920 [06:46<37:14, 2.39it/s, tok/s=13674.6]
Inference: 10%|▉ | 578/5920 [06:46<37:14, 2.39it/s, tok/s=13672.6]
Inference: 10%|▉ | 579/5920 [06:46<49:07, 1.81it/s, tok/s=13672.6]
Inference: 10%|▉ | 579/5920 [06:46<49:07, 1.81it/s, tok/s=13687.8]
Inference: 10%|▉ | 580/5920 [06:47<49:06, 1.81it/s, tok/s=13706.3]
Inference: 10%|▉ | 581/5920 [06:47<28:54, 3.08it/s, tok/s=13706.3]
Inference: 10%|▉ | 581/5920 [06:47<28:54, 3.08it/s, tok/s=13712.9]
Inference: 10%|▉ | 582/5920 [06:47<30:06, 2.95it/s, tok/s=13712.9]
Inference: 10%|▉ | 582/5920 [06:47<30:06, 2.95it/s, tok/s=13718.0]
Inference: 10%|▉ | 583/5920 [06:47<27:03, 3.29it/s, tok/s=13718.0]
Inference: 10%|▉ | 583/5920 [06:47<27:03, 3.29it/s, tok/s=13718.8]
Inference: 10%|▉ | 584/5920 [06:48<27:45, 3.20it/s, tok/s=13718.8]
Inference: 10%|▉ | 584/5920 [06:48<27:45, 3.20it/s, tok/s=13751.3]
Inference: 10%|▉ | 585/5920 [06:49<46:20, 1.92it/s, tok/s=13751.3]
Inference: 10%|▉ | 585/5920 [06:49<46:20, 1.92it/s, tok/s=13726.4]
Inference: 10%|▉ | 586/5920 [06:49<43:49, 2.03it/s, tok/s=13726.4]
Inference: 10%|▉ | 586/5920 [06:49<43:49, 2.03it/s, tok/s=13738.6]
Inference: 10%|▉ | 587/5920 [06:49<43:48, 2.03it/s, tok/s=13778.5]
Inference: 10%|▉ | 588/5920 [06:49<28:29, 3.12it/s, tok/s=13778.5]
Inference: 10%|▉ | 588/5920 [06:49<28:29, 3.12it/s, tok/s=13782.2]
Inference: 10%|▉ | 589/5920 [06:49<23:47, 3.73it/s, tok/s=13782.2]
Inference: 10%|▉ | 589/5920 [06:49<23:47, 3.73it/s, tok/s=13787.9]
Inference: 10%|▉ | 590/5920 [06:51<56:58, 1.56it/s, tok/s=13787.9]
Inference: 10%|▉ | 590/5920 [06:51<56:58, 1.56it/s, tok/s=13775.1]
Inference: 10%|▉ | 591/5920 [06:51<56:57, 1.56it/s, tok/s=13780.6]
Inference: 10%|█ | 592/5920 [06:52<42:45, 2.08it/s, tok/s=13780.6]
Inference: 10%|█ | 592/5920 [06:52<42:45, 2.08it/s, tok/s=13798.8]
Inference: 10%|█ | 593/5920 [06:52<42:58, 2.07it/s, tok/s=13798.8]
Inference: 10%|█ | 593/5920 [06:52<42:58, 2.07it/s, tok/s=13813.0]
Inference: 10%|█ | 594/5920 [06:52<34:48, 2.55it/s, tok/s=13813.0]
Inference: 10%|█ | 594/5920 [06:52<34:48, 2.55it/s, tok/s=13830.8]
Inference: 10%|█ | 595/5920 [06:53<36:13, 2.45it/s, tok/s=13830.8]
Inference: 10%|█ | 595/5920 [06:53<36:13, 2.45it/s, tok/s=13847.8]
Inference: 10%|█ | 596/5920 [06:53<31:51, 2.78it/s, tok/s=13847.8]
Inference: 10%|█ | 596/5920 [06:53<31:51, 2.78it/s, tok/s=13848.9]
Inference: 10%|█ | 597/5920 [06:53<25:25, 3.49it/s, tok/s=13848.9]
Inference: 10%|█ | 597/5920 [06:53<25:25, 3.49it/s, tok/s=13854.4]
Inference: 10%|█ | 598/5920 [06:53<22:07, 4.01it/s, tok/s=13854.4]
Inference: 10%|█ | 598/5920 [06:53<22:07, 4.01it/s, tok/s=13893.7]
Inference: 10%|█ | 599/5920 [06:53<22:04, 4.02it/s, tok/s=13893.7]
Inference: 10%|█ | 599/5920 [06:53<22:04, 4.02it/s, tok/s=13930.9]
Inference: 10%|█ | 600/5920 [06:54<39:22, 2.25it/s, tok/s=13930.9]
Inference: 10%|█ | 600/5920 [06:54<39:22, 2.25it/s, tok/s=13908.3]
Inference: 10%|█ | 601/5920 [06:56<1:02:43, 1.41it/s, tok/s=13908.3]
Inference: 10%|█ | 601/5920 [06:56<1:02:43, 1.41it/s, tok/s=13902.0]
Inference: 10%|█ | 602/5920 [06:56<51:49, 1.71it/s, tok/s=13902.0]
Inference: 10%|█ | 602/5920 [06:56<51:49, 1.71it/s, tok/s=13936.9]
Inference: 10%|█ | 603/5920 [06:56<51:49, 1.71it/s, tok/s=13952.7]
Inference: 10%|█ | 604/5920 [06:56<36:27, 2.43it/s, tok/s=13952.7]
Inference: 10%|█ | 604/5920 [06:56<36:27, 2.43it/s, tok/s=13951.6]
Inference: 10%|█ | 605/5920 [06:57<33:58, 2.61it/s, tok/s=13951.6]
Inference: 10%|█ | 605/5920 [06:57<33:58, 2.61it/s, tok/s=13968.1]
Inference: 10%|█ | 606/5920 [06:58<1:03:59, 1.38it/s, tok/s=13968.1]
Inference: 10%|█ | 606/5920 [06:58<1:03:59, 1.38it/s, tok/s=13923.3]
Inference: 10%|█ | 607/5920 [06:59<1:06:52, 1.32it/s, tok/s=13923.3]
Inference: 10%|█ | 607/5920 [06:59<1:06:52, 1.32it/s, tok/s=13904.6]
Inference: 10%|█ | 608/5920 [06:59<51:29, 1.72it/s, tok/s=13904.6]
Inference: 10%|█ | 608/5920 [06:59<51:29, 1.72it/s, tok/s=13909.1]
Inference: 10%|█ | 609/5920 [07:00<45:07, 1.96it/s, tok/s=13909.1]
Inference: 10%|█ | 609/5920 [07:00<45:07, 1.96it/s, tok/s=13913.4]
Inference: 10%|█ | 610/5920 [07:00<45:06, 1.96it/s, tok/s=13958.0]
Inference: 10%|█ | 611/5920 [07:01<47:24, 1.87it/s, tok/s=13958.0]
Inference: 10%|█ | 611/5920 [07:01<47:24, 1.87it/s, tok/s=13964.4]
Inference: 10%|█ | 612/5920 [07:04<1:49:11, 1.23s/it, tok/s=13964.4]
Inference: 10%|█ | 612/5920 [07:04<1:49:11, 1.23s/it, tok/s=13883.4]
Inference: 10%|█ | 613/5920 [07:05<1:40:31, 1.14s/it, tok/s=13883.4]
Inference: 10%|█ | 613/5920 [07:05<1:40:31, 1.14s/it, tok/s=13890.1]
Inference: 10%|█ | 614/5920 [07:06<1:26:03, 1.03it/s, tok/s=13890.1]
Inference: 10%|█ | 614/5920 [07:06<1:26:03, 1.03it/s, tok/s=13888.2]
Inference: 10%|█ | 615/5920 [07:07<1:25:59, 1.03it/s, tok/s=13888.2]
Inference: 10%|█ | 615/5920 [07:07<1:25:59, 1.03it/s, tok/s=13872.7]
Inference: 10%|█ | 616/5920 [07:07<1:17:01, 1.15it/s, tok/s=13872.7]
Inference: 10%|█ | 616/5920 [07:07<1:17:01, 1.15it/s, tok/s=13896.2]
Inference: 10%|█ | 617/5920 [07:07<1:02:46, 1.41it/s, tok/s=13896.2]
Inference: 10%|█ | 617/5920 [07:07<1:02:46, 1.41it/s, tok/s=13929.6]
Inference: 10%|█ | 618/5920 [07:08<48:08, 1.84it/s, tok/s=13929.6]
Inference: 10%|█ | 618/5920 [07:08<48:08, 1.84it/s, tok/s=13934.3]
Inference: 10%|█ | 619/5920 [07:09<1:07:16, 1.31it/s, tok/s=13934.3]
Inference: 10%|█ | 619/5920 [07:09<1:07:16, 1.31it/s, tok/s=13902.8]
Inference: 10%|█ | 620/5920 [07:09<58:15, 1.52it/s, tok/s=13902.8]
Inference: 10%|█ | 620/5920 [07:09<58:15, 1.52it/s, tok/s=13936.9]
Inference: 10%|█ | 621/5920 [07:10<1:00:16, 1.47it/s, tok/s=13936.9]
Inference: 10%|█ | 621/5920 [07:10<1:00:16, 1.47it/s, tok/s=13949.9]
Inference: 11%|█ | 622/5920 [07:12<1:34:01, 1.06s/it, tok/s=13949.9]
Inference: 11%|█ | 622/5920 [07:12<1:34:01, 1.06s/it, tok/s=13923.7]
Inference: 11%|█ | 623/5920 [07:13<1:21:00, 1.09it/s, tok/s=13923.7]
Inference: 11%|█ | 623/5920 [07:13<1:21:00, 1.09it/s, tok/s=13948.1]
Inference: 11%|█ | 624/5920 [07:13<1:08:19, 1.29it/s, tok/s=13948.1]
Inference: 11%|█ | 624/5920 [07:13<1:08:19, 1.29it/s, tok/s=13977.8]
Inference: 11%|█ | 625/5920 [07:14<1:02:17, 1.42it/s, tok/s=13977.8]
Inference: 11%|█ | 625/5920 [07:14<1:02:17, 1.42it/s, tok/s=13994.6]
Inference: 11%|█ | 626/5920 [07:14<1:07:25, 1.31it/s, tok/s=13994.6]
Inference: 11%|█ | 626/5920 [07:14<1:07:25, 1.31it/s, tok/s=14005.8]
Inference: 11%|█ | 627/5920 [07:15<1:07:59, 1.30it/s, tok/s=14005.8]
Inference: 11%|█ | 627/5920 [07:15<1:07:59, 1.30it/s, tok/s=14025.6]
Inference: 11%|█ | 628/5920 [07:16<1:11:40, 1.23it/s, tok/s=14025.6]
Inference: 11%|█ | 628/5920 [07:16<1:11:40, 1.23it/s, tok/s=14013.3]
Inference: 11%|█ | 629/5920 [07:17<1:07:45, 1.30it/s, tok/s=14013.3]
Inference: 11%|█ | 629/5920 [07:17<1:07:45, 1.30it/s, tok/s=14008.5]
Inference: 11%|█ | 630/5920 [07:18<1:10:48, 1.25it/s, tok/s=14008.5]
Inference: 11%|█ | 630/5920 [07:18<1:10:48, 1.25it/s, tok/s=13999.9]
Inference: 11%|█ | 631/5920 [07:18<1:04:32, 1.37it/s, tok/s=13999.9]
Inference: 11%|█ | 631/5920 [07:18<1:04:32, 1.37it/s, tok/s=13991.7]
Inference: 11%|█ | 632/5920 [07:18<1:04:31, 1.37it/s, tok/s=14033.2]
Inference: 11%|█ | 633/5920 [07:20<1:05:32, 1.34it/s, tok/s=14033.2]
Inference: 11%|█ | 633/5920 [07:20<1:05:32, 1.34it/s, tok/s=14021.2]
Inference: 11%|█ | 634/5920 [07:20<52:42, 1.67it/s, tok/s=14021.2]
Inference: 11%|█ | 634/5920 [07:20<52:42, 1.67it/s, tok/s=14029.2]
Inference: 11%|█ | 635/5920 [07:20<48:58, 1.80it/s, tok/s=14029.2]
Inference: 11%|█ | 635/5920 [07:20<48:58, 1.80it/s, tok/s=14059.2]
Inference: 11%|█ | 636/5920 [07:21<55:46, 1.58it/s, tok/s=14059.2]
Inference: 11%|█ | 636/5920 [07:21<55:46, 1.58it/s, tok/s=14067.5]
Inference: 11%|█ | 637/5920 [07:22<50:33, 1.74it/s, tok/s=14067.5]
Inference: 11%|█ | 637/5920 [07:22<50:33, 1.74it/s, tok/s=14101.3]
Inference: 11%|█ | 638/5920 [07:22<51:06, 1.72it/s, tok/s=14101.3]
Inference: 11%|█ | 638/5920 [07:22<51:06, 1.72it/s, tok/s=14093.5]
Inference: 11%|█ | 639/5920 [07:23<52:10, 1.69it/s, tok/s=14093.5]
Inference: 11%|█ | 639/5920 [07:23<52:10, 1.69it/s, tok/s=14102.1]
Inference: 11%|█ | 640/5920 [07:24<1:16:09, 1.16it/s, tok/s=14102.1]
Inference: 11%|█ | 640/5920 [07:24<1:16:09, 1.16it/s, tok/s=14096.3]
Inference: 11%|█ | 641/5920 [07:25<1:01:20, 1.43it/s, tok/s=14096.3]
Inference: 11%|█ | 641/5920 [07:25<1:01:20, 1.43it/s, tok/s=14120.6]
Inference: 11%|█ | 642/5920 [07:25<1:01:19, 1.43it/s, tok/s=14163.0]
Inference: 11%|█ | 643/5920 [07:25<40:28, 2.17it/s, tok/s=14163.0]
Inference: 11%|█ | 643/5920 [07:25<40:28, 2.17it/s, tok/s=14196.9]
Inference: 11%|█ | 644/5920 [07:25<34:50, 2.52it/s, tok/s=14196.9]
Inference: 11%|█ | 644/5920 [07:25<34:50, 2.52it/s, tok/s=14200.6]
Inference: 11%|█ | 645/5920 [07:26<42:17, 2.08it/s, tok/s=14200.6]
Inference: 11%|█ | 645/5920 [07:26<42:17, 2.08it/s, tok/s=14220.6]
Inference: 11%|█ | 646/5920 [07:26<42:17, 2.08it/s, tok/s=14263.4]
Inference: 11%|█ | 647/5920 [07:28<1:08:05, 1.29it/s, tok/s=14263.4]
Inference: 11%|█ | 647/5920 [07:28<1:08:05, 1.29it/s, tok/s=14233.7]
Inference: 11%|█ | 648/5920 [07:28<1:08:04, 1.29it/s, tok/s=14274.5]
Inference: 11%|█ | 649/5920 [07:29<58:34, 1.50it/s, tok/s=14274.5]
Inference: 11%|█ | 649/5920 [07:29<58:34, 1.50it/s, tok/s=14293.8]
Inference: 11%|█ | 650/5920 [07:30<1:01:28, 1.43it/s, tok/s=14293.8]
Inference: 11%|█ | 650/5920 [07:30<1:01:28, 1.43it/s, tok/s=14311.3]
Inference: 11%|█ | 651/5920 [07:30<54:50, 1.60it/s, tok/s=14311.3]
Inference: 11%|█ | 651/5920 [07:30<54:50, 1.60it/s, tok/s=14343.1]
Inference: 11%|█ | 652/5920 [07:31<1:01:32, 1.43it/s, tok/s=14343.1]
Inference: 11%|█ | 652/5920 [07:31<1:01:32, 1.43it/s, tok/s=14357.0]
Inference: 11%|█ | 653/5920 [07:32<55:31, 1.58it/s, tok/s=14357.0]
Inference: 11%|█ | 653/5920 [07:32<55:31, 1.58it/s, tok/s=14376.1]
Inference: 11%|█ | 654/5920 [07:32<44:18, 1.98it/s, tok/s=14376.1]
Inference: 11%|█ | 654/5920 [07:32<44:18, 1.98it/s, tok/s=14386.3]
Inference: 11%|█ | 655/5920 [07:32<43:28, 2.02it/s, tok/s=14386.3]
Inference: 11%|█ | 655/5920 [07:32<43:28, 2.02it/s, tok/s=14408.0]
Inference: 11%|█ | 656/5920 [07:33<38:26, 2.28it/s, tok/s=14408.0]
Inference: 11%|█ | 656/5920 [07:33<38:26, 2.28it/s, tok/s=14417.2]
Inference: 11%|█ | 657/5920 [07:33<32:15, 2.72it/s, tok/s=14417.2]
Inference: 11%|█ | 657/5920 [07:33<32:15, 2.72it/s, tok/s=14447.0]
Inference: 11%|█ | 658/5920 [07:34<37:23, 2.35it/s, tok/s=14447.0]
Inference: 11%|█ | 658/5920 [07:34<37:23, 2.35it/s, tok/s=14441.7]
Inference: 11%|█ | 659/5920 [07:35<1:03:18, 1.39it/s, tok/s=14441.7]
Inference: 11%|█ | 659/5920 [07:35<1:03:18, 1.39it/s, tok/s=14428.9]
Inference: 11%|█ | 660/5920 [07:35<51:35, 1.70it/s, tok/s=14428.9]
Inference: 11%|█ | 660/5920 [07:35<51:35, 1.70it/s, tok/s=14431.5]
Inference: 11%|█ | 661/5920 [07:37<1:14:56, 1.17it/s, tok/s=14431.5]
Inference: 11%|█ | 661/5920 [07:37<1:14:56, 1.17it/s, tok/s=14420.2]
Inference: 11%|█ | 662/5920 [07:37<56:02, 1.56it/s, tok/s=14420.2]
Inference: 11%|█ | 662/5920 [07:37<56:02, 1.56it/s, tok/s=14445.4]
Inference: 11%|█ | 663/5920 [07:38<1:05:28, 1.34it/s, tok/s=14445.4]
Inference: 11%|█ | 663/5920 [07:38<1:05:28, 1.34it/s, tok/s=14456.6]
Inference: 11%|█ | 664/5920 [07:38<48:49, 1.79it/s, tok/s=14456.6]
Inference: 11%|█ | 664/5920 [07:38<48:49, 1.79it/s, tok/s=14474.0]
Inference: 11%|█ | 665/5920 [07:39<51:38, 1.70it/s, tok/s=14474.0]
Inference: 11%|█ | 665/5920 [07:39<51:38, 1.70it/s, tok/s=14475.0]
Inference: 11%|█▏ | 666/5920 [07:39<51:38, 1.70it/s, tok/s=14489.2]
Inference: 11%|█▏ | 667/5920 [07:39<29:58, 2.92it/s, tok/s=14489.2]
Inference: 11%|█▏ | 667/5920 [07:39<29:58, 2.92it/s, tok/s=14512.6]
Inference: 11%|█▏ | 668/5920 [07:40<50:38, 1.73it/s, tok/s=14512.6]
Inference: 11%|█▏ | 668/5920 [07:40<50:38, 1.73it/s, tok/s=14493.7]
Inference: 11%|█▏ | 669/5920 [07:41<1:02:28, 1.40it/s, tok/s=14493.7]
Inference: 11%|█▏ | 669/5920 [07:41<1:02:28, 1.40it/s, tok/s=14479.2]
Inference: 11%|█▏ | 670/5920 [07:41<53:16, 1.64it/s, tok/s=14479.2]
Inference: 11%|█▏ | 670/5920 [07:41<53:16, 1.64it/s, tok/s=14504.1]
Inference: 11%|█▏ | 671/5920 [07:41<53:15, 1.64it/s, tok/s=14524.8]
Inference: 11%|█▏ | 672/5920 [07:42<34:16, 2.55it/s, tok/s=14524.8]
Inference: 11%|█▏ | 672/5920 [07:42<34:16, 2.55it/s, tok/s=14551.3]
Inference: 11%|█▏ | 673/5920 [07:44<1:11:25, 1.22it/s, tok/s=14551.3]
Inference: 11%|█▏ | 673/5920 [07:44<1:11:25, 1.22it/s, tok/s=14525.6]
Inference: 11%|█▏ | 674/5920 [07:44<59:28, 1.47it/s, tok/s=14525.6]
Inference: 11%|█▏ | 674/5920 [07:44<59:28, 1.47it/s, tok/s=14542.1]
Inference: 11%|█▏ | 675/5920 [07:44<59:27, 1.47it/s, tok/s=14559.8]
Inference: 11%|█▏ | 676/5920 [07:45<46:31, 1.88it/s, tok/s=14559.8]
Inference: 11%|█▏ | 676/5920 [07:45<46:31, 1.88it/s, tok/s=14573.0]
Inference: 11%|█▏ | 677/5920 [07:47<1:18:23, 1.11it/s, tok/s=14573.0]
Inference: 11%|█▏ | 677/5920 [07:47<1:18:23, 1.11it/s, tok/s=14525.5]
Inference: 11%|█▏ | 678/5920 [07:48<1:27:13, 1.00it/s, tok/s=14525.5]
Inference: 11%|█▏ | 678/5920 [07:48<1:27:13, 1.00it/s, tok/s=14528.1]
Inference: 11%|█▏ | 679/5920 [07:49<1:15:41, 1.15it/s, tok/s=14528.1]
Inference: 11%|█▏ | 679/5920 [07:49<1:15:41, 1.15it/s, tok/s=14532.9]
Inference: 11%|█▏ | 680/5920 [07:51<1:44:12, 1.19s/it, tok/s=14532.9]
Inference: 11%|█▏ | 680/5920 [07:51<1:44:12, 1.19s/it, tok/s=14505.2]
Inference: 12%|█▏ | 681/5920 [07:51<1:22:11, 1.06it/s, tok/s=14505.2]
Inference: 12%|█▏ | 681/5920 [07:51<1:22:11, 1.06it/s, tok/s=14512.8]
Inference: 12%|█▏ | 682/5920 [07:53<1:59:34, 1.37s/it, tok/s=14512.8]
Inference: 12%|█▏ | 682/5920 [07:53<1:59:34, 1.37s/it, tok/s=14480.5]
Inference: 12%|█▏ | 683/5920 [07:55<1:53:50, 1.30s/it, tok/s=14480.5]
Inference: 12%|█▏ | 683/5920 [07:55<1:53:50, 1.30s/it, tok/s=14479.8]
Inference: 12%|█▏ | 684/5920 [07:55<1:53:48, 1.30s/it, tok/s=14497.5]
Inference: 12%|█▏ | 685/5920 [07:55<1:06:11, 1.32it/s, tok/s=14497.5]
Inference: 12%|█▏ | 685/5920 [07:55<1:06:11, 1.32it/s, tok/s=14535.3]
Inference: 12%|█▏ | 686/5920 [07:55<1:06:11, 1.32it/s, tok/s=14553.1]
Inference: 12%|█▏ | 687/5920 [07:55<47:24, 1.84it/s, tok/s=14553.1]
Inference: 12%|█▏ | 687/5920 [07:55<47:24, 1.84it/s, tok/s=14567.1]
Inference: 12%|█▏ | 688/5920 [07:55<41:25, 2.11it/s, tok/s=14567.1]
Inference: 12%|█▏ | 688/5920 [07:55<41:25, 2.11it/s, tok/s=14598.9]
Inference: 12%|█▏ | 689/5920 [07:56<47:43, 1.83it/s, tok/s=14598.9]
Inference: 12%|█▏ | 689/5920 [07:56<47:43, 1.83it/s, tok/s=14613.0]
Inference: 12%|█▏ | 690/5920 [07:57<1:02:25, 1.40it/s, tok/s=14613.0]
Inference: 12%|█▏ | 690/5920 [07:57<1:02:25, 1.40it/s, tok/s=14608.1]
Inference: 12%|█▏ | 691/5920 [07:58<1:05:52, 1.32it/s, tok/s=14608.1]
Inference: 12%|█▏ | 691/5920 [07:58<1:05:52, 1.32it/s, tok/s=14600.5]
Inference: 12%|█▏ | 692/5920 [08:01<1:43:06, 1.18s/it, tok/s=14600.5]
Inference: 12%|█▏ | 692/5920 [08:01<1:43:06, 1.18s/it, tok/s=14577.6]
Inference: 12%|█▏ | 693/5920 [08:01<1:32:53, 1.07s/it, tok/s=14577.6]
Inference: 12%|█▏ | 693/5920 [08:01<1:32:53, 1.07s/it, tok/s=14578.6]
Inference: 12%|█▏ | 694/5920 [08:03<1:49:41, 1.26s/it, tok/s=14578.6]
Inference: 12%|█▏ | 694/5920 [08:03<1:49:41, 1.26s/it, tok/s=14560.3]
Inference: 12%|█▏ | 695/5920 [08:04<1:26:28, 1.01it/s, tok/s=14560.3]
Inference: 12%|█▏ | 695/5920 [08:04<1:26:28, 1.01it/s, tok/s=14577.6]
Inference: 12%|█▏ | 696/5920 [08:04<1:16:30, 1.14it/s, tok/s=14577.6]
Inference: 12%|█▏ | 696/5920 [08:04<1:16:30, 1.14it/s, tok/s=14580.7]
Inference: 12%|█▏ | 697/5920 [08:06<1:48:41, 1.25s/it, tok/s=14580.7]
Inference: 12%|█▏ | 697/5920 [08:06<1:48:41, 1.25s/it, tok/s=14536.8]
Inference: 12%|█▏ | 698/5920 [08:07<1:29:21, 1.03s/it, tok/s=14536.8]
Inference: 12%|█▏ | 698/5920 [08:07<1:29:21, 1.03s/it, tok/s=14548.0]
Inference: 12%|█▏ | 699/5920 [08:09<2:06:07, 1.45s/it, tok/s=14548.0]
Inference: 12%|█▏ | 699/5920 [08:09<2:06:07, 1.45s/it, tok/s=14507.4]
Inference: 12%|█▏ | 700/5920 [08:11<2:09:10, 1.48s/it, tok/s=14507.4]
Inference: 12%|█▏ | 700/5920 [08:11<2:09:10, 1.48s/it, tok/s=14504.9]
Inference: 12%|█▏ | 701/5920 [08:11<1:34:02, 1.08s/it, tok/s=14504.9]
Inference: 12%|█▏ | 701/5920 [08:11<1:34:02, 1.08s/it, tok/s=14524.5]
Inference: 12%|█▏ | 702/5920 [08:13<2:01:51, 1.40s/it, tok/s=14524.5]
Inference: 12%|█▏ | 702/5920 [08:13<2:01:51, 1.40s/it, tok/s=14482.3]
Inference: 12%|█▏ | 703/5920 [08:13<1:35:59, 1.10s/it, tok/s=14482.3]
Inference: 12%|█▏ | 703/5920 [08:13<1:35:59, 1.10s/it, tok/s=14513.5]
Inference: 12%|█▏ | 704/5920 [08:15<1:52:45, 1.30s/it, tok/s=14513.5]
Inference: 12%|█▏ | 704/5920 [08:15<1:52:45, 1.30s/it, tok/s=14505.1]
Inference: 12%|█▏ | 705/5920 [08:19<2:51:15, 1.97s/it, tok/s=14505.1]
Inference: 12%|█▏ | 705/5920 [08:19<2:51:15, 1.97s/it, tok/s=14449.1]
Inference: 12%|█▏ | 706/5920 [08:19<2:03:09, 1.42s/it, tok/s=14449.1]
Inference: 12%|█▏ | 706/5920 [08:19<2:03:09, 1.42s/it, tok/s=14470.2]
Inference: 12%|█▏ | 707/5920 [08:20<1:50:47, 1.28s/it, tok/s=14470.2]
Inference: 12%|█▏ | 707/5920 [08:20<1:50:47, 1.28s/it, tok/s=14487.0]
Inference: 12%|█▏ | 708/5920 [08:22<2:04:42, 1.44s/it, tok/s=14487.0]
Inference: 12%|█▏ | 708/5920 [08:22<2:04:42, 1.44s/it, tok/s=14478.0]
Inference: 12%|█▏ | 709/5920 [08:22<1:30:40, 1.04s/it, tok/s=14478.0]
Inference: 12%|█▏ | 709/5920 [08:22<1:30:40, 1.04s/it, tok/s=14498.4]
Inference: 12%|█▏ | 710/5920 [08:22<1:22:54, 1.05it/s, tok/s=14498.4]
Inference: 12%|█▏ | 710/5920 [08:22<1:22:54, 1.05it/s, tok/s=14498.6]
Inference: 12%|█▏ | 711/5920 [08:25<1:59:31, 1.38s/it, tok/s=14498.6]
Inference: 12%|█▏ | 711/5920 [08:25<1:59:31, 1.38s/it, tok/s=14474.6]
Inference: 12%|█▏ | 712/5920 [08:26<2:03:59, 1.43s/it, tok/s=14474.6]
Inference: 12%|█▏ | 712/5920 [08:26<2:03:59, 1.43s/it, tok/s=14459.9]
Inference: 12%|█▏ | 713/5920 [08:27<1:33:47, 1.08s/it, tok/s=14459.9]
Inference: 12%|█▏ | 713/5920 [08:27<1:33:47, 1.08s/it, tok/s=14494.1]
Inference: 12%|█▏ | 714/5920 [08:27<1:08:48, 1.26it/s, tok/s=14494.1]
Inference: 12%|█▏ | 714/5920 [08:27<1:08:48, 1.26it/s, tok/s=14533.7]
Inference: 12%|█▏ | 715/5920 [08:27<1:01:41, 1.41it/s, tok/s=14533.7]
Inference: 12%|█▏ | 715/5920 [08:27<1:01:41, 1.41it/s, tok/s=14539.1]
Inference: 12%|█▏ | 716/5920 [08:32<2:52:44, 1.99s/it, tok/s=14539.1]
Inference: 12%|█▏ | 716/5920 [08:32<2:52:44, 1.99s/it, tok/s=14432.4]
Inference: 12%|█▏ | 717/5920 [08:33<2:16:36, 1.58s/it, tok/s=14432.4]
Inference: 12%|█▏ | 717/5920 [08:33<2:16:36, 1.58s/it, tok/s=14457.4]
Inference: 12%|█▏ | 718/5920 [08:33<1:49:02, 1.26s/it, tok/s=14457.4]
Inference: 12%|█▏ | 718/5920 [08:33<1:49:02, 1.26s/it, tok/s=14481.3]
Inference: 12%|█▏ | 719/5920 [08:37<2:50:54, 1.97s/it, tok/s=14481.3]
Inference: 12%|█▏ | 719/5920 [08:37<2:50:54, 1.97s/it, tok/s=14423.7]
Inference: 12%|█▏ | 720/5920 [08:38<2:14:50, 1.56s/it, tok/s=14423.7]
Inference: 12%|█▏ | 720/5920 [08:38<2:14:50, 1.56s/it, tok/s=14438.8]
Inference: 12%|█▏ | 721/5920 [08:39<2:12:39, 1.53s/it, tok/s=14438.8]
Inference: 12%|█▏ | 721/5920 [08:39<2:12:39, 1.53s/it, tok/s=14440.2]
Inference: 12%|█▏ | 722/5920 [08:40<1:44:30, 1.21s/it, tok/s=14440.2]
Inference: 12%|█▏ | 722/5920 [08:40<1:44:30, 1.21s/it, tok/s=14447.8]
Inference: 12%|█▏ | 723/5920 [08:41<1:46:47, 1.23s/it, tok/s=14447.8]
Inference: 12%|█▏ | 723/5920 [08:41<1:46:47, 1.23s/it, tok/s=14444.6]
Inference: 12%|█▏ | 724/5920 [08:42<1:34:28, 1.09s/it, tok/s=14444.6]
Inference: 12%|█▏ | 724/5920 [08:42<1:34:28, 1.09s/it, tok/s=14461.3]
Inference: 12%|█▏ | 725/5920 [08:43<1:36:55, 1.12s/it, tok/s=14461.3]
Inference: 12%|█▏ | 725/5920 [08:43<1:36:55, 1.12s/it, tok/s=14471.1]
Inference: 12%|█▏ | 726/5920 [08:43<1:17:31, 1.12it/s, tok/s=14471.1]
Inference: 12%|█▏ | 726/5920 [08:43<1:17:31, 1.12it/s, tok/s=14485.4]
Inference: 12%|█▏ | 727/5920 [08:43<1:17:31, 1.12it/s, tok/s=14502.8]
Inference: 12%|█▏ | 728/5920 [08:46<1:46:24, 1.23s/it, tok/s=14502.8]
Inference: 12%|█▏ | 728/5920 [08:46<1:46:24, 1.23s/it, tok/s=14420.4]
Inference: 12%|█▏ | 729/5920 [08:47<1:27:14, 1.01s/it, tok/s=14420.4]
Inference: 12%|█▏ | 729/5920 [08:47<1:27:14, 1.01s/it, tok/s=14453.3]
Inference: 12%|█▏ | 730/5920 [08:47<1:20:01, 1.08it/s, tok/s=14453.3]
Inference: 12%|█▏ | 730/5920 [08:47<1:20:01, 1.08it/s, tok/s=14461.9]
Inference: 12%|█▏ | 731/5920 [08:48<1:19:14, 1.09it/s, tok/s=14461.9]
Inference: 12%|█▏ | 731/5920 [08:48<1:19:14, 1.09it/s, tok/s=14479.1]
Inference: 12%|█▏ | 732/5920 [08:49<1:08:12, 1.27it/s, tok/s=14479.1]
Inference: 12%|█▏ | 732/5920 [08:49<1:08:12, 1.27it/s, tok/s=14498.8]
Inference: 12%|█▏ | 733/5920 [08:50<1:08:36, 1.26it/s, tok/s=14498.8]
Inference: 12%|█▏ | 733/5920 [08:50<1:08:36, 1.26it/s, tok/s=14494.7]
Inference: 12%|█▏ | 734/5920 [08:52<1:49:09, 1.26s/it, tok/s=14494.7]
Inference: 12%|█▏ | 734/5920 [08:52<1:49:09, 1.26s/it, tok/s=14455.0]
Inference: 12%|█▏ | 735/5920 [08:53<1:37:43, 1.13s/it, tok/s=14455.0]
Inference: 12%|█▏ | 735/5920 [08:53<1:37:43, 1.13s/it, tok/s=14475.8]
Inference: 12%|█▏ | 736/5920 [08:53<1:19:49, 1.08it/s, tok/s=14475.8]
Inference: 12%|█▏ | 736/5920 [08:53<1:19:49, 1.08it/s, tok/s=14486.5]
Inference: 12%|█▏ | 737/5920 [08:54<1:17:26, 1.12it/s, tok/s=14486.5]
Inference: 12%|█▏ | 737/5920 [08:54<1:17:26, 1.12it/s, tok/s=14507.8]
Inference: 12%|█▏ | 738/5920 [08:54<57:04, 1.51it/s, tok/s=14507.8]
Inference: 12%|█▏ | 738/5920 [08:54<57:04, 1.51it/s, tok/s=14551.5]
Inference: 12%|█▏ | 739/5920 [08:55<1:12:24, 1.19it/s, tok/s=14551.5]
Inference: 12%|█▏ | 739/5920 [08:55<1:12:24, 1.19it/s, tok/s=14554.2]
Inference: 12%|█▎ | 740/5920 [08:56<1:05:59, 1.31it/s, tok/s=14554.2]
Inference: 12%|█▎ | 740/5920 [08:56<1:05:59, 1.31it/s, tok/s=14580.2]
Inference: 13%|█▎ | 741/5920 [08:56<50:59, 1.69it/s, tok/s=14580.2]
Inference: 13%|█▎ | 741/5920 [08:56<50:59, 1.69it/s, tok/s=14595.3]
Inference: 13%|█▎ | 742/5920 [08:58<1:22:10, 1.05it/s, tok/s=14595.3]
Inference: 13%|█▎ | 742/5920 [08:58<1:22:10, 1.05it/s, tok/s=14566.3]
Inference: 13%|█▎ | 743/5920 [09:00<1:57:35, 1.36s/it, tok/s=14566.3]
Inference: 13%|█▎ | 743/5920 [09:00<1:57:35, 1.36s/it, tok/s=14532.7]
Inference: 13%|█▎ | 744/5920 [09:03<2:29:58, 1.74s/it, tok/s=14532.7]
Inference: 13%|█▎ | 744/5920 [09:03<2:29:58, 1.74s/it, tok/s=14482.4]
Inference: 13%|█▎ | 745/5920 [09:04<2:15:57, 1.58s/it, tok/s=14482.4]
Inference: 13%|█▎ | 745/5920 [09:04<2:15:57, 1.58s/it, tok/s=14457.3]
Inference: 13%|█▎ | 746/5920 [09:05<1:53:15, 1.31s/it, tok/s=14457.3]
Inference: 13%|█▎ | 746/5920 [09:05<1:53:15, 1.31s/it, tok/s=14458.5]
Inference: 13%|█▎ | 747/5920 [09:07<2:13:32, 1.55s/it, tok/s=14458.5]
Inference: 13%|█▎ | 747/5920 [09:07<2:13:32, 1.55s/it, tok/s=14432.1]
Inference: 13%|█▎ | 748/5920 [09:07<1:47:08, 1.24s/it, tok/s=14432.1]
Inference: 13%|█▎ | 748/5920 [09:07<1:47:08, 1.24s/it, tok/s=14461.8]
Inference: 13%|█▎ | 749/5920 [09:09<1:59:57, 1.39s/it, tok/s=14461.8]
Inference: 13%|█▎ | 749/5920 [09:09<1:59:57, 1.39s/it, tok/s=14457.3]
Inference: 13%|█▎ | 750/5920 [09:10<1:50:27, 1.28s/it, tok/s=14457.3]
Inference: 13%|█▎ | 750/5920 [09:10<1:50:27, 1.28s/it, tok/s=14470.2]
Inference: 13%|█▎ | 751/5920 [09:11<1:38:34, 1.14s/it, tok/s=14470.2]
Inference: 13%|█▎ | 751/5920 [09:11<1:38:34, 1.14s/it, tok/s=14455.9]
Inference: 13%|█▎ | 752/5920 [09:12<1:21:17, 1.06it/s, tok/s=14455.9]
Inference: 13%|█▎ | 752/5920 [09:12<1:21:17, 1.06it/s, tok/s=14462.2]
Inference: 13%|█▎ | 753/5920 [09:12<1:18:45, 1.09it/s, tok/s=14462.2]
Inference: 13%|█▎ | 753/5920 [09:12<1:18:45, 1.09it/s, tok/s=14459.3]
Inference: 13%|█▎ | 754/5920 [09:13<1:23:34, 1.03it/s, tok/s=14459.3]
Inference: 13%|█▎ | 754/5920 [09:13<1:23:34, 1.03it/s, tok/s=14472.9]
Inference: 13%|█▎ | 755/5920 [09:14<1:03:22, 1.36it/s, tok/s=14472.9]
Inference: 13%|█▎ | 755/5920 [09:14<1:03:22, 1.36it/s, tok/s=14509.1]
Inference: 13%|█▎ | 756/5920 [09:16<1:33:51, 1.09s/it, tok/s=14509.1]
Inference: 13%|█▎ | 756/5920 [09:16<1:33:51, 1.09s/it, tok/s=14496.5]
Inference: 13%|█▎ | 757/5920 [09:18<2:14:13, 1.56s/it, tok/s=14496.5]
Inference: 13%|█▎ | 757/5920 [09:18<2:14:13, 1.56s/it, tok/s=14469.5]
Inference: 13%|█▎ | 758/5920 [09:19<2:04:27, 1.45s/it, tok/s=14469.5]
Inference: 13%|█▎ | 758/5920 [09:19<2:04:27, 1.45s/it, tok/s=14473.2]
Inference: 13%|█▎ | 759/5920 [09:20<1:43:24, 1.20s/it, tok/s=14473.2]
Inference: 13%|█▎ | 759/5920 [09:20<1:43:24, 1.20s/it, tok/s=14477.4]
Inference: 13%|█▎ | 760/5920 [09:24<3:04:30, 2.15s/it, tok/s=14477.4]
Inference: 13%|█▎ | 760/5920 [09:24<3:04:30, 2.15s/it, tok/s=14406.9]
Inference: 13%|█▎ | 761/5920 [09:24<3:04:28, 2.15s/it, tok/s=14424.6]
Inference: 13%|█▎ | 762/5920 [09:25<1:53:55, 1.33s/it, tok/s=14424.6]
Inference: 13%|█▎ | 762/5920 [09:25<1:53:55, 1.33s/it, tok/s=14429.3]
Inference: 13%|█▎ | 763/5920 [09:26<1:40:38, 1.17s/it, tok/s=14429.3]
Inference: 13%|█▎ | 763/5920 [09:26<1:40:38, 1.17s/it, tok/s=14432.6]
Inference: 13%|█▎ | 764/5920 [09:27<1:38:30, 1.15s/it, tok/s=14432.6]
Inference: 13%|█▎ | 764/5920 [09:27<1:38:30, 1.15s/it, tok/s=14425.2]
Inference: 13%|█▎ | 765/5920 [09:29<2:00:22, 1.40s/it, tok/s=14425.2]
Inference: 13%|█▎ | 765/5920 [09:29<2:00:22, 1.40s/it, tok/s=14415.9]
Inference: 13%|█▎ | 766/5920 [09:34<3:16:47, 2.29s/it, tok/s=14415.9]
Inference: 13%|█▎ | 766/5920 [09:34<3:16:47, 2.29s/it, tok/s=14322.2]
Inference: 13%|█▎ | 767/5920 [09:35<2:43:03, 1.90s/it, tok/s=14322.2]
Inference: 13%|█▎ | 767/5920 [09:35<2:43:03, 1.90s/it, tok/s=14337.5]
Inference: 13%|█▎ | 768/5920 [09:35<2:43:01, 1.90s/it, tok/s=14380.5]
Inference: 13%|█▎ | 769/5920 [09:37<2:15:01, 1.57s/it, tok/s=14380.5]
Inference: 13%|█▎ | 769/5920 [09:37<2:15:01, 1.57s/it, tok/s=14351.8]
Inference: 13%|█▎ | 770/5920 [09:38<2:16:16, 1.59s/it, tok/s=14351.8]
Inference: 13%|█▎ | 770/5920 [09:38<2:16:16, 1.59s/it, tok/s=14320.6]
Inference: 13%|█▎ | 771/5920 [09:39<1:58:35, 1.38s/it, tok/s=14320.6]
Inference: 13%|█▎ | 771/5920 [09:39<1:58:35, 1.38s/it, tok/s=14320.4]
Inference: 13%|█▎ | 772/5920 [09:39<1:28:59, 1.04s/it, tok/s=14320.4]
Inference: 13%|█▎ | 772/5920 [09:39<1:28:59, 1.04s/it, tok/s=14355.1]
Inference: 13%|█▎ | 773/5920 [09:40<1:18:16, 1.10it/s, tok/s=14355.1]
Inference: 13%|█▎ | 773/5920 [09:40<1:18:16, 1.10it/s, tok/s=14348.7]
Inference: 13%|█▎ | 774/5920 [09:40<1:18:15, 1.10it/s, tok/s=14354.3]
Inference: 13%|█▎ | 775/5920 [09:42<1:26:58, 1.01s/it, tok/s=14354.3]
Inference: 13%|█▎ | 775/5920 [09:42<1:26:58, 1.01s/it, tok/s=14334.6]
Inference: 13%|█▎ | 776/5920 [09:42<1:08:17, 1.26it/s, tok/s=14334.6]
Inference: 13%|█▎ | 776/5920 [09:42<1:08:17, 1.26it/s, tok/s=14353.2]
Inference: 13%|█▎ | 777/5920 [09:45<1:48:03, 1.26s/it, tok/s=14353.2]
Inference: 13%|█▎ | 777/5920 [09:45<1:48:03, 1.26s/it, tok/s=14309.0]
Inference: 13%|█▎ | 778/5920 [09:46<1:53:10, 1.32s/it, tok/s=14309.0]
Inference: 13%|█▎ | 778/5920 [09:46<1:53:10, 1.32s/it, tok/s=14299.9]
Inference: 13%|█▎ | 779/5920 [09:47<1:29:23, 1.04s/it, tok/s=14299.9]
Inference: 13%|█▎ | 779/5920 [09:47<1:29:23, 1.04s/it, tok/s=14334.2]
Inference: 13%|█▎ | 780/5920 [09:47<1:13:10, 1.17it/s, tok/s=14334.2]
Inference: 13%|█▎ | 780/5920 [09:47<1:13:10, 1.17it/s, tok/s=14368.4]
Inference: 13%|█▎ | 781/5920 [09:49<1:32:09, 1.08s/it, tok/s=14368.4]
Inference: 13%|█▎ | 781/5920 [09:49<1:32:09, 1.08s/it, tok/s=14362.3]
Inference: 13%|█▎ | 782/5920 [09:49<1:18:35, 1.09it/s, tok/s=14362.3]
Inference: 13%|█▎ | 782/5920 [09:49<1:18:35, 1.09it/s, tok/s=14372.9]
Inference: 13%|█▎ | 783/5920 [09:50<1:23:25, 1.03it/s, tok/s=14372.9]
Inference: 13%|█▎ | 783/5920 [09:50<1:23:25, 1.03it/s, tok/s=14352.6]
Inference: 13%|█▎ | 784/5920 [09:51<1:15:38, 1.13it/s, tok/s=14352.6]
Inference: 13%|█▎ | 784/5920 [09:51<1:15:38, 1.13it/s, tok/s=14354.4]
Inference: 13%|█▎ | 785/5920 [09:53<1:28:57, 1.04s/it, tok/s=14354.4]
Inference: 13%|█▎ | 785/5920 [09:53<1:28:57, 1.04s/it, tok/s=14329.5]
Inference: 13%|█▎ | 786/5920 [09:53<1:10:34, 1.21it/s, tok/s=14329.5]
Inference: 13%|█▎ | 786/5920 [09:53<1:10:34, 1.21it/s, tok/s=14329.1]
Inference: 13%|█▎ | 787/5920 [09:53<1:10:33, 1.21it/s, tok/s=14351.5]
Inference: 13%|█▎ | 788/5920 [09:53<41:53, 2.04it/s, tok/s=14351.5]
Inference: 13%|█▎ | 788/5920 [09:53<41:53, 2.04it/s, tok/s=14390.1]
Inference: 13%|█▎ | 789/5920 [09:54<43:34, 1.96it/s, tok/s=14390.1]
Inference: 13%|█▎ | 789/5920 [09:54<43:34, 1.96it/s, tok/s=14400.5]
Inference: 13%|█▎ | 790/5920 [09:54<45:36, 1.87it/s, tok/s=14400.5]
Inference: 13%|█▎ | 790/5920 [09:54<45:36, 1.87it/s, tok/s=14407.0]
Inference: 13%|█▎ | 791/5920 [09:57<1:48:21, 1.27s/it, tok/s=14407.0]
Inference: 13%|█▎ | 791/5920 [09:57<1:48:21, 1.27s/it, tok/s=14341.6]
Inference: 13%|█▎ | 792/5920 [09:58<1:29:19, 1.05s/it, tok/s=14341.6]
Inference: 13%|█▎ | 792/5920 [09:58<1:29:19, 1.05s/it, tok/s=14340.9]
Inference: 13%|█▎ | 793/5920 [09:59<1:23:15, 1.03it/s, tok/s=14340.9]
Inference: 13%|█▎ | 793/5920 [09:59<1:23:15, 1.03it/s, tok/s=14362.5]
Inference: 13%|█▎ | 794/5920 [10:00<1:23:07, 1.03it/s, tok/s=14362.5]
Inference: 13%|█▎ | 794/5920 [10:00<1:23:07, 1.03it/s, tok/s=14384.9]
Inference: 13%|█▎ | 795/5920 [10:01<1:25:14, 1.00it/s, tok/s=14384.9]
Inference: 13%|█▎ | 795/5920 [10:01<1:25:14, 1.00it/s, tok/s=14368.9]
Inference: 13%|█▎ | 796/5920 [10:01<1:04:11, 1.33it/s, tok/s=14368.9]
Inference: 13%|█▎ | 796/5920 [10:01<1:04:11, 1.33it/s, tok/s=14386.3]
Inference: 13%|█▎ | 797/5920 [10:01<58:21, 1.46it/s, tok/s=14386.3]
Inference: 13%|█▎ | 797/5920 [10:01<58:21, 1.46it/s, tok/s=14414.3]
Inference: 13%|█▎ | 798/5920 [10:02<53:15, 1.60it/s, tok/s=14414.3]
Inference: 13%|█▎ | 798/5920 [10:02<53:15, 1.60it/s, tok/s=14413.8]
Inference: 13%|█▎ | 799/5920 [10:02<45:11, 1.89it/s, tok/s=14413.8]
Inference: 13%|█▎ | 799/5920 [10:02<45:11, 1.89it/s, tok/s=14415.7]
Inference: 14%|█▎ | 800/5920 [10:04<1:10:21, 1.21it/s, tok/s=14415.7]
Inference: 14%|█▎ | 800/5920 [10:04<1:10:21, 1.21it/s, tok/s=14394.2]
Inference: 14%|█▎ | 801/5920 [10:05<1:17:33, 1.10it/s, tok/s=14394.2]
Inference: 14%|█▎ | 801/5920 [10:05<1:17:33, 1.10it/s, tok/s=14393.4]
Inference: 14%|█▎ | 802/5920 [10:05<1:17:32, 1.10it/s, tok/s=14411.8]
Inference: 14%|█▎ | 803/5920 [10:06<1:12:36, 1.17it/s, tok/s=14411.8]
Inference: 14%|█▎ | 803/5920 [10:06<1:12:36, 1.17it/s, tok/s=14396.0]
Inference: 14%|█▎ | 804/5920 [10:06<1:12:35, 1.17it/s, tok/s=14426.2]
Inference: 14%|█▎ | 805/5920 [10:07<57:21, 1.49it/s, tok/s=14426.2]
Inference: 14%|█▎ | 805/5920 [10:07<57:21, 1.49it/s, tok/s=14446.5]
Inference: 14%|█▎ | 806/5920 [10:07<57:20, 1.49it/s, tok/s=14453.9]
Inference: 14%|█▎ | 807/5920 [10:09<59:39, 1.43it/s, tok/s=14453.9]
Inference: 14%|█▎ | 807/5920 [10:09<59:39, 1.43it/s, tok/s=14438.4]
Inference: 14%|█▎ | 808/5920 [10:09<1:01:15, 1.39it/s, tok/s=14438.4]
Inference: 14%|█▎ | 808/5920 [10:09<1:01:15, 1.39it/s, tok/s=14438.6]
Inference: 14%|█▎ | 809/5920 [10:12<1:29:40, 1.05s/it, tok/s=14438.6]
Inference: 14%|█▎ | 809/5920 [10:12<1:29:40, 1.05s/it, tok/s=14428.9]
Inference: 14%|█▎ | 810/5920 [10:13<1:32:40, 1.09s/it, tok/s=14428.9]
Inference: 14%|█▎ | 810/5920 [10:13<1:32:40, 1.09s/it, tok/s=14409.6]
Inference: 14%|█▎ | 811/5920 [10:13<1:32:39, 1.09s/it, tok/s=14415.9]
Inference: 14%|█▎ | 812/5920 [10:13<1:00:07, 1.42it/s, tok/s=14415.9]
Inference: 14%|█▎ | 812/5920 [10:13<1:00:07, 1.42it/s, tok/s=14433.0]
Inference: 14%|█▎ | 813/5920 [10:14<57:11, 1.49it/s, tok/s=14433.0]
Inference: 14%|█▎ | 813/5920 [10:14<57:11, 1.49it/s, tok/s=14430.8]
Inference: 14%|█▍ | 814/5920 [10:14<51:17, 1.66it/s, tok/s=14430.8]
Inference: 14%|█▍ | 814/5920 [10:14<51:17, 1.66it/s, tok/s=14429.2]
Inference: 14%|█▍ | 815/5920 [10:15<1:04:02, 1.33it/s, tok/s=14429.2]
Inference: 14%|█▍ | 815/5920 [10:15<1:04:02, 1.33it/s, tok/s=14408.2]
Inference: 14%|█▍ | 816/5920 [10:15<1:04:01, 1.33it/s, tok/s=14414.4]
Inference: 14%|█▍ | 817/5920 [10:17<1:00:38, 1.40it/s, tok/s=14414.4]
Inference: 14%|█▍ | 817/5920 [10:17<1:00:38, 1.40it/s, tok/s=14392.5]
Inference: 14%|█▍ | 818/5920 [10:18<1:13:47, 1.15it/s, tok/s=14392.5]
Inference: 14%|█▍ | 818/5920 [10:18<1:13:47, 1.15it/s, tok/s=14375.3]
Inference: 14%|█▍ | 819/5920 [10:19<1:13:15, 1.16it/s, tok/s=14375.3]
Inference: 14%|█▍ | 819/5920 [10:19<1:13:15, 1.16it/s, tok/s=14373.9]
Inference: 14%|█▍ | 820/5920 [10:19<1:00:17, 1.41it/s, tok/s=14373.9]
Inference: 14%|█▍ | 820/5920 [10:19<1:00:17, 1.41it/s, tok/s=14375.0]
Inference: 14%|█▍ | 821/5920 [10:20<58:26, 1.45it/s, tok/s=14375.0]
Inference: 14%|█▍ | 821/5920 [10:20<58:26, 1.45it/s, tok/s=14368.4]
Inference: 14%|█▍ | 822/5920 [10:20<49:24, 1.72it/s, tok/s=14368.4]
Inference: 14%|█▍ | 822/5920 [10:20<49:24, 1.72it/s, tok/s=14402.1]
Inference: 14%|█▍ | 823/5920 [10:23<1:53:28, 1.34s/it, tok/s=14402.1]
Inference: 14%|█▍ | 823/5920 [10:23<1:53:28, 1.34s/it, tok/s=14340.2]
Inference: 14%|█▍ | 824/5920 [10:23<1:53:26, 1.34s/it, tok/s=14346.8]
Inference: 14%|█▍ | 825/5920 [10:24<1:13:55, 1.15it/s, tok/s=14346.8]
Inference: 14%|█▍ | 825/5920 [10:24<1:13:55, 1.15it/s, tok/s=14375.6]
Inference: 14%|█▍ | 826/5920 [10:24<1:02:30, 1.36it/s, tok/s=14375.6]
Inference: 14%|█▍ | 826/5920 [10:24<1:02:30, 1.36it/s, tok/s=14383.6]
Inference: 14%|█▍ | 827/5920 [10:27<1:46:50, 1.26s/it, tok/s=14383.6]
Inference: 14%|█▍ | 827/5920 [10:27<1:46:50, 1.26s/it, tok/s=14329.7]
Inference: 14%|█▍ | 828/5920 [10:27<1:46:49, 1.26s/it, tok/s=14369.5]
Inference: 14%|█▍ | 829/5920 [10:30<1:47:37, 1.27s/it, tok/s=14369.5]
Inference: 14%|█▍ | 829/5920 [10:30<1:47:37, 1.27s/it, tok/s=14326.8]
Inference: 14%|█▍ | 830/5920 [10:30<1:33:21, 1.10s/it, tok/s=14326.8]
Inference: 14%|█▍ | 830/5920 [10:30<1:33:21, 1.10s/it, tok/s=14354.7]
Inference: 14%|█▍ | 831/5920 [10:32<1:48:02, 1.27s/it, tok/s=14354.7]
Inference: 14%|█▍ | 831/5920 [10:32<1:48:02, 1.27s/it, tok/s=14335.2]
Inference: 14%|█▍ | 832/5920 [10:33<1:39:53, 1.18s/it, tok/s=14335.2]
Inference: 14%|█▍ | 832/5920 [10:33<1:39:53, 1.18s/it, tok/s=14323.0]
Inference: 14%|█▍ | 833/5920 [10:33<1:20:04, 1.06it/s, tok/s=14323.0]
Inference: 14%|█▍ | 833/5920 [10:33<1:20:04, 1.06it/s, tok/s=14327.7]
Inference: 14%|█▍ | 834/5920 [10:33<1:02:01, 1.37it/s, tok/s=14327.7]
Inference: 14%|█▍ | 834/5920 [10:33<1:02:01, 1.37it/s, tok/s=14334.6]
Inference: 14%|█▍ | 835/5920 [10:35<1:31:24, 1.08s/it, tok/s=14334.6]
Inference: 14%|█▍ | 835/5920 [10:35<1:31:24, 1.08s/it, tok/s=14315.3]
Inference: 14%|█▍ | 836/5920 [10:37<1:41:59, 1.20s/it, tok/s=14315.3]
Inference: 14%|█▍ | 836/5920 [10:37<1:41:59, 1.20s/it, tok/s=14290.5]
Inference: 14%|█▍ | 837/5920 [10:37<1:21:07, 1.04it/s, tok/s=14290.5]
Inference: 14%|█▍ | 837/5920 [10:37<1:21:07, 1.04it/s, tok/s=14295.5]
Inference: 14%|█▍ | 838/5920 [10:38<1:10:50, 1.20it/s, tok/s=14295.5]
Inference: 14%|█▍ | 838/5920 [10:38<1:10:50, 1.20it/s, tok/s=14296.8]
Inference: 14%|█▍ | 839/5920 [10:38<54:49, 1.54it/s, tok/s=14296.8]
Inference: 14%|█▍ | 839/5920 [10:38<54:49, 1.54it/s, tok/s=14304.4]
Inference: 14%|█▍ | 840/5920 [10:38<43:49, 1.93it/s, tok/s=14304.4]
Inference: 14%|█▍ | 840/5920 [10:38<43:49, 1.93it/s, tok/s=14313.7]
Inference: 14%|█▍ | 841/5920 [10:41<1:49:00, 1.29s/it, tok/s=14313.7]
Inference: 14%|█▍ | 841/5920 [10:41<1:49:00, 1.29s/it, tok/s=14253.5]
Inference: 14%|█▍ | 842/5920 [10:42<1:27:30, 1.03s/it, tok/s=14253.5]
Inference: 14%|█▍ | 842/5920 [10:42<1:27:30, 1.03s/it, tok/s=14256.5]
Inference: 14%|█▍ | 843/5920 [10:42<1:12:50, 1.16it/s, tok/s=14256.5]
Inference: 14%|█▍ | 843/5920 [10:42<1:12:50, 1.16it/s, tok/s=14255.3]
Inference: 14%|█▍ | 844/5920 [10:43<1:15:37, 1.12it/s, tok/s=14255.3]
Inference: 14%|█▍ | 844/5920 [10:43<1:15:37, 1.12it/s, tok/s=14242.7]
Inference: 14%|█▍ | 845/5920 [10:44<1:05:56, 1.28it/s, tok/s=14242.7]
Inference: 14%|█▍ | 845/5920 [10:44<1:05:56, 1.28it/s, tok/s=14239.4]
Inference: 14%|█▍ | 846/5920 [10:45<1:14:12, 1.14it/s, tok/s=14239.4]
Inference: 14%|█▍ | 846/5920 [10:45<1:14:12, 1.14it/s, tok/s=14235.2]
Inference: 14%|█▍ | 847/5920 [10:46<1:23:07, 1.02it/s, tok/s=14235.2]
Inference: 14%|█▍ | 847/5920 [10:46<1:23:07, 1.02it/s, tok/s=14219.9]
Inference: 14%|█▍ | 848/5920 [10:46<1:13:21, 1.15it/s, tok/s=14219.9]
Inference: 14%|█▍ | 848/5920 [10:46<1:13:21, 1.15it/s, tok/s=14247.8]
Inference: 14%|█▍ | 849/5920 [10:48<1:20:50, 1.05it/s, tok/s=14247.8]
Inference: 14%|█▍ | 849/5920 [10:48<1:20:50, 1.05it/s, tok/s=14232.9]
Inference: 14%|█▍ | 850/5920 [10:50<1:50:21, 1.31s/it, tok/s=14232.9]
Inference: 14%|█▍ | 850/5920 [10:50<1:50:21, 1.31s/it, tok/s=14198.6]
Inference: 14%|█▍ | 851/5920 [10:51<1:47:50, 1.28s/it, tok/s=14198.6]
Inference: 14%|█▍ | 851/5920 [10:51<1:47:50, 1.28s/it, tok/s=14180.4]
Inference: 14%|█▍ | 852/5920 [10:52<1:37:05, 1.15s/it, tok/s=14180.4]
Inference: 14%|█▍ | 852/5920 [10:52<1:37:05, 1.15s/it, tok/s=14172.9]
Inference: 14%|█▍ | 853/5920 [10:53<1:32:23, 1.09s/it, tok/s=14172.9]
Inference: 14%|█▍ | 853/5920 [10:53<1:32:23, 1.09s/it, tok/s=14166.8]
Inference: 14%|█▍ | 854/5920 [10:54<1:37:25, 1.15s/it, tok/s=14166.8]
Inference: 14%|█▍ | 854/5920 [10:54<1:37:25, 1.15s/it, tok/s=14147.8]
Inference: 14%|█▍ | 855/5920 [10:54<1:37:24, 1.15s/it, tok/s=14168.2]
Inference: 14%|█▍ | 856/5920 [10:54<58:01, 1.45it/s, tok/s=14168.2]
Inference: 14%|█▍ | 856/5920 [10:54<58:01, 1.45it/s, tok/s=14170.5]
Inference: 14%|█▍ | 857/5920 [10:56<1:09:09, 1.22it/s, tok/s=14170.5]
Inference: 14%|█▍ | 857/5920 [10:56<1:09:09, 1.22it/s, tok/s=14152.9]
Inference: 14%|█▍ | 858/5920 [10:56<1:04:57, 1.30it/s, tok/s=14152.9]
Inference: 14%|█▍ | 858/5920 [10:56<1:04:57, 1.30it/s, tok/s=14153.4]
Inference: 15%|█▍ | 859/5920 [10:56<1:04:56, 1.30it/s, tok/s=14174.3]
Inference: 15%|█▍ | 860/5920 [10:59<1:26:09, 1.02s/it, tok/s=14174.3]
Inference: 15%|█▍ | 860/5920 [10:59<1:26:09, 1.02s/it, tok/s=14132.2]
Inference: 15%|█▍ | 861/5920 [10:59<1:08:37, 1.23it/s, tok/s=14132.2]
Inference: 15%|█▍ | 861/5920 [10:59<1:08:37, 1.23it/s, tok/s=14143.7]
Inference: 15%|█▍ | 862/5920 [11:02<1:59:43, 1.42s/it, tok/s=14143.7]
Inference: 15%|█▍ | 862/5920 [11:02<1:59:43, 1.42s/it, tok/s=14085.7]
Inference: 15%|█▍ | 863/5920 [11:03<1:47:40, 1.28s/it, tok/s=14085.7]
Inference: 15%|█▍ | 863/5920 [11:03<1:47:40, 1.28s/it, tok/s=14077.7]
Inference: 15%|█▍ | 864/5920 [11:06<2:23:12, 1.70s/it, tok/s=14077.7]
Inference: 15%|█▍ | 864/5920 [11:06<2:23:12, 1.70s/it, tok/s=14029.0]
Inference: 15%|█▍ | 865/5920 [11:06<1:45:20, 1.25s/it, tok/s=14029.0]
Inference: 15%|█▍ | 865/5920 [11:06<1:45:20, 1.25s/it, tok/s=14036.1]
Inference: 15%|█▍ | 866/5920 [11:06<1:22:00, 1.03it/s, tok/s=14036.1]
Inference: 15%|█▍ | 866/5920 [11:06<1:22:00, 1.03it/s, tok/s=14069.7]
Inference: 15%|█▍ | 867/5920 [11:08<1:41:08, 1.20s/it, tok/s=14069.7]
Inference: 15%|█▍ | 867/5920 [11:08<1:41:08, 1.20s/it, tok/s=14041.2]
Inference: 15%|█▍ | 868/5920 [11:09<1:38:12, 1.17s/it, tok/s=14041.2]
Inference: 15%|█▍ | 868/5920 [11:09<1:38:12, 1.17s/it, tok/s=14031.0]
Inference: 15%|█▍ | 869/5920 [11:10<1:33:02, 1.11s/it, tok/s=14031.0]
Inference: 15%|█▍ | 869/5920 [11:10<1:33:02, 1.11s/it, tok/s=14020.4]
Inference: 15%|█▍ | 870/5920 [11:11<1:31:46, 1.09s/it, tok/s=14020.4]
Inference: 15%|█▍ | 870/5920 [11:11<1:31:46, 1.09s/it, tok/s=14018.4]
Inference: 15%|█▍ | 871/5920 [11:11<1:31:45, 1.09s/it, tok/s=14030.5]
Inference: 15%|█▍ | 872/5920 [11:12<1:05:38, 1.28it/s, tok/s=14030.5]
Inference: 15%|█▍ | 872/5920 [11:12<1:05:38, 1.28it/s, tok/s=14026.9]
Inference: 15%|█▍ | 873/5920 [11:12<1:05:37, 1.28it/s, tok/s=14039.5]
Inference: 15%|█▍ | 874/5920 [11:13<52:51, 1.59it/s, tok/s=14039.5]
Inference: 15%|█▍ | 874/5920 [11:13<52:51, 1.59it/s, tok/s=14063.5]
Inference: 15%|█▍ | 875/5920 [11:14<53:42, 1.57it/s, tok/s=14063.5]
Inference: 15%|█▍ | 875/5920 [11:14<53:42, 1.57it/s, tok/s=14062.9]
Inference: 15%|█▍ | 876/5920 [11:15<1:06:16, 1.27it/s, tok/s=14062.9]
Inference: 15%|█▍ | 876/5920 [11:15<1:06:16, 1.27it/s, tok/s=14077.0]
Inference: 15%|█▍ | 877/5920 [11:15<1:03:30, 1.32it/s, tok/s=14077.0]
Inference: 15%|█▍ | 877/5920 [11:15<1:03:30, 1.32it/s, tok/s=14071.3]
Inference: 15%|█▍ | 878/5920 [11:16<59:33, 1.41it/s, tok/s=14071.3]
Inference: 15%|█▍ | 878/5920 [11:16<59:33, 1.41it/s, tok/s=14079.0]
Inference: 15%|█▍ | 879/5920 [11:16<59:32, 1.41it/s, tok/s=14101.1]
Inference: 15%|█▍ | 880/5920 [11:17<51:01, 1.65it/s, tok/s=14101.1]
Inference: 15%|█▍ | 880/5920 [11:17<51:01, 1.65it/s, tok/s=14094.8]
Inference: 15%|█▍ | 881/5920 [11:20<1:34:53, 1.13s/it, tok/s=14094.8]
Inference: 15%|█▍ | 881/5920 [11:20<1:34:53, 1.13s/it, tok/s=14045.0]
Inference: 15%|█▍ | 882/5920 [11:21<1:36:16, 1.15s/it, tok/s=14045.0]
Inference: 15%|█▍ | 882/5920 [11:21<1:36:16, 1.15s/it, tok/s=14059.9]
Inference: 15%|█▍ | 883/5920 [11:21<1:13:35, 1.14it/s, tok/s=14059.9]
Inference: 15%|█▍ | 883/5920 [11:21<1:13:35, 1.14it/s, tok/s=14068.8]
Inference: 15%|█▍ | 884/5920 [11:22<1:13:28, 1.14it/s, tok/s=14068.8]
Inference: 15%|█▍ | 884/5920 [11:22<1:13:28, 1.14it/s, tok/s=14060.5]
Inference: 15%|█▍ | 885/5920 [11:22<59:54, 1.40it/s, tok/s=14060.5]
Inference: 15%|█▍ | 885/5920 [11:22<59:54, 1.40it/s, tok/s=14078.2]
Inference: 15%|█▍ | 886/5920 [11:23<49:04, 1.71it/s, tok/s=14078.2]
Inference: 15%|█▍ | 886/5920 [11:23<49:04, 1.71it/s, tok/s=14101.6]
Inference: 15%|█▍ | 887/5920 [11:24<1:19:45, 1.05it/s, tok/s=14101.6]
Inference: 15%|█▍ | 887/5920 [11:24<1:19:45, 1.05it/s, tok/s=14075.1]
Inference: 15%|█▌ | 888/5920 [11:25<1:03:55, 1.31it/s, tok/s=14075.1]
Inference: 15%|█▌ | 888/5920 [11:25<1:03:55, 1.31it/s, tok/s=14090.3]
Inference: 15%|█▌ | 889/5920 [11:25<1:03:54, 1.31it/s, tok/s=14100.3]
Inference: 15%|█▌ | 890/5920 [11:27<1:19:27, 1.06it/s, tok/s=14100.3]
Inference: 15%|█▌ | 890/5920 [11:27<1:19:27, 1.06it/s, tok/s=14074.8]
Inference: 15%|█▌ | 891/5920 [11:28<1:28:45, 1.06s/it, tok/s=14074.8]
Inference: 15%|█▌ | 891/5920 [11:28<1:28:45, 1.06s/it, tok/s=14062.3]
Inference: 15%|█▌ | 892/5920 [11:29<1:14:06, 1.13it/s, tok/s=14062.3]
Inference: 15%|█▌ | 892/5920 [11:29<1:14:06, 1.13it/s, tok/s=14066.4]
Inference: 15%|█▌ | 893/5920 [11:30<1:30:28, 1.08s/it, tok/s=14066.4]
Inference: 15%|█▌ | 893/5920 [11:30<1:30:28, 1.08s/it, tok/s=14042.7]
Inference: 15%|█▌ | 894/5920 [11:33<1:56:01, 1.39s/it, tok/s=14042.7]
Inference: 15%|█▌ | 894/5920 [11:33<1:56:01, 1.39s/it, tok/s=14011.4]
Inference: 15%|█▌ | 895/5920 [11:33<1:27:13, 1.04s/it, tok/s=14011.4]
Inference: 15%|█▌ | 895/5920 [11:33<1:27:13, 1.04s/it, tok/s=14021.8]
Inference: 15%|█▌ | 896/5920 [11:33<1:04:53, 1.29it/s, tok/s=14021.8]
Inference: 15%|█▌ | 896/5920 [11:33<1:04:53, 1.29it/s, tok/s=14032.7]
Inference: 15%|█▌ | 897/5920 [11:33<1:04:53, 1.29it/s, tok/s=14047.4]
Inference: 15%|█▌ | 898/5920 [11:34<50:21, 1.66it/s, tok/s=14047.4]
Inference: 15%|█▌ | 898/5920 [11:34<50:21, 1.66it/s, tok/s=14073.9]
Inference: 15%|█▌ | 899/5920 [11:35<55:28, 1.51it/s, tok/s=14073.9]
Inference: 15%|█▌ | 899/5920 [11:35<55:28, 1.51it/s, tok/s=14081.9]
Inference: 15%|█▌ | 900/5920 [11:35<55:27, 1.51it/s, tok/s=14091.2]
Inference: 15%|█▌ | 901/5920 [11:35<39:46, 2.10it/s, tok/s=14091.2]
Inference: 15%|█▌ | 901/5920 [11:35<39:46, 2.10it/s, tok/s=14096.6]
Inference: 15%|█▌ | 902/5920 [11:35<34:30, 2.42it/s, tok/s=14096.6]
Inference: 15%|█▌ | 902/5920 [11:35<34:30, 2.42it/s, tok/s=14106.5]
Inference: 15%|█▌ | 903/5920 [11:36<49:42, 1.68it/s, tok/s=14106.5]
Inference: 15%|█▌ | 903/5920 [11:36<49:42, 1.68it/s, tok/s=14095.1]
Inference: 15%|█▌ | 904/5920 [11:37<52:24, 1.60it/s, tok/s=14095.1]
Inference: 15%|█▌ | 904/5920 [11:37<52:24, 1.60it/s, tok/s=14094.3]
Inference: 15%|█▌ | 905/5920 [11:37<52:23, 1.60it/s, tok/s=14114.5]
Inference: 15%|█▌ | 906/5920 [11:40<1:18:39, 1.06it/s, tok/s=14114.5]
Inference: 15%|█▌ | 906/5920 [11:40<1:18:39, 1.06it/s, tok/s=14085.3]
Inference: 15%|█▌ | 907/5920 [11:41<1:24:38, 1.01s/it, tok/s=14085.3]
Inference: 15%|█▌ | 907/5920 [11:41<1:24:38, 1.01s/it, tok/s=14099.4]
Inference: 15%|█▌ | 908/5920 [11:41<1:07:12, 1.24it/s, tok/s=14099.4]
Inference: 15%|█▌ | 908/5920 [11:41<1:07:12, 1.24it/s, tok/s=14106.9]
Inference: 15%|█▌ | 909/5920 [11:43<1:32:21, 1.11s/it, tok/s=14106.9]
Inference: 15%|█▌ | 909/5920 [11:43<1:32:21, 1.11s/it, tok/s=14085.0]
Inference: 15%|█▌ | 910/5920 [11:44<1:26:23, 1.03s/it, tok/s=14085.0]
Inference: 15%|█▌ | 910/5920 [11:44<1:26:23, 1.03s/it, tok/s=14086.0]
Inference: 15%|█▌ | 911/5920 [11:47<2:23:11, 1.72s/it, tok/s=14086.0]
Inference: 15%|█▌ | 911/5920 [11:47<2:23:11, 1.72s/it, tok/s=14029.7]
Inference: 15%|█▌ | 912/5920 [11:48<1:45:37, 1.27s/it, tok/s=14029.7]
Inference: 15%|█▌ | 912/5920 [11:48<1:45:37, 1.27s/it, tok/s=14039.5]
Inference: 15%|█▌ | 913/5920 [11:49<1:51:45, 1.34s/it, tok/s=14039.5]
Inference: 15%|█▌ | 913/5920 [11:49<1:51:45, 1.34s/it, tok/s=14032.3]
Inference: 15%|█▌ | 914/5920 [11:49<1:51:43, 1.34s/it, tok/s=14047.2]
Inference: 15%|█▌ | 915/5920 [11:49<1:06:20, 1.26it/s, tok/s=14047.2]
Inference: 15%|█▌ | 915/5920 [11:49<1:06:20, 1.26it/s, tok/s=14054.6]
Inference: 15%|█▌ | 916/5920 [11:50<53:05, 1.57it/s, tok/s=14054.6]
Inference: 15%|█▌ | 916/5920 [11:50<53:05, 1.57it/s, tok/s=14076.0]
Inference: 15%|█▌ | 917/5920 [11:52<1:38:18, 1.18s/it, tok/s=14076.0]
Inference: 15%|█▌ | 917/5920 [11:52<1:38:18, 1.18s/it, tok/s=14034.0]
Inference: 16%|█▌ | 918/5920 [11:53<1:20:21, 1.04it/s, tok/s=14034.0]
Inference: 16%|█▌ | 918/5920 [11:53<1:20:21, 1.04it/s, tok/s=14044.3]
Inference: 16%|█▌ | 919/5920 [11:55<1:51:02, 1.33s/it, tok/s=14044.3]
Inference: 16%|█▌ | 919/5920 [11:55<1:51:02, 1.33s/it, tok/s=14018.4]
Inference: 16%|█▌ | 920/5920 [11:55<1:23:53, 1.01s/it, tok/s=14018.4]
Inference: 16%|█▌ | 920/5920 [11:55<1:23:53, 1.01s/it, tok/s=14037.0]
Inference: 16%|█▌ | 921/5920 [11:59<2:30:37, 1.81s/it, tok/s=14037.0]
Inference: 16%|█▌ | 921/5920 [11:59<2:30:37, 1.81s/it, tok/s=13976.5]
Inference: 16%|█▌ | 922/5920 [12:00<2:07:31, 1.53s/it, tok/s=13976.5]
Inference: 16%|█▌ | 922/5920 [12:00<2:07:31, 1.53s/it, tok/s=13976.3]
Inference: 16%|█▌ | 923/5920 [12:01<1:56:14, 1.40s/it, tok/s=13976.3]
Inference: 16%|█▌ | 923/5920 [12:01<1:56:14, 1.40s/it, tok/s=13977.5]
Inference: 16%|█▌ | 924/5920 [12:01<1:31:36, 1.10s/it, tok/s=13977.5]
Inference: 16%|█▌ | 924/5920 [12:01<1:31:36, 1.10s/it, tok/s=14008.8]
Inference: 16%|█▌ | 925/5920 [12:01<1:31:35, 1.10s/it, tok/s=14019.7]
Inference: 16%|█▌ | 926/5920 [12:05<1:52:30, 1.35s/it, tok/s=14019.7]
Inference: 16%|█▌ | 926/5920 [12:05<1:52:30, 1.35s/it, tok/s=13970.4]
Inference: 16%|█▌ | 927/5920 [12:06<1:53:26, 1.36s/it, tok/s=13970.4]
Inference: 16%|█▌ | 927/5920 [12:06<1:53:26, 1.36s/it, tok/s=13967.1]
Inference: 16%|█▌ | 928/5920 [12:06<1:30:51, 1.09s/it, tok/s=13967.1]
Inference: 16%|█▌ | 928/5920 [12:06<1:30:51, 1.09s/it, tok/s=13974.8]
Inference: 16%|█▌ | 929/5920 [12:06<1:10:58, 1.17it/s, tok/s=13974.8]
Inference: 16%|█▌ | 929/5920 [12:06<1:10:58, 1.17it/s, tok/s=13991.9]
Inference: 16%|█▌ | 930/5920 [12:07<55:06, 1.51it/s, tok/s=13991.9]
Inference: 16%|█▌ | 930/5920 [12:07<55:06, 1.51it/s, tok/s=14008.6]
Inference: 16%|█▌ | 931/5920 [12:08<1:09:21, 1.20it/s, tok/s=14008.6]
Inference: 16%|█▌ | 931/5920 [12:08<1:09:21, 1.20it/s, tok/s=14000.6]
Inference: 16%|█▌ | 932/5920 [12:08<1:09:20, 1.20it/s, tok/s=14029.9]
Inference: 16%|█▌ | 933/5920 [12:13<2:08:38, 1.55s/it, tok/s=14029.9]
Inference: 16%|█▌ | 933/5920 [12:13<2:08:38, 1.55s/it, tok/s=13960.1]
Inference: 16%|█▌ | 934/5920 [12:17<3:11:15, 2.30s/it, tok/s=13960.1]
Inference: 16%|█▌ | 934/5920 [12:17<3:11:15, 2.30s/it, tok/s=13884.1]
Inference: 16%|█▌ | 935/5920 [12:19<2:55:07, 2.11s/it, tok/s=13884.1]
Inference: 16%|█▌ | 935/5920 [12:19<2:55:07, 2.11s/it, tok/s=13873.9]
Inference: 16%|█▌ | 936/5920 [12:19<2:11:52, 1.59s/it, tok/s=13873.9]
Inference: 16%|█▌ | 936/5920 [12:19<2:11:52, 1.59s/it, tok/s=13881.4]
Inference: 16%|█▌ | 937/5920 [12:20<1:51:33, 1.34s/it, tok/s=13881.4]
Inference: 16%|█▌ | 937/5920 [12:20<1:51:33, 1.34s/it, tok/s=13876.0]
Inference: 16%|█▌ | 938/5920 [12:21<1:37:00, 1.17s/it, tok/s=13876.0]
Inference: 16%|█▌ | 938/5920 [12:21<1:37:00, 1.17s/it, tok/s=13883.2]
Inference: 16%|█▌ | 939/5920 [12:21<1:16:47, 1.08it/s, tok/s=13883.2]
Inference: 16%|█▌ | 939/5920 [12:21<1:16:47, 1.08it/s, tok/s=13917.1]
Inference: 16%|█▌ | 940/5920 [12:22<1:23:21, 1.00s/it, tok/s=13917.1]
Inference: 16%|█▌ | 940/5920 [12:22<1:23:21, 1.00s/it, tok/s=13916.3]
Inference: 16%|█▌ | 941/5920 [12:22<1:01:27, 1.35it/s, tok/s=13916.3]
Inference: 16%|█▌ | 941/5920 [12:22<1:01:27, 1.35it/s, tok/s=13928.6]
Inference: 16%|█▌ | 942/5920 [12:23<1:03:28, 1.31it/s, tok/s=13928.6]
Inference: 16%|█▌ | 942/5920 [12:23<1:03:28, 1.31it/s, tok/s=13952.6]
Inference: 16%|█▌ | 943/5920 [12:24<1:12:16, 1.15it/s, tok/s=13952.6]
Inference: 16%|█▌ | 943/5920 [12:24<1:12:16, 1.15it/s, tok/s=13957.6]
Inference: 16%|█▌ | 944/5920 [12:25<1:05:20, 1.27it/s, tok/s=13957.6]
Inference: 16%|█▌ | 944/5920 [12:25<1:05:20, 1.27it/s, tok/s=13959.1]
Inference: 16%|█▌ | 945/5920 [12:26<1:15:24, 1.10it/s, tok/s=13959.1]
Inference: 16%|█▌ | 945/5920 [12:26<1:15:24, 1.10it/s, tok/s=13958.0]
Inference: 16%|█▌ | 946/5920 [12:27<1:10:55, 1.17it/s, tok/s=13958.0]
Inference: 16%|█▌ | 946/5920 [12:27<1:10:55, 1.17it/s, tok/s=13960.3]
Inference: 16%|█▌ | 947/5920 [12:27<1:10:54, 1.17it/s, tok/s=13973.5]
Inference: 16%|█▌ | 948/5920 [12:30<1:39:12, 1.20s/it, tok/s=13973.5]
Inference: 16%|█▌ | 948/5920 [12:30<1:39:12, 1.20s/it, tok/s=13929.3]
Inference: 16%|█▌ | 949/5920 [12:30<1:39:11, 1.20s/it, tok/s=13943.9]
Inference: 16%|█▌ | 950/5920 [12:34<2:03:53, 1.50s/it, tok/s=13943.9]
Inference: 16%|█▌ | 950/5920 [12:34<2:03:53, 1.50s/it, tok/s=13886.0]
Inference: 16%|█▌ | 951/5920 [12:35<1:55:40, 1.40s/it, tok/s=13886.0]
Inference: 16%|█▌ | 951/5920 [12:35<1:55:40, 1.40s/it, tok/s=13889.5]
Inference: 16%|█▌ | 952/5920 [12:37<2:11:11, 1.58s/it, tok/s=13889.5]
Inference: 16%|█▌ | 952/5920 [12:37<2:11:11, 1.58s/it, tok/s=13858.7]
Inference: 16%|█▌ | 953/5920 [12:38<2:00:10, 1.45s/it, tok/s=13858.7]
Inference: 16%|█▌ | 953/5920 [12:38<2:00:10, 1.45s/it, tok/s=13858.5]
Inference: 16%|█▌ | 954/5920 [12:39<1:44:40, 1.26s/it, tok/s=13858.5]
Inference: 16%|█▌ | 954/5920 [12:39<1:44:40, 1.26s/it, tok/s=13851.8]
Inference: 16%|█▌ | 955/5920 [12:39<1:31:12, 1.10s/it, tok/s=13851.8]
Inference: 16%|█▌ | 955/5920 [12:39<1:31:12, 1.10s/it, tok/s=13859.9]
Inference: 16%|█▌ | 956/5920 [12:40<1:08:29, 1.21it/s, tok/s=13859.9]
Inference: 16%|█▌ | 956/5920 [12:40<1:08:29, 1.21it/s, tok/s=13872.8]
Inference: 16%|█▌ | 957/5920 [12:41<1:26:05, 1.04s/it, tok/s=13872.8]
Inference: 16%|█▌ | 957/5920 [12:41<1:26:05, 1.04s/it, tok/s=13850.8]
Inference: 16%|█▌ | 958/5920 [12:43<1:35:24, 1.15s/it, tok/s=13850.8]
Inference: 16%|█▌ | 958/5920 [12:43<1:35:24, 1.15s/it, tok/s=13846.8]
Inference: 16%|█▌ | 959/5920 [12:43<1:14:08, 1.12it/s, tok/s=13846.8]
Inference: 16%|█▌ | 959/5920 [12:43<1:14:08, 1.12it/s, tok/s=13863.3]
Inference: 16%|█▌ | 960/5920 [12:43<55:11, 1.50it/s, tok/s=13863.3]
Inference: 16%|█▌ | 960/5920 [12:43<55:11, 1.50it/s, tok/s=13877.4]
Inference: 16%|█▌ | 961/5920 [12:43<46:28, 1.78it/s, tok/s=13877.4]
Inference: 16%|█▌ | 961/5920 [12:43<46:28, 1.78it/s, tok/s=13888.3]
Inference: 16%|█▋ | 962/5920 [12:44<1:01:28, 1.34it/s, tok/s=13888.3]
Inference: 16%|█▋ | 962/5920 [12:44<1:01:28, 1.34it/s, tok/s=13887.1]
Inference: 16%|█▋ | 963/5920 [12:45<1:04:21, 1.28it/s, tok/s=13887.1]
Inference: 16%|█▋ | 963/5920 [12:45<1:04:21, 1.28it/s, tok/s=13888.4]
Inference: 16%|█▋ | 964/5920 [12:47<1:35:07, 1.15s/it, tok/s=13888.4]
Inference: 16%|█▋ | 964/5920 [12:47<1:35:07, 1.15s/it, tok/s=13870.4]
Inference: 16%|█▋ | 965/5920 [12:48<1:11:18, 1.16it/s, tok/s=13870.4]
Inference: 16%|█▋ | 965/5920 [12:48<1:11:18, 1.16it/s, tok/s=13875.5]
Inference: 16%|█▋ | 966/5920 [12:48<1:03:27, 1.30it/s, tok/s=13875.5]
Inference: 16%|█▋ | 966/5920 [12:48<1:03:27, 1.30it/s, tok/s=13883.1]
Inference: 16%|█▋ | 967/5920 [12:48<52:39, 1.57it/s, tok/s=13883.1]
Inference: 16%|█▋ | 967/5920 [12:48<52:39, 1.57it/s, tok/s=13890.6]
Inference: 16%|█▋ | 968/5920 [12:49<41:59, 1.97it/s, tok/s=13890.6]
Inference: 16%|█▋ | 968/5920 [12:49<41:59, 1.97it/s, tok/s=13902.1]
Inference: 16%|█▋ | 969/5920 [12:49<35:35, 2.32it/s, tok/s=13902.1]
Inference: 16%|█▋ | 969/5920 [12:49<35:35, 2.32it/s, tok/s=13916.7]
Inference: 16%|█▋ | 970/5920 [12:50<43:28, 1.90it/s, tok/s=13916.7]
Inference: 16%|█▋ | 970/5920 [12:50<43:28, 1.90it/s, tok/s=13924.2]
Inference: 16%|█▋ | 971/5920 [12:50<42:34, 1.94it/s, tok/s=13924.2]
Inference: 16%|█▋ | 971/5920 [12:50<42:34, 1.94it/s, tok/s=13930.0]
Inference: 16%|█▋ | 972/5920 [12:50<33:08, 2.49it/s, tok/s=13930.0]
Inference: 16%|█▋ | 972/5920 [12:50<33:08, 2.49it/s, tok/s=13935.4]
Inference: 16%|█▋ | 973/5920 [12:51<32:36, 2.53it/s, tok/s=13935.4]
Inference: 16%|█▋ | 973/5920 [12:51<32:36, 2.53it/s, tok/s=13944.1]
Inference: 16%|█▋ | 974/5920 [12:51<25:22, 3.25it/s, tok/s=13944.1]
Inference: 16%|█▋ | 974/5920 [12:51<25:22, 3.25it/s, tok/s=13956.8]
Inference: 16%|█▋ | 975/5920 [12:51<25:22, 3.25it/s, tok/s=13978.1]
Inference: 16%|█▋ | 976/5920 [12:52<27:41, 2.97it/s, tok/s=13978.1]
Inference: 16%|█▋ | 976/5920 [12:52<27:41, 2.97it/s, tok/s=13982.1]
Inference: 17%|█▋ | 977/5920 [12:54<1:09:12, 1.19it/s, tok/s=13982.1]
Inference: 17%|█▋ | 977/5920 [12:54<1:09:12, 1.19it/s, tok/s=13948.3]
Inference: 17%|█▋ | 978/5920 [12:54<58:21, 1.41it/s, tok/s=13948.3]
Inference: 17%|█▋ | 978/5920 [12:54<58:21, 1.41it/s, tok/s=13955.5]
Inference: 17%|█▋ | 979/5920 [12:54<45:28, 1.81it/s, tok/s=13955.5]
Inference: 17%|█▋ | 979/5920 [12:54<45:28, 1.81it/s, tok/s=13967.7]
Inference: 17%|█▋ | 980/5920 [12:54<45:28, 1.81it/s, tok/s=13980.0]
Inference: 17%|█▋ | 981/5920 [12:58<1:37:19, 1.18s/it, tok/s=13980.0]
Inference: 17%|█▋ | 981/5920 [12:58<1:37:19, 1.18s/it, tok/s=13925.4]
Inference: 17%|█▋ | 982/5920 [12:59<1:30:35, 1.10s/it, tok/s=13925.4]
Inference: 17%|█▋ | 982/5920 [12:59<1:30:35, 1.10s/it, tok/s=13926.4]
Inference: 17%|█▋ | 983/5920 [13:00<1:19:05, 1.04it/s, tok/s=13926.4]
Inference: 17%|█▋ | 983/5920 [13:00<1:19:05, 1.04it/s, tok/s=13942.3]
Inference: 17%|█▋ | 984/5920 [13:00<1:19:04, 1.04it/s, tok/s=13950.0]
Inference: 17%|█▋ | 985/5920 [13:00<56:47, 1.45it/s, tok/s=13950.0]
Inference: 17%|█▋ | 985/5920 [13:00<56:47, 1.45it/s, tok/s=13945.8]
Inference: 17%|█▋ | 986/5920 [13:01<51:32, 1.60it/s, tok/s=13945.8]
Inference: 17%|█▋ | 986/5920 [13:01<51:32, 1.60it/s, tok/s=13963.9]
Inference: 17%|█▋ | 987/5920 [13:03<1:15:05, 1.09it/s, tok/s=13963.9]
Inference: 17%|█▋ | 987/5920 [13:03<1:15:05, 1.09it/s, tok/s=13940.5]
Inference: 17%|█▋ | 988/5920 [13:03<1:03:10, 1.30it/s, tok/s=13940.5]
Inference: 17%|█▋ | 988/5920 [13:03<1:03:10, 1.30it/s, tok/s=13944.6]
Inference: 17%|█▋ | 989/5920 [13:04<1:05:23, 1.26it/s, tok/s=13944.6]
Inference: 17%|█▋ | 989/5920 [13:04<1:05:23, 1.26it/s, tok/s=13945.2]
Inference: 17%|█▋ | 990/5920 [13:05<1:04:49, 1.27it/s, tok/s=13945.2]
Inference: 17%|█▋ | 990/5920 [13:05<1:04:49, 1.27it/s, tok/s=13947.5]
Inference: 17%|█▋ | 991/5920 [13:05<1:08:04, 1.21it/s, tok/s=13947.5]
Inference: 17%|█▋ | 991/5920 [13:05<1:08:04, 1.21it/s, tok/s=13945.5]
Inference: 17%|█▋ | 992/5920 [13:07<1:19:02, 1.04it/s, tok/s=13945.5]
Inference: 17%|█▋ | 992/5920 [13:07<1:19:02, 1.04it/s, tok/s=13948.1]
Inference: 17%|█▋ | 993/5920 [13:07<1:19:02, 1.04it/s, tok/s=13985.9]
Inference: 17%|█▋ | 994/5920 [13:09<1:26:47, 1.06s/it, tok/s=13985.9]
Inference: 17%|█▋ | 994/5920 [13:09<1:26:47, 1.06s/it, tok/s=13953.3]
Inference: 17%|█▋ | 995/5920 [13:09<1:26:46, 1.06s/it, tok/s=13970.5]
Inference: 17%|█▋ | 996/5920 [13:11<1:20:29, 1.02it/s, tok/s=13970.5]
Inference: 17%|█▋ | 996/5920 [13:11<1:20:29, 1.02it/s, tok/s=13947.7]
Inference: 17%|█▋ | 997/5920 [13:13<1:39:18, 1.21s/it, tok/s=13947.7]
Inference: 17%|█▋ | 997/5920 [13:13<1:39:18, 1.21s/it, tok/s=13921.6]
Inference: 17%|█▋ | 998/5920 [13:13<1:39:17, 1.21s/it, tok/s=13936.4]
Inference: 17%|█▋ | 999/5920 [13:14<1:21:10, 1.01it/s, tok/s=13936.4]
Inference: 17%|█▋ | 999/5920 [13:14<1:21:10, 1.01it/s, tok/s=13923.7]
Inference: 17%|█▋ | 1000/5920 [13:14<1:07:48, 1.21it/s, tok/s=13923.7]
Inference: 17%|█▋ | 1000/5920 [13:14<1:07:48, 1.21it/s, tok/s=13944.7]
Inference: 17%|█▋ | 1001/5920 [13:18<2:04:34, 1.52s/it, tok/s=13944.7]
Inference: 17%|█▋ | 1001/5920 [13:18<2:04:34, 1.52s/it, tok/s=13903.7]
Inference: 17%|█▋ | 1002/5920 [13:18<1:37:43, 1.19s/it, tok/s=13903.7]
Inference: 17%|█▋ | 1002/5920 [13:18<1:37:43, 1.19s/it, tok/s=13907.7]
Inference: 17%|█▋ | 1003/5920 [13:19<1:22:54, 1.01s/it, tok/s=13907.7]
Inference: 17%|█▋ | 1003/5920 [13:19<1:22:54, 1.01s/it, tok/s=13920.2]
Inference: 17%|█▋ | 1004/5920 [13:19<1:03:46, 1.28it/s, tok/s=13920.2]
Inference: 17%|█▋ | 1004/5920 [13:19<1:03:46, 1.28it/s, tok/s=13926.8]
Inference: 17%|█▋ | 1005/5920 [13:20<59:02, 1.39it/s, tok/s=13926.8]
Inference: 17%|█▋ | 1005/5920 [13:20<59:02, 1.39it/s, tok/s=13924.8]
Inference: 17%|█▋ | 1006/5920 [13:20<53:03, 1.54it/s, tok/s=13924.8]
Inference: 17%|█▋ | 1006/5920 [13:20<53:03, 1.54it/s, tok/s=13924.8]
Inference: 17%|█▋ | 1007/5920 [13:21<48:01, 1.70it/s, tok/s=13924.8]
Inference: 17%|█▋ | 1007/5920 [13:21<48:01, 1.70it/s, tok/s=13934.7]
Inference: 17%|█▋ | 1008/5920 [13:21<36:51, 2.22it/s, tok/s=13934.7]
Inference: 17%|█▋ | 1008/5920 [13:21<36:51, 2.22it/s, tok/s=13958.6]
Inference: 17%|█▋ | 1009/5920 [13:21<29:59, 2.73it/s, tok/s=13958.6]
Inference: 17%|█▋ | 1009/5920 [13:21<29:59, 2.73it/s, tok/s=13963.9]
Inference: 17%|█▋ | 1010/5920 [13:21<24:24, 3.35it/s, tok/s=13963.9]
Inference: 17%|█▋ | 1010/5920 [13:21<24:24, 3.35it/s, tok/s=13981.1]
Inference: 17%|█▋ | 1011/5920 [13:22<36:02, 2.27it/s, tok/s=13981.1]
Inference: 17%|█▋ | 1011/5920 [13:22<36:02, 2.27it/s, tok/s=13984.2]
Inference: 17%|█▋ | 1012/5920 [13:22<30:53, 2.65it/s, tok/s=13984.2]
Inference: 17%|█▋ | 1012/5920 [13:22<30:53, 2.65it/s, tok/s=13989.6]
Inference: 17%|█▋ | 1013/5920 [13:22<27:03, 3.02it/s, tok/s=13989.6]
Inference: 17%|█▋ | 1013/5920 [13:22<27:03, 3.02it/s, tok/s=14011.5]
Inference: 17%|█▋ | 1014/5920 [13:23<30:53, 2.65it/s, tok/s=14011.5]
Inference: 17%|█▋ | 1014/5920 [13:23<30:53, 2.65it/s, tok/s=14022.7]
Inference: 17%|█▋ | 1015/5920 [13:26<1:46:11, 1.30s/it, tok/s=14022.7]
Inference: 17%|█▋ | 1015/5920 [13:26<1:46:11, 1.30s/it, tok/s=13977.3]
Inference: 17%|█▋ | 1016/5920 [13:26<1:22:56, 1.01s/it, tok/s=13977.3]
Inference: 17%|█▋ | 1016/5920 [13:26<1:22:56, 1.01s/it, tok/s=13981.4]
Inference: 17%|█▋ | 1017/5920 [13:29<1:56:50, 1.43s/it, tok/s=13981.4]
Inference: 17%|█▋ | 1017/5920 [13:29<1:56:50, 1.43s/it, tok/s=13957.7]
Inference: 17%|█▋ | 1018/5920 [13:30<1:48:39, 1.33s/it, tok/s=13957.7]
Inference: 17%|█▋ | 1018/5920 [13:30<1:48:39, 1.33s/it, tok/s=13947.3]
Inference: 17%|█▋ | 1019/5920 [13:30<1:48:37, 1.33s/it, tok/s=13963.9]
Inference: 17%|█▋ | 1020/5920 [13:30<1:00:29, 1.35it/s, tok/s=13963.9]
Inference: 17%|█▋ | 1020/5920 [13:30<1:00:29, 1.35it/s, tok/s=13969.8]
Inference: 17%|█▋ | 1021/5920 [13:33<1:47:18, 1.31s/it, tok/s=13969.8]
Inference: 17%|█▋ | 1021/5920 [13:33<1:47:18, 1.31s/it, tok/s=13924.9]
Inference: 17%|█▋ | 1022/5920 [13:35<2:02:32, 1.50s/it, tok/s=13924.9]
Inference: 17%|█▋ | 1022/5920 [13:35<2:02:32, 1.50s/it, tok/s=13898.9]
Inference: 17%|█▋ | 1023/5920 [13:37<2:14:29, 1.65s/it, tok/s=13898.9]
Inference: 17%|█▋ | 1023/5920 [13:37<2:14:29, 1.65s/it, tok/s=13873.3]
Inference: 17%|█▋ | 1024/5920 [13:38<1:50:47, 1.36s/it, tok/s=13873.3]
Inference: 17%|█▋ | 1024/5920 [13:38<1:50:47, 1.36s/it, tok/s=13870.8]
Inference: 17%|█▋ | 1025/5920 [13:39<1:46:05, 1.30s/it, tok/s=13870.8]
Inference: 17%|█▋ | 1025/5920 [13:39<1:46:05, 1.30s/it, tok/s=13861.1]
Inference: 17%|█▋ | 1026/5920 [13:39<1:46:03, 1.30s/it, tok/s=13869.7]
Inference: 17%|█▋ | 1027/5920 [13:39<1:04:46, 1.26it/s, tok/s=13869.7]
Inference: 17%|█▋ | 1027/5920 [13:39<1:04:46, 1.26it/s, tok/s=13876.0]
Inference: 17%|█▋ | 1028/5920 [13:39<1:04:45, 1.26it/s, tok/s=13883.9]
Inference: 17%|█▋ | 1029/5920 [13:40<48:12, 1.69it/s, tok/s=13883.9]
Inference: 17%|█▋ | 1029/5920 [13:40<48:12, 1.69it/s, tok/s=13883.9]
Inference: 17%|█▋ | 1030/5920 [13:40<42:08, 1.93it/s, tok/s=13883.9]
Inference: 17%|█▋ | 1030/5920 [13:40<42:08, 1.93it/s, tok/s=13899.3]
Inference: 17%|█▋ | 1031/5920 [13:43<1:20:43, 1.01it/s, tok/s=13899.3]
Inference: 17%|█▋ | 1031/5920 [13:43<1:20:43, 1.01it/s, tok/s=13877.3]
Inference: 17%|█▋ | 1032/5920 [13:43<1:11:11, 1.14it/s, tok/s=13877.3]
Inference: 17%|█▋ | 1032/5920 [13:43<1:11:11, 1.14it/s, tok/s=13891.7]
Inference: 17%|█▋ | 1033/5920 [13:43<1:11:11, 1.14it/s, tok/s=13900.8]
Inference: 17%|█▋ | 1034/5920 [13:43<44:05, 1.85it/s, tok/s=13900.8]
Inference: 17%|█▋ | 1034/5920 [13:43<44:05, 1.85it/s, tok/s=13907.2]
Inference: 17%|█▋ | 1035/5920 [13:45<1:03:05, 1.29it/s, tok/s=13907.2]
Inference: 17%|█▋ | 1035/5920 [13:45<1:03:05, 1.29it/s, tok/s=13891.2]
Inference: 18%|█▊ | 1036/5920 [13:46<1:05:53, 1.24it/s, tok/s=13891.2]
Inference: 18%|█▊ | 1036/5920 [13:46<1:05:53, 1.24it/s, tok/s=13901.0]
Inference: 18%|█▊ | 1037/5920 [13:46<53:24, 1.52it/s, tok/s=13901.0]
Inference: 18%|█▊ | 1037/5920 [13:46<53:24, 1.52it/s, tok/s=13906.0]
Inference: 18%|█▊ | 1038/5920 [13:47<1:09:42, 1.17it/s, tok/s=13906.0]
Inference: 18%|█▊ | 1038/5920 [13:47<1:09:42, 1.17it/s, tok/s=13891.9]
Inference: 18%|█▊ | 1039/5920 [13:48<1:02:23, 1.30it/s, tok/s=13891.9]
Inference: 18%|█▊ | 1039/5920 [13:48<1:02:23, 1.30it/s, tok/s=13891.1]
Inference: 18%|█▊ | 1040/5920 [13:48<47:40, 1.71it/s, tok/s=13891.1]
Inference: 18%|█▊ | 1040/5920 [13:48<47:40, 1.71it/s, tok/s=13909.3]
Inference: 18%|█▊ | 1041/5920 [13:48<38:06, 2.13it/s, tok/s=13909.3]
Inference: 18%|█▊ | 1041/5920 [13:48<38:06, 2.13it/s, tok/s=13924.0]
Inference: 18%|█▊ | 1042/5920 [13:49<35:58, 2.26it/s, tok/s=13924.0]
Inference: 18%|█▊ | 1042/5920 [13:49<35:58, 2.26it/s, tok/s=13940.1]
Inference: 18%|█▊ | 1043/5920 [13:49<35:58, 2.26it/s, tok/s=13958.6]
Inference: 18%|█▊ | 1044/5920 [13:49<26:18, 3.09it/s, tok/s=13958.6]
Inference: 18%|█▊ | 1044/5920 [13:49<26:18, 3.09it/s, tok/s=13963.2]
Inference: 18%|█▊ | 1045/5920 [13:50<34:39, 2.34it/s, tok/s=13963.2]
Inference: 18%|█▊ | 1045/5920 [13:50<34:39, 2.34it/s, tok/s=13975.8]
Inference: 18%|█▊ | 1046/5920 [13:50<35:39, 2.28it/s, tok/s=13975.8]
Inference: 18%|█▊ | 1046/5920 [13:50<35:39, 2.28it/s, tok/s=13992.8]
Inference: 18%|█▊ | 1047/5920 [13:51<45:44, 1.78it/s, tok/s=13992.8]
Inference: 18%|█▊ | 1047/5920 [13:51<45:44, 1.78it/s, tok/s=13987.8]
Inference: 18%|█▊ | 1048/5920 [13:52<47:12, 1.72it/s, tok/s=13987.8]
Inference: 18%|█▊ | 1048/5920 [13:52<47:12, 1.72it/s, tok/s=13986.4]
Inference: 18%|█▊ | 1049/5920 [13:53<1:04:59, 1.25it/s, tok/s=13986.4]
Inference: 18%|█▊ | 1049/5920 [13:53<1:04:59, 1.25it/s, tok/s=13979.3]
Inference: 18%|█▊ | 1050/5920 [13:53<49:22, 1.64it/s, tok/s=13979.3]
Inference: 18%|█▊ | 1050/5920 [13:53<49:22, 1.64it/s, tok/s=13994.3]
Inference: 18%|█▊ | 1051/5920 [13:53<40:50, 1.99it/s, tok/s=13994.3]
Inference: 18%|█▊ | 1051/5920 [13:53<40:50, 1.99it/s, tok/s=14007.1]
Inference: 18%|█▊ | 1052/5920 [13:54<36:37, 2.21it/s, tok/s=14007.1]
Inference: 18%|█▊ | 1052/5920 [13:54<36:37, 2.21it/s, tok/s=14016.6]
Inference: 18%|█▊ | 1053/5920 [13:55<44:02, 1.84it/s, tok/s=14016.6]
Inference: 18%|█▊ | 1053/5920 [13:55<44:02, 1.84it/s, tok/s=14024.1]
Inference: 18%|█▊ | 1054/5920 [13:56<54:06, 1.50it/s, tok/s=14024.1]
Inference: 18%|█▊ | 1054/5920 [13:56<54:06, 1.50it/s, tok/s=14016.0]
Inference: 18%|█▊ | 1055/5920 [13:57<1:19:06, 1.03it/s, tok/s=14016.0]
Inference: 18%|█▊ | 1055/5920 [13:57<1:19:06, 1.03it/s, tok/s=13997.6]
Inference: 18%|█▊ | 1056/5920 [13:58<1:05:49, 1.23it/s, tok/s=13997.6]
Inference: 18%|█▊ | 1056/5920 [13:58<1:05:49, 1.23it/s, tok/s=13999.3]
Inference: 18%|█▊ | 1057/5920 [13:58<52:57, 1.53it/s, tok/s=13999.3]
Inference: 18%|█▊ | 1057/5920 [13:58<52:57, 1.53it/s, tok/s=14020.7]
Inference: 18%|█▊ | 1058/5920 [13:59<59:59, 1.35it/s, tok/s=14020.7]
Inference: 18%|█▊ | 1058/5920 [13:59<59:59, 1.35it/s, tok/s=14014.0]
Inference: 18%|█▊ | 1059/5920 [13:59<46:24, 1.75it/s, tok/s=14014.0]
Inference: 18%|█▊ | 1059/5920 [13:59<46:24, 1.75it/s, tok/s=14026.2]
Inference: 18%|█▊ | 1060/5920 [14:00<56:36, 1.43it/s, tok/s=14026.2]
Inference: 18%|█▊ | 1060/5920 [14:00<56:36, 1.43it/s, tok/s=14018.4]
Inference: 18%|█▊ | 1061/5920 [14:00<56:35, 1.43it/s, tok/s=14050.1]
Inference: 18%|█▊ | 1062/5920 [14:00<34:31, 2.34it/s, tok/s=14050.1]
Inference: 18%|█▊ | 1062/5920 [14:00<34:31, 2.34it/s, tok/s=14070.1]
Inference: 18%|█▊ | 1063/5920 [14:02<1:09:24, 1.17it/s, tok/s=14070.1]
Inference: 18%|█▊ | 1063/5920 [14:02<1:09:24, 1.17it/s, tok/s=14051.7]
Inference: 18%|█▊ | 1064/5920 [14:04<1:17:15, 1.05it/s, tok/s=14051.7]
Inference: 18%|█▊ | 1064/5920 [14:04<1:17:15, 1.05it/s, tok/s=14046.3]
Inference: 18%|█▊ | 1065/5920 [14:07<2:20:44, 1.74s/it, tok/s=14046.3]
Inference: 18%|█▊ | 1065/5920 [14:07<2:20:44, 1.74s/it, tok/s=13993.4]
Inference: 18%|█▊ | 1066/5920 [14:10<2:40:56, 1.99s/it, tok/s=13993.4]
Inference: 18%|█▊ | 1066/5920 [14:10<2:40:56, 1.99s/it, tok/s=13970.1]
Inference: 18%|█▊ | 1067/5920 [14:11<2:09:37, 1.60s/it, tok/s=13970.1]
Inference: 18%|█▊ | 1067/5920 [14:11<2:09:37, 1.60s/it, tok/s=13969.7]
Inference: 18%|█▊ | 1068/5920 [14:11<1:42:39, 1.27s/it, tok/s=13969.7]
Inference: 18%|█▊ | 1068/5920 [14:11<1:42:39, 1.27s/it, tok/s=13971.0]
Inference: 18%|█▊ | 1069/5920 [14:12<1:23:10, 1.03s/it, tok/s=13971.0]
Inference: 18%|█▊ | 1069/5920 [14:12<1:23:10, 1.03s/it, tok/s=13992.4]
Inference: 18%|█▊ | 1070/5920 [14:16<2:35:32, 1.92s/it, tok/s=13992.4]
Inference: 18%|█▊ | 1070/5920 [14:16<2:35:32, 1.92s/it, tok/s=13935.7]
Inference: 18%|█▊ | 1071/5920 [14:17<2:17:57, 1.71s/it, tok/s=13935.7]
Inference: 18%|█▊ | 1071/5920 [14:17<2:17:57, 1.71s/it, tok/s=13922.7]
Inference: 18%|█▊ | 1072/5920 [14:18<2:05:34, 1.55s/it, tok/s=13922.7]
Inference: 18%|█▊ | 1072/5920 [14:18<2:05:34, 1.55s/it, tok/s=13915.2]
Inference: 18%|█▊ | 1073/5920 [14:18<2:05:32, 1.55s/it, tok/s=13932.1]
Inference: 18%|█▊ | 1074/5920 [14:20<1:34:10, 1.17s/it, tok/s=13932.1]
Inference: 18%|█▊ | 1074/5920 [14:20<1:34:10, 1.17s/it, tok/s=13929.5]
Inference: 18%|█▊ | 1075/5920 [14:21<1:39:16, 1.23s/it, tok/s=13929.5]
Inference: 18%|█▊ | 1075/5920 [14:21<1:39:16, 1.23s/it, tok/s=13938.8]
Inference: 18%|█▊ | 1076/5920 [14:22<1:29:14, 1.11s/it, tok/s=13938.8]
Inference: 18%|█▊ | 1076/5920 [14:22<1:29:14, 1.11s/it, tok/s=13949.9]
Inference: 18%|█▊ | 1077/5920 [14:23<1:29:58, 1.11s/it, tok/s=13949.9]
Inference: 18%|█▊ | 1077/5920 [14:23<1:29:58, 1.11s/it, tok/s=13943.3]
Inference: 18%|█▊ | 1078/5920 [14:23<1:09:03, 1.17it/s, tok/s=13943.3]
Inference: 18%|█▊ | 1078/5920 [14:23<1:09:03, 1.17it/s, tok/s=13957.4]
Inference: 18%|█▊ | 1079/5920 [14:23<57:20, 1.41it/s, tok/s=13957.4]
Inference: 18%|█▊ | 1079/5920 [14:23<57:20, 1.41it/s, tok/s=13967.2]
Inference: 18%|█▊ | 1080/5920 [14:26<1:31:21, 1.13s/it, tok/s=13967.2]
Inference: 18%|█▊ | 1080/5920 [14:26<1:31:21, 1.13s/it, tok/s=13958.5]
Inference: 18%|█▊ | 1081/5920 [14:27<1:46:32, 1.32s/it, tok/s=13958.5]
Inference: 18%|█▊ | 1081/5920 [14:27<1:46:32, 1.32s/it, tok/s=13950.3]
Inference: 18%|█▊ | 1082/5920 [14:28<1:20:53, 1.00s/it, tok/s=13950.3]
Inference: 18%|█▊ | 1082/5920 [14:28<1:20:53, 1.00s/it, tok/s=13955.5]
Inference: 18%|█▊ | 1083/5920 [14:28<1:17:52, 1.04it/s, tok/s=13955.5]
Inference: 18%|█▊ | 1083/5920 [14:28<1:17:52, 1.04it/s, tok/s=13971.4]
Inference: 18%|█▊ | 1084/5920 [14:29<1:00:49, 1.32it/s, tok/s=13971.4]
Inference: 18%|█▊ | 1084/5920 [14:29<1:00:49, 1.32it/s, tok/s=13994.1]
Inference: 18%|█▊ | 1085/5920 [14:30<1:22:13, 1.02s/it, tok/s=13994.1]
Inference: 18%|█▊ | 1085/5920 [14:30<1:22:13, 1.02s/it, tok/s=13985.3]
Inference: 18%|█▊ | 1086/5920 [14:32<1:29:34, 1.11s/it, tok/s=13985.3]
Inference: 18%|█▊ | 1086/5920 [14:32<1:29:34, 1.11s/it, tok/s=13983.9]
Inference: 18%|█▊ | 1087/5920 [14:34<2:03:15, 1.53s/it, tok/s=13983.9]
Inference: 18%|█▊ | 1087/5920 [14:34<2:03:15, 1.53s/it, tok/s=13968.3]
Inference: 18%|█▊ | 1088/5920 [14:36<2:01:53, 1.51s/it, tok/s=13968.3]
Inference: 18%|█▊ | 1088/5920 [14:36<2:01:53, 1.51s/it, tok/s=13956.0]
Inference: 18%|█▊ | 1089/5920 [14:38<2:30:42, 1.87s/it, tok/s=13956.0]
Inference: 18%|█▊ | 1089/5920 [14:38<2:30:42, 1.87s/it, tok/s=13924.5]
Inference: 18%|█▊ | 1090/5920 [14:39<1:56:09, 1.44s/it, tok/s=13924.5]
Inference: 18%|█▊ | 1090/5920 [14:39<1:56:09, 1.44s/it, tok/s=13938.6]
Inference: 18%|█▊ | 1091/5920 [14:40<1:45:47, 1.31s/it, tok/s=13938.6]
Inference: 18%|█▊ | 1091/5920 [14:40<1:45:47, 1.31s/it, tok/s=13939.9]
Inference: 18%|█▊ | 1092/5920 [14:40<1:22:20, 1.02s/it, tok/s=13939.9]
Inference: 18%|█▊ | 1092/5920 [14:40<1:22:20, 1.02s/it, tok/s=13952.2]
Inference: 18%|█▊ | 1093/5920 [14:41<1:21:07, 1.01s/it, tok/s=13952.2]
Inference: 18%|█▊ | 1093/5920 [14:41<1:21:07, 1.01s/it, tok/s=13943.2]
Inference: 18%|█▊ | 1094/5920 [14:41<1:03:50, 1.26it/s, tok/s=13943.2]
Inference: 18%|█▊ | 1094/5920 [14:41<1:03:50, 1.26it/s, tok/s=13949.7]
Inference: 18%|█▊ | 1095/5920 [14:44<1:46:50, 1.33s/it, tok/s=13949.7]
Inference: 18%|█▊ | 1095/5920 [14:44<1:46:50, 1.33s/it, tok/s=13932.6]
Inference: 19%|█▊ | 1096/5920 [14:44<1:18:35, 1.02it/s, tok/s=13932.6]
Inference: 19%|█▊ | 1096/5920 [14:44<1:18:35, 1.02it/s, tok/s=13954.5]
Inference: 19%|█▊ | 1097/5920 [14:45<1:03:37, 1.26it/s, tok/s=13954.5]
Inference: 19%|█▊ | 1097/5920 [14:45<1:03:37, 1.26it/s, tok/s=13970.7]
Inference: 19%|█▊ | 1098/5920 [14:45<57:07, 1.41it/s, tok/s=13970.7]
Inference: 19%|█▊ | 1098/5920 [14:45<57:07, 1.41it/s, tok/s=13979.6]
Inference: 19%|█▊ | 1099/5920 [14:45<44:30, 1.81it/s, tok/s=13979.6]
Inference: 19%|█▊ | 1099/5920 [14:45<44:30, 1.81it/s, tok/s=13989.9]
Inference: 19%|█▊ | 1100/5920 [14:46<39:48, 2.02it/s, tok/s=13989.9]
Inference: 19%|█▊ | 1100/5920 [14:46<39:48, 2.02it/s, tok/s=14001.8]
Inference: 19%|█▊ | 1101/5920 [14:47<1:10:13, 1.14it/s, tok/s=14001.8]
Inference: 19%|█▊ | 1101/5920 [14:47<1:10:13, 1.14it/s, tok/s=13984.3]
Inference: 19%|█▊ | 1102/5920 [14:47<1:10:13, 1.14it/s, tok/s=14006.0]
Inference: 19%|█▊ | 1103/5920 [14:50<1:24:22, 1.05s/it, tok/s=14006.0]
Inference: 19%|█▊ | 1103/5920 [14:50<1:24:22, 1.05s/it, tok/s=13990.5]
Inference: 19%|█▊ | 1104/5920 [14:51<1:25:49, 1.07s/it, tok/s=13990.5]
Inference: 19%|█▊ | 1104/5920 [14:51<1:25:49, 1.07s/it, tok/s=13983.7]
Inference: 19%|█▊ | 1105/5920 [14:53<1:37:44, 1.22s/it, tok/s=13983.7]
Inference: 19%|█▊ | 1105/5920 [14:53<1:37:44, 1.22s/it, tok/s=13976.5]
Inference: 19%|█▊ | 1106/5920 [14:55<2:06:00, 1.57s/it, tok/s=13976.5]
Inference: 19%|█▊ | 1106/5920 [14:55<2:06:00, 1.57s/it, tok/s=13949.2]
Inference: 19%|█▊ | 1107/5920 [14:56<1:52:10, 1.40s/it, tok/s=13949.2]
Inference: 19%|█▊ | 1107/5920 [14:56<1:52:10, 1.40s/it, tok/s=13965.2]
Inference: 19%|█▊ | 1108/5920 [14:57<1:48:54, 1.36s/it, tok/s=13965.2]
Inference: 19%|█▊ | 1108/5920 [14:57<1:48:54, 1.36s/it, tok/s=13963.8]
Inference: 19%|█▊ | 1109/5920 [14:58<1:38:28, 1.23s/it, tok/s=13963.8]
Inference: 19%|█▊ | 1109/5920 [14:58<1:38:28, 1.23s/it, tok/s=13968.4]
Inference: 19%|█▉ | 1110/5920 [14:59<1:29:46, 1.12s/it, tok/s=13968.4]
Inference: 19%|█▉ | 1110/5920 [14:59<1:29:46, 1.12s/it, tok/s=13968.3]
Inference: 19%|█▉ | 1111/5920 [15:01<1:53:17, 1.41s/it, tok/s=13968.3]
Inference: 19%|█▉ | 1111/5920 [15:01<1:53:17, 1.41s/it, tok/s=13954.1]
Inference: 19%|█▉ | 1112/5920 [15:07<3:39:17, 2.74s/it, tok/s=13954.1]
Inference: 19%|█▉ | 1112/5920 [15:07<3:39:17, 2.74s/it, tok/s=13885.1]
Inference: 19%|█▉ | 1113/5920 [15:08<3:05:38, 2.32s/it, tok/s=13885.1]
Inference: 19%|█▉ | 1113/5920 [15:08<3:05:38, 2.32s/it, tok/s=13890.8]
Inference: 19%|█▉ | 1114/5920 [15:09<2:24:54, 1.81s/it, tok/s=13890.8]
Inference: 19%|█▉ | 1114/5920 [15:09<2:24:54, 1.81s/it, tok/s=13900.9]
Inference: 19%|█▉ | 1115/5920 [15:11<2:28:38, 1.86s/it, tok/s=13900.9]
Inference: 19%|█▉ | 1115/5920 [15:11<2:28:38, 1.86s/it, tok/s=13884.4]
Inference: 19%|█▉ | 1116/5920 [15:11<2:28:36, 1.86s/it, tok/s=13904.2]
Inference: 19%|█▉ | 1117/5920 [15:12<1:31:06, 1.14s/it, tok/s=13904.2]
Inference: 19%|█▉ | 1117/5920 [15:12<1:31:06, 1.14s/it, tok/s=13920.0]
Inference: 19%|█▉ | 1118/5920 [15:13<1:32:17, 1.15s/it, tok/s=13920.0]
Inference: 19%|█▉ | 1118/5920 [15:13<1:32:17, 1.15s/it, tok/s=13924.8]
Inference: 19%|█▉ | 1119/5920 [15:14<1:43:15, 1.29s/it, tok/s=13924.8]
Inference: 19%|█▉ | 1119/5920 [15:14<1:43:15, 1.29s/it, tok/s=13924.9]
Inference: 19%|█▉ | 1120/5920 [15:16<1:41:27, 1.27s/it, tok/s=13924.9]
Inference: 19%|█▉ | 1120/5920 [15:16<1:41:27, 1.27s/it, tok/s=13920.0]
Inference: 19%|█▉ | 1121/5920 [15:20<2:48:39, 2.11s/it, tok/s=13920.0]
Inference: 19%|█▉ | 1121/5920 [15:20<2:48:39, 2.11s/it, tok/s=13880.6]
Inference: 19%|█▉ | 1122/5920 [15:20<2:48:37, 2.11s/it, tok/s=13899.2]
Inference: 19%|█▉ | 1123/5920 [15:22<2:12:57, 1.66s/it, tok/s=13899.2]
Inference: 19%|█▉ | 1123/5920 [15:22<2:12:57, 1.66s/it, tok/s=13888.2]
Inference: 19%|█▉ | 1124/5920 [15:24<2:15:56, 1.70s/it, tok/s=13888.2]
Inference: 19%|█▉ | 1124/5920 [15:24<2:15:56, 1.70s/it, tok/s=13887.2]
Inference: 19%|█▉ | 1125/5920 [15:25<2:01:29, 1.52s/it, tok/s=13887.2]
Inference: 19%|█▉ | 1125/5920 [15:25<2:01:29, 1.52s/it, tok/s=13879.3]
Inference: 19%|█▉ | 1126/5920 [15:26<1:57:54, 1.48s/it, tok/s=13879.3]
Inference: 19%|█▉ | 1126/5920 [15:26<1:57:54, 1.48s/it, tok/s=13874.2]
Inference: 19%|█▉ | 1127/5920 [15:26<1:57:53, 1.48s/it, tok/s=13886.0]
Inference: 19%|█▉ | 1128/5920 [15:27<1:21:59, 1.03s/it, tok/s=13886.0]
Inference: 19%|█▉ | 1128/5920 [15:27<1:21:59, 1.03s/it, tok/s=13894.4]
Inference: 19%|█▉ | 1129/5920 [15:28<1:09:53, 1.14it/s, tok/s=13894.4]
Inference: 19%|█▉ | 1129/5920 [15:28<1:09:53, 1.14it/s, tok/s=13907.0]
Inference: 19%|█▉ | 1130/5920 [15:28<1:03:46, 1.25it/s, tok/s=13907.0]
Inference: 19%|█▉ | 1130/5920 [15:28<1:03:46, 1.25it/s, tok/s=13923.9]
Inference: 19%|█▉ | 1131/5920 [15:29<57:13, 1.39it/s, tok/s=13923.9]
Inference: 19%|█▉ | 1131/5920 [15:29<57:13, 1.39it/s, tok/s=13940.8]
Inference: 19%|█▉ | 1132/5920 [15:30<1:18:41, 1.01it/s, tok/s=13940.8]
Inference: 19%|█▉ | 1132/5920 [15:30<1:18:41, 1.01it/s, tok/s=13934.7]
Inference: 19%|█▉ | 1133/5920 [15:32<1:36:04, 1.20s/it, tok/s=13934.7]
Inference: 19%|█▉ | 1133/5920 [15:32<1:36:04, 1.20s/it, tok/s=13929.1]
Inference: 19%|█▉ | 1134/5920 [15:35<2:20:20, 1.76s/it, tok/s=13929.1]
Inference: 19%|█▉ | 1134/5920 [15:35<2:20:20, 1.76s/it, tok/s=13909.7]
Inference: 19%|█▉ | 1135/5920 [15:36<1:48:32, 1.36s/it, tok/s=13909.7]
Inference: 19%|█▉ | 1135/5920 [15:36<1:48:32, 1.36s/it, tok/s=13921.0]
Inference: 19%|█▉ | 1136/5920 [15:36<1:28:43, 1.11s/it, tok/s=13921.0]
Inference: 19%|█▉ | 1136/5920 [15:36<1:28:43, 1.11s/it, tok/s=13921.0]
Inference: 19%|█▉ | 1137/5920 [15:37<1:13:50, 1.08it/s, tok/s=13921.0]
Inference: 19%|█▉ | 1137/5920 [15:37<1:13:50, 1.08it/s, tok/s=13924.9]
Inference: 19%|█▉ | 1138/5920 [15:38<1:25:02, 1.07s/it, tok/s=13924.9]
Inference: 19%|█▉ | 1138/5920 [15:38<1:25:02, 1.07s/it, tok/s=13916.1]
Inference: 19%|█▉ | 1139/5920 [15:39<1:32:02, 1.15s/it, tok/s=13916.1]
Inference: 19%|█▉ | 1139/5920 [15:39<1:32:02, 1.15s/it, tok/s=13903.6]
Inference: 19%|█▉ | 1140/5920 [15:40<1:16:13, 1.05it/s, tok/s=13903.6]
Inference: 19%|█▉ | 1140/5920 [15:40<1:16:13, 1.05it/s, tok/s=13920.3]
Inference: 19%|█▉ | 1141/5920 [15:40<1:00:11, 1.32it/s, tok/s=13920.3]
Inference: 19%|█▉ | 1141/5920 [15:40<1:00:11, 1.32it/s, tok/s=13929.8]
Inference: 19%|█▉ | 1142/5920 [15:42<1:31:21, 1.15s/it, tok/s=13929.8]
Inference: 19%|█▉ | 1142/5920 [15:42<1:31:21, 1.15s/it, tok/s=13920.4]
Inference: 19%|█▉ | 1143/5920 [15:43<1:10:28, 1.13it/s, tok/s=13920.4]
Inference: 19%|█▉ | 1143/5920 [15:43<1:10:28, 1.13it/s, tok/s=13929.6]
Inference: 19%|█▉ | 1144/5920 [15:43<54:23, 1.46it/s, tok/s=13929.6]
Inference: 19%|█▉ | 1144/5920 [15:43<54:23, 1.46it/s, tok/s=13941.6]
Inference: 19%|█▉ | 1145/5920 [15:45<1:22:25, 1.04s/it, tok/s=13941.6]
Inference: 19%|█▉ | 1145/5920 [15:45<1:22:25, 1.04s/it, tok/s=13937.0]
Inference: 19%|█▉ | 1146/5920 [15:49<2:31:54, 1.91s/it, tok/s=13937.0]
Inference: 19%|█▉ | 1146/5920 [15:49<2:31:54, 1.91s/it, tok/s=13895.0]
Inference: 19%|█▉ | 1147/5920 [15:49<1:54:01, 1.43s/it, tok/s=13895.0]
Inference: 19%|█▉ | 1147/5920 [15:49<1:54:01, 1.43s/it, tok/s=13904.0]
Inference: 19%|█▉ | 1148/5920 [15:49<1:54:00, 1.43s/it, tok/s=13916.9]
Inference: 19%|█▉ | 1149/5920 [15:49<1:10:00, 1.14it/s, tok/s=13916.9]
Inference: 19%|█▉ | 1149/5920 [15:49<1:10:00, 1.14it/s, tok/s=13932.1]
Inference: 19%|█▉ | 1150/5920 [15:51<1:18:01, 1.02it/s, tok/s=13932.1]
Inference: 19%|█▉ | 1150/5920 [15:51<1:18:01, 1.02it/s, tok/s=13927.6]
Inference: 19%|█▉ | 1151/5920 [15:52<1:35:24, 1.20s/it, tok/s=13927.6]
Inference: 19%|█▉ | 1151/5920 [15:52<1:35:24, 1.20s/it, tok/s=13919.7]
Inference: 19%|█▉ | 1152/5920 [15:53<1:15:48, 1.05it/s, tok/s=13919.7]
Inference: 19%|█▉ | 1152/5920 [15:53<1:15:48, 1.05it/s, tok/s=13941.8]
Inference: 19%|█▉ | 1153/5920 [15:53<1:05:57, 1.20it/s, tok/s=13941.8]
Inference: 19%|█▉ | 1153/5920 [15:53<1:05:57, 1.20it/s, tok/s=13954.7]
Inference: 19%|█▉ | 1154/5920 [15:56<1:55:17, 1.45s/it, tok/s=13954.7]
Inference: 19%|█▉ | 1154/5920 [15:56<1:55:17, 1.45s/it, tok/s=13931.7]
Inference: 20%|█▉ | 1155/5920 [15:58<2:04:11, 1.56s/it, tok/s=13931.7]
Inference: 20%|█▉ | 1155/5920 [15:58<2:04:11, 1.56s/it, tok/s=13922.0]
Inference: 20%|█▉ | 1156/5920 [15:58<1:33:03, 1.17s/it, tok/s=13922.0]
Inference: 20%|█▉ | 1156/5920 [15:58<1:33:03, 1.17s/it, tok/s=13946.0]
Inference: 20%|█▉ | 1157/5920 [16:00<1:35:08, 1.20s/it, tok/s=13946.0]
Inference: 20%|█▉ | 1157/5920 [16:00<1:35:08, 1.20s/it, tok/s=13936.9]
Inference: 20%|█▉ | 1158/5920 [16:01<1:31:56, 1.16s/it, tok/s=13936.9]
Inference: 20%|█▉ | 1158/5920 [16:01<1:31:56, 1.16s/it, tok/s=13930.8]
Inference: 20%|█▉ | 1159/5920 [16:06<2:59:46, 2.27s/it, tok/s=13930.8]
Inference: 20%|█▉ | 1159/5920 [16:06<2:59:46, 2.27s/it, tok/s=13887.6]
Inference: 20%|█▉ | 1160/5920 [16:08<2:52:41, 2.18s/it, tok/s=13887.6]
Inference: 20%|█▉ | 1160/5920 [16:08<2:52:41, 2.18s/it, tok/s=13868.3]
Inference: 20%|█▉ | 1161/5920 [16:09<2:35:35, 1.96s/it, tok/s=13868.3]
Inference: 20%|█▉ | 1161/5920 [16:09<2:35:35, 1.96s/it, tok/s=13866.8]
Inference: 20%|█▉ | 1162/5920 [16:09<1:51:34, 1.41s/it, tok/s=13866.8]
Inference: 20%|█▉ | 1162/5920 [16:09<1:51:34, 1.41s/it, tok/s=13884.7]
Inference: 20%|█▉ | 1163/5920 [16:09<1:51:32, 1.41s/it, tok/s=13893.9]
Inference: 20%|█▉ | 1164/5920 [16:09<1:02:53, 1.26it/s, tok/s=13893.9]
Inference: 20%|█▉ | 1164/5920 [16:09<1:02:53, 1.26it/s, tok/s=13919.5]
Inference: 20%|█▉ | 1165/5920 [16:15<2:38:22, 2.00s/it, tok/s=13919.5]
Inference: 20%|█▉ | 1165/5920 [16:15<2:38:22, 2.00s/it, tok/s=13857.6]
Inference: 20%|█▉ | 1166/5920 [16:18<3:09:17, 2.39s/it, tok/s=13857.6]
Inference: 20%|█▉ | 1166/5920 [16:18<3:09:17, 2.39s/it, tok/s=13827.9]
Inference: 20%|█▉ | 1167/5920 [16:20<2:58:30, 2.25s/it, tok/s=13827.9]
Inference: 20%|█▉ | 1167/5920 [16:20<2:58:30, 2.25s/it, tok/s=13825.0]
Inference: 20%|█▉ | 1168/5920 [16:22<2:52:40, 2.18s/it, tok/s=13825.0]
Inference: 20%|█▉ | 1168/5920 [16:22<2:52:40, 2.18s/it, tok/s=13823.7]
Inference: 20%|█▉ | 1169/5920 [16:22<2:06:23, 1.60s/it, tok/s=13823.7]
Inference: 20%|█▉ | 1169/5920 [16:22<2:06:23, 1.60s/it, tok/s=13830.7]
Inference: 20%|█▉ | 1170/5920 [16:23<1:34:29, 1.19s/it, tok/s=13830.7]
Inference: 20%|█▉ | 1170/5920 [16:23<1:34:29, 1.19s/it, tok/s=13841.6]
Inference: 20%|█▉ | 1171/5920 [16:25<1:58:16, 1.49s/it, tok/s=13841.6]
Inference: 20%|█▉ | 1171/5920 [16:25<1:58:16, 1.49s/it, tok/s=13832.4]
Inference: 20%|█▉ | 1172/5920 [16:26<1:45:33, 1.33s/it, tok/s=13832.4]
Inference: 20%|█▉ | 1172/5920 [16:26<1:45:33, 1.33s/it, tok/s=13828.5]
Inference: 20%|█▉ | 1173/5920 [16:26<1:21:39, 1.03s/it, tok/s=13828.5]
Inference: 20%|█▉ | 1173/5920 [16:26<1:21:39, 1.03s/it, tok/s=13849.9]
Inference: 20%|█▉ | 1174/5920 [16:27<1:24:39, 1.07s/it, tok/s=13849.9]
Inference: 20%|█▉ | 1174/5920 [16:27<1:24:39, 1.07s/it, tok/s=13842.8]
Inference: 20%|█▉ | 1175/5920 [16:29<1:31:58, 1.16s/it, tok/s=13842.8]
Inference: 20%|█▉ | 1175/5920 [16:29<1:31:58, 1.16s/it, tok/s=13832.9]
Inference: 20%|█▉ | 1176/5920 [16:29<1:31:57, 1.16s/it, tok/s=13840.9]
Inference: 20%|█▉ | 1177/5920 [16:32<1:44:59, 1.33s/it, tok/s=13840.9]
Inference: 20%|█▉ | 1177/5920 [16:32<1:44:59, 1.33s/it, tok/s=13813.0]
Inference: 20%|█▉ | 1178/5920 [16:34<2:08:43, 1.63s/it, tok/s=13813.0]
Inference: 20%|█▉ | 1178/5920 [16:34<2:08:43, 1.63s/it, tok/s=13798.1]
Inference: 20%|█▉ | 1179/5920 [16:34<2:08:41, 1.63s/it, tok/s=13821.0]
Inference: 20%|█▉ | 1180/5920 [16:34<1:18:16, 1.01it/s, tok/s=13821.0]
Inference: 20%|█▉ | 1180/5920 [16:34<1:18:16, 1.01it/s, tok/s=13828.0]
Inference: 20%|█▉ | 1181/5920 [16:35<1:18:19, 1.01it/s, tok/s=13828.0]
Inference: 20%|█▉ | 1181/5920 [16:35<1:18:19, 1.01it/s, tok/s=13823.5]
Inference: 20%|█▉ | 1182/5920 [16:39<2:11:24, 1.66s/it, tok/s=13823.5]
Inference: 20%|█▉ | 1182/5920 [16:39<2:11:24, 1.66s/it, tok/s=13780.1]
Inference: 20%|█▉ | 1183/5920 [16:39<1:41:47, 1.29s/it, tok/s=13780.1]
Inference: 20%|█▉ | 1183/5920 [16:39<1:41:47, 1.29s/it, tok/s=13802.6]
Inference: 20%|██ | 1184/5920 [16:39<1:41:45, 1.29s/it, tok/s=13831.4]
Inference: 20%|██ | 1185/5920 [16:41<1:29:28, 1.13s/it, tok/s=13831.4]
Inference: 20%|██ | 1185/5920 [16:41<1:29:28, 1.13s/it, tok/s=13814.7]
Inference: 20%|██ | 1186/5920 [16:41<1:29:27, 1.13s/it, tok/s=13829.1]
Inference: 20%|██ | 1187/5920 [16:42<1:03:26, 1.24it/s, tok/s=13829.1]
Inference: 20%|██ | 1187/5920 [16:42<1:03:26, 1.24it/s, tok/s=13835.8]
Inference: 20%|██ | 1188/5920 [16:43<1:13:33, 1.07it/s, tok/s=13835.8]
Inference: 20%|██ | 1188/5920 [16:43<1:13:33, 1.07it/s, tok/s=13830.9]
Inference: 20%|██ | 1189/5920 [16:43<1:00:10, 1.31it/s, tok/s=13830.9]
Inference: 20%|██ | 1189/5920 [16:43<1:00:10, 1.31it/s, tok/s=13846.5]
Inference: 20%|██ | 1190/5920 [16:45<1:27:34, 1.11s/it, tok/s=13846.5]
Inference: 20%|██ | 1190/5920 [16:45<1:27:34, 1.11s/it, tok/s=13839.3]
Inference: 20%|██ | 1191/5920 [16:47<1:39:48, 1.27s/it, tok/s=13839.3]
Inference: 20%|██ | 1191/5920 [16:47<1:39:48, 1.27s/it, tok/s=13832.4]
Inference: 20%|██ | 1192/5920 [16:48<1:21:56, 1.04s/it, tok/s=13832.4]
Inference: 20%|██ | 1192/5920 [16:48<1:21:56, 1.04s/it, tok/s=13848.8]
Inference: 20%|██ | 1193/5920 [16:50<1:52:35, 1.43s/it, tok/s=13848.8]
Inference: 20%|██ | 1193/5920 [16:50<1:52:35, 1.43s/it, tok/s=13824.3]
Inference: 20%|██ | 1194/5920 [16:51<1:46:19, 1.35s/it, tok/s=13824.3]
Inference: 20%|██ | 1194/5920 [16:51<1:46:19, 1.35s/it, tok/s=13832.1]
Inference: 20%|██ | 1195/5920 [16:53<1:57:25, 1.49s/it, tok/s=13832.1]
Inference: 20%|██ | 1195/5920 [16:53<1:57:25, 1.49s/it, tok/s=13823.6]
Inference: 20%|██ | 1196/5920 [16:57<3:00:59, 2.30s/it, tok/s=13823.6]
Inference: 20%|██ | 1196/5920 [16:57<3:00:59, 2.30s/it, tok/s=13790.0]
Inference: 20%|██ | 1197/5920 [17:03<4:24:18, 3.36s/it, tok/s=13790.0]
Inference: 20%|██ | 1197/5920 [17:03<4:24:18, 3.36s/it, tok/s=13733.7]
Inference: 20%|██ | 1198/5920 [17:05<3:41:27, 2.81s/it, tok/s=13733.7]
Inference: 20%|██ | 1198/5920 [17:05<3:41:27, 2.81s/it, tok/s=13723.0]
Inference: 20%|██ | 1199/5920 [17:07<3:38:07, 2.77s/it, tok/s=13723.0]
Inference: 20%|██ | 1199/5920 [17:07<3:38:07, 2.77s/it, tok/s=13707.6]
Inference: 20%|██ | 1200/5920 [17:08<2:37:31, 2.00s/it, tok/s=13707.6]
Inference: 20%|██ | 1200/5920 [17:08<2:37:31, 2.00s/it, tok/s=13724.8]
Inference: 20%|██ | 1201/5920 [17:09<2:22:25, 1.81s/it, tok/s=13724.8]
Inference: 20%|██ | 1201/5920 [17:09<2:22:25, 1.81s/it, tok/s=13727.2]
Inference: 20%|██ | 1202/5920 [17:10<1:53:49, 1.45s/it, tok/s=13727.2]
Inference: 20%|██ | 1202/5920 [17:10<1:53:49, 1.45s/it, tok/s=13740.1]
Inference: 20%|██ | 1203/5920 [17:11<1:43:46, 1.32s/it, tok/s=13740.1]
Inference: 20%|██ | 1203/5920 [17:11<1:43:46, 1.32s/it, tok/s=13735.0]
Inference: 20%|██ | 1204/5920 [17:11<1:21:14, 1.03s/it, tok/s=13735.0]
Inference: 20%|██ | 1204/5920 [17:11<1:21:14, 1.03s/it, tok/s=13740.9]
Inference: 20%|██ | 1205/5920 [17:14<1:59:09, 1.52s/it, tok/s=13740.9]
Inference: 20%|██ | 1205/5920 [17:14<1:59:09, 1.52s/it, tok/s=13729.7]
Inference: 20%|██ | 1206/5920 [17:17<2:44:29, 2.09s/it, tok/s=13729.7]
Inference: 20%|██ | 1206/5920 [17:17<2:44:29, 2.09s/it, tok/s=13695.7]
Inference: 20%|██ | 1207/5920 [17:17<2:01:07, 1.54s/it, tok/s=13695.7]
Inference: 20%|██ | 1207/5920 [17:17<2:01:07, 1.54s/it, tok/s=13702.3]
Inference: 20%|██ | 1208/5920 [17:18<1:39:52, 1.27s/it, tok/s=13702.3]
Inference: 20%|██ | 1208/5920 [17:18<1:39:52, 1.27s/it, tok/s=13702.5]
Inference: 20%|██ | 1209/5920 [17:18<1:23:44, 1.07s/it, tok/s=13702.5]
Inference: 20%|██ | 1209/5920 [17:18<1:23:44, 1.07s/it, tok/s=13713.2]
Inference: 20%|██ | 1210/5920 [17:23<2:36:36, 1.99s/it, tok/s=13713.2]
Inference: 20%|██ | 1210/5920 [17:23<2:36:36, 1.99s/it, tok/s=13669.0]
Inference: 20%|██ | 1211/5920 [17:25<2:40:01, 2.04s/it, tok/s=13669.0]
Inference: 20%|██ | 1211/5920 [17:25<2:40:01, 2.04s/it, tok/s=13667.4]
Inference: 20%|██ | 1212/5920 [17:25<2:39:59, 2.04s/it, tok/s=13692.8]
Inference: 20%|██ | 1213/5920 [17:28<2:22:02, 1.81s/it, tok/s=13692.8]
Inference: 20%|██ | 1213/5920 [17:28<2:22:02, 1.81s/it, tok/s=13662.6]
Inference: 21%|██ | 1214/5920 [17:29<2:08:18, 1.64s/it, tok/s=13662.6]
Inference: 21%|██ | 1214/5920 [17:29<2:08:18, 1.64s/it, tok/s=13657.6]
Inference: 21%|██ | 1215/5920 [17:30<2:04:59, 1.59s/it, tok/s=13657.6]
Inference: 21%|██ | 1215/5920 [17:30<2:04:59, 1.59s/it, tok/s=13651.2]
Inference: 21%|██ | 1216/5920 [17:31<1:46:56, 1.36s/it, tok/s=13651.2]
Inference: 21%|██ | 1216/5920 [17:31<1:46:56, 1.36s/it, tok/s=13666.6]
Inference: 21%|██ | 1217/5920 [17:34<2:22:37, 1.82s/it, tok/s=13666.6]
Inference: 21%|██ | 1217/5920 [17:34<2:22:37, 1.82s/it, tok/s=13656.4]
Inference: 21%|██ | 1218/5920 [17:35<1:49:30, 1.40s/it, tok/s=13656.4]
Inference: 21%|██ | 1218/5920 [17:35<1:49:30, 1.40s/it, tok/s=13676.1]
Inference: 21%|██ | 1219/5920 [17:37<2:16:22, 1.74s/it, tok/s=13676.1]
Inference: 21%|██ | 1219/5920 [17:37<2:16:22, 1.74s/it, tok/s=13649.6]
Inference: 21%|██ | 1220/5920 [17:38<1:49:26, 1.40s/it, tok/s=13649.6]
Inference: 21%|██ | 1220/5920 [17:38<1:49:26, 1.40s/it, tok/s=13653.0]
Inference: 21%|██ | 1221/5920 [17:38<1:35:06, 1.21s/it, tok/s=13653.0]
Inference: 21%|██ | 1221/5920 [17:38<1:35:06, 1.21s/it, tok/s=13670.5]
Inference: 21%|██ | 1222/5920 [17:38<1:35:04, 1.21s/it, tok/s=13682.2]
Inference: 21%|██ | 1223/5920 [17:42<1:48:31, 1.39s/it, tok/s=13682.2]
Inference: 21%|██ | 1223/5920 [17:42<1:48:31, 1.39s/it, tok/s=13662.7]
Inference: 21%|██ | 1224/5920 [17:44<2:12:37, 1.69s/it, tok/s=13662.7]
Inference: 21%|██ | 1224/5920 [17:44<2:12:37, 1.69s/it, tok/s=13657.2]
Inference: 21%|██ | 1225/5920 [17:45<1:45:20, 1.35s/it, tok/s=13657.2]
Inference: 21%|██ | 1225/5920 [17:45<1:45:20, 1.35s/it, tok/s=13664.1]
Inference: 21%|██ | 1226/5920 [17:45<1:19:26, 1.02s/it, tok/s=13664.1]
Inference: 21%|██ | 1226/5920 [17:45<1:19:26, 1.02s/it, tok/s=13672.7]
Inference: 21%|██ | 1227/5920 [17:45<1:05:55, 1.19it/s, tok/s=13672.7]
Inference: 21%|██ | 1227/5920 [17:45<1:05:55, 1.19it/s, tok/s=13675.4]
Inference: 21%|██ | 1228/5920 [17:47<1:20:41, 1.03s/it, tok/s=13675.4]
Inference: 21%|██ | 1228/5920 [17:47<1:20:41, 1.03s/it, tok/s=13663.3]
Inference: 21%|██ | 1229/5920 [17:47<1:15:15, 1.04it/s, tok/s=13663.3]
Inference: 21%|██ | 1229/5920 [17:47<1:15:15, 1.04it/s, tok/s=13678.3]
Inference: 21%|██ | 1230/5920 [17:49<1:22:51, 1.06s/it, tok/s=13678.3]
Inference: 21%|██ | 1230/5920 [17:49<1:22:51, 1.06s/it, tok/s=13671.5]
Inference: 21%|██ | 1231/5920 [17:50<1:27:34, 1.12s/it, tok/s=13671.5]
Inference: 21%|██ | 1231/5920 [17:50<1:27:34, 1.12s/it, tok/s=13678.2]
Inference: 21%|██ | 1232/5920 [17:51<1:26:53, 1.11s/it, tok/s=13678.2]
Inference: 21%|██ | 1232/5920 [17:51<1:26:53, 1.11s/it, tok/s=13691.9]
Inference: 21%|██ | 1233/5920 [17:51<1:07:42, 1.15it/s, tok/s=13691.9]
Inference: 21%|██ | 1233/5920 [17:51<1:07:42, 1.15it/s, tok/s=13716.4]
Inference: 21%|██ | 1234/5920 [17:52<58:18, 1.34it/s, tok/s=13716.4]
Inference: 21%|██ | 1234/5920 [17:52<58:18, 1.34it/s, tok/s=13721.7]
Inference: 21%|██ | 1235/5920 [17:53<56:26, 1.38it/s, tok/s=13721.7]
Inference: 21%|██ | 1235/5920 [17:53<56:26, 1.38it/s, tok/s=13739.5]
Inference: 21%|██ | 1236/5920 [17:56<1:58:29, 1.52s/it, tok/s=13739.5]
Inference: 21%|██ | 1236/5920 [17:56<1:58:29, 1.52s/it, tok/s=13711.7]
Inference: 21%|██ | 1237/5920 [17:57<1:43:39, 1.33s/it, tok/s=13711.7]
Inference: 21%|██ | 1237/5920 [17:57<1:43:39, 1.33s/it, tok/s=13707.0]
Inference: 21%|██ | 1238/5920 [17:57<1:43:38, 1.33s/it, tok/s=13715.7]
Inference: 21%|██ | 1239/5920 [17:57<1:05:46, 1.19it/s, tok/s=13715.7]
Inference: 21%|██ | 1239/5920 [17:57<1:05:46, 1.19it/s, tok/s=13717.9]
Inference: 21%|██ | 1240/5920 [17:58<1:00:36, 1.29it/s, tok/s=13717.9]
Inference: 21%|██ | 1240/5920 [17:58<1:00:36, 1.29it/s, tok/s=13722.1]
Inference: 21%|██ | 1241/5920 [18:00<1:34:38, 1.21s/it, tok/s=13722.1]
Inference: 21%|██ | 1241/5920 [18:00<1:34:38, 1.21s/it, tok/s=13702.4]
Inference: 21%|██ | 1242/5920 [18:01<1:14:21, 1.05it/s, tok/s=13702.4]
Inference: 21%|██ | 1242/5920 [18:01<1:14:21, 1.05it/s, tok/s=13727.0]
Inference: 21%|██ | 1243/5920 [18:01<57:50, 1.35it/s, tok/s=13727.0]
Inference: 21%|██ | 1243/5920 [18:01<57:50, 1.35it/s, tok/s=13735.4]
Inference: 21%|██ | 1244/5920 [18:01<48:51, 1.60it/s, tok/s=13735.4]
Inference: 21%|██ | 1244/5920 [18:01<48:51, 1.60it/s, tok/s=13754.6]
Inference: 21%|██ | 1245/5920 [18:01<41:36, 1.87it/s, tok/s=13754.6]
Inference: 21%|██ | 1245/5920 [18:01<41:36, 1.87it/s, tok/s=13765.9]
Inference: 21%|██ | 1246/5920 [18:05<1:43:12, 1.32s/it, tok/s=13765.9]
Inference: 21%|██ | 1246/5920 [18:05<1:43:12, 1.32s/it, tok/s=13745.7]
Inference: 21%|██ | 1247/5920 [18:06<1:42:29, 1.32s/it, tok/s=13745.7]
Inference: 21%|██ | 1247/5920 [18:06<1:42:29, 1.32s/it, tok/s=13754.4]
Inference: 21%|██ | 1248/5920 [18:06<1:42:27, 1.32s/it, tok/s=13763.3]
Inference: 21%|██ | 1249/5920 [18:06<58:45, 1.33it/s, tok/s=13763.3]
Inference: 21%|██ | 1249/5920 [18:06<58:45, 1.33it/s, tok/s=13782.5]
Inference: 21%|██ | 1250/5920 [18:07<1:00:46, 1.28it/s, tok/s=13782.5]
Inference: 21%|██ | 1250/5920 [18:07<1:00:46, 1.28it/s, tok/s=13799.3]
Inference: 21%|██ | 1251/5920 [18:09<1:23:01, 1.07s/it, tok/s=13799.3]
Inference: 21%|██ | 1251/5920 [18:09<1:23:01, 1.07s/it, tok/s=13791.1]
Inference: 21%|██ | 1252/5920 [18:13<2:23:36, 1.85s/it, tok/s=13791.1]
Inference: 21%|██ | 1252/5920 [18:13<2:23:36, 1.85s/it, tok/s=13762.9]
Inference: 21%|██ | 1253/5920 [18:14<2:17:16, 1.76s/it, tok/s=13762.9]
Inference: 21%|██ | 1253/5920 [18:14<2:17:16, 1.76s/it, tok/s=13764.2]
Inference: 21%|██ | 1254/5920 [18:15<1:42:16, 1.32s/it, tok/s=13764.2]
Inference: 21%|██ | 1254/5920 [18:15<1:42:16, 1.32s/it, tok/s=13784.1]
Inference: 21%|██ | 1255/5920 [18:18<2:35:50, 2.00s/it, tok/s=13784.1]
Inference: 21%|██ | 1255/5920 [18:18<2:35:50, 2.00s/it, tok/s=13763.8]
Inference: 21%|██ | 1256/5920 [18:19<2:06:16, 1.62s/it, tok/s=13763.8]
Inference: 21%|██ | 1256/5920 [18:19<2:06:16, 1.62s/it, tok/s=13774.9]
Inference: 21%|██ | 1257/5920 [18:20<1:46:56, 1.38s/it, tok/s=13774.9]
Inference: 21%|██ | 1257/5920 [18:20<1:46:56, 1.38s/it, tok/s=13789.6]
Inference: 21%|██▏ | 1258/5920 [18:21<1:33:23, 1.20s/it, tok/s=13789.6]
Inference: 21%|██▏ | 1258/5920 [18:21<1:33:23, 1.20s/it, tok/s=13787.6]
Inference: 21%|██▏ | 1259/5920 [18:21<1:21:08, 1.04s/it, tok/s=13787.6]
Inference: 21%|██▏ | 1259/5920 [18:21<1:21:08, 1.04s/it, tok/s=13790.6]
Inference: 21%|██▏ | 1260/5920 [18:23<1:27:00, 1.12s/it, tok/s=13790.6]
Inference: 21%|██▏ | 1260/5920 [18:23<1:27:00, 1.12s/it, tok/s=13782.7]
Inference: 21%|██▏ | 1261/5920 [18:23<1:26:59, 1.12s/it, tok/s=13805.7]
Inference: 21%|██▏ | 1262/5920 [18:24<1:11:35, 1.08it/s, tok/s=13805.7]
Inference: 21%|██▏ | 1262/5920 [18:24<1:11:35, 1.08it/s, tok/s=13800.9]
Inference: 21%|██▏ | 1263/5920 [18:24<1:11:35, 1.08it/s, tok/s=13821.1]
Inference: 21%|██▏ | 1264/5920 [18:25<1:03:34, 1.22it/s, tok/s=13821.1]
Inference: 21%|██▏ | 1264/5920 [18:25<1:03:34, 1.22it/s, tok/s=13830.9]
Inference: 21%|██▏ | 1265/5920 [18:25<1:03:34, 1.22it/s, tok/s=13840.6]
Inference: 21%|██▏ | 1266/5920 [18:25<42:42, 1.82it/s, tok/s=13840.6]
Inference: 21%|██▏ | 1266/5920 [18:25<42:42, 1.82it/s, tok/s=13863.7]
Inference: 21%|██▏ | 1267/5920 [18:25<42:41, 1.82it/s, tok/s=13883.7]
Inference: 21%|██▏ | 1268/5920 [18:26<38:48, 2.00it/s, tok/s=13883.7]
Inference: 21%|██▏ | 1268/5920 [18:26<38:48, 2.00it/s, tok/s=13898.1]
Inference: 21%|██▏ | 1269/5920 [18:27<46:38, 1.66it/s, tok/s=13898.1]
Inference: 21%|██▏ | 1269/5920 [18:27<46:38, 1.66it/s, tok/s=13896.1]
Inference: 21%|██▏ | 1270/5920 [18:28<52:27, 1.48it/s, tok/s=13896.1]
Inference: 21%|██▏ | 1270/5920 [18:28<52:27, 1.48it/s, tok/s=13905.2]
Inference: 21%|██▏ | 1271/5920 [18:28<45:20, 1.71it/s, tok/s=13905.2]
Inference: 21%|██▏ | 1271/5920 [18:28<45:20, 1.71it/s, tok/s=13924.0]
Inference: 21%|██▏ | 1272/5920 [18:29<36:58, 2.10it/s, tok/s=13924.0]
Inference: 21%|██▏ | 1272/5920 [18:29<36:58, 2.10it/s, tok/s=13938.8]
Inference: 22%|██▏ | 1273/5920 [18:29<35:08, 2.20it/s, tok/s=13938.8]
Inference: 22%|██▏ | 1273/5920 [18:29<35:08, 2.20it/s, tok/s=13949.2]
Inference: 22%|██▏ | 1274/5920 [18:30<45:33, 1.70it/s, tok/s=13949.2]
Inference: 22%|██▏ | 1274/5920 [18:30<45:33, 1.70it/s, tok/s=13948.9]
Inference: 22%|██▏ | 1275/5920 [18:30<42:58, 1.80it/s, tok/s=13948.9]
Inference: 22%|██▏ | 1275/5920 [18:30<42:58, 1.80it/s, tok/s=13965.4]
Inference: 22%|██▏ | 1276/5920 [18:30<42:57, 1.80it/s, tok/s=13988.0]
Inference: 22%|██▏ | 1277/5920 [18:31<33:24, 2.32it/s, tok/s=13988.0]
Inference: 22%|██▏ | 1277/5920 [18:31<33:24, 2.32it/s, tok/s=13995.1]
Inference: 22%|██▏ | 1278/5920 [18:32<35:35, 2.17it/s, tok/s=13995.1]
Inference: 22%|██▏ | 1278/5920 [18:32<35:35, 2.17it/s, tok/s=14009.0]
Inference: 22%|██▏ | 1279/5920 [18:33<54:20, 1.42it/s, tok/s=14009.0]
Inference: 22%|██▏ | 1279/5920 [18:33<54:20, 1.42it/s, tok/s=14003.7]
Inference: 22%|██▏ | 1280/5920 [18:33<54:20, 1.42it/s, tok/s=14015.0]
Inference: 22%|██▏ | 1281/5920 [18:34<55:11, 1.40it/s, tok/s=14015.0]
Inference: 22%|██▏ | 1281/5920 [18:34<55:11, 1.40it/s, tok/s=14009.9]
Inference: 22%|██▏ | 1282/5920 [18:36<1:03:53, 1.21it/s, tok/s=14009.9]
Inference: 22%|██▏ | 1282/5920 [18:36<1:03:53, 1.21it/s, tok/s=14002.6]
Inference: 22%|██▏ | 1283/5920 [18:37<1:12:27, 1.07it/s, tok/s=14002.6]
Inference: 22%|██▏ | 1283/5920 [18:37<1:12:27, 1.07it/s, tok/s=13994.5]
Inference: 22%|██▏ | 1284/5920 [18:37<1:12:26, 1.07it/s, tok/s=14002.7]
Inference: 22%|██▏ | 1285/5920 [18:37<44:17, 1.74it/s, tok/s=14002.7]
Inference: 22%|██▏ | 1285/5920 [18:37<44:17, 1.74it/s, tok/s=14022.8]
Inference: 22%|██▏ | 1286/5920 [18:37<37:21, 2.07it/s, tok/s=14022.8]
Inference: 22%|██▏ | 1286/5920 [18:37<37:21, 2.07it/s, tok/s=14041.8]
Inference: 22%|██▏ | 1287/5920 [18:38<45:41, 1.69it/s, tok/s=14041.8]
Inference: 22%|██▏ | 1287/5920 [18:38<45:41, 1.69it/s, tok/s=14046.8]
Inference: 22%|██▏ | 1288/5920 [18:40<1:19:53, 1.03s/it, tok/s=14046.8]
Inference: 22%|██▏ | 1288/5920 [18:40<1:19:53, 1.03s/it, tok/s=14043.7]
Inference: 22%|██▏ | 1289/5920 [18:41<1:07:54, 1.14it/s, tok/s=14043.7]
Inference: 22%|██▏ | 1289/5920 [18:41<1:07:54, 1.14it/s, tok/s=14060.1]
Inference: 22%|██▏ | 1290/5920 [18:42<1:16:32, 1.01it/s, tok/s=14060.1]
Inference: 22%|██▏ | 1290/5920 [18:42<1:16:32, 1.01it/s, tok/s=14070.9]
Inference: 22%|██▏ | 1291/5920 [18:43<1:22:53, 1.07s/it, tok/s=14070.9]
Inference: 22%|██▏ | 1291/5920 [18:43<1:22:53, 1.07s/it, tok/s=14064.1]
Inference: 22%|██▏ | 1292/5920 [18:45<1:31:35, 1.19s/it, tok/s=14064.1]
Inference: 22%|██▏ | 1292/5920 [18:45<1:31:35, 1.19s/it, tok/s=14070.0]
Inference: 22%|██▏ | 1293/5920 [18:46<1:30:09, 1.17s/it, tok/s=14070.0]
Inference: 22%|██▏ | 1293/5920 [18:46<1:30:09, 1.17s/it, tok/s=14076.9]
Inference: 22%|██▏ | 1294/5920 [18:48<1:55:38, 1.50s/it, tok/s=14076.9]
Inference: 22%|██▏ | 1294/5920 [18:48<1:55:38, 1.50s/it, tok/s=14073.5]
Inference: 22%|██▏ | 1295/5920 [18:48<1:25:20, 1.11s/it, tok/s=14073.5]
Inference: 22%|██▏ | 1295/5920 [18:48<1:25:20, 1.11s/it, tok/s=14085.0]
Inference: 22%|██▏ | 1296/5920 [18:50<1:34:00, 1.22s/it, tok/s=14085.0]
Inference: 22%|██▏ | 1296/5920 [18:50<1:34:00, 1.22s/it, tok/s=14079.3]
Inference: 22%|██▏ | 1297/5920 [18:51<1:36:12, 1.25s/it, tok/s=14079.3]
Inference: 22%|██▏ | 1297/5920 [18:51<1:36:12, 1.25s/it, tok/s=14085.2]
Inference: 22%|██▏ | 1298/5920 [18:53<1:36:24, 1.25s/it, tok/s=14085.2]
Inference: 22%|██▏ | 1298/5920 [18:53<1:36:24, 1.25s/it, tok/s=14081.7]
Inference: 22%|██▏ | 1299/5920 [18:54<1:31:11, 1.18s/it, tok/s=14081.7]
Inference: 22%|██▏ | 1299/5920 [18:54<1:31:11, 1.18s/it, tok/s=14081.3]
Inference: 22%|██▏ | 1300/5920 [18:54<1:09:26, 1.11it/s, tok/s=14081.3]
Inference: 22%|██▏ | 1300/5920 [18:54<1:09:26, 1.11it/s, tok/s=14087.1]
Inference: 22%|██▏ | 1301/5920 [18:54<1:03:53, 1.20it/s, tok/s=14087.1]
Inference: 22%|██▏ | 1301/5920 [18:54<1:03:53, 1.20it/s, tok/s=14099.2]
Inference: 22%|██▏ | 1302/5920 [18:55<1:06:39, 1.15it/s, tok/s=14099.2]
Inference: 22%|██▏ | 1302/5920 [18:55<1:06:39, 1.15it/s, tok/s=14099.6]
Inference: 22%|██▏ | 1303/5920 [18:57<1:21:38, 1.06s/it, tok/s=14099.6]
Inference: 22%|██▏ | 1303/5920 [18:57<1:21:38, 1.06s/it, tok/s=14104.1]
Inference: 22%|██▏ | 1304/5920 [18:58<1:30:06, 1.17s/it, tok/s=14104.1]
Inference: 22%|██▏ | 1304/5920 [18:58<1:30:06, 1.17s/it, tok/s=14098.1]
Inference: 22%|██▏ | 1305/5920 [18:59<1:22:22, 1.07s/it, tok/s=14098.1]
Inference: 22%|██▏ | 1305/5920 [18:59<1:22:22, 1.07s/it, tok/s=14099.7]
Inference: 22%|██▏ | 1306/5920 [19:01<1:36:28, 1.25s/it, tok/s=14099.7]
Inference: 22%|██▏ | 1306/5920 [19:01<1:36:28, 1.25s/it, tok/s=14087.1]
Inference: 22%|██▏ | 1307/5920 [19:02<1:23:59, 1.09s/it, tok/s=14087.1]
Inference: 22%|██▏ | 1307/5920 [19:02<1:23:59, 1.09s/it, tok/s=14094.9]
Inference: 22%|██▏ | 1308/5920 [19:02<1:12:58, 1.05it/s, tok/s=14094.9]
Inference: 22%|██▏ | 1308/5920 [19:02<1:12:58, 1.05it/s, tok/s=14109.6]
Inference: 22%|██▏ | 1309/5920 [19:06<2:26:16, 1.90s/it, tok/s=14109.6]
Inference: 22%|██▏ | 1309/5920 [19:06<2:26:16, 1.90s/it, tok/s=14085.2]
Inference: 22%|██▏ | 1310/5920 [19:07<1:52:16, 1.46s/it, tok/s=14085.2]
Inference: 22%|██▏ | 1310/5920 [19:07<1:52:16, 1.46s/it, tok/s=14088.6]
Inference: 22%|██▏ | 1311/5920 [19:08<1:43:05, 1.34s/it, tok/s=14088.6]
Inference: 22%|██▏ | 1311/5920 [19:08<1:43:05, 1.34s/it, tok/s=14097.8]
Inference: 22%|██▏ | 1312/5920 [19:10<1:56:46, 1.52s/it, tok/s=14097.8]
Inference: 22%|██▏ | 1312/5920 [19:10<1:56:46, 1.52s/it, tok/s=14079.8]
Inference: 22%|██▏ | 1313/5920 [19:12<2:11:06, 1.71s/it, tok/s=14079.8]
Inference: 22%|██▏ | 1313/5920 [19:12<2:11:06, 1.71s/it, tok/s=14069.4]
Inference: 22%|██▏ | 1314/5920 [19:12<1:40:58, 1.32s/it, tok/s=14069.4]
Inference: 22%|██▏ | 1314/5920 [19:12<1:40:58, 1.32s/it, tok/s=14082.5]
Inference: 22%|██▏ | 1315/5920 [19:15<2:17:21, 1.79s/it, tok/s=14082.5]
Inference: 22%|██▏ | 1315/5920 [19:15<2:17:21, 1.79s/it, tok/s=14056.3]
Inference: 22%|██▏ | 1316/5920 [19:18<2:38:23, 2.06s/it, tok/s=14056.3]
Inference: 22%|██▏ | 1316/5920 [19:18<2:38:23, 2.06s/it, tok/s=14028.5]
Inference: 22%|██▏ | 1317/5920 [19:19<2:10:54, 1.71s/it, tok/s=14028.5]
Inference: 22%|██▏ | 1317/5920 [19:19<2:10:54, 1.71s/it, tok/s=14028.4]
Inference: 22%|██▏ | 1318/5920 [19:19<2:10:53, 1.71s/it, tok/s=14031.5]
Inference: 22%|██▏ | 1319/5920 [19:19<1:14:50, 1.02it/s, tok/s=14031.5]
Inference: 22%|██▏ | 1319/5920 [19:19<1:14:50, 1.02it/s, tok/s=14038.8]
Inference: 22%|██▏ | 1320/5920 [19:20<1:19:52, 1.04s/it, tok/s=14038.8]
Inference: 22%|██▏ | 1320/5920 [19:20<1:19:52, 1.04s/it, tok/s=14036.1]
Inference: 22%|██▏ | 1321/5920 [19:22<1:27:09, 1.14s/it, tok/s=14036.1]
Inference: 22%|██▏ | 1321/5920 [19:22<1:27:09, 1.14s/it, tok/s=14040.9]
Inference: 22%|██▏ | 1322/5920 [19:23<1:21:52, 1.07s/it, tok/s=14040.9]
Inference: 22%|██▏ | 1322/5920 [19:23<1:21:52, 1.07s/it, tok/s=14038.7]
Inference: 22%|██▏ | 1323/5920 [19:23<1:13:36, 1.04it/s, tok/s=14038.7]
Inference: 22%|██▏ | 1323/5920 [19:23<1:13:36, 1.04it/s, tok/s=14043.8]
Inference: 22%|██▏ | 1324/5920 [19:24<1:12:18, 1.06it/s, tok/s=14043.8]
Inference: 22%|██▏ | 1324/5920 [19:24<1:12:18, 1.06it/s, tok/s=14038.7]
Inference: 22%|██▏ | 1325/5920 [19:24<54:33, 1.40it/s, tok/s=14038.7]
Inference: 22%|██▏ | 1325/5920 [19:24<54:33, 1.40it/s, tok/s=14058.4]
Inference: 22%|██▏ | 1326/5920 [19:24<54:32, 1.40it/s, tok/s=14063.4]
Inference: 22%|██▏ | 1327/5920 [19:24<31:34, 2.42it/s, tok/s=14063.4]
Inference: 22%|██▏ | 1327/5920 [19:24<31:34, 2.42it/s, tok/s=14084.9]
Inference: 22%|██▏ | 1328/5920 [19:26<53:10, 1.44it/s, tok/s=14084.9]
Inference: 22%|██▏ | 1328/5920 [19:26<53:10, 1.44it/s, tok/s=14074.7]
Inference: 22%|██▏ | 1329/5920 [19:27<50:32, 1.51it/s, tok/s=14074.7]
Inference: 22%|██▏ | 1329/5920 [19:27<50:32, 1.51it/s, tok/s=14093.7]
Inference: 22%|██▏ | 1330/5920 [19:27<50:31, 1.51it/s, tok/s=14097.5]
Inference: 22%|██▏ | 1331/5920 [19:27<33:02, 2.31it/s, tok/s=14097.5]
Inference: 22%|██▏ | 1331/5920 [19:27<33:02, 2.31it/s, tok/s=14107.4]
Inference: 22%|██▎ | 1332/5920 [19:27<30:21, 2.52it/s, tok/s=14107.4]
Inference: 22%|██▎ | 1332/5920 [19:27<30:21, 2.52it/s, tok/s=14114.4]
Inference: 23%|██▎ | 1333/5920 [19:27<26:37, 2.87it/s, tok/s=14114.4]
Inference: 23%|██▎ | 1333/5920 [19:27<26:37, 2.87it/s, tok/s=14116.4]
Inference: 23%|██▎ | 1334/5920 [19:27<26:37, 2.87it/s, tok/s=14136.9]
Inference: 23%|██▎ | 1335/5920 [19:28<27:31, 2.78it/s, tok/s=14136.9]
Inference: 23%|██▎ | 1335/5920 [19:28<27:31, 2.78it/s, tok/s=14139.8]
Inference: 23%|██▎ | 1336/5920 [19:28<27:30, 2.78it/s, tok/s=14152.3]
Inference: 23%|██▎ | 1337/5920 [19:29<27:52, 2.74it/s, tok/s=14152.3]
Inference: 23%|██▎ | 1337/5920 [19:29<27:52, 2.74it/s, tok/s=14154.3]
Inference: 23%|██▎ | 1338/5920 [19:29<26:57, 2.83it/s, tok/s=14154.3]
Inference: 23%|██▎ | 1338/5920 [19:29<26:57, 2.83it/s, tok/s=14165.0]
Inference: 23%|██▎ | 1339/5920 [19:29<26:14, 2.91it/s, tok/s=14165.0]
Inference: 23%|██▎ | 1339/5920 [19:29<26:14, 2.91it/s, tok/s=14185.7]
Inference: 23%|██▎ | 1340/5920 [19:34<1:48:24, 1.42s/it, tok/s=14185.7]
Inference: 23%|██▎ | 1340/5920 [19:34<1:48:24, 1.42s/it, tok/s=14152.3]
Inference: 23%|██▎ | 1341/5920 [19:35<1:34:29, 1.24s/it, tok/s=14152.3]
Inference: 23%|██▎ | 1341/5920 [19:35<1:34:29, 1.24s/it, tok/s=14148.3]
Inference: 23%|██▎ | 1342/5920 [19:35<1:20:12, 1.05s/it, tok/s=14148.3]
Inference: 23%|██▎ | 1342/5920 [19:35<1:20:12, 1.05s/it, tok/s=14153.3]
Inference: 23%|██▎ | 1343/5920 [19:35<1:00:28, 1.26it/s, tok/s=14153.3]
Inference: 23%|██▎ | 1343/5920 [19:35<1:00:28, 1.26it/s, tok/s=14175.7]
Inference: 23%|██▎ | 1344/5920 [19:37<1:10:16, 1.09it/s, tok/s=14175.7]
Inference: 23%|██▎ | 1344/5920 [19:37<1:10:16, 1.09it/s, tok/s=14172.8]
Inference: 23%|██▎ | 1345/5920 [19:37<1:05:03, 1.17it/s, tok/s=14172.8]
Inference: 23%|██▎ | 1345/5920 [19:37<1:05:03, 1.17it/s, tok/s=14173.7]
Inference: 23%|██▎ | 1346/5920 [19:39<1:13:14, 1.04it/s, tok/s=14173.7]
Inference: 23%|██▎ | 1346/5920 [19:39<1:13:14, 1.04it/s, tok/s=14181.6]
Inference: 23%|██▎ | 1347/5920 [19:39<1:07:38, 1.13it/s, tok/s=14181.6]
Inference: 23%|██▎ | 1347/5920 [19:39<1:07:38, 1.13it/s, tok/s=14183.0]
Inference: 23%|██▎ | 1348/5920 [19:40<52:31, 1.45it/s, tok/s=14183.0]
Inference: 23%|██▎ | 1348/5920 [19:40<52:31, 1.45it/s, tok/s=14202.0]
Inference: 23%|██▎ | 1349/5920 [19:41<1:14:32, 1.02it/s, tok/s=14202.0]
Inference: 23%|██▎ | 1349/5920 [19:41<1:14:32, 1.02it/s, tok/s=14202.9]
Inference: 23%|██▎ | 1350/5920 [19:41<55:16, 1.38it/s, tok/s=14202.9]
Inference: 23%|██▎ | 1350/5920 [19:41<55:16, 1.38it/s, tok/s=14226.8]
Inference: 23%|██▎ | 1351/5920 [19:42<52:29, 1.45it/s, tok/s=14226.8]
Inference: 23%|██▎ | 1351/5920 [19:42<52:29, 1.45it/s, tok/s=14224.8]
Inference: 23%|██▎ | 1352/5920 [19:42<52:28, 1.45it/s, tok/s=14249.2]
Inference: 23%|██▎ | 1353/5920 [19:43<54:50, 1.39it/s, tok/s=14249.2]
Inference: 23%|██▎ | 1353/5920 [19:43<54:50, 1.39it/s, tok/s=14254.2]
Inference: 23%|██▎ | 1354/5920 [19:45<1:09:19, 1.10it/s, tok/s=14254.2]
Inference: 23%|██▎ | 1354/5920 [19:45<1:09:19, 1.10it/s, tok/s=14250.7]
Inference: 23%|██▎ | 1355/5920 [19:47<1:39:34, 1.31s/it, tok/s=14250.7]
Inference: 23%|██▎ | 1355/5920 [19:47<1:39:34, 1.31s/it, tok/s=14233.0]
Inference: 23%|██▎ | 1356/5920 [19:48<1:15:32, 1.01it/s, tok/s=14233.0]
Inference: 23%|██▎ | 1356/5920 [19:48<1:15:32, 1.01it/s, tok/s=14255.5]
Inference: 23%|██▎ | 1357/5920 [19:48<1:04:37, 1.18it/s, tok/s=14255.5]
Inference: 23%|██▎ | 1357/5920 [19:48<1:04:37, 1.18it/s, tok/s=14274.6]
Inference: 23%|██▎ | 1358/5920 [19:48<51:56, 1.46it/s, tok/s=14274.6]
Inference: 23%|██▎ | 1358/5920 [19:48<51:56, 1.46it/s, tok/s=14282.7]
Inference: 23%|██▎ | 1359/5920 [19:49<51:46, 1.47it/s, tok/s=14282.7]
Inference: 23%|██▎ | 1359/5920 [19:49<51:46, 1.47it/s, tok/s=14281.4]
Inference: 23%|██▎ | 1360/5920 [19:49<44:43, 1.70it/s, tok/s=14281.4]
Inference: 23%|██▎ | 1360/5920 [19:49<44:43, 1.70it/s, tok/s=14299.7]
Inference: 23%|██▎ | 1361/5920 [19:49<44:43, 1.70it/s, tok/s=14309.9]
Inference: 23%|██▎ | 1362/5920 [19:51<52:18, 1.45it/s, tok/s=14309.9]
Inference: 23%|██▎ | 1362/5920 [19:51<52:18, 1.45it/s, tok/s=14314.3]
Inference: 23%|██▎ | 1363/5920 [19:51<48:28, 1.57it/s, tok/s=14314.3]
Inference: 23%|██▎ | 1363/5920 [19:51<48:28, 1.57it/s, tok/s=14333.8]
Inference: 23%|██▎ | 1364/5920 [19:52<46:09, 1.64it/s, tok/s=14333.8]
Inference: 23%|██▎ | 1364/5920 [19:52<46:09, 1.64it/s, tok/s=14352.8]
Inference: 23%|██▎ | 1365/5920 [19:52<36:15, 2.09it/s, tok/s=14352.8]
Inference: 23%|██▎ | 1365/5920 [19:52<36:15, 2.09it/s, tok/s=14356.6]
Inference: 23%|██▎ | 1366/5920 [19:53<44:37, 1.70it/s, tok/s=14356.6]
Inference: 23%|██▎ | 1366/5920 [19:53<44:37, 1.70it/s, tok/s=14352.6]
Inference: 23%|██▎ | 1367/5920 [19:53<41:40, 1.82it/s, tok/s=14352.6]
Inference: 23%|██▎ | 1367/5920 [19:53<41:40, 1.82it/s, tok/s=14355.8]
Inference: 23%|██▎ | 1368/5920 [19:55<55:48, 1.36it/s, tok/s=14355.8]
Inference: 23%|██▎ | 1368/5920 [19:55<55:48, 1.36it/s, tok/s=14347.6]
Inference: 23%|██▎ | 1369/5920 [19:56<1:05:44, 1.15it/s, tok/s=14347.6]
Inference: 23%|██▎ | 1369/5920 [19:56<1:05:44, 1.15it/s, tok/s=14358.1]
Inference: 23%|██▎ | 1370/5920 [19:57<1:20:40, 1.06s/it, tok/s=14358.1]
Inference: 23%|██▎ | 1370/5920 [19:57<1:20:40, 1.06s/it, tok/s=14344.7]
Inference: 23%|██▎ | 1371/5920 [19:57<1:20:39, 1.06s/it, tok/s=14349.5]
Inference: 23%|██▎ | 1372/5920 [19:58<53:21, 1.42it/s, tok/s=14349.5]
Inference: 23%|██▎ | 1372/5920 [19:58<53:21, 1.42it/s, tok/s=14357.3]
Inference: 23%|██▎ | 1373/5920 [19:58<50:08, 1.51it/s, tok/s=14357.3]
Inference: 23%|██▎ | 1373/5920 [19:58<50:08, 1.51it/s, tok/s=14356.4]
Inference: 23%|██▎ | 1374/5920 [19:59<43:12, 1.75it/s, tok/s=14356.4]
Inference: 23%|██▎ | 1374/5920 [19:59<43:12, 1.75it/s, tok/s=14371.1]
Inference: 23%|██▎ | 1375/5920 [19:59<38:51, 1.95it/s, tok/s=14371.1]
Inference: 23%|██▎ | 1375/5920 [19:59<38:51, 1.95it/s, tok/s=14387.1]
Inference: 23%|██▎ | 1376/5920 [20:00<44:52, 1.69it/s, tok/s=14387.1]
Inference: 23%|██▎ | 1376/5920 [20:00<44:52, 1.69it/s, tok/s=14388.5]
Inference: 23%|██▎ | 1377/5920 [20:01<1:00:44, 1.25it/s, tok/s=14388.5]
Inference: 23%|██▎ | 1377/5920 [20:01<1:00:44, 1.25it/s, tok/s=14377.7]
Inference: 23%|██▎ | 1378/5920 [20:02<54:17, 1.39it/s, tok/s=14377.7]
Inference: 23%|██▎ | 1378/5920 [20:02<54:17, 1.39it/s, tok/s=14377.2]
Inference: 23%|██▎ | 1379/5920 [20:02<54:16, 1.39it/s, tok/s=14382.7]
Inference: 23%|██▎ | 1380/5920 [20:02<35:05, 2.16it/s, tok/s=14382.7]
Inference: 23%|██▎ | 1380/5920 [20:02<35:05, 2.16it/s, tok/s=14389.7]
Inference: 23%|██▎ | 1381/5920 [20:03<39:21, 1.92it/s, tok/s=14389.7]
Inference: 23%|██▎ | 1381/5920 [20:03<39:21, 1.92it/s, tok/s=14391.4]
Inference: 23%|██▎ | 1382/5920 [20:03<33:38, 2.25it/s, tok/s=14391.4]
Inference: 23%|██▎ | 1382/5920 [20:03<33:38, 2.25it/s, tok/s=14394.0]
Inference: 23%|██▎ | 1383/5920 [20:03<27:24, 2.76it/s, tok/s=14394.0]
Inference: 23%|██▎ | 1383/5920 [20:03<27:24, 2.76it/s, tok/s=14401.5]
Inference: 23%|██▎ | 1384/5920 [20:03<28:09, 2.68it/s, tok/s=14401.5]
Inference: 23%|██▎ | 1384/5920 [20:03<28:09, 2.68it/s, tok/s=14402.3]
Inference: 23%|██▎ | 1385/5920 [20:04<39:27, 1.92it/s, tok/s=14402.3]
Inference: 23%|██▎ | 1385/5920 [20:04<39:27, 1.92it/s, tok/s=14415.9]
Inference: 23%|██▎ | 1386/5920 [20:05<32:36, 2.32it/s, tok/s=14415.9]
Inference: 23%|██▎ | 1386/5920 [20:05<32:36, 2.32it/s, tok/s=14418.7]
Inference: 23%|██▎ | 1387/5920 [20:05<32:36, 2.32it/s, tok/s=14432.1]
Inference: 23%|██▎ | 1388/5920 [20:07<1:06:35, 1.13it/s, tok/s=14432.1]
Inference: 23%|██▎ | 1388/5920 [20:07<1:06:35, 1.13it/s, tok/s=14411.4]
Inference: 23%|██▎ | 1389/5920 [20:09<1:19:22, 1.05s/it, tok/s=14411.4]
Inference: 23%|██▎ | 1389/5920 [20:09<1:19:22, 1.05s/it, tok/s=14402.9]
Inference: 23%|██▎ | 1390/5920 [20:10<1:12:52, 1.04it/s, tok/s=14402.9]
Inference: 23%|██▎ | 1390/5920 [20:10<1:12:52, 1.04it/s, tok/s=14419.2]
Inference: 23%|██▎ | 1391/5920 [20:12<1:46:29, 1.41s/it, tok/s=14419.2]
Inference: 23%|██▎ | 1391/5920 [20:12<1:46:29, 1.41s/it, tok/s=14411.0]
Inference: 24%|██▎ | 1392/5920 [20:14<1:53:59, 1.51s/it, tok/s=14411.0]
Inference: 24%|██▎ | 1392/5920 [20:14<1:53:59, 1.51s/it, tok/s=14395.8]
Inference: 24%|██▎ | 1393/5920 [20:14<1:24:09, 1.12s/it, tok/s=14395.8]
Inference: 24%|██▎ | 1393/5920 [20:14<1:24:09, 1.12s/it, tok/s=14398.1]
Inference: 24%|██▎ | 1394/5920 [20:15<1:10:49, 1.07it/s, tok/s=14398.1]
Inference: 24%|██▎ | 1394/5920 [20:15<1:10:49, 1.07it/s, tok/s=14416.9]
Inference: 24%|██▎ | 1395/5920 [20:16<1:12:17, 1.04it/s, tok/s=14416.9]
Inference: 24%|██▎ | 1395/5920 [20:16<1:12:17, 1.04it/s, tok/s=14429.7]
Inference: 24%|██▎ | 1396/5920 [20:16<1:02:33, 1.21it/s, tok/s=14429.7]
Inference: 24%|██▎ | 1396/5920 [20:16<1:02:33, 1.21it/s, tok/s=14445.2]
Inference: 24%|██▎ | 1397/5920 [20:16<1:02:32, 1.21it/s, tok/s=14468.8]
Inference: 24%|██▎ | 1398/5920 [20:17<46:57, 1.60it/s, tok/s=14468.8]
Inference: 24%|██▎ | 1398/5920 [20:17<46:57, 1.60it/s, tok/s=14476.6]
Inference: 24%|██▎ | 1399/5920 [20:18<1:00:52, 1.24it/s, tok/s=14476.6]
Inference: 24%|██▎ | 1399/5920 [20:18<1:00:52, 1.24it/s, tok/s=14484.8]
Inference: 24%|██▎ | 1400/5920 [20:19<47:53, 1.57it/s, tok/s=14484.8]
Inference: 24%|██▎ | 1400/5920 [20:19<47:53, 1.57it/s, tok/s=14490.2]
Inference: 24%|██▎ | 1401/5920 [20:19<42:35, 1.77it/s, tok/s=14490.2]
Inference: 24%|██▎ | 1401/5920 [20:19<42:35, 1.77it/s, tok/s=14510.3]
Inference: 24%|██▎ | 1402/5920 [20:19<37:35, 2.00it/s, tok/s=14510.3]
Inference: 24%|██▎ | 1402/5920 [20:19<37:35, 2.00it/s, tok/s=14518.1]
Inference: 24%|██▎ | 1403/5920 [20:20<42:00, 1.79it/s, tok/s=14518.1]
Inference: 24%|██▎ | 1403/5920 [20:20<42:00, 1.79it/s, tok/s=14520.7]
Inference: 24%|██▎ | 1404/5920 [20:21<46:25, 1.62it/s, tok/s=14520.7]
Inference: 24%|██▎ | 1404/5920 [20:21<46:25, 1.62it/s, tok/s=14518.5]
Inference: 24%|██▎ | 1405/5920 [20:23<1:14:44, 1.01it/s, tok/s=14518.5]
Inference: 24%|██▎ | 1405/5920 [20:23<1:14:44, 1.01it/s, tok/s=14504.2]
Inference: 24%|██▍ | 1406/5920 [20:23<58:38, 1.28it/s, tok/s=14504.2]
Inference: 24%|██▍ | 1406/5920 [20:23<58:38, 1.28it/s, tok/s=14525.6]
Inference: 24%|██▍ | 1407/5920 [20:24<57:56, 1.30it/s, tok/s=14525.6]
Inference: 24%|██▍ | 1407/5920 [20:24<57:56, 1.30it/s, tok/s=14529.5]
Inference: 24%|██▍ | 1408/5920 [20:24<43:10, 1.74it/s, tok/s=14529.5]
Inference: 24%|██▍ | 1408/5920 [20:24<43:10, 1.74it/s, tok/s=14552.8]
Inference: 24%|██▍ | 1409/5920 [20:24<42:36, 1.76it/s, tok/s=14552.8]
Inference: 24%|██▍ | 1409/5920 [20:24<42:36, 1.76it/s, tok/s=14570.8]
Inference: 24%|██▍ | 1410/5920 [20:25<43:43, 1.72it/s, tok/s=14570.8]
Inference: 24%|██▍ | 1410/5920 [20:25<43:43, 1.72it/s, tok/s=14588.1]
Inference: 24%|██▍ | 1411/5920 [20:25<38:06, 1.97it/s, tok/s=14588.1]
Inference: 24%|██▍ | 1411/5920 [20:25<38:06, 1.97it/s, tok/s=14595.7]
Inference: 24%|██▍ | 1412/5920 [20:26<49:01, 1.53it/s, tok/s=14595.7]
Inference: 24%|██▍ | 1412/5920 [20:26<49:01, 1.53it/s, tok/s=14597.0]
Inference: 24%|██▍ | 1413/5920 [20:27<55:51, 1.34it/s, tok/s=14597.0]
Inference: 24%|██▍ | 1413/5920 [20:27<55:51, 1.34it/s, tok/s=14595.1]
Inference: 24%|██▍ | 1414/5920 [20:28<48:59, 1.53it/s, tok/s=14595.1]
Inference: 24%|██▍ | 1414/5920 [20:28<48:59, 1.53it/s, tok/s=14595.4]
Inference: 24%|██▍ | 1415/5920 [20:28<42:29, 1.77it/s, tok/s=14595.4]
Inference: 24%|██▍ | 1415/5920 [20:28<42:29, 1.77it/s, tok/s=14601.4]
Inference: 24%|██▍ | 1416/5920 [20:28<41:32, 1.81it/s, tok/s=14601.4]
Inference: 24%|██▍ | 1416/5920 [20:28<41:32, 1.81it/s, tok/s=14619.7]
Inference: 24%|██▍ | 1417/5920 [20:29<46:51, 1.60it/s, tok/s=14619.7]
Inference: 24%|██▍ | 1417/5920 [20:29<46:51, 1.60it/s, tok/s=14622.5]
Inference: 24%|██▍ | 1418/5920 [20:31<1:07:07, 1.12it/s, tok/s=14622.5]
Inference: 24%|██▍ | 1418/5920 [20:31<1:07:07, 1.12it/s, tok/s=14613.9]
Inference: 24%|██▍ | 1419/5920 [20:31<59:38, 1.26it/s, tok/s=14613.9]
Inference: 24%|██▍ | 1419/5920 [20:31<59:38, 1.26it/s, tok/s=14628.5]
Inference: 24%|██▍ | 1420/5920 [20:32<51:31, 1.46it/s, tok/s=14628.5]
Inference: 24%|██▍ | 1420/5920 [20:32<51:31, 1.46it/s, tok/s=14628.9]
Inference: 24%|██▍ | 1421/5920 [20:32<41:34, 1.80it/s, tok/s=14628.9]
Inference: 24%|██▍ | 1421/5920 [20:32<41:34, 1.80it/s, tok/s=14649.9]
Inference: 24%|██▍ | 1422/5920 [20:35<1:33:03, 1.24s/it, tok/s=14649.9]
Inference: 24%|██▍ | 1422/5920 [20:35<1:33:03, 1.24s/it, tok/s=14633.3]
Inference: 24%|██▍ | 1423/5920 [20:36<1:19:10, 1.06s/it, tok/s=14633.3]
Inference: 24%|██▍ | 1423/5920 [20:36<1:19:10, 1.06s/it, tok/s=14635.9]
Inference: 24%|██▍ | 1424/5920 [20:36<1:00:45, 1.23it/s, tok/s=14635.9]
Inference: 24%|██▍ | 1424/5920 [20:36<1:00:45, 1.23it/s, tok/s=14647.0]
Inference: 24%|██▍ | 1425/5920 [20:36<1:00:44, 1.23it/s, tok/s=14671.2]
Inference: 24%|██▍ | 1426/5920 [20:36<34:27, 2.17it/s, tok/s=14671.2]
Inference: 24%|██▍ | 1426/5920 [20:36<34:27, 2.17it/s, tok/s=14677.9]
Inference: 24%|██▍ | 1427/5920 [20:37<49:21, 1.52it/s, tok/s=14677.9]
Inference: 24%|██▍ | 1427/5920 [20:37<49:21, 1.52it/s, tok/s=14678.5]
Inference: 24%|██▍ | 1428/5920 [20:37<42:39, 1.75it/s, tok/s=14678.5]
Inference: 24%|██▍ | 1428/5920 [20:37<42:39, 1.75it/s, tok/s=14679.2]
Inference: 24%|██▍ | 1429/5920 [20:38<42:31, 1.76it/s, tok/s=14679.2]
Inference: 24%|██▍ | 1429/5920 [20:38<42:31, 1.76it/s, tok/s=14694.6]
Inference: 24%|██▍ | 1430/5920 [20:40<1:13:17, 1.02it/s, tok/s=14694.6]
Inference: 24%|██▍ | 1430/5920 [20:40<1:13:17, 1.02it/s, tok/s=14678.6]
Inference: 24%|██▍ | 1431/5920 [20:41<1:12:22, 1.03it/s, tok/s=14678.6]
Inference: 24%|██▍ | 1431/5920 [20:41<1:12:22, 1.03it/s, tok/s=14689.4]
Inference: 24%|██▍ | 1432/5920 [20:42<1:09:02, 1.08it/s, tok/s=14689.4]
Inference: 24%|██▍ | 1432/5920 [20:42<1:09:02, 1.08it/s, tok/s=14703.5]
Inference: 24%|██▍ | 1433/5920 [20:42<1:09:01, 1.08it/s, tok/s=14707.3]
Inference: 24%|██▍ | 1434/5920 [20:42<42:08, 1.77it/s, tok/s=14707.3]
Inference: 24%|██▍ | 1434/5920 [20:42<42:08, 1.77it/s, tok/s=14716.0]
Inference: 24%|██▍ | 1435/5920 [20:45<1:20:11, 1.07s/it, tok/s=14716.0]
Inference: 24%|██▍ | 1435/5920 [20:45<1:20:11, 1.07s/it, tok/s=14690.2]
Inference: 24%|██▍ | 1436/5920 [20:45<1:05:02, 1.15it/s, tok/s=14690.2]
Inference: 24%|██▍ | 1436/5920 [20:45<1:05:02, 1.15it/s, tok/s=14691.2]
Inference: 24%|██▍ | 1437/5920 [20:46<1:05:43, 1.14it/s, tok/s=14691.2]
Inference: 24%|██▍ | 1437/5920 [20:46<1:05:43, 1.14it/s, tok/s=14685.4]
Inference: 24%|██▍ | 1438/5920 [20:46<57:31, 1.30it/s, tok/s=14685.4]
Inference: 24%|██▍ | 1438/5920 [20:46<57:31, 1.30it/s, tok/s=14703.8]
Inference: 24%|██▍ | 1439/5920 [20:46<57:30, 1.30it/s, tok/s=14707.5]
Inference: 24%|██▍ | 1440/5920 [20:47<39:31, 1.89it/s, tok/s=14707.5]
Inference: 24%|██▍ | 1440/5920 [20:47<39:31, 1.89it/s, tok/s=14717.1]
Inference: 24%|██▍ | 1441/5920 [20:47<34:10, 2.18it/s, tok/s=14717.1]
Inference: 24%|██▍ | 1441/5920 [20:47<34:10, 2.18it/s, tok/s=14721.9]
Inference: 24%|██▍ | 1442/5920 [20:47<27:32, 2.71it/s, tok/s=14721.9]
Inference: 24%|██▍ | 1442/5920 [20:47<27:32, 2.71it/s, tok/s=14725.6]
Inference: 24%|██▍ | 1443/5920 [20:48<43:28, 1.72it/s, tok/s=14725.6]
Inference: 24%|██▍ | 1443/5920 [20:48<43:28, 1.72it/s, tok/s=14736.0]
Inference: 24%|██▍ | 1444/5920 [20:51<1:19:10, 1.06s/it, tok/s=14736.0]
Inference: 24%|██▍ | 1444/5920 [20:51<1:19:10, 1.06s/it, tok/s=14729.8]
Inference: 24%|██▍ | 1445/5920 [20:51<1:08:00, 1.10it/s, tok/s=14729.8]
Inference: 24%|██▍ | 1445/5920 [20:51<1:08:00, 1.10it/s, tok/s=14731.6]
Inference: 24%|██▍ | 1446/5920 [20:53<1:23:59, 1.13s/it, tok/s=14731.6]
Inference: 24%|██▍ | 1446/5920 [20:53<1:23:59, 1.13s/it, tok/s=14721.6]
Inference: 24%|██▍ | 1447/5920 [20:53<1:05:32, 1.14it/s, tok/s=14721.6]
Inference: 24%|██▍ | 1447/5920 [20:53<1:05:32, 1.14it/s, tok/s=14730.2]
Inference: 24%|██▍ | 1448/5920 [20:54<1:10:42, 1.05it/s, tok/s=14730.2]
Inference: 24%|██▍ | 1448/5920 [20:54<1:10:42, 1.05it/s, tok/s=14728.9]
Inference: 24%|██▍ | 1449/5920 [20:54<53:47, 1.39it/s, tok/s=14728.9]
Inference: 24%|██▍ | 1449/5920 [20:54<53:47, 1.39it/s, tok/s=14736.6]
Inference: 24%|██▍ | 1450/5920 [20:56<1:11:09, 1.05it/s, tok/s=14736.6]
Inference: 24%|██▍ | 1450/5920 [20:56<1:11:09, 1.05it/s, tok/s=14742.7]
Inference: 25%|██▍ | 1451/5920 [20:56<59:00, 1.26it/s, tok/s=14742.7]
Inference: 25%|██▍ | 1451/5920 [20:56<59:00, 1.26it/s, tok/s=14752.6]
Inference: 25%|██▍ | 1452/5920 [20:57<57:43, 1.29it/s, tok/s=14752.6]
Inference: 25%|██▍ | 1452/5920 [20:57<57:43, 1.29it/s, tok/s=14747.9]
Inference: 25%|██▍ | 1453/5920 [20:58<1:04:41, 1.15it/s, tok/s=14747.9]
Inference: 25%|██▍ | 1453/5920 [20:58<1:04:41, 1.15it/s, tok/s=14759.0]
Inference: 25%|██▍ | 1454/5920 [21:00<1:23:57, 1.13s/it, tok/s=14759.0]
Inference: 25%|██▍ | 1454/5920 [21:00<1:23:57, 1.13s/it, tok/s=14752.1]
Inference: 25%|██▍ | 1455/5920 [21:00<1:07:58, 1.09it/s, tok/s=14752.1]
Inference: 25%|██▍ | 1455/5920 [21:00<1:07:58, 1.09it/s, tok/s=14755.5]
Inference: 25%|██▍ | 1456/5920 [21:01<1:03:46, 1.17it/s, tok/s=14755.5]
Inference: 25%|██▍ | 1456/5920 [21:01<1:03:46, 1.17it/s, tok/s=14752.3]
Inference: 25%|██▍ | 1457/5920 [21:04<1:45:47, 1.42s/it, tok/s=14752.3]
Inference: 25%|██▍ | 1457/5920 [21:04<1:45:47, 1.42s/it, tok/s=14729.3]
Inference: 25%|██▍ | 1458/5920 [21:05<1:35:32, 1.28s/it, tok/s=14729.3]
Inference: 25%|██▍ | 1458/5920 [21:05<1:35:32, 1.28s/it, tok/s=14722.7]
Inference: 25%|██▍ | 1459/5920 [21:07<1:51:16, 1.50s/it, tok/s=14722.7]
Inference: 25%|██▍ | 1459/5920 [21:07<1:51:16, 1.50s/it, tok/s=14708.4]
Inference: 25%|██▍ | 1460/5920 [21:08<1:37:40, 1.31s/it, tok/s=14708.4]
Inference: 25%|██▍ | 1460/5920 [21:08<1:37:40, 1.31s/it, tok/s=14721.9]
Inference: 25%|██▍ | 1461/5920 [21:08<1:37:39, 1.31s/it, tok/s=14727.0]
Inference: 25%|██▍ | 1462/5920 [21:08<1:05:42, 1.13it/s, tok/s=14727.0]
Inference: 25%|██▍ | 1462/5920 [21:08<1:05:42, 1.13it/s, tok/s=14740.5]
Inference: 25%|██▍ | 1463/5920 [21:09<1:07:47, 1.10it/s, tok/s=14740.5]
Inference: 25%|██▍ | 1463/5920 [21:09<1:07:47, 1.10it/s, tok/s=14752.6]
Inference: 25%|██▍ | 1464/5920 [21:10<54:33, 1.36it/s, tok/s=14752.6]
Inference: 25%|██▍ | 1464/5920 [21:10<54:33, 1.36it/s, tok/s=14754.6]
Inference: 25%|██▍ | 1465/5920 [21:10<54:32, 1.36it/s, tok/s=14776.5]
Inference: 25%|██▍ | 1466/5920 [21:14<1:46:10, 1.43s/it, tok/s=14776.5]
Inference: 25%|██▍ | 1466/5920 [21:14<1:46:10, 1.43s/it, tok/s=14725.0]
Inference: 25%|██▍ | 1467/5920 [21:14<1:23:44, 1.13s/it, tok/s=14725.0]
Inference: 25%|██▍ | 1467/5920 [21:14<1:23:44, 1.13s/it, tok/s=14728.1]
Inference: 25%|██▍ | 1468/5920 [21:15<1:10:41, 1.05it/s, tok/s=14728.1]
Inference: 25%|██▍ | 1468/5920 [21:15<1:10:41, 1.05it/s, tok/s=14735.4]
Inference: 25%|██▍ | 1469/5920 [21:15<55:26, 1.34it/s, tok/s=14735.4]
Inference: 25%|██▍ | 1469/5920 [21:15<55:26, 1.34it/s, tok/s=14746.6]
Inference: 25%|██▍ | 1470/5920 [21:15<55:25, 1.34it/s, tok/s=14768.0]
Inference: 25%|██▍ | 1471/5920 [21:16<47:54, 1.55it/s, tok/s=14768.0]
Inference: 25%|██▍ | 1471/5920 [21:16<47:54, 1.55it/s, tok/s=14767.3]
Inference: 25%|██▍ | 1472/5920 [21:16<38:26, 1.93it/s, tok/s=14767.3]
Inference: 25%|██▍ | 1472/5920 [21:16<38:26, 1.93it/s, tok/s=14789.7]
Inference: 25%|██▍ | 1473/5920 [21:17<35:02, 2.11it/s, tok/s=14789.7]
Inference: 25%|██▍ | 1473/5920 [21:17<35:02, 2.11it/s, tok/s=14791.0]
Inference: 25%|██▍ | 1474/5920 [21:17<44:13, 1.68it/s, tok/s=14791.0]
Inference: 25%|██▍ | 1474/5920 [21:17<44:13, 1.68it/s, tok/s=14785.6]
Inference: 25%|██▍ | 1475/5920 [21:18<36:40, 2.02it/s, tok/s=14785.6]
Inference: 25%|██▍ | 1475/5920 [21:18<36:40, 2.02it/s, tok/s=14806.6]
Inference: 25%|██▍ | 1476/5920 [21:19<53:31, 1.38it/s, tok/s=14806.6]
Inference: 25%|██▍ | 1476/5920 [21:19<53:31, 1.38it/s, tok/s=14815.0]
Inference: 25%|██▍ | 1477/5920 [21:20<1:01:31, 1.20it/s, tok/s=14815.0]
Inference: 25%|██▍ | 1477/5920 [21:20<1:01:31, 1.20it/s, tok/s=14810.5]
Inference: 25%|██▍ | 1478/5920 [21:22<1:15:18, 1.02s/it, tok/s=14810.5]
Inference: 25%|██▍ | 1478/5920 [21:22<1:15:18, 1.02s/it, tok/s=14800.9]
Inference: 25%|██▍ | 1479/5920 [21:22<59:39, 1.24it/s, tok/s=14800.9]
Inference: 25%|██▍ | 1479/5920 [21:22<59:39, 1.24it/s, tok/s=14809.1]
Inference: 25%|██▌ | 1480/5920 [21:23<1:01:07, 1.21it/s, tok/s=14809.1]
Inference: 25%|██▌ | 1480/5920 [21:23<1:01:07, 1.21it/s, tok/s=14807.6]
Inference: 25%|██▌ | 1481/5920 [21:23<1:01:06, 1.21it/s, tok/s=14823.6]
Inference: 25%|██▌ | 1482/5920 [21:23<1:01:05, 1.21it/s, tok/s=14825.3]
Inference: 25%|██▌ | 1483/5920 [21:23<30:19, 2.44it/s, tok/s=14825.3]
Inference: 25%|██▌ | 1483/5920 [21:23<30:19, 2.44it/s, tok/s=14826.9]
Inference: 25%|██▌ | 1484/5920 [21:23<26:27, 2.79it/s, tok/s=14826.9]
Inference: 25%|██▌ | 1484/5920 [21:23<26:27, 2.79it/s, tok/s=14830.3]
Inference: 25%|██▌ | 1485/5920 [21:24<32:06, 2.30it/s, tok/s=14830.3]
Inference: 25%|██▌ | 1485/5920 [21:24<32:06, 2.30it/s, tok/s=14843.7]
Inference: 25%|██▌ | 1486/5920 [21:25<38:21, 1.93it/s, tok/s=14843.7]
Inference: 25%|██▌ | 1486/5920 [21:25<38:21, 1.93it/s, tok/s=14839.7]
Inference: 25%|██▌ | 1487/5920 [21:26<50:28, 1.46it/s, tok/s=14839.7]
Inference: 25%|██▌ | 1487/5920 [21:26<50:28, 1.46it/s, tok/s=14836.4]
Inference: 25%|██▌ | 1488/5920 [21:26<48:51, 1.51it/s, tok/s=14836.4]
Inference: 25%|██▌ | 1488/5920 [21:26<48:51, 1.51it/s, tok/s=14852.9]
Inference: 25%|██▌ | 1489/5920 [21:27<47:16, 1.56it/s, tok/s=14852.9]
Inference: 25%|██▌ | 1489/5920 [21:27<47:16, 1.56it/s, tok/s=14852.1]
Inference: 25%|██▌ | 1490/5920 [21:27<38:32, 1.92it/s, tok/s=14852.1]
Inference: 25%|██▌ | 1490/5920 [21:27<38:32, 1.92it/s, tok/s=14856.6]
Inference: 25%|██▌ | 1491/5920 [21:27<30:34, 2.41it/s, tok/s=14856.6]
Inference: 25%|██▌ | 1491/5920 [21:27<30:34, 2.41it/s, tok/s=14857.7]
Inference: 25%|██▌ | 1492/5920 [21:28<38:27, 1.92it/s, tok/s=14857.7]
Inference: 25%|██▌ | 1492/5920 [21:28<38:27, 1.92it/s, tok/s=14872.1]
Inference: 25%|██▌ | 1493/5920 [21:30<1:12:56, 1.01it/s, tok/s=14872.1]
Inference: 25%|██▌ | 1493/5920 [21:30<1:12:56, 1.01it/s, tok/s=14858.2]
Inference: 25%|██▌ | 1494/5920 [21:30<1:12:55, 1.01it/s, tok/s=14860.8]
Inference: 25%|██▌ | 1495/5920 [21:31<49:44, 1.48it/s, tok/s=14860.8]
Inference: 25%|██▌ | 1495/5920 [21:31<49:44, 1.48it/s, tok/s=14858.7]
Inference: 25%|██▌ | 1496/5920 [21:31<40:06, 1.84it/s, tok/s=14858.7]
Inference: 25%|██▌ | 1496/5920 [21:31<40:06, 1.84it/s, tok/s=14866.5]
Inference: 25%|██▌ | 1497/5920 [21:31<38:27, 1.92it/s, tok/s=14866.5]
Inference: 25%|██▌ | 1497/5920 [21:31<38:27, 1.92it/s, tok/s=14866.1]
Inference: 25%|██▌ | 1498/5920 [21:32<35:31, 2.07it/s, tok/s=14866.1]
Inference: 25%|██▌ | 1498/5920 [21:32<35:31, 2.07it/s, tok/s=14864.9]
Inference: 25%|██▌ | 1499/5920 [21:32<29:47, 2.47it/s, tok/s=14864.9]
Inference: 25%|██▌ | 1499/5920 [21:32<29:47, 2.47it/s, tok/s=14868.1]
Inference: 25%|██▌ | 1500/5920 [21:33<32:13, 2.29it/s, tok/s=14868.1]
Inference: 25%|██▌ | 1500/5920 [21:33<32:13, 2.29it/s, tok/s=14876.0]
Inference: 25%|██▌ | 1501/5920 [21:35<1:15:50, 1.03s/it, tok/s=14876.0]
Inference: 25%|██▌ | 1501/5920 [21:35<1:15:50, 1.03s/it, tok/s=14857.5]
Inference: 25%|██▌ | 1502/5920 [21:35<1:03:26, 1.16it/s, tok/s=14857.5]
Inference: 25%|██▌ | 1502/5920 [21:35<1:03:26, 1.16it/s, tok/s=14875.1]
Inference: 25%|██▌ | 1503/5920 [21:36<50:00, 1.47it/s, tok/s=14875.1]
Inference: 25%|██▌ | 1503/5920 [21:36<50:00, 1.47it/s, tok/s=14874.8]
Inference: 25%|██▌ | 1504/5920 [21:36<50:22, 1.46it/s, tok/s=14874.8]
Inference: 25%|██▌ | 1504/5920 [21:36<50:22, 1.46it/s, tok/s=14888.0]
Inference: 25%|██▌ | 1505/5920 [21:37<46:38, 1.58it/s, tok/s=14888.0]
Inference: 25%|██▌ | 1505/5920 [21:37<46:38, 1.58it/s, tok/s=14885.1]
Inference: 25%|██▌ | 1506/5920 [21:37<35:24, 2.08it/s, tok/s=14885.1]
Inference: 25%|██▌ | 1506/5920 [21:37<35:24, 2.08it/s, tok/s=14890.9]
Inference: 25%|██▌ | 1507/5920 [21:37<35:23, 2.08it/s, tok/s=14896.2]
Inference: 25%|██▌ | 1508/5920 [21:38<27:49, 2.64it/s, tok/s=14896.2]
Inference: 25%|██▌ | 1508/5920 [21:38<27:49, 2.64it/s, tok/s=14893.9]
Inference: 25%|██▌ | 1509/5920 [21:38<23:53, 3.08it/s, tok/s=14893.9]
Inference: 25%|██▌ | 1509/5920 [21:38<23:53, 3.08it/s, tok/s=14905.9]
Inference: 26%|██▌ | 1510/5920 [21:39<49:41, 1.48it/s, tok/s=14905.9]
Inference: 26%|██▌ | 1510/5920 [21:39<49:41, 1.48it/s, tok/s=14897.5]
Inference: 26%|██▌ | 1511/5920 [21:39<49:40, 1.48it/s, tok/s=14899.7]
Inference: 26%|██▌ | 1512/5920 [21:40<33:55, 2.17it/s, tok/s=14899.7]
Inference: 26%|██▌ | 1512/5920 [21:40<33:55, 2.17it/s, tok/s=14909.5]
Inference: 26%|██▌ | 1513/5920 [21:40<33:54, 2.17it/s, tok/s=14915.0]
Inference: 26%|██▌ | 1514/5920 [21:40<30:10, 2.43it/s, tok/s=14915.0]
Inference: 26%|██▌ | 1514/5920 [21:40<30:10, 2.43it/s, tok/s=14910.2]
Inference: 26%|██▌ | 1515/5920 [21:40<30:10, 2.43it/s, tok/s=14917.8]
Inference: 26%|██▌ | 1516/5920 [21:41<23:52, 3.07it/s, tok/s=14917.8]
Inference: 26%|██▌ | 1516/5920 [21:41<23:52, 3.07it/s, tok/s=14938.0]
Inference: 26%|██▌ | 1517/5920 [21:41<23:52, 3.07it/s, tok/s=14942.4]
Inference: 26%|██▌ | 1518/5920 [21:44<49:53, 1.47it/s, tok/s=14942.4]
Inference: 26%|██▌ | 1518/5920 [21:44<49:53, 1.47it/s, tok/s=14922.9]
Inference: 26%|██▌ | 1519/5920 [21:44<43:07, 1.70it/s, tok/s=14922.9]
Inference: 26%|██▌ | 1519/5920 [21:44<43:07, 1.70it/s, tok/s=14922.9]
Inference: 26%|██▌ | 1520/5920 [21:44<40:53, 1.79it/s, tok/s=14922.9]
Inference: 26%|██▌ | 1520/5920 [21:44<40:53, 1.79it/s, tok/s=14935.1]
Inference: 26%|██▌ | 1521/5920 [21:46<1:02:43, 1.17it/s, tok/s=14935.1]
Inference: 26%|██▌ | 1521/5920 [21:46<1:02:43, 1.17it/s, tok/s=14937.0]
Inference: 26%|██▌ | 1522/5920 [21:46<1:02:42, 1.17it/s, tok/s=14942.7]
Inference: 26%|██▌ | 1523/5920 [21:46<42:30, 1.72it/s, tok/s=14942.7]
Inference: 26%|██▌ | 1523/5920 [21:46<42:30, 1.72it/s, tok/s=14961.4]
Inference: 26%|██▌ | 1524/5920 [21:47<39:19, 1.86it/s, tok/s=14961.4]
Inference: 26%|██▌ | 1524/5920 [21:47<39:19, 1.86it/s, tok/s=14959.5]
Inference: 26%|██▌ | 1525/5920 [21:48<44:34, 1.64it/s, tok/s=14959.5]
Inference: 26%|██▌ | 1525/5920 [21:48<44:34, 1.64it/s, tok/s=14954.8]
Inference: 26%|██▌ | 1526/5920 [21:48<38:58, 1.88it/s, tok/s=14954.8]
Inference: 26%|██▌ | 1526/5920 [21:48<38:58, 1.88it/s, tok/s=14971.4]
Inference: 26%|██▌ | 1527/5920 [21:49<51:25, 1.42it/s, tok/s=14971.4]
Inference: 26%|██▌ | 1527/5920 [21:49<51:25, 1.42it/s, tok/s=14963.2]
Inference: 26%|██▌ | 1528/5920 [21:50<1:01:26, 1.19it/s, tok/s=14963.2]
Inference: 26%|██▌ | 1528/5920 [21:50<1:01:26, 1.19it/s, tok/s=14952.4]
Inference: 26%|██▌ | 1529/5920 [21:53<1:34:55, 1.30s/it, tok/s=14952.4]
Inference: 26%|██▌ | 1529/5920 [21:53<1:34:55, 1.30s/it, tok/s=14929.8]
Inference: 26%|██▌ | 1530/5920 [21:55<1:47:01, 1.46s/it, tok/s=14929.8]
Inference: 26%|██▌ | 1530/5920 [21:55<1:47:01, 1.46s/it, tok/s=14920.2]
Inference: 26%|██▌ | 1531/5920 [21:55<1:25:51, 1.17s/it, tok/s=14920.2]
Inference: 26%|██▌ | 1531/5920 [21:55<1:25:51, 1.17s/it, tok/s=14931.1]
Inference: 26%|██▌ | 1532/5920 [21:55<1:25:50, 1.17s/it, tok/s=14936.8]
Inference: 26%|██▌ | 1533/5920 [21:55<51:49, 1.41it/s, tok/s=14936.8]
Inference: 26%|██▌ | 1533/5920 [21:55<51:49, 1.41it/s, tok/s=14953.7]
Inference: 26%|██▌ | 1534/5920 [21:56<50:24, 1.45it/s, tok/s=14953.7]
Inference: 26%|██▌ | 1534/5920 [21:56<50:24, 1.45it/s, tok/s=14969.4]
Inference: 26%|██▌ | 1535/5920 [21:56<41:35, 1.76it/s, tok/s=14969.4]
Inference: 26%|██▌ | 1535/5920 [21:56<41:35, 1.76it/s, tok/s=14973.5]
Inference: 26%|██▌ | 1536/5920 [21:56<34:17, 2.13it/s, tok/s=14973.5]
Inference: 26%|██▌ | 1536/5920 [21:56<34:17, 2.13it/s, tok/s=14974.0]
Inference: 26%|██▌ | 1537/5920 [21:57<30:36, 2.39it/s, tok/s=14974.0]
Inference: 26%|██▌ | 1537/5920 [21:57<30:36, 2.39it/s, tok/s=14993.1]
Inference: 26%|██▌ | 1538/5920 [21:57<29:03, 2.51it/s, tok/s=14993.1]
Inference: 26%|██▌ | 1538/5920 [21:57<29:03, 2.51it/s, tok/s=14992.2]
Inference: 26%|██▌ | 1539/5920 [21:58<33:49, 2.16it/s, tok/s=14992.2]
Inference: 26%|██▌ | 1539/5920 [21:58<33:49, 2.16it/s, tok/s=14991.0]
Inference: 26%|██▌ | 1540/5920 [21:59<59:39, 1.22it/s, tok/s=14991.0]
Inference: 26%|██▌ | 1540/5920 [21:59<59:39, 1.22it/s, tok/s=14981.2]
Inference: 26%|██▌ | 1541/5920 [22:00<49:11, 1.48it/s, tok/s=14981.2]
Inference: 26%|██▌ | 1541/5920 [22:00<49:11, 1.48it/s, tok/s=14999.1]
Inference: 26%|██▌ | 1542/5920 [22:00<40:50, 1.79it/s, tok/s=14999.1]
Inference: 26%|██▌ | 1542/5920 [22:00<40:50, 1.79it/s, tok/s=15011.5]
Inference: 26%|██▌ | 1543/5920 [22:01<1:00:09, 1.21it/s, tok/s=15011.5]
Inference: 26%|██▌ | 1543/5920 [22:01<1:00:09, 1.21it/s, tok/s=15003.8]
Inference: 26%|██▌ | 1544/5920 [22:02<49:05, 1.49it/s, tok/s=15003.8]
Inference: 26%|██▌ | 1544/5920 [22:02<49:05, 1.49it/s, tok/s=15011.4]
Inference: 26%|██▌ | 1545/5920 [22:02<49:04, 1.49it/s, tok/s=15027.2]
Inference: 26%|██▌ | 1546/5920 [22:02<31:21, 2.32it/s, tok/s=15027.2]
Inference: 26%|██▌ | 1546/5920 [22:02<31:21, 2.32it/s, tok/s=15028.3]
Inference: 26%|██▌ | 1547/5920 [22:03<42:02, 1.73it/s, tok/s=15028.3]
Inference: 26%|██▌ | 1547/5920 [22:03<42:02, 1.73it/s, tok/s=15027.5]
Inference: 26%|██▌ | 1548/5920 [22:04<54:11, 1.34it/s, tok/s=15027.5]
Inference: 26%|██▌ | 1548/5920 [22:04<54:11, 1.34it/s, tok/s=15017.9]
Inference: 26%|██▌ | 1549/5920 [22:05<50:13, 1.45it/s, tok/s=15017.9]
Inference: 26%|██▌ | 1549/5920 [22:05<50:13, 1.45it/s, tok/s=15028.8]
Inference: 26%|██▌ | 1550/5920 [22:06<1:00:45, 1.20it/s, tok/s=15028.8]
Inference: 26%|██▌ | 1550/5920 [22:06<1:00:45, 1.20it/s, tok/s=15025.6]
Inference: 26%|██▌ | 1551/5920 [22:07<1:02:31, 1.16it/s, tok/s=15025.6]
Inference: 26%|██▌ | 1551/5920 [22:07<1:02:31, 1.16it/s, tok/s=15022.2]
Inference: 26%|██▌ | 1552/5920 [22:07<53:48, 1.35it/s, tok/s=15022.2]
Inference: 26%|██▌ | 1552/5920 [22:07<53:48, 1.35it/s, tok/s=15022.8]
Inference: 26%|██▌ | 1553/5920 [22:08<46:55, 1.55it/s, tok/s=15022.8]
Inference: 26%|██▌ | 1553/5920 [22:08<46:55, 1.55it/s, tok/s=15028.8]
Inference: 26%|██▋ | 1554/5920 [22:08<45:06, 1.61it/s, tok/s=15028.8]
Inference: 26%|██▋ | 1554/5920 [22:08<45:06, 1.61it/s, tok/s=15027.1]
Inference: 26%|██▋ | 1555/5920 [22:09<40:21, 1.80it/s, tok/s=15027.1]
Inference: 26%|██▋ | 1555/5920 [22:09<40:21, 1.80it/s, tok/s=15028.3]
Inference: 26%|██▋ | 1556/5920 [22:10<1:04:43, 1.12it/s, tok/s=15028.3]
Inference: 26%|██▋ | 1556/5920 [22:10<1:04:43, 1.12it/s, tok/s=15012.9]
Inference: 26%|██▋ | 1557/5920 [22:11<50:59, 1.43it/s, tok/s=15012.9]
Inference: 26%|██▋ | 1557/5920 [22:11<50:59, 1.43it/s, tok/s=15020.6]
Inference: 26%|██▋ | 1558/5920 [22:11<50:58, 1.43it/s, tok/s=15023.8]
Inference: 26%|██▋ | 1559/5920 [22:11<39:59, 1.82it/s, tok/s=15023.8]
Inference: 26%|██▋ | 1559/5920 [22:11<39:59, 1.82it/s, tok/s=15038.3]
Inference: 26%|██▋ | 1560/5920 [22:12<32:19, 2.25it/s, tok/s=15038.3]
Inference: 26%|██▋ | 1560/5920 [22:12<32:19, 2.25it/s, tok/s=15044.2]
Inference: 26%|██▋ | 1561/5920 [22:12<32:18, 2.25it/s, tok/s=15046.5]
Inference: 26%|██▋ | 1562/5920 [22:12<30:44, 2.36it/s, tok/s=15046.5]
Inference: 26%|██▋ | 1562/5920 [22:12<30:44, 2.36it/s, tok/s=15040.9]
Inference: 26%|██▋ | 1563/5920 [22:14<50:53, 1.43it/s, tok/s=15040.9]
Inference: 26%|██▋ | 1563/5920 [22:14<50:53, 1.43it/s, tok/s=15030.1]
Inference: 26%|██▋ | 1564/5920 [22:15<1:01:11, 1.19it/s, tok/s=15030.1]
Inference: 26%|██▋ | 1564/5920 [22:15<1:01:11, 1.19it/s, tok/s=15024.6]
Inference: 26%|██▋ | 1565/5920 [22:16<1:01:45, 1.18it/s, tok/s=15024.6]
Inference: 26%|██▋ | 1565/5920 [22:16<1:01:45, 1.18it/s, tok/s=15029.0]
Inference: 26%|██▋ | 1566/5920 [22:16<1:01:44, 1.18it/s, tok/s=15050.5]
Inference: 26%|██▋ | 1567/5920 [22:18<1:01:14, 1.18it/s, tok/s=15050.5]
Inference: 26%|██▋ | 1567/5920 [22:18<1:01:14, 1.18it/s, tok/s=15043.9]
Inference: 26%|██▋ | 1568/5920 [22:18<51:46, 1.40it/s, tok/s=15043.9]
Inference: 26%|██▋ | 1568/5920 [22:18<51:46, 1.40it/s, tok/s=15059.1]
Inference: 27%|██▋ | 1569/5920 [22:18<51:45, 1.40it/s, tok/s=15070.4]
Inference: 27%|██▋ | 1570/5920 [22:18<34:26, 2.10it/s, tok/s=15070.4]
Inference: 27%|██▋ | 1570/5920 [22:18<34:26, 2.10it/s, tok/s=15083.1]
Inference: 27%|██▋ | 1571/5920 [22:18<34:26, 2.10it/s, tok/s=15092.4]
Inference: 27%|██▋ | 1572/5920 [22:19<28:03, 2.58it/s, tok/s=15092.4]
Inference: 27%|██▋ | 1572/5920 [22:19<28:03, 2.58it/s, tok/s=15091.7]
Inference: 27%|██▋ | 1573/5920 [22:19<28:02, 2.58it/s, tok/s=15095.0]
Inference: 27%|██▋ | 1574/5920 [22:19<21:57, 3.30it/s, tok/s=15095.0]
Inference: 27%|██▋ | 1574/5920 [22:19<21:57, 3.30it/s, tok/s=15114.7]
Inference: 27%|██▋ | 1575/5920 [22:20<26:27, 2.74it/s, tok/s=15114.7]
Inference: 27%|██▋ | 1575/5920 [22:20<26:27, 2.74it/s, tok/s=15111.9]
Inference: 27%|██▋ | 1576/5920 [22:20<26:26, 2.74it/s, tok/s=15115.1]
Inference: 27%|██▋ | 1577/5920 [22:20<22:35, 3.20it/s, tok/s=15115.1]
Inference: 27%|██▋ | 1577/5920 [22:20<22:35, 3.20it/s, tok/s=15133.6]
Inference: 27%|██▋ | 1578/5920 [22:21<29:42, 2.44it/s, tok/s=15133.6]
Inference: 27%|██▋ | 1578/5920 [22:21<29:42, 2.44it/s, tok/s=15127.3]
Inference: 27%|██▋ | 1579/5920 [22:22<45:40, 1.58it/s, tok/s=15127.3]
Inference: 27%|██▋ | 1579/5920 [22:22<45:40, 1.58it/s, tok/s=15129.6]
Inference: 27%|██▋ | 1580/5920 [22:25<1:18:33, 1.09s/it, tok/s=15129.6]
Inference: 27%|██▋ | 1580/5920 [22:25<1:18:33, 1.09s/it, tok/s=15117.1]
Inference: 27%|██▋ | 1581/5920 [22:26<1:17:32, 1.07s/it, tok/s=15117.1]
Inference: 27%|██▋ | 1581/5920 [22:26<1:17:32, 1.07s/it, tok/s=15119.7]
Inference: 27%|██▋ | 1582/5920 [22:26<1:07:28, 1.07it/s, tok/s=15119.7]
Inference: 27%|██▋ | 1582/5920 [22:26<1:07:28, 1.07it/s, tok/s=15118.9]
Inference: 27%|██▋ | 1583/5920 [22:27<1:00:20, 1.20it/s, tok/s=15118.9]
Inference: 27%|██▋ | 1583/5920 [22:27<1:00:20, 1.20it/s, tok/s=15124.5]
Inference: 27%|██▋ | 1584/5920 [22:28<1:07:16, 1.07it/s, tok/s=15124.5]
Inference: 27%|██▋ | 1584/5920 [22:28<1:07:16, 1.07it/s, tok/s=15116.2]
Inference: 27%|██▋ | 1585/5920 [22:30<1:21:18, 1.13s/it, tok/s=15116.2]
Inference: 27%|██▋ | 1585/5920 [22:30<1:21:18, 1.13s/it, tok/s=15113.9]
Inference: 27%|██▋ | 1586/5920 [22:31<1:18:40, 1.09s/it, tok/s=15113.9]
Inference: 27%|██▋ | 1586/5920 [22:31<1:18:40, 1.09s/it, tok/s=15125.0]
Inference: 27%|██▋ | 1587/5920 [22:33<1:48:43, 1.51s/it, tok/s=15125.0]
Inference: 27%|██▋ | 1587/5920 [22:33<1:48:43, 1.51s/it, tok/s=15103.5]
Inference: 27%|██▋ | 1588/5920 [22:33<1:22:27, 1.14s/it, tok/s=15103.5]
Inference: 27%|██▋ | 1588/5920 [22:33<1:22:27, 1.14s/it, tok/s=15105.9]
Inference: 27%|██▋ | 1589/5920 [22:34<1:00:52, 1.19it/s, tok/s=15105.9]
Inference: 27%|██▋ | 1589/5920 [22:34<1:00:52, 1.19it/s, tok/s=15111.4]
Inference: 27%|██▋ | 1590/5920 [22:34<49:21, 1.46it/s, tok/s=15111.4]
Inference: 27%|██▋ | 1590/5920 [22:34<49:21, 1.46it/s, tok/s=15120.7]
Inference: 27%|██▋ | 1591/5920 [22:34<40:56, 1.76it/s, tok/s=15120.7]
Inference: 27%|██▋ | 1591/5920 [22:34<40:56, 1.76it/s, tok/s=15124.8]
Inference: 27%|██▋ | 1592/5920 [22:34<30:53, 2.34it/s, tok/s=15124.8]
Inference: 27%|██▋ | 1592/5920 [22:34<30:53, 2.34it/s, tok/s=15131.0]
Inference: 27%|██▋ | 1593/5920 [22:35<25:40, 2.81it/s, tok/s=15131.0]
Inference: 27%|██▋ | 1593/5920 [22:35<25:40, 2.81it/s, tok/s=15140.9]
Inference: 27%|██▋ | 1594/5920 [22:35<26:35, 2.71it/s, tok/s=15140.9]
Inference: 27%|██▋ | 1594/5920 [22:35<26:35, 2.71it/s, tok/s=15140.2]
Inference: 27%|██▋ | 1595/5920 [22:35<26:35, 2.71it/s, tok/s=15142.7]
Inference: 27%|██▋ | 1596/5920 [22:36<39:54, 1.81it/s, tok/s=15142.7]
Inference: 27%|██▋ | 1596/5920 [22:36<39:54, 1.81it/s, tok/s=15143.2]
Inference: 27%|██▋ | 1597/5920 [22:39<1:17:26, 1.07s/it, tok/s=15143.2]
Inference: 27%|██▋ | 1597/5920 [22:39<1:17:26, 1.07s/it, tok/s=15127.2]
Inference: 27%|██▋ | 1598/5920 [22:39<59:22, 1.21it/s, tok/s=15127.2]
Inference: 27%|██▋ | 1598/5920 [22:39<59:22, 1.21it/s, tok/s=15129.1]
Inference: 27%|██▋ | 1599/5920 [22:40<59:12, 1.22it/s, tok/s=15129.1]
Inference: 27%|██▋ | 1599/5920 [22:40<59:12, 1.22it/s, tok/s=15130.3]
Inference: 27%|██▋ | 1600/5920 [22:41<1:05:49, 1.09it/s, tok/s=15130.3]
Inference: 27%|██▋ | 1600/5920 [22:41<1:05:49, 1.09it/s, tok/s=15135.1]
Inference: 27%|██▋ | 1601/5920 [22:43<1:30:58, 1.26s/it, tok/s=15135.1]
Inference: 27%|██▋ | 1601/5920 [22:43<1:30:58, 1.26s/it, tok/s=15128.8]
Inference: 27%|██▋ | 1602/5920 [22:45<1:34:10, 1.31s/it, tok/s=15128.8]
Inference: 27%|██▋ | 1602/5920 [22:45<1:34:10, 1.31s/it, tok/s=15119.0]
Inference: 27%|██▋ | 1603/5920 [22:49<2:34:47, 2.15s/it, tok/s=15119.0]
Inference: 27%|██▋ | 1603/5920 [22:49<2:34:47, 2.15s/it, tok/s=15087.6]
Inference: 27%|██▋ | 1604/5920 [22:49<1:53:20, 1.58s/it, tok/s=15087.6]
Inference: 27%|██▋ | 1604/5920 [22:49<1:53:20, 1.58s/it, tok/s=15088.7]
Inference: 27%|██▋ | 1605/5920 [22:49<1:24:39, 1.18s/it, tok/s=15088.7]
Inference: 27%|██▋ | 1605/5920 [22:49<1:24:39, 1.18s/it, tok/s=15108.1]
Inference: 27%|██▋ | 1606/5920 [22:54<2:28:22, 2.06s/it, tok/s=15108.1]
Inference: 27%|██▋ | 1606/5920 [22:54<2:28:22, 2.06s/it, tok/s=15067.0]
Inference: 27%|██▋ | 1607/5920 [22:54<2:01:36, 1.69s/it, tok/s=15067.0]
Inference: 27%|██▋ | 1607/5920 [22:54<2:01:36, 1.69s/it, tok/s=15076.1]
Inference: 27%|██▋ | 1608/5920 [22:54<2:01:35, 1.69s/it, tok/s=15084.3]
Inference: 27%|██▋ | 1609/5920 [22:56<1:30:34, 1.26s/it, tok/s=15084.3]
Inference: 27%|██▋ | 1609/5920 [22:56<1:30:34, 1.26s/it, tok/s=15071.2]
Inference: 27%|██▋ | 1610/5920 [22:56<1:19:18, 1.10s/it, tok/s=15071.2]
Inference: 27%|██▋ | 1610/5920 [22:56<1:19:18, 1.10s/it, tok/s=15067.8]
Inference: 27%|██▋ | 1611/5920 [22:57<1:00:28, 1.19it/s, tok/s=15067.8]
Inference: 27%|██▋ | 1611/5920 [22:57<1:00:28, 1.19it/s, tok/s=15085.2]
Inference: 27%|██▋ | 1612/5920 [22:57<48:50, 1.47it/s, tok/s=15085.2]
Inference: 27%|██▋ | 1612/5920 [22:57<48:50, 1.47it/s, tok/s=15099.9]
Inference: 27%|██▋ | 1613/5920 [23:01<1:54:33, 1.60s/it, tok/s=15099.9]
Inference: 27%|██▋ | 1613/5920 [23:01<1:54:33, 1.60s/it, tok/s=15069.2]
Inference: 27%|██▋ | 1614/5920 [23:02<1:52:49, 1.57s/it, tok/s=15069.2]
Inference: 27%|██▋ | 1614/5920 [23:02<1:52:49, 1.57s/it, tok/s=15060.4]
Inference: 27%|██▋ | 1615/5920 [23:06<2:38:02, 2.20s/it, tok/s=15060.4]
Inference: 27%|██▋ | 1615/5920 [23:06<2:38:02, 2.20s/it, tok/s=15026.2]
Inference: 27%|██▋ | 1616/5920 [23:06<1:55:42, 1.61s/it, tok/s=15026.2]
Inference: 27%|██▋ | 1616/5920 [23:06<1:55:42, 1.61s/it, tok/s=15027.8]
Inference: 27%|██▋ | 1617/5920 [23:07<1:29:20, 1.25s/it, tok/s=15027.8]
Inference: 27%|██▋ | 1617/5920 [23:07<1:29:20, 1.25s/it, tok/s=15037.0]
Inference: 27%|██▋ | 1618/5920 [23:08<1:35:42, 1.33s/it, tok/s=15037.0]
Inference: 27%|██▋ | 1618/5920 [23:08<1:35:42, 1.33s/it, tok/s=15038.7]
Inference: 27%|██▋ | 1619/5920 [23:08<1:10:47, 1.01it/s, tok/s=15038.7]
Inference: 27%|██▋ | 1619/5920 [23:08<1:10:47, 1.01it/s, tok/s=15051.4]
Inference: 27%|██▋ | 1620/5920 [23:10<1:25:46, 1.20s/it, tok/s=15051.4]
Inference: 27%|██▋ | 1620/5920 [23:10<1:25:46, 1.20s/it, tok/s=15040.8]
Inference: 27%|██▋ | 1621/5920 [23:13<2:02:05, 1.70s/it, tok/s=15040.8]
Inference: 27%|██▋ | 1621/5920 [23:13<2:02:05, 1.70s/it, tok/s=15011.8]
Inference: 27%|██▋ | 1622/5920 [23:14<1:44:12, 1.45s/it, tok/s=15011.8]
Inference: 27%|██▋ | 1622/5920 [23:14<1:44:12, 1.45s/it, tok/s=15010.0]
Inference: 27%|██▋ | 1623/5920 [23:15<1:30:56, 1.27s/it, tok/s=15010.0]
Inference: 27%|██▋ | 1623/5920 [23:15<1:30:56, 1.27s/it, tok/s=15009.5]
Inference: 27%|██▋ | 1624/5920 [23:17<1:51:42, 1.56s/it, tok/s=15009.5]
Inference: 27%|██▋ | 1624/5920 [23:17<1:51:42, 1.56s/it, tok/s=14992.8]
Inference: 27%|██▋ | 1625/5920 [23:22<3:15:09, 2.73s/it, tok/s=14992.8]
Inference: 27%|██▋ | 1625/5920 [23:22<3:15:09, 2.73s/it, tok/s=14944.2]
Inference: 27%|██▋ | 1626/5920 [23:24<3:02:07, 2.54s/it, tok/s=14944.2]
Inference: 27%|██▋ | 1626/5920 [23:24<3:02:07, 2.54s/it, tok/s=14924.1]
Inference: 27%|██▋ | 1627/5920 [23:25<2:16:34, 1.91s/it, tok/s=14924.1]
Inference: 27%|██▋ | 1627/5920 [23:25<2:16:34, 1.91s/it, tok/s=14934.3]
Inference: 28%|██▊ | 1628/5920 [23:27<2:16:14, 1.90s/it, tok/s=14934.3]
Inference: 28%|██▊ | 1628/5920 [23:27<2:16:14, 1.90s/it, tok/s=14931.7]
Inference: 28%|██▊ | 1629/5920 [23:30<2:42:36, 2.27s/it, tok/s=14931.7]
Inference: 28%|██▊ | 1629/5920 [23:30<2:42:36, 2.27s/it, tok/s=14908.7]
Inference: 28%|██▊ | 1630/5920 [23:30<2:02:04, 1.71s/it, tok/s=14908.7]
Inference: 28%|██▊ | 1630/5920 [23:30<2:02:04, 1.71s/it, tok/s=14918.4]
Inference: 28%|██▊ | 1631/5920 [23:30<1:27:53, 1.23s/it, tok/s=14918.4]
Inference: 28%|██▊ | 1631/5920 [23:30<1:27:53, 1.23s/it, tok/s=14926.3]
Inference: 28%|██▊ | 1632/5920 [23:30<1:03:53, 1.12it/s, tok/s=14926.3]
Inference: 28%|██▊ | 1632/5920 [23:30<1:03:53, 1.12it/s, tok/s=14927.6]
Inference: 28%|██▊ | 1633/5920 [23:31<1:03:53, 1.12it/s, tok/s=14939.6]
Inference: 28%|██▊ | 1634/5920 [23:31<42:21, 1.69it/s, tok/s=14939.6]
Inference: 28%|██▊ | 1634/5920 [23:31<42:21, 1.69it/s, tok/s=14938.4]
Inference: 28%|██▊ | 1635/5920 [23:31<42:21, 1.69it/s, tok/s=14951.5]
Inference: 28%|██▊ | 1636/5920 [23:33<50:37, 1.41it/s, tok/s=14951.5]
Inference: 28%|██▊ | 1636/5920 [23:33<50:37, 1.41it/s, tok/s=14950.4]
Inference: 28%|██▊ | 1637/5920 [23:35<1:10:03, 1.02it/s, tok/s=14950.4]
Inference: 28%|██▊ | 1637/5920 [23:35<1:10:03, 1.02it/s, tok/s=14945.8]
Inference: 28%|██▊ | 1638/5920 [23:37<1:28:26, 1.24s/it, tok/s=14945.8]
Inference: 28%|██▊ | 1638/5920 [23:37<1:28:26, 1.24s/it, tok/s=14929.9]
Inference: 28%|██▊ | 1639/5920 [23:38<1:24:35, 1.19s/it, tok/s=14929.9]
Inference: 28%|██▊ | 1639/5920 [23:38<1:24:35, 1.19s/it, tok/s=14934.9]
Inference: 28%|██▊ | 1640/5920 [23:38<1:05:12, 1.09it/s, tok/s=14934.9]
Inference: 28%|██▊ | 1640/5920 [23:38<1:05:12, 1.09it/s, tok/s=14942.5]
Inference: 28%|██▊ | 1641/5920 [23:40<1:22:04, 1.15s/it, tok/s=14942.5]
Inference: 28%|██▊ | 1641/5920 [23:40<1:22:04, 1.15s/it, tok/s=14930.4]
Inference: 28%|██▊ | 1642/5920 [23:41<1:25:09, 1.19s/it, tok/s=14930.4]
Inference: 28%|██▊ | 1642/5920 [23:41<1:25:09, 1.19s/it, tok/s=14927.8]
Inference: 28%|██▊ | 1643/5920 [23:44<1:54:26, 1.61s/it, tok/s=14927.8]
Inference: 28%|██▊ | 1643/5920 [23:44<1:54:26, 1.61s/it, tok/s=14904.5]
Inference: 28%|██▊ | 1644/5920 [23:45<1:46:07, 1.49s/it, tok/s=14904.5]
Inference: 28%|██▊ | 1644/5920 [23:45<1:46:07, 1.49s/it, tok/s=14895.7]
Inference: 28%|██▊ | 1645/5920 [23:46<1:32:11, 1.29s/it, tok/s=14895.7]
Inference: 28%|██▊ | 1645/5920 [23:46<1:32:11, 1.29s/it, tok/s=14890.2]
Inference: 28%|██▊ | 1646/5920 [23:46<1:14:57, 1.05s/it, tok/s=14890.2]
Inference: 28%|██▊ | 1646/5920 [23:46<1:14:57, 1.05s/it, tok/s=14901.2]
Inference: 28%|██▊ | 1647/5920 [23:48<1:22:06, 1.15s/it, tok/s=14901.2]
Inference: 28%|██▊ | 1647/5920 [23:48<1:22:06, 1.15s/it, tok/s=14889.5]
Inference: 28%|██▊ | 1648/5920 [23:49<1:33:44, 1.32s/it, tok/s=14889.5]
Inference: 28%|██▊ | 1648/5920 [23:49<1:33:44, 1.32s/it, tok/s=14876.0]
Inference: 28%|██▊ | 1649/5920 [23:50<1:25:06, 1.20s/it, tok/s=14876.0]
Inference: 28%|██▊ | 1649/5920 [23:50<1:25:06, 1.20s/it, tok/s=14870.5]
Inference: 28%|██▊ | 1650/5920 [23:51<1:09:06, 1.03it/s, tok/s=14870.5]
Inference: 28%|██▊ | 1650/5920 [23:51<1:09:06, 1.03it/s, tok/s=14880.6]
Inference: 28%|██▊ | 1651/5920 [23:51<51:30, 1.38it/s, tok/s=14880.6]
Inference: 28%|██▊ | 1651/5920 [23:51<51:30, 1.38it/s, tok/s=14900.1]
Inference: 28%|██▊ | 1652/5920 [23:52<1:04:07, 1.11it/s, tok/s=14900.1]
Inference: 28%|██▊ | 1652/5920 [23:52<1:04:07, 1.11it/s, tok/s=14889.4]
Inference: 28%|██▊ | 1653/5920 [23:52<52:52, 1.35it/s, tok/s=14889.4]
Inference: 28%|██▊ | 1653/5920 [23:52<52:52, 1.35it/s, tok/s=14897.8]
Inference: 28%|██▊ | 1654/5920 [23:52<52:51, 1.35it/s, tok/s=14900.7]
Inference: 28%|██▊ | 1655/5920 [23:53<36:48, 1.93it/s, tok/s=14900.7]
Inference: 28%|██▊ | 1655/5920 [23:53<36:48, 1.93it/s, tok/s=14911.5]
Inference: 28%|██▊ | 1656/5920 [23:53<36:48, 1.93it/s, tok/s=14921.1]
Inference: 28%|██▊ | 1657/5920 [23:53<28:31, 2.49it/s, tok/s=14921.1]
Inference: 28%|██▊ | 1657/5920 [23:53<28:31, 2.49it/s, tok/s=14920.0]
Inference: 28%|██▊ | 1658/5920 [23:54<27:17, 2.60it/s, tok/s=14920.0]
Inference: 28%|██▊ | 1658/5920 [23:54<27:17, 2.60it/s, tok/s=14920.3]
Inference: 28%|██▊ | 1659/5920 [23:55<35:56, 1.98it/s, tok/s=14920.3]
Inference: 28%|██▊ | 1659/5920 [23:55<35:56, 1.98it/s, tok/s=14913.7]
Inference: 28%|██▊ | 1660/5920 [23:55<35:55, 1.98it/s, tok/s=14920.5]
Inference: 28%|██▊ | 1661/5920 [23:56<47:46, 1.49it/s, tok/s=14920.5]
Inference: 28%|██▊ | 1661/5920 [23:56<47:46, 1.49it/s, tok/s=14911.4]
Inference: 28%|██▊ | 1662/5920 [23:58<57:45, 1.23it/s, tok/s=14911.4]
Inference: 28%|██▊ | 1662/5920 [23:58<57:45, 1.23it/s, tok/s=14900.9]
Inference: 28%|██▊ | 1663/5920 [23:58<53:49, 1.32it/s, tok/s=14900.9]
Inference: 28%|██▊ | 1663/5920 [23:58<53:49, 1.32it/s, tok/s=14897.7]
Inference: 28%|██▊ | 1664/5920 [23:59<45:14, 1.57it/s, tok/s=14897.7]
Inference: 28%|██▊ | 1664/5920 [23:59<45:14, 1.57it/s, tok/s=14903.5]
Inference: 28%|██▊ | 1665/5920 [23:59<38:21, 1.85it/s, tok/s=14903.5]
Inference: 28%|██▊ | 1665/5920 [23:59<38:21, 1.85it/s, tok/s=14905.3]
Inference: 28%|██▊ | 1666/5920 [24:04<1:59:43, 1.69s/it, tok/s=14905.3]
Inference: 28%|██▊ | 1666/5920 [24:04<1:59:43, 1.69s/it, tok/s=14870.0]
Inference: 28%|██▊ | 1667/5920 [24:04<1:32:38, 1.31s/it, tok/s=14870.0]
Inference: 28%|██▊ | 1667/5920 [24:04<1:32:38, 1.31s/it, tok/s=14880.5]
Inference: 28%|██▊ | 1668/5920 [24:04<1:32:37, 1.31s/it, tok/s=14884.0]
Inference: 28%|██▊ | 1669/5920 [24:04<1:32:35, 1.31s/it, tok/s=14886.6]
Inference: 28%|██▊ | 1670/5920 [24:07<1:17:55, 1.10s/it, tok/s=14886.6]
Inference: 28%|██▊ | 1670/5920 [24:07<1:17:55, 1.10s/it, tok/s=14868.4]
Inference: 28%|██▊ | 1671/5920 [24:07<1:07:49, 1.04it/s, tok/s=14868.4]
Inference: 28%|██▊ | 1671/5920 [24:07<1:07:49, 1.04it/s, tok/s=14881.0]
Inference: 28%|██▊ | 1672/5920 [24:08<59:49, 1.18it/s, tok/s=14881.0]
Inference: 28%|██▊ | 1672/5920 [24:08<59:49, 1.18it/s, tok/s=14879.6]
Inference: 28%|██▊ | 1673/5920 [24:08<47:19, 1.50it/s, tok/s=14879.6]
Inference: 28%|██▊ | 1673/5920 [24:08<47:19, 1.50it/s, tok/s=14891.5]
Inference: 28%|██▊ | 1674/5920 [24:08<39:28, 1.79it/s, tok/s=14891.5]
Inference: 28%|██▊ | 1674/5920 [24:08<39:28, 1.79it/s, tok/s=14893.0]
Inference: 28%|██▊ | 1675/5920 [24:10<1:09:18, 1.02it/s, tok/s=14893.0]
Inference: 28%|██▊ | 1675/5920 [24:10<1:09:18, 1.02it/s, tok/s=14884.6]
Inference: 28%|██▊ | 1676/5920 [24:11<1:09:09, 1.02it/s, tok/s=14884.6]
Inference: 28%|██▊ | 1676/5920 [24:11<1:09:09, 1.02it/s, tok/s=14880.1]
Inference: 28%|██▊ | 1677/5920 [24:14<1:41:12, 1.43s/it, tok/s=14880.1]
Inference: 28%|██▊ | 1677/5920 [24:14<1:41:12, 1.43s/it, tok/s=14857.1]
Inference: 28%|██▊ | 1678/5920 [24:15<1:41:52, 1.44s/it, tok/s=14857.1]
Inference: 28%|██▊ | 1678/5920 [24:15<1:41:52, 1.44s/it, tok/s=14845.4]
Inference: 28%|██▊ | 1679/5920 [24:15<1:41:50, 1.44s/it, tok/s=14851.9]
Inference: 28%|██▊ | 1680/5920 [24:16<1:03:14, 1.12it/s, tok/s=14851.9]
Inference: 28%|██▊ | 1680/5920 [24:16<1:03:14, 1.12it/s, tok/s=14851.5]
Inference: 28%|██▊ | 1681/5920 [24:17<1:07:10, 1.05it/s, tok/s=14851.5]
Inference: 28%|██▊ | 1681/5920 [24:17<1:07:10, 1.05it/s, tok/s=14854.1]
Inference: 28%|██▊ | 1682/5920 [24:19<1:29:20, 1.26s/it, tok/s=14854.1]
Inference: 28%|██▊ | 1682/5920 [24:19<1:29:20, 1.26s/it, tok/s=14835.0]
Inference: 28%|██▊ | 1683/5920 [24:19<1:11:05, 1.01s/it, tok/s=14835.0]
Inference: 28%|██▊ | 1683/5920 [24:19<1:11:05, 1.01s/it, tok/s=14837.0]
Inference: 28%|██▊ | 1684/5920 [24:20<1:13:20, 1.04s/it, tok/s=14837.0]
Inference: 28%|██▊ | 1684/5920 [24:20<1:13:20, 1.04s/it, tok/s=14828.7]
Inference: 28%|██▊ | 1685/5920 [24:21<1:02:46, 1.12it/s, tok/s=14828.7]
Inference: 28%|██▊ | 1685/5920 [24:21<1:02:46, 1.12it/s, tok/s=14826.9]
Inference: 28%|██▊ | 1686/5920 [24:21<57:28, 1.23it/s, tok/s=14826.9]
Inference: 28%|██▊ | 1686/5920 [24:21<57:28, 1.23it/s, tok/s=14824.6]
Inference: 28%|██▊ | 1687/5920 [24:24<1:24:06, 1.19s/it, tok/s=14824.6]
Inference: 28%|██▊ | 1687/5920 [24:24<1:24:06, 1.19s/it, tok/s=14806.5]
Inference: 29%|██▊ | 1688/5920 [24:25<1:38:20, 1.39s/it, tok/s=14806.5]
Inference: 29%|██▊ | 1688/5920 [24:25<1:38:20, 1.39s/it, tok/s=14792.6]
Inference: 29%|██▊ | 1689/5920 [24:25<1:38:19, 1.39s/it, tok/s=14796.2]
Inference: 29%|██▊ | 1690/5920 [24:27<1:13:52, 1.05s/it, tok/s=14796.2]
Inference: 29%|██▊ | 1690/5920 [24:27<1:13:52, 1.05s/it, tok/s=14787.4]
Inference: 29%|██▊ | 1691/5920 [24:29<1:31:51, 1.30s/it, tok/s=14787.4]
Inference: 29%|██▊ | 1691/5920 [24:29<1:31:51, 1.30s/it, tok/s=14783.8]
Inference: 29%|██▊ | 1692/5920 [24:29<1:15:07, 1.07s/it, tok/s=14783.8]
Inference: 29%|██▊ | 1692/5920 [24:29<1:15:07, 1.07s/it, tok/s=14799.0]
Inference: 29%|██▊ | 1693/5920 [24:30<1:05:39, 1.07it/s, tok/s=14799.0]
Inference: 29%|██▊ | 1693/5920 [24:30<1:05:39, 1.07it/s, tok/s=14797.7]
Inference: 29%|██▊ | 1694/5920 [24:31<1:06:06, 1.07it/s, tok/s=14797.7]
Inference: 29%|██▊ | 1694/5920 [24:31<1:06:06, 1.07it/s, tok/s=14792.7]
Inference: 29%|██▊ | 1695/5920 [24:31<1:06:05, 1.07it/s, tok/s=14796.3]
Inference: 29%|██▊ | 1696/5920 [24:32<58:33, 1.20it/s, tok/s=14796.3]
Inference: 29%|██▊ | 1696/5920 [24:32<58:33, 1.20it/s, tok/s=14800.9]
Inference: 29%|██▊ | 1697/5920 [24:32<49:50, 1.41it/s, tok/s=14800.9]
Inference: 29%|██▊ | 1697/5920 [24:32<49:50, 1.41it/s, tok/s=14811.6]
Inference: 29%|██▊ | 1698/5920 [24:35<1:25:02, 1.21s/it, tok/s=14811.6]
Inference: 29%|██▊ | 1698/5920 [24:35<1:25:02, 1.21s/it, tok/s=14798.8]
Inference: 29%|██▊ | 1699/5920 [24:36<1:15:17, 1.07s/it, tok/s=14798.8]
Inference: 29%|██▊ | 1699/5920 [24:36<1:15:17, 1.07s/it, tok/s=14796.8]
Inference: 29%|██▊ | 1700/5920 [24:36<58:18, 1.21it/s, tok/s=14796.8]
Inference: 29%|██▊ | 1700/5920 [24:36<58:18, 1.21it/s, tok/s=14799.2]
Inference: 29%|██▊ | 1701/5920 [24:37<1:10:17, 1.00it/s, tok/s=14799.2]
Inference: 29%|██▊ | 1701/5920 [24:37<1:10:17, 1.00it/s, tok/s=14790.0]
Inference: 29%|██▉ | 1702/5920 [24:38<58:50, 1.19it/s, tok/s=14790.0]
Inference: 29%|██▉ | 1702/5920 [24:38<58:50, 1.19it/s, tok/s=14788.9]
Inference: 29%|██▉ | 1703/5920 [24:38<51:21, 1.37it/s, tok/s=14788.9]
Inference: 29%|██▉ | 1703/5920 [24:38<51:21, 1.37it/s, tok/s=14787.4]
Inference: 29%|██▉ | 1704/5920 [24:39<41:07, 1.71it/s, tok/s=14787.4]
Inference: 29%|██▉ | 1704/5920 [24:39<41:07, 1.71it/s, tok/s=14793.6]
Inference: 29%|██▉ | 1705/5920 [24:39<37:08, 1.89it/s, tok/s=14793.6]
Inference: 29%|██▉ | 1705/5920 [24:39<37:08, 1.89it/s, tok/s=14796.1]
Inference: 29%|██▉ | 1706/5920 [24:40<47:08, 1.49it/s, tok/s=14796.1]
Inference: 29%|██▉ | 1706/5920 [24:40<47:08, 1.49it/s, tok/s=14789.2]
Inference: 29%|██▉ | 1707/5920 [24:40<42:55, 1.64it/s, tok/s=14789.2]
Inference: 29%|██▉ | 1707/5920 [24:40<42:55, 1.64it/s, tok/s=14788.1]
Inference: 29%|██▉ | 1708/5920 [24:41<41:30, 1.69it/s, tok/s=14788.1]
Inference: 29%|██▉ | 1708/5920 [24:41<41:30, 1.69it/s, tok/s=14786.8]
Inference: 29%|██▉ | 1709/5920 [24:42<49:28, 1.42it/s, tok/s=14786.8]
Inference: 29%|██▉ | 1709/5920 [24:42<49:28, 1.42it/s, tok/s=14782.0]
Inference: 29%|██▉ | 1710/5920 [24:44<1:18:04, 1.11s/it, tok/s=14782.0]
Inference: 29%|██▉ | 1710/5920 [24:44<1:18:04, 1.11s/it, tok/s=14765.4]
Inference: 29%|██▉ | 1711/5920 [24:45<1:11:23, 1.02s/it, tok/s=14765.4]
Inference: 29%|██▉ | 1711/5920 [24:45<1:11:23, 1.02s/it, tok/s=14773.5]
Inference: 29%|██▉ | 1712/5920 [24:45<1:11:22, 1.02s/it, tok/s=14777.7]
Inference: 29%|██▉ | 1713/5920 [24:45<42:40, 1.64it/s, tok/s=14777.7]
Inference: 29%|██▉ | 1713/5920 [24:45<42:40, 1.64it/s, tok/s=14794.8]
Inference: 29%|██▉ | 1714/5920 [24:45<36:26, 1.92it/s, tok/s=14794.8]
Inference: 29%|██▉ | 1714/5920 [24:45<36:26, 1.92it/s, tok/s=14796.0]
Inference: 29%|██▉ | 1715/5920 [24:46<37:27, 1.87it/s, tok/s=14796.0]
Inference: 29%|██▉ | 1715/5920 [24:46<37:27, 1.87it/s, tok/s=14793.9]
Inference: 29%|██▉ | 1716/5920 [24:47<53:08, 1.32it/s, tok/s=14793.9]
Inference: 29%|██▉ | 1716/5920 [24:47<53:08, 1.32it/s, tok/s=14785.8]
Inference: 29%|██▉ | 1717/5920 [24:50<1:34:31, 1.35s/it, tok/s=14785.8]
Inference: 29%|██▉ | 1717/5920 [24:50<1:34:31, 1.35s/it, tok/s=14763.3]
Inference: 29%|██▉ | 1718/5920 [24:51<1:28:40, 1.27s/it, tok/s=14763.3]
Inference: 29%|██▉ | 1718/5920 [24:51<1:28:40, 1.27s/it, tok/s=14769.7]
Inference: 29%|██▉ | 1719/5920 [24:51<1:28:39, 1.27s/it, tok/s=14775.9]
Inference: 29%|██▉ | 1720/5920 [24:52<1:05:38, 1.07it/s, tok/s=14775.9]
Inference: 29%|██▉ | 1720/5920 [24:52<1:05:38, 1.07it/s, tok/s=14769.3]
Inference: 29%|██▉ | 1721/5920 [24:54<1:28:26, 1.26s/it, tok/s=14769.3]
Inference: 29%|██▉ | 1721/5920 [24:54<1:28:26, 1.26s/it, tok/s=14750.4]
Inference: 29%|██▉ | 1722/5920 [24:55<1:23:16, 1.19s/it, tok/s=14750.4]
Inference: 29%|██▉ | 1722/5920 [24:55<1:23:16, 1.19s/it, tok/s=14744.7]
Inference: 29%|██▉ | 1723/5920 [24:57<1:32:43, 1.33s/it, tok/s=14744.7]
Inference: 29%|██▉ | 1723/5920 [24:57<1:32:43, 1.33s/it, tok/s=14733.4]
Inference: 29%|██▉ | 1724/5920 [24:57<1:10:02, 1.00s/it, tok/s=14733.4]
Inference: 29%|██▉ | 1724/5920 [24:57<1:10:02, 1.00s/it, tok/s=14743.2]
Inference: 29%|██▉ | 1725/5920 [24:58<1:10:36, 1.01s/it, tok/s=14743.2]
Inference: 29%|██▉ | 1725/5920 [24:58<1:10:36, 1.01s/it, tok/s=14738.8]
Inference: 29%|██▉ | 1726/5920 [24:59<57:05, 1.22it/s, tok/s=14738.8]
Inference: 29%|██▉ | 1726/5920 [24:59<57:05, 1.22it/s, tok/s=14740.2]
Inference: 29%|██▉ | 1727/5920 [24:59<57:04, 1.22it/s, tok/s=14746.3]
Inference: 29%|██▉ | 1728/5920 [25:01<1:00:05, 1.16it/s, tok/s=14746.3]
Inference: 29%|██▉ | 1728/5920 [25:01<1:00:05, 1.16it/s, tok/s=14734.9]
Inference: 29%|██▉ | 1729/5920 [25:01<47:27, 1.47it/s, tok/s=14734.9]
Inference: 29%|██▉ | 1729/5920 [25:01<47:27, 1.47it/s, tok/s=14737.1]
Inference: 29%|██▉ | 1730/5920 [25:02<1:07:57, 1.03it/s, tok/s=14737.1]
Inference: 29%|██▉ | 1730/5920 [25:02<1:07:57, 1.03it/s, tok/s=14724.7]
Inference: 29%|██▉ | 1731/5920 [25:03<54:54, 1.27it/s, tok/s=14724.7]
Inference: 29%|██▉ | 1731/5920 [25:03<54:54, 1.27it/s, tok/s=14727.2]
Inference: 29%|██▉ | 1732/5920 [25:05<1:21:22, 1.17s/it, tok/s=14727.2]
Inference: 29%|██▉ | 1732/5920 [25:05<1:21:22, 1.17s/it, tok/s=14715.4]
Inference: 29%|██▉ | 1733/5920 [25:06<1:18:13, 1.12s/it, tok/s=14715.4]
Inference: 29%|██▉ | 1733/5920 [25:06<1:18:13, 1.12s/it, tok/s=14708.4]
Inference: 29%|██▉ | 1734/5920 [25:06<58:08, 1.20it/s, tok/s=14708.4]
Inference: 29%|██▉ | 1734/5920 [25:06<58:08, 1.20it/s, tok/s=14710.4]
Inference: 29%|██▉ | 1735/5920 [25:06<44:44, 1.56it/s, tok/s=14710.4]
Inference: 29%|██▉ | 1735/5920 [25:06<44:44, 1.56it/s, tok/s=14711.8]
Inference: 29%|██▉ | 1736/5920 [25:08<1:11:27, 1.02s/it, tok/s=14711.8]
Inference: 29%|██▉ | 1736/5920 [25:08<1:11:27, 1.02s/it, tok/s=14695.8]
Inference: 29%|██▉ | 1737/5920 [25:08<53:08, 1.31it/s, tok/s=14695.8]
Inference: 29%|██▉ | 1737/5920 [25:08<53:08, 1.31it/s, tok/s=14700.8]
Inference: 29%|██▉ | 1738/5920 [25:09<59:58, 1.16it/s, tok/s=14700.8]
Inference: 29%|██▉ | 1738/5920 [25:09<59:58, 1.16it/s, tok/s=14697.2]
Inference: 29%|██▉ | 1739/5920 [25:10<50:21, 1.38it/s, tok/s=14697.2]
Inference: 29%|██▉ | 1739/5920 [25:10<50:21, 1.38it/s, tok/s=14698.5]
Inference: 29%|██▉ | 1740/5920 [25:10<41:39, 1.67it/s, tok/s=14698.5]
Inference: 29%|██▉ | 1740/5920 [25:10<41:39, 1.67it/s, tok/s=14699.3]
Inference: 29%|██▉ | 1741/5920 [25:12<59:32, 1.17it/s, tok/s=14699.3]
Inference: 29%|██▉ | 1741/5920 [25:12<59:32, 1.17it/s, tok/s=14699.6]
Inference: 29%|██▉ | 1742/5920 [25:12<43:55, 1.59it/s, tok/s=14699.6]
Inference: 29%|██▉ | 1742/5920 [25:12<43:55, 1.59it/s, tok/s=14704.9]
Inference: 29%|██▉ | 1743/5920 [25:13<1:01:05, 1.14it/s, tok/s=14704.9]
Inference: 29%|██▉ | 1743/5920 [25:13<1:01:05, 1.14it/s, tok/s=14699.2]
Inference: 29%|██▉ | 1744/5920 [25:13<45:55, 1.52it/s, tok/s=14699.2]
Inference: 29%|██▉ | 1744/5920 [25:13<45:55, 1.52it/s, tok/s=14714.3]
Inference: 29%|██▉ | 1745/5920 [25:13<34:39, 2.01it/s, tok/s=14714.3]
Inference: 29%|██▉ | 1745/5920 [25:13<34:39, 2.01it/s, tok/s=14729.7]
Inference: 29%|██▉ | 1746/5920 [25:13<34:39, 2.01it/s, tok/s=14738.3]
Inference: 30%|██▉ | 1747/5920 [25:14<30:08, 2.31it/s, tok/s=14738.3]
Inference: 30%|██▉ | 1747/5920 [25:14<30:08, 2.31it/s, tok/s=14735.3]
Inference: 30%|██▉ | 1748/5920 [25:15<42:07, 1.65it/s, tok/s=14735.3]
Inference: 30%|██▉ | 1748/5920 [25:15<42:07, 1.65it/s, tok/s=14730.0]
Inference: 30%|██▉ | 1749/5920 [25:17<56:56, 1.22it/s, tok/s=14730.0]
Inference: 30%|██▉ | 1749/5920 [25:17<56:56, 1.22it/s, tok/s=14732.3]
Inference: 30%|██▉ | 1750/5920 [25:17<56:55, 1.22it/s, tok/s=14735.7]
Inference: 30%|██▉ | 1751/5920 [25:17<42:40, 1.63it/s, tok/s=14735.7]
Inference: 30%|██▉ | 1751/5920 [25:17<42:40, 1.63it/s, tok/s=14732.5]
Inference: 30%|██▉ | 1752/5920 [25:17<35:04, 1.98it/s, tok/s=14732.5]
Inference: 30%|██▉ | 1752/5920 [25:17<35:04, 1.98it/s, tok/s=14744.8]
Inference: 30%|██▉ | 1753/5920 [25:17<35:04, 1.98it/s, tok/s=14750.1]
Inference: 30%|██▉ | 1754/5920 [25:19<40:49, 1.70it/s, tok/s=14750.1]
Inference: 30%|██▉ | 1754/5920 [25:19<40:49, 1.70it/s, tok/s=14742.4]
Inference: 30%|██▉ | 1755/5920 [25:19<37:22, 1.86it/s, tok/s=14742.4]
Inference: 30%|██▉ | 1755/5920 [25:19<37:22, 1.86it/s, tok/s=14742.2]
Inference: 30%|██▉ | 1756/5920 [25:20<36:09, 1.92it/s, tok/s=14742.2]
Inference: 30%|██▉ | 1756/5920 [25:20<36:09, 1.92it/s, tok/s=14748.8]
Inference: 30%|██▉ | 1757/5920 [25:20<36:08, 1.92it/s, tok/s=14754.0]
Inference: 30%|██▉ | 1758/5920 [25:23<1:13:55, 1.07s/it, tok/s=14754.0]
Inference: 30%|██▉ | 1758/5920 [25:23<1:13:55, 1.07s/it, tok/s=14721.8]
Inference: 30%|██▉ | 1759/5920 [25:24<1:13:14, 1.06s/it, tok/s=14721.8]
Inference: 30%|██▉ | 1759/5920 [25:24<1:13:14, 1.06s/it, tok/s=14715.2]
Inference: 30%|██▉ | 1760/5920 [25:25<58:41, 1.18it/s, tok/s=14715.2]
Inference: 30%|██▉ | 1760/5920 [25:25<58:41, 1.18it/s, tok/s=14716.6]
Inference: 30%|██▉ | 1761/5920 [25:25<58:40, 1.18it/s, tok/s=14719.5]
Inference: 30%|██▉ | 1762/5920 [25:26<48:24, 1.43it/s, tok/s=14719.5]
Inference: 30%|██▉ | 1762/5920 [25:26<48:24, 1.43it/s, tok/s=14714.5]
Inference: 30%|██▉ | 1763/5920 [25:26<48:53, 1.42it/s, tok/s=14714.5]
Inference: 30%|██▉ | 1763/5920 [25:26<48:53, 1.42it/s, tok/s=14714.6]
Inference: 30%|██▉ | 1764/5920 [25:27<52:20, 1.32it/s, tok/s=14714.6]
Inference: 30%|██▉ | 1764/5920 [25:27<52:20, 1.32it/s, tok/s=14711.6]
Inference: 30%|██▉ | 1765/5920 [25:27<42:00, 1.65it/s, tok/s=14711.6]
Inference: 30%|██▉ | 1765/5920 [25:27<42:00, 1.65it/s, tok/s=14716.3]
Inference: 30%|██▉ | 1766/5920 [25:28<50:31, 1.37it/s, tok/s=14716.3]
Inference: 30%|██▉ | 1766/5920 [25:28<50:31, 1.37it/s, tok/s=14710.0]
Inference: 30%|██▉ | 1767/5920 [25:29<41:42, 1.66it/s, tok/s=14710.0]
Inference: 30%|██▉ | 1767/5920 [25:29<41:42, 1.66it/s, tok/s=14723.8]
Inference: 30%|██▉ | 1768/5920 [25:29<43:44, 1.58it/s, tok/s=14723.8]
Inference: 30%|██▉ | 1768/5920 [25:29<43:44, 1.58it/s, tok/s=14723.3]
Inference: 30%|██▉ | 1769/5920 [25:30<34:40, 2.00it/s, tok/s=14723.3]
Inference: 30%|██▉ | 1769/5920 [25:30<34:40, 2.00it/s, tok/s=14727.3]
Inference: 30%|██▉ | 1770/5920 [25:30<30:51, 2.24it/s, tok/s=14727.3]
Inference: 30%|██▉ | 1770/5920 [25:30<30:51, 2.24it/s, tok/s=14741.8]
Inference: 30%|██▉ | 1771/5920 [25:30<24:03, 2.87it/s, tok/s=14741.8]
Inference: 30%|██▉ | 1771/5920 [25:30<24:03, 2.87it/s, tok/s=14746.7]
Inference: 30%|██▉ | 1772/5920 [25:31<33:01, 2.09it/s, tok/s=14746.7]
Inference: 30%|██▉ | 1772/5920 [25:31<33:01, 2.09it/s, tok/s=14750.3]
Inference: 30%|██▉ | 1773/5920 [25:31<36:10, 1.91it/s, tok/s=14750.3]
Inference: 30%|██▉ | 1773/5920 [25:31<36:10, 1.91it/s, tok/s=14749.3]
Inference: 30%|██▉ | 1774/5920 [25:32<36:10, 1.91it/s, tok/s=14753.9]
Inference: 30%|██▉ | 1775/5920 [25:32<31:48, 2.17it/s, tok/s=14753.9]
Inference: 30%|██▉ | 1775/5920 [25:32<31:48, 2.17it/s, tok/s=14750.8]
Inference: 30%|███ | 1776/5920 [25:33<33:35, 2.06it/s, tok/s=14750.8]
Inference: 30%|███ | 1776/5920 [25:33<33:35, 2.06it/s, tok/s=14748.8]
Inference: 30%|███ | 1777/5920 [25:33<34:41, 1.99it/s, tok/s=14748.8]
Inference: 30%|███ | 1777/5920 [25:33<34:41, 1.99it/s, tok/s=14757.8]
Inference: 30%|███ | 1778/5920 [25:34<38:52, 1.78it/s, tok/s=14757.8]
Inference: 30%|███ | 1778/5920 [25:34<38:52, 1.78it/s, tok/s=14757.4]
Inference: 30%|███ | 1779/5920 [25:34<30:10, 2.29it/s, tok/s=14757.4]
Inference: 30%|███ | 1779/5920 [25:34<30:10, 2.29it/s, tok/s=14766.5]
Inference: 30%|███ | 1780/5920 [25:34<30:10, 2.29it/s, tok/s=14784.5]
Inference: 30%|███ | 1781/5920 [25:35<25:02, 2.75it/s, tok/s=14784.5]
Inference: 30%|███ | 1781/5920 [25:35<25:02, 2.75it/s, tok/s=14787.3]
Inference: 30%|███ | 1782/5920 [25:35<22:41, 3.04it/s, tok/s=14787.3]
Inference: 30%|███ | 1782/5920 [25:35<22:41, 3.04it/s, tok/s=14791.1]
Inference: 30%|███ | 1783/5920 [25:36<28:18, 2.44it/s, tok/s=14791.1]
Inference: 30%|███ | 1783/5920 [25:36<28:18, 2.44it/s, tok/s=14789.0]
Inference: 30%|███ | 1784/5920 [25:36<26:41, 2.58it/s, tok/s=14789.0]
Inference: 30%|███ | 1784/5920 [25:36<26:41, 2.58it/s, tok/s=14793.9]
Inference: 30%|███ | 1785/5920 [25:37<32:57, 2.09it/s, tok/s=14793.9]
Inference: 30%|███ | 1785/5920 [25:37<32:57, 2.09it/s, tok/s=14795.3]
Inference: 30%|███ | 1786/5920 [25:37<33:18, 2.07it/s, tok/s=14795.3]
Inference: 30%|███ | 1786/5920 [25:37<33:18, 2.07it/s, tok/s=14800.3]
Inference: 30%|███ | 1787/5920 [25:37<33:17, 2.07it/s, tok/s=14803.0]
Inference: 30%|███ | 1788/5920 [25:38<32:59, 2.09it/s, tok/s=14803.0]
Inference: 30%|███ | 1788/5920 [25:38<32:59, 2.09it/s, tok/s=14799.6]
Inference: 30%|███ | 1789/5920 [25:40<50:43, 1.36it/s, tok/s=14799.6]
Inference: 30%|███ | 1789/5920 [25:40<50:43, 1.36it/s, tok/s=14791.0]
Inference: 30%|███ | 1790/5920 [25:41<1:06:41, 1.03it/s, tok/s=14791.0]
Inference: 30%|███ | 1790/5920 [25:41<1:06:41, 1.03it/s, tok/s=14792.5]
Inference: 30%|███ | 1791/5920 [25:43<1:17:11, 1.12s/it, tok/s=14792.5]
Inference: 30%|███ | 1791/5920 [25:43<1:17:11, 1.12s/it, tok/s=14783.7]
Inference: 30%|███ | 1792/5920 [25:44<1:23:17, 1.21s/it, tok/s=14783.7]
Inference: 30%|███ | 1792/5920 [25:44<1:23:17, 1.21s/it, tok/s=14776.5]
Inference: 30%|███ | 1793/5920 [25:46<1:32:06, 1.34s/it, tok/s=14776.5]
Inference: 30%|███ | 1793/5920 [25:46<1:32:06, 1.34s/it, tok/s=14777.8]
Inference: 30%|███ | 1794/5920 [25:47<1:32:36, 1.35s/it, tok/s=14777.8]
Inference: 30%|███ | 1794/5920 [25:47<1:32:36, 1.35s/it, tok/s=14771.6]
Inference: 30%|███ | 1795/5920 [25:49<1:30:03, 1.31s/it, tok/s=14771.6]
Inference: 30%|███ | 1795/5920 [25:49<1:30:03, 1.31s/it, tok/s=14777.7]
Inference: 30%|███ | 1796/5920 [25:49<1:11:40, 1.04s/it, tok/s=14777.7]
Inference: 30%|███ | 1796/5920 [25:49<1:11:40, 1.04s/it, tok/s=14781.2]
Inference: 30%|███ | 1797/5920 [25:49<53:27, 1.29it/s, tok/s=14781.2]
Inference: 30%|███ | 1797/5920 [25:49<53:27, 1.29it/s, tok/s=14796.3]
Inference: 30%|███ | 1798/5920 [25:51<1:12:01, 1.05s/it, tok/s=14796.3]
Inference: 30%|███ | 1798/5920 [25:51<1:12:01, 1.05s/it, tok/s=14787.8]
Inference: 30%|███ | 1799/5920 [25:51<53:33, 1.28it/s, tok/s=14787.8]
Inference: 30%|███ | 1799/5920 [25:51<53:33, 1.28it/s, tok/s=14796.2]
Inference: 30%|███ | 1800/5920 [25:52<50:14, 1.37it/s, tok/s=14796.2]
Inference: 30%|███ | 1800/5920 [25:52<50:14, 1.37it/s, tok/s=14797.6]
Inference: 30%|███ | 1801/5920 [25:52<50:59, 1.35it/s, tok/s=14797.6]
Inference: 30%|███ | 1801/5920 [25:52<50:59, 1.35it/s, tok/s=14793.6]
Inference: 30%|███ | 1802/5920 [25:52<50:58, 1.35it/s, tok/s=14796.5]
Inference: 30%|███ | 1803/5920 [25:53<31:32, 2.18it/s, tok/s=14796.5]
Inference: 30%|███ | 1803/5920 [25:53<31:32, 2.18it/s, tok/s=14809.7]
Inference: 30%|███ | 1804/5920 [25:53<34:07, 2.01it/s, tok/s=14809.7]
Inference: 30%|███ | 1804/5920 [25:53<34:07, 2.01it/s, tok/s=14807.4]
Inference: 30%|███ | 1805/5920 [25:56<1:09:44, 1.02s/it, tok/s=14807.4]
Inference: 30%|███ | 1805/5920 [25:56<1:09:44, 1.02s/it, tok/s=14798.3]
Inference: 31%|███ | 1806/5920 [25:56<1:09:43, 1.02s/it, tok/s=14812.5]
Inference: 31%|███ | 1807/5920 [25:56<44:00, 1.56it/s, tok/s=14812.5]
Inference: 31%|███ | 1807/5920 [25:56<44:00, 1.56it/s, tok/s=14817.2]
Inference: 31%|███ | 1808/5920 [25:56<36:57, 1.85it/s, tok/s=14817.2]
Inference: 31%|███ | 1808/5920 [25:56<36:57, 1.85it/s, tok/s=14821.1]
Inference: 31%|███ | 1809/5920 [25:57<38:10, 1.80it/s, tok/s=14821.1]
Inference: 31%|███ | 1809/5920 [25:57<38:10, 1.80it/s, tok/s=14822.0]
Inference: 31%|███ | 1810/5920 [25:57<38:21, 1.79it/s, tok/s=14822.0]
Inference: 31%|███ | 1810/5920 [25:57<38:21, 1.79it/s, tok/s=14823.6]
Inference: 31%|███ | 1811/5920 [25:58<34:09, 2.00it/s, tok/s=14823.6]
Inference: 31%|███ | 1811/5920 [25:58<34:09, 2.00it/s, tok/s=14829.1]
Inference: 31%|███ | 1812/5920 [26:00<1:17:32, 1.13s/it, tok/s=14829.1]
Inference: 31%|███ | 1812/5920 [26:00<1:17:32, 1.13s/it, tok/s=14808.7]
Inference: 31%|███ | 1813/5920 [26:04<2:14:00, 1.96s/it, tok/s=14808.7]
Inference: 31%|███ | 1813/5920 [26:04<2:14:00, 1.96s/it, tok/s=14782.9]
Inference: 31%|███ | 1814/5920 [26:05<1:44:17, 1.52s/it, tok/s=14782.9]
Inference: 31%|███ | 1814/5920 [26:05<1:44:17, 1.52s/it, tok/s=14782.5]
Inference: 31%|███ | 1815/5920 [26:06<1:27:39, 1.28s/it, tok/s=14782.5]
Inference: 31%|███ | 1815/5920 [26:06<1:27:39, 1.28s/it, tok/s=14790.2]
Inference: 31%|███ | 1816/5920 [26:07<1:25:03, 1.24s/it, tok/s=14790.2]
Inference: 31%|███ | 1816/5920 [26:07<1:25:03, 1.24s/it, tok/s=14784.7]
Inference: 31%|███ | 1817/5920 [26:08<1:27:52, 1.29s/it, tok/s=14784.7]
Inference: 31%|███ | 1817/5920 [26:08<1:27:52, 1.29s/it, tok/s=14777.1]
Inference: 31%|███ | 1818/5920 [26:12<2:20:58, 2.06s/it, tok/s=14777.1]
Inference: 31%|███ | 1818/5920 [26:12<2:20:58, 2.06s/it, tok/s=14746.5]
Inference: 31%|███ | 1819/5920 [26:13<1:58:56, 1.74s/it, tok/s=14746.5]
Inference: 31%|███ | 1819/5920 [26:13<1:58:56, 1.74s/it, tok/s=14743.4]
Inference: 31%|███ | 1820/5920 [26:15<2:12:32, 1.94s/it, tok/s=14743.4]
Inference: 31%|███ | 1820/5920 [26:15<2:12:32, 1.94s/it, tok/s=14738.0]
Inference: 31%|███ | 1821/5920 [26:16<1:54:21, 1.67s/it, tok/s=14738.0]
Inference: 31%|███ | 1821/5920 [26:16<1:54:21, 1.67s/it, tok/s=14732.4]
Inference: 31%|███ | 1822/5920 [26:17<1:23:58, 1.23s/it, tok/s=14732.4]
Inference: 31%|███ | 1822/5920 [26:17<1:23:58, 1.23s/it, tok/s=14740.2]
Inference: 31%|███ | 1823/5920 [26:17<1:02:13, 1.10it/s, tok/s=14740.2]
Inference: 31%|███ | 1823/5920 [26:17<1:02:13, 1.10it/s, tok/s=14745.7]
Inference: 31%|███ | 1824/5920 [26:17<49:57, 1.37it/s, tok/s=14745.7]
Inference: 31%|███ | 1824/5920 [26:17<49:57, 1.37it/s, tok/s=14749.1]
Inference: 31%|███ | 1825/5920 [26:22<2:16:07, 1.99s/it, tok/s=14749.1]
Inference: 31%|███ | 1825/5920 [26:22<2:16:07, 1.99s/it, tok/s=14709.1]
Inference: 31%|███ | 1826/5920 [26:24<2:07:21, 1.87s/it, tok/s=14709.1]
Inference: 31%|███ | 1826/5920 [26:24<2:07:21, 1.87s/it, tok/s=14707.6]
Inference: 31%|███ | 1827/5920 [26:24<1:40:06, 1.47s/it, tok/s=14707.6]
Inference: 31%|███ | 1827/5920 [26:24<1:40:06, 1.47s/it, tok/s=14717.5]
Inference: 31%|███ | 1828/5920 [26:25<1:20:52, 1.19s/it, tok/s=14717.5]
Inference: 31%|███ | 1828/5920 [26:25<1:20:52, 1.19s/it, tok/s=14723.2]
Inference: 31%|███ | 1829/5920 [26:30<2:41:19, 2.37s/it, tok/s=14723.2]
Inference: 31%|███ | 1829/5920 [26:30<2:41:19, 2.37s/it, tok/s=14683.8]
Inference: 31%|███ | 1830/5920 [26:33<2:48:50, 2.48s/it, tok/s=14683.8]
Inference: 31%|███ | 1830/5920 [26:33<2:48:50, 2.48s/it, tok/s=14672.2]
Inference: 31%|███ | 1831/5920 [26:37<3:26:54, 3.04s/it, tok/s=14672.2]
Inference: 31%|███ | 1831/5920 [26:37<3:26:54, 3.04s/it, tok/s=14644.9]
Inference: 31%|███ | 1832/5920 [26:39<3:03:26, 2.69s/it, tok/s=14644.9]
Inference: 31%|███ | 1832/5920 [26:39<3:03:26, 2.69s/it, tok/s=14634.2]
Inference: 31%|███ | 1833/5920 [26:39<3:03:23, 2.69s/it, tok/s=14646.2]
Inference: 31%|███ | 1834/5920 [26:39<1:41:09, 1.49s/it, tok/s=14646.2]
Inference: 31%|███ | 1834/5920 [26:39<1:41:09, 1.49s/it, tok/s=14657.1]
Inference: 31%|███ | 1835/5920 [26:40<1:34:18, 1.39s/it, tok/s=14657.1]
Inference: 31%|███ | 1835/5920 [26:40<1:34:18, 1.39s/it, tok/s=14664.9]
Inference: 31%|███ | 1836/5920 [26:42<1:45:43, 1.55s/it, tok/s=14664.9]
Inference: 31%|███ | 1836/5920 [26:42<1:45:43, 1.55s/it, tok/s=14654.1]
Inference: 31%|███ | 1837/5920 [26:46<2:29:07, 2.19s/it, tok/s=14654.1]
Inference: 31%|███ | 1837/5920 [26:46<2:29:07, 2.19s/it, tok/s=14636.7]
Inference: 31%|███ | 1838/5920 [26:46<1:56:59, 1.72s/it, tok/s=14636.7]
Inference: 31%|███ | 1838/5920 [26:46<1:56:59, 1.72s/it, tok/s=14648.1]
Inference: 31%|███ | 1839/5920 [26:50<2:34:29, 2.27s/it, tok/s=14648.1]
Inference: 31%|███ | 1839/5920 [26:50<2:34:29, 2.27s/it, tok/s=14622.6]
Inference: 31%|███ | 1840/5920 [26:51<1:59:11, 1.75s/it, tok/s=14622.6]
Inference: 31%|███ | 1840/5920 [26:51<1:59:11, 1.75s/it, tok/s=14636.9]
Inference: 31%|███ | 1841/5920 [26:53<2:16:23, 2.01s/it, tok/s=14636.9]
Inference: 31%|███ | 1841/5920 [26:53<2:16:23, 2.01s/it, tok/s=14629.3]
Inference: 31%|███ | 1842/5920 [26:54<1:44:20, 1.54s/it, tok/s=14629.3]
Inference: 31%|███ | 1842/5920 [26:54<1:44:20, 1.54s/it, tok/s=14635.1]
Inference: 31%|███ | 1843/5920 [26:55<1:38:01, 1.44s/it, tok/s=14635.1]
Inference: 31%|███ | 1843/5920 [26:55<1:38:01, 1.44s/it, tok/s=14630.8]
Inference: 31%|███ | 1844/5920 [26:55<1:12:43, 1.07s/it, tok/s=14630.8]
Inference: 31%|███ | 1844/5920 [26:55<1:12:43, 1.07s/it, tok/s=14636.0]
Inference: 31%|███ | 1845/5920 [26:59<2:12:40, 1.95s/it, tok/s=14636.0]
Inference: 31%|███ | 1845/5920 [26:59<2:12:40, 1.95s/it, tok/s=14609.3]
Inference: 31%|███ | 1846/5920 [27:02<2:26:05, 2.15s/it, tok/s=14609.3]
Inference: 31%|███ | 1846/5920 [27:02<2:26:05, 2.15s/it, tok/s=14594.7]
Inference: 31%|███ | 1847/5920 [27:02<1:58:16, 1.74s/it, tok/s=14594.7]
Inference: 31%|███ | 1847/5920 [27:02<1:58:16, 1.74s/it, tok/s=14594.9]
Inference: 31%|███ | 1848/5920 [27:06<2:37:14, 2.32s/it, tok/s=14594.9]
Inference: 31%|███ | 1848/5920 [27:06<2:37:14, 2.32s/it, tok/s=14579.2]
Inference: 31%|███ | 1849/5920 [27:10<3:05:00, 2.73s/it, tok/s=14579.2]
Inference: 31%|███ | 1849/5920 [27:10<3:05:00, 2.73s/it, tok/s=14557.9]
Inference: 31%|███▏ | 1850/5920 [27:10<2:14:44, 1.99s/it, tok/s=14557.9]
Inference: 31%|███▏ | 1850/5920 [27:10<2:14:44, 1.99s/it, tok/s=14570.2]
Inference: 31%|███▏ | 1851/5920 [27:14<2:54:51, 2.58s/it, tok/s=14570.2]
Inference: 31%|███▏ | 1851/5920 [27:14<2:54:51, 2.58s/it, tok/s=14545.5]
Inference: 31%|███▏ | 1852/5920 [27:14<2:11:56, 1.95s/it, tok/s=14545.5]
Inference: 31%|███▏ | 1852/5920 [27:14<2:11:56, 1.95s/it, tok/s=14553.4]
Inference: 31%|███▏ | 1853/5920 [27:15<1:48:36, 1.60s/it, tok/s=14553.4]
Inference: 31%|███▏ | 1853/5920 [27:15<1:48:36, 1.60s/it, tok/s=14557.3]
Inference: 31%|███▏ | 1854/5920 [27:18<2:14:45, 1.99s/it, tok/s=14557.3]
Inference: 31%|███▏ | 1854/5920 [27:18<2:14:45, 1.99s/it, tok/s=14547.5]
Inference: 31%|███▏ | 1855/5920 [27:22<2:56:50, 2.61s/it, tok/s=14547.5]
Inference: 31%|███▏ | 1855/5920 [27:22<2:56:50, 2.61s/it, tok/s=14524.2]
Inference: 31%|███▏ | 1856/5920 [27:23<2:11:35, 1.94s/it, tok/s=14524.2]
Inference: 31%|███▏ | 1856/5920 [27:23<2:11:35, 1.94s/it, tok/s=14529.4]
Inference: 31%|███▏ | 1857/5920 [27:23<1:40:05, 1.48s/it, tok/s=14529.4]
Inference: 31%|███▏ | 1857/5920 [27:23<1:40:05, 1.48s/it, tok/s=14533.4]
Inference: 31%|███▏ | 1858/5920 [27:23<1:40:03, 1.48s/it, tok/s=14543.1]
Inference: 31%|███▏ | 1859/5920 [27:24<1:11:29, 1.06s/it, tok/s=14543.1]
Inference: 31%|███▏ | 1859/5920 [27:24<1:11:29, 1.06s/it, tok/s=14550.2]
Inference: 31%|███▏ | 1860/5920 [27:26<1:18:11, 1.16s/it, tok/s=14550.2]
Inference: 31%|███▏ | 1860/5920 [27:26<1:18:11, 1.16s/it, tok/s=14555.0]
Inference: 31%|███▏ | 1861/5920 [27:26<1:18:10, 1.16s/it, tok/s=14562.1]
Inference: 31%|███▏ | 1862/5920 [27:27<1:00:27, 1.12it/s, tok/s=14562.1]
Inference: 31%|███▏ | 1862/5920 [27:27<1:00:27, 1.12it/s, tok/s=14562.2]
Inference: 31%|███▏ | 1863/5920 [27:27<57:08, 1.18it/s, tok/s=14562.2]
Inference: 31%|███▏ | 1863/5920 [27:27<57:08, 1.18it/s, tok/s=14562.9]
Inference: 31%|███▏ | 1864/5920 [27:28<56:17, 1.20it/s, tok/s=14562.9]
Inference: 31%|███▏ | 1864/5920 [27:28<56:17, 1.20it/s, tok/s=14563.3]
Inference: 32%|███▏ | 1865/5920 [27:29<56:42, 1.19it/s, tok/s=14563.3]
Inference: 32%|███▏ | 1865/5920 [27:29<56:42, 1.19it/s, tok/s=14564.7]
Inference: 32%|███▏ | 1866/5920 [27:37<3:03:18, 2.71s/it, tok/s=14564.7]
Inference: 32%|███▏ | 1866/5920 [27:37<3:03:18, 2.71s/it, tok/s=14503.6]
Inference: 32%|███▏ | 1867/5920 [27:37<2:15:19, 2.00s/it, tok/s=14503.6]
Inference: 32%|███▏ | 1867/5920 [27:37<2:15:19, 2.00s/it, tok/s=14513.0]
Inference: 32%|███▏ | 1868/5920 [27:41<2:56:34, 2.61s/it, tok/s=14513.0]
Inference: 32%|███▏ | 1868/5920 [27:41<2:56:34, 2.61s/it, tok/s=14492.0]
Inference: 32%|███▏ | 1869/5920 [27:41<2:11:55, 1.95s/it, tok/s=14492.0]
Inference: 32%|███▏ | 1869/5920 [27:41<2:11:55, 1.95s/it, tok/s=14505.4]
Inference: 32%|███▏ | 1870/5920 [27:43<2:04:42, 1.85s/it, tok/s=14505.4]
Inference: 32%|███▏ | 1870/5920 [27:43<2:04:42, 1.85s/it, tok/s=14498.8]
Inference: 32%|███▏ | 1871/5920 [27:47<2:38:30, 2.35s/it, tok/s=14498.8]
Inference: 32%|███▏ | 1871/5920 [27:47<2:38:30, 2.35s/it, tok/s=14474.9]
Inference: 32%|███▏ | 1872/5920 [27:47<2:04:30, 1.85s/it, tok/s=14474.9]
Inference: 32%|███▏ | 1872/5920 [27:47<2:04:30, 1.85s/it, tok/s=14480.0]
Inference: 32%|███▏ | 1873/5920 [27:51<2:46:24, 2.47s/it, tok/s=14480.0]
Inference: 32%|███▏ | 1873/5920 [27:51<2:46:24, 2.47s/it, tok/s=14453.4]
Inference: 32%|███▏ | 1874/5920 [27:54<2:55:15, 2.60s/it, tok/s=14453.4]
Inference: 32%|███▏ | 1874/5920 [27:54<2:55:15, 2.60s/it, tok/s=14435.9]
Inference: 32%|███▏ | 1875/5920 [27:58<3:20:25, 2.97s/it, tok/s=14435.9]
Inference: 32%|███▏ | 1875/5920 [27:58<3:20:25, 2.97s/it, tok/s=14410.0]
Inference: 32%|███▏ | 1876/5920 [27:59<2:51:41, 2.55s/it, tok/s=14410.0]
Inference: 32%|███▏ | 1876/5920 [27:59<2:51:41, 2.55s/it, tok/s=14413.0]
Inference: 32%|███▏ | 1877/5920 [28:01<2:35:01, 2.30s/it, tok/s=14413.0]
Inference: 32%|███▏ | 1877/5920 [28:01<2:35:01, 2.30s/it, tok/s=14414.5]
Inference: 32%|███▏ | 1878/5920 [28:02<2:00:01, 1.78s/it, tok/s=14414.5]
Inference: 32%|███▏ | 1878/5920 [28:02<2:00:01, 1.78s/it, tok/s=14426.7]
Inference: 32%|███▏ | 1879/5920 [28:02<1:29:37, 1.33s/it, tok/s=14426.7]
Inference: 32%|███▏ | 1879/5920 [28:02<1:29:37, 1.33s/it, tok/s=14432.2]
Inference: 32%|███▏ | 1880/5920 [28:02<1:05:24, 1.03it/s, tok/s=14432.2]
Inference: 32%|███▏ | 1880/5920 [28:02<1:05:24, 1.03it/s, tok/s=14438.2]
Inference: 32%|███▏ | 1881/5920 [28:07<2:28:34, 2.21s/it, tok/s=14438.2]
Inference: 32%|███▏ | 1881/5920 [28:07<2:28:34, 2.21s/it, tok/s=14409.5]
Inference: 32%|███▏ | 1882/5920 [28:11<2:59:50, 2.67s/it, tok/s=14409.5]
Inference: 32%|███▏ | 1882/5920 [28:11<2:59:50, 2.67s/it, tok/s=14394.3]
Inference: 32%|███▏ | 1883/5920 [28:12<2:33:19, 2.28s/it, tok/s=14394.3]
Inference: 32%|███▏ | 1883/5920 [28:12<2:33:19, 2.28s/it, tok/s=14391.1]
Inference: 32%|███▏ | 1884/5920 [28:15<2:45:29, 2.46s/it, tok/s=14391.1]
Inference: 32%|███▏ | 1884/5920 [28:15<2:45:29, 2.46s/it, tok/s=14377.7]
Inference: 32%|███▏ | 1885/5920 [28:16<2:16:56, 2.04s/it, tok/s=14377.7]
Inference: 32%|███▏ | 1885/5920 [28:16<2:16:56, 2.04s/it, tok/s=14376.5]
Inference: 32%|███▏ | 1886/5920 [28:21<3:04:51, 2.75s/it, tok/s=14376.5]
Inference: 32%|███▏ | 1886/5920 [28:21<3:04:51, 2.75s/it, tok/s=14347.8]
Inference: 32%|███▏ | 1887/5920 [28:22<2:35:35, 2.31s/it, tok/s=14347.8]
Inference: 32%|███▏ | 1887/5920 [28:22<2:35:35, 2.31s/it, tok/s=14348.1]
Inference: 32%|███▏ | 1888/5920 [28:24<2:28:07, 2.20s/it, tok/s=14348.1]
Inference: 32%|███▏ | 1888/5920 [28:24<2:28:07, 2.20s/it, tok/s=14339.1]
Inference: 32%|███▏ | 1889/5920 [28:25<2:13:06, 1.98s/it, tok/s=14339.1]
Inference: 32%|███▏ | 1889/5920 [28:25<2:13:06, 1.98s/it, tok/s=14341.8]
Inference: 32%|███▏ | 1890/5920 [28:26<1:37:23, 1.45s/it, tok/s=14341.8]
Inference: 32%|███▏ | 1890/5920 [28:26<1:37:23, 1.45s/it, tok/s=14350.0]
Inference: 32%|███▏ | 1891/5920 [28:28<1:59:28, 1.78s/it, tok/s=14350.0]
Inference: 32%|███▏ | 1891/5920 [28:28<1:59:28, 1.78s/it, tok/s=14346.2]
Inference: 32%|███▏ | 1892/5920 [28:31<2:12:34, 1.97s/it, tok/s=14346.2]
Inference: 32%|███▏ | 1892/5920 [28:31<2:12:34, 1.97s/it, tok/s=14335.4]
Inference: 32%|███▏ | 1893/5920 [28:32<2:08:23, 1.91s/it, tok/s=14335.4]
Inference: 32%|███▏ | 1893/5920 [28:32<2:08:23, 1.91s/it, tok/s=14328.3]
Inference: 32%|███▏ | 1894/5920 [28:34<1:58:20, 1.76s/it, tok/s=14328.3]
Inference: 32%|███▏ | 1894/5920 [28:34<1:58:20, 1.76s/it, tok/s=14330.5]
Inference: 32%|███▏ | 1895/5920 [28:35<1:47:16, 1.60s/it, tok/s=14330.5]
Inference: 32%|███▏ | 1895/5920 [28:35<1:47:16, 1.60s/it, tok/s=14330.5]
Inference: 32%|███▏ | 1896/5920 [28:37<1:58:30, 1.77s/it, tok/s=14330.5]
Inference: 32%|███▏ | 1896/5920 [28:37<1:58:30, 1.77s/it, tok/s=14321.3]
Inference: 32%|███▏ | 1897/5920 [28:38<1:36:54, 1.45s/it, tok/s=14321.3]
Inference: 32%|███▏ | 1897/5920 [28:38<1:36:54, 1.45s/it, tok/s=14330.9]
Inference: 32%|███▏ | 1898/5920 [28:39<1:33:00, 1.39s/it, tok/s=14330.9]
Inference: 32%|███▏ | 1898/5920 [28:39<1:33:00, 1.39s/it, tok/s=14327.1]
Inference: 32%|███▏ | 1899/5920 [28:42<1:57:14, 1.75s/it, tok/s=14327.1]
Inference: 32%|███▏ | 1899/5920 [28:42<1:57:14, 1.75s/it, tok/s=14323.1]
Inference: 32%|███▏ | 1900/5920 [28:44<2:05:23, 1.87s/it, tok/s=14323.1]
Inference: 32%|███▏ | 1900/5920 [28:44<2:05:23, 1.87s/it, tok/s=14322.6]
Inference: 32%|███▏ | 1901/5920 [28:47<2:28:15, 2.21s/it, tok/s=14322.6]
Inference: 32%|███▏ | 1901/5920 [28:47<2:28:15, 2.21s/it, tok/s=14315.1]
Inference: 32%|███▏ | 1902/5920 [28:48<2:16:04, 2.03s/it, tok/s=14315.1]
Inference: 32%|███▏ | 1902/5920 [28:48<2:16:04, 2.03s/it, tok/s=14318.8]
Inference: 32%|███▏ | 1903/5920 [28:48<2:16:02, 2.03s/it, tok/s=14332.5]
Inference: 32%|███▏ | 1904/5920 [28:50<1:33:16, 1.39s/it, tok/s=14332.5]
Inference: 32%|███▏ | 1904/5920 [28:50<1:33:16, 1.39s/it, tok/s=14337.7]
Inference: 32%|███▏ | 1905/5920 [28:51<1:30:07, 1.35s/it, tok/s=14337.7]
Inference: 32%|███▏ | 1905/5920 [28:51<1:30:07, 1.35s/it, tok/s=14337.8]
Inference: 32%|███▏ | 1906/5920 [28:52<1:17:36, 1.16s/it, tok/s=14337.8]
Inference: 32%|███▏ | 1906/5920 [28:52<1:17:36, 1.16s/it, tok/s=14349.5]
Inference: 32%|███▏ | 1907/5920 [28:52<59:27, 1.12it/s, tok/s=14349.5]
Inference: 32%|███▏ | 1907/5920 [28:52<59:27, 1.12it/s, tok/s=14359.4]
Inference: 32%|███▏ | 1908/5920 [28:53<59:43, 1.12it/s, tok/s=14359.4]
Inference: 32%|███▏ | 1908/5920 [28:53<59:43, 1.12it/s, tok/s=14359.4]
Inference: 32%|███▏ | 1909/5920 [28:53<59:43, 1.12it/s, tok/s=14376.6]
Inference: 32%|███▏ | 1910/5920 [28:53<34:45, 1.92it/s, tok/s=14376.6]
Inference: 32%|███▏ | 1910/5920 [28:53<34:45, 1.92it/s, tok/s=14393.3]
Inference: 32%|███▏ | 1911/5920 [28:54<44:24, 1.50it/s, tok/s=14393.3]
Inference: 32%|███▏ | 1911/5920 [28:54<44:24, 1.50it/s, tok/s=14389.8]
Inference: 32%|███▏ | 1912/5920 [28:54<44:24, 1.50it/s, tok/s=14396.2]
Inference: 32%|███▏ | 1913/5920 [28:54<44:23, 1.50it/s, tok/s=14404.0]
Inference: 32%|███▏ | 1914/5920 [28:54<44:22, 1.50it/s, tok/s=14412.5]
Inference: 32%|███▏ | 1915/5920 [28:54<21:18, 3.13it/s, tok/s=14412.5]
Inference: 32%|███▏ | 1915/5920 [28:54<21:18, 3.13it/s, tok/s=14420.0]
Inference: 32%|███▏ | 1916/5920 [28:56<36:18, 1.84it/s, tok/s=14420.0]
Inference: 32%|███▏ | 1916/5920 [28:56<36:18, 1.84it/s, tok/s=14421.5]
Inference: 32%|███▏ | 1917/5920 [28:57<40:08, 1.66it/s, tok/s=14421.5]
Inference: 32%|███▏ | 1917/5920 [28:57<40:08, 1.66it/s, tok/s=14431.2]
Inference: 32%|███▏ | 1918/5920 [28:59<1:03:18, 1.05it/s, tok/s=14431.2]
Inference: 32%|███▏ | 1918/5920 [28:59<1:03:18, 1.05it/s, tok/s=14415.2]
Inference: 32%|███▏ | 1919/5920 [29:03<1:51:43, 1.68s/it, tok/s=14415.2]
Inference: 32%|███▏ | 1919/5920 [29:03<1:51:43, 1.68s/it, tok/s=14399.5]
Inference: 32%|███▏ | 1920/5920 [29:06<2:20:10, 2.10s/it, tok/s=14399.5]
Inference: 32%|███▏ | 1920/5920 [29:06<2:20:10, 2.10s/it, tok/s=14389.0]
Inference: 32%|███▏ | 1921/5920 [29:07<2:00:07, 1.80s/it, tok/s=14389.0]
Inference: 32%|███▏ | 1921/5920 [29:07<2:00:07, 1.80s/it, tok/s=14396.1]
Inference: 32%|███▏ | 1922/5920 [29:08<1:38:20, 1.48s/it, tok/s=14396.1]
Inference: 32%|███▏ | 1922/5920 [29:08<1:38:20, 1.48s/it, tok/s=14408.3]
Inference: 32%|███▏ | 1923/5920 [29:08<1:20:41, 1.21s/it, tok/s=14408.3]
Inference: 32%|███▏ | 1923/5920 [29:08<1:20:41, 1.21s/it, tok/s=14418.3]
Inference: 32%|███▎ | 1924/5920 [29:09<1:09:40, 1.05s/it, tok/s=14418.3]
Inference: 32%|███▎ | 1924/5920 [29:09<1:09:40, 1.05s/it, tok/s=14415.6]
Inference: 33%|███▎ | 1925/5920 [29:10<1:11:42, 1.08s/it, tok/s=14415.6]
Inference: 33%|███▎ | 1925/5920 [29:10<1:11:42, 1.08s/it, tok/s=14414.9]
Inference: 33%|███▎ | 1926/5920 [29:11<1:10:48, 1.06s/it, tok/s=14414.9]
Inference: 33%|███▎ | 1926/5920 [29:11<1:10:48, 1.06s/it, tok/s=14414.6]
Inference: 33%|███▎ | 1927/5920 [29:13<1:29:37, 1.35s/it, tok/s=14414.6]
Inference: 33%|███▎ | 1927/5920 [29:13<1:29:37, 1.35s/it, tok/s=14400.5]
Inference: 33%|███▎ | 1928/5920 [29:14<1:18:50, 1.19s/it, tok/s=14400.5]
Inference: 33%|███▎ | 1928/5920 [29:14<1:18:50, 1.19s/it, tok/s=14411.1]
Inference: 33%|███▎ | 1929/5920 [29:15<1:21:11, 1.22s/it, tok/s=14411.1]
Inference: 33%|███▎ | 1929/5920 [29:15<1:21:11, 1.22s/it, tok/s=14402.9]
Inference: 33%|███▎ | 1930/5920 [29:17<1:37:17, 1.46s/it, tok/s=14402.9]
Inference: 33%|███▎ | 1930/5920 [29:17<1:37:17, 1.46s/it, tok/s=14403.2]
Inference: 33%|███▎ | 1931/5920 [29:18<1:22:04, 1.23s/it, tok/s=14403.2]
Inference: 33%|███▎ | 1931/5920 [29:18<1:22:04, 1.23s/it, tok/s=14406.9]
Inference: 33%|███▎ | 1932/5920 [29:19<1:16:39, 1.15s/it, tok/s=14406.9]
Inference: 33%|███▎ | 1932/5920 [29:19<1:16:39, 1.15s/it, tok/s=14406.5]
Inference: 33%|███▎ | 1933/5920 [29:20<1:18:40, 1.18s/it, tok/s=14406.5]
Inference: 33%|███▎ | 1933/5920 [29:20<1:18:40, 1.18s/it, tok/s=14411.2]
Inference: 33%|███▎ | 1934/5920 [29:24<2:02:25, 1.84s/it, tok/s=14411.2]
Inference: 33%|███▎ | 1934/5920 [29:24<2:02:25, 1.84s/it, tok/s=14392.8]
Inference: 33%|███▎ | 1935/5920 [29:26<2:14:26, 2.02s/it, tok/s=14392.8]
Inference: 33%|███▎ | 1935/5920 [29:26<2:14:26, 2.02s/it, tok/s=14390.0]
Inference: 33%|███▎ | 1936/5920 [29:26<1:38:00, 1.48s/it, tok/s=14390.0]
Inference: 33%|███▎ | 1936/5920 [29:26<1:38:00, 1.48s/it, tok/s=14405.5]
Inference: 33%|███▎ | 1937/5920 [29:27<1:33:47, 1.41s/it, tok/s=14405.5]
Inference: 33%|███▎ | 1937/5920 [29:27<1:33:47, 1.41s/it, tok/s=14397.4]
Inference: 33%|███▎ | 1938/5920 [29:31<2:11:41, 1.98s/it, tok/s=14397.4]
Inference: 33%|███▎ | 1938/5920 [29:31<2:11:41, 1.98s/it, tok/s=14372.9]
Inference: 33%|███▎ | 1939/5920 [29:33<2:17:56, 2.08s/it, tok/s=14372.9]
Inference: 33%|███▎ | 1939/5920 [29:33<2:17:56, 2.08s/it, tok/s=14356.1]
Inference: 33%|███▎ | 1940/5920 [29:34<2:00:28, 1.82s/it, tok/s=14356.1]
Inference: 33%|███▎ | 1940/5920 [29:34<2:00:28, 1.82s/it, tok/s=14356.0]
Inference: 33%|███▎ | 1941/5920 [29:34<2:00:26, 1.82s/it, tok/s=14366.1]
Inference: 33%|███▎ | 1942/5920 [29:36<1:33:36, 1.41s/it, tok/s=14366.1]
Inference: 33%|███▎ | 1942/5920 [29:36<1:33:36, 1.41s/it, tok/s=14353.8]
Inference: 33%|███▎ | 1943/5920 [29:37<1:20:24, 1.21s/it, tok/s=14353.8]
Inference: 33%|███▎ | 1943/5920 [29:37<1:20:24, 1.21s/it, tok/s=14351.2]
Inference: 33%|███▎ | 1944/5920 [29:38<1:17:19, 1.17s/it, tok/s=14351.2]
Inference: 33%|███▎ | 1944/5920 [29:38<1:17:19, 1.17s/it, tok/s=14355.7]
Inference: 33%|███▎ | 1945/5920 [29:38<1:01:58, 1.07it/s, tok/s=14355.7]
Inference: 33%|███▎ | 1945/5920 [29:38<1:01:58, 1.07it/s, tok/s=14362.1]
Inference: 33%|███▎ | 1946/5920 [29:39<1:07:32, 1.02s/it, tok/s=14362.1]
Inference: 33%|███▎ | 1946/5920 [29:39<1:07:32, 1.02s/it, tok/s=14359.5]
Inference: 33%|███▎ | 1947/5920 [29:41<1:12:27, 1.09s/it, tok/s=14359.5]
Inference: 33%|███▎ | 1947/5920 [29:41<1:12:27, 1.09s/it, tok/s=14362.8]
Inference: 33%|███▎ | 1948/5920 [29:41<1:02:49, 1.05it/s, tok/s=14362.8]
Inference: 33%|███▎ | 1948/5920 [29:41<1:02:49, 1.05it/s, tok/s=14360.3]
Inference: 33%|███▎ | 1949/5920 [29:43<1:18:19, 1.18s/it, tok/s=14360.3]
Inference: 33%|███▎ | 1949/5920 [29:43<1:18:19, 1.18s/it, tok/s=14363.1]
Inference: 33%|███▎ | 1950/5920 [29:43<58:29, 1.13it/s, tok/s=14363.1]
Inference: 33%|███▎ | 1950/5920 [29:43<58:29, 1.13it/s, tok/s=14364.0]
Inference: 33%|███▎ | 1951/5920 [29:44<1:04:36, 1.02it/s, tok/s=14364.0]
Inference: 33%|███▎ | 1951/5920 [29:44<1:04:36, 1.02it/s, tok/s=14361.4]
Inference: 33%|███▎ | 1952/5920 [29:46<1:09:35, 1.05s/it, tok/s=14361.4]
Inference: 33%|███▎ | 1952/5920 [29:46<1:09:35, 1.05s/it, tok/s=14353.7]
Inference: 33%|███▎ | 1953/5920 [29:47<1:12:09, 1.09s/it, tok/s=14353.7]
Inference: 33%|███▎ | 1953/5920 [29:47<1:12:09, 1.09s/it, tok/s=14346.7]
Inference: 33%|███▎ | 1954/5920 [29:47<1:12:08, 1.09s/it, tok/s=14363.4]
Inference: 33%|███▎ | 1955/5920 [29:48<49:50, 1.33it/s, tok/s=14363.4]
Inference: 33%|███▎ | 1955/5920 [29:48<49:50, 1.33it/s, tok/s=14374.6]
Inference: 33%|███▎ | 1956/5920 [29:48<46:25, 1.42it/s, tok/s=14374.6]
Inference: 33%|███▎ | 1956/5920 [29:48<46:25, 1.42it/s, tok/s=14387.1]
Inference: 33%|███▎ | 1957/5920 [29:49<47:51, 1.38it/s, tok/s=14387.1]
Inference: 33%|███▎ | 1957/5920 [29:49<47:51, 1.38it/s, tok/s=14388.6]
Inference: 33%|███▎ | 1958/5920 [29:49<39:42, 1.66it/s, tok/s=14388.6]
Inference: 33%|███▎ | 1958/5920 [29:49<39:42, 1.66it/s, tok/s=14388.9]
Inference: 33%|███▎ | 1959/5920 [29:50<37:03, 1.78it/s, tok/s=14388.9]
Inference: 33%|███▎ | 1959/5920 [29:50<37:03, 1.78it/s, tok/s=14399.5]
Inference: 33%|███▎ | 1960/5920 [29:53<1:39:17, 1.50s/it, tok/s=14399.5]
Inference: 33%|███▎ | 1960/5920 [29:53<1:39:17, 1.50s/it, tok/s=14380.6]
Inference: 33%|███▎ | 1961/5920 [29:54<1:13:55, 1.12s/it, tok/s=14380.6]
Inference: 33%|███▎ | 1961/5920 [29:54<1:13:55, 1.12s/it, tok/s=14387.5]
Inference: 33%|███▎ | 1962/5920 [29:54<1:07:25, 1.02s/it, tok/s=14387.5]
Inference: 33%|███▎ | 1962/5920 [29:54<1:07:25, 1.02s/it, tok/s=14398.0]
Inference: 33%|███▎ | 1963/5920 [29:57<1:31:27, 1.39s/it, tok/s=14398.0]
Inference: 33%|███▎ | 1963/5920 [29:57<1:31:27, 1.39s/it, tok/s=14382.3]
Inference: 33%|███▎ | 1964/5920 [29:57<1:17:13, 1.17s/it, tok/s=14382.3]
Inference: 33%|███▎ | 1964/5920 [29:57<1:17:13, 1.17s/it, tok/s=14393.8]
Inference: 33%|███▎ | 1965/5920 [29:57<1:17:11, 1.17s/it, tok/s=14395.8]
Inference: 33%|███▎ | 1966/5920 [29:58<57:30, 1.15it/s, tok/s=14395.8]
Inference: 33%|███▎ | 1966/5920 [29:58<57:30, 1.15it/s, tok/s=14403.7]
Inference: 33%|███▎ | 1967/5920 [29:59<51:59, 1.27it/s, tok/s=14403.7]
Inference: 33%|███▎ | 1967/5920 [29:59<51:59, 1.27it/s, tok/s=14401.5]
Inference: 33%|███▎ | 1968/5920 [30:00<1:04:55, 1.01it/s, tok/s=14401.5]
Inference: 33%|███▎ | 1968/5920 [30:00<1:04:55, 1.01it/s, tok/s=14399.7]
Inference: 33%|███▎ | 1969/5920 [30:01<50:48, 1.30it/s, tok/s=14399.7]
Inference: 33%|███▎ | 1969/5920 [30:01<50:48, 1.30it/s, tok/s=14414.9]
Inference: 33%|███▎ | 1970/5920 [30:01<50:21, 1.31it/s, tok/s=14414.9]
Inference: 33%|███▎ | 1970/5920 [30:01<50:21, 1.31it/s, tok/s=14416.8]
Inference: 33%|███▎ | 1971/5920 [30:02<48:01, 1.37it/s, tok/s=14416.8]
Inference: 33%|███▎ | 1971/5920 [30:02<48:01, 1.37it/s, tok/s=14427.6]
Inference: 33%|███▎ | 1972/5920 [30:02<40:52, 1.61it/s, tok/s=14427.6]
Inference: 33%|███▎ | 1972/5920 [30:02<40:52, 1.61it/s, tok/s=14434.3]
Inference: 33%|███▎ | 1973/5920 [30:04<52:32, 1.25it/s, tok/s=14434.3]
Inference: 33%|███▎ | 1973/5920 [30:04<52:32, 1.25it/s, tok/s=14441.2]
Inference: 33%|███▎ | 1974/5920 [30:05<1:03:48, 1.03it/s, tok/s=14441.2]
Inference: 33%|███▎ | 1974/5920 [30:05<1:03:48, 1.03it/s, tok/s=14433.3]
Inference: 33%|███▎ | 1975/5920 [30:05<53:57, 1.22it/s, tok/s=14433.3]
Inference: 33%|███▎ | 1975/5920 [30:05<53:57, 1.22it/s, tok/s=14432.6]
Inference: 33%|███▎ | 1976/5920 [30:07<1:11:47, 1.09s/it, tok/s=14432.6]
Inference: 33%|███▎ | 1976/5920 [30:07<1:11:47, 1.09s/it, tok/s=14434.0]
Inference: 33%|███▎ | 1977/5920 [30:09<1:33:33, 1.42s/it, tok/s=14434.0]
Inference: 33%|███▎ | 1977/5920 [30:09<1:33:33, 1.42s/it, tok/s=14433.1]
Inference: 33%|███▎ | 1978/5920 [30:12<1:54:31, 1.74s/it, tok/s=14433.1]
Inference: 33%|███▎ | 1978/5920 [30:12<1:54:31, 1.74s/it, tok/s=14425.1]
Inference: 33%|███▎ | 1979/5920 [30:12<1:54:29, 1.74s/it, tok/s=14435.5]
Inference: 33%|███▎ | 1980/5920 [30:13<1:25:21, 1.30s/it, tok/s=14435.5]
Inference: 33%|███▎ | 1980/5920 [30:13<1:25:21, 1.30s/it, tok/s=14440.3]
Inference: 33%|███▎ | 1981/5920 [30:17<1:56:13, 1.77s/it, tok/s=14440.3]
Inference: 33%|███▎ | 1981/5920 [30:17<1:56:13, 1.77s/it, tok/s=14417.3]
Inference: 33%|███▎ | 1982/5920 [30:17<1:29:06, 1.36s/it, tok/s=14417.3]
Inference: 33%|███▎ | 1982/5920 [30:17<1:29:06, 1.36s/it, tok/s=14428.7]
Inference: 33%|███▎ | 1983/5920 [30:19<1:36:11, 1.47s/it, tok/s=14428.7]
Inference: 33%|███▎ | 1983/5920 [30:19<1:36:11, 1.47s/it, tok/s=14417.4]
Inference: 34%|███▎ | 1984/5920 [30:20<1:36:18, 1.47s/it, tok/s=14417.4]
Inference: 34%|███▎ | 1984/5920 [30:20<1:36:18, 1.47s/it, tok/s=14422.3]
Inference: 34%|███▎ | 1985/5920 [30:21<1:18:00, 1.19s/it, tok/s=14422.3]
Inference: 34%|███▎ | 1985/5920 [30:21<1:18:00, 1.19s/it, tok/s=14434.9]
Inference: 34%|███▎ | 1986/5920 [30:21<1:04:18, 1.02it/s, tok/s=14434.9]
Inference: 34%|███▎ | 1986/5920 [30:21<1:04:18, 1.02it/s, tok/s=14447.7]
Inference: 34%|███▎ | 1987/5920 [30:22<1:04:45, 1.01it/s, tok/s=14447.7]
Inference: 34%|███▎ | 1987/5920 [30:22<1:04:45, 1.01it/s, tok/s=14442.3]
Inference: 34%|███▎ | 1988/5920 [30:24<1:14:16, 1.13s/it, tok/s=14442.3]
Inference: 34%|███▎ | 1988/5920 [30:24<1:14:16, 1.13s/it, tok/s=14433.0]
Inference: 34%|███▎ | 1989/5920 [30:25<1:22:23, 1.26s/it, tok/s=14433.0]
Inference: 34%|███▎ | 1989/5920 [30:25<1:22:23, 1.26s/it, tok/s=14437.3]
Inference: 34%|███▎ | 1990/5920 [30:27<1:29:04, 1.36s/it, tok/s=14437.3]
Inference: 34%|███▎ | 1990/5920 [30:27<1:29:04, 1.36s/it, tok/s=14436.9]
Inference: 34%|███▎ | 1991/5920 [30:27<1:29:02, 1.36s/it, tok/s=14439.9]
Inference: 34%|███▎ | 1992/5920 [30:28<1:01:35, 1.06it/s, tok/s=14439.9]
Inference: 34%|███▎ | 1992/5920 [30:28<1:01:35, 1.06it/s, tok/s=14439.8]
Inference: 34%|███▎ | 1993/5920 [30:28<1:01:34, 1.06it/s, tok/s=14454.0]
Inference: 34%|███▎ | 1994/5920 [30:28<44:03, 1.49it/s, tok/s=14454.0]
Inference: 34%|███▎ | 1994/5920 [30:28<44:03, 1.49it/s, tok/s=14452.4]
Inference: 34%|███▎ | 1995/5920 [30:32<1:24:32, 1.29s/it, tok/s=14452.4]
Inference: 34%|███▎ | 1995/5920 [30:32<1:24:32, 1.29s/it, tok/s=14441.4]
Inference: 34%|███▎ | 1996/5920 [30:34<1:40:20, 1.53s/it, tok/s=14441.4]
Inference: 34%|███▎ | 1996/5920 [30:34<1:40:20, 1.53s/it, tok/s=14425.8]
Inference: 34%|███▎ | 1997/5920 [30:34<1:18:13, 1.20s/it, tok/s=14425.8]
Inference: 34%|███▎ | 1997/5920 [30:34<1:18:13, 1.20s/it, tok/s=14440.6]
Inference: 34%|███▍ | 1998/5920 [30:35<1:15:27, 1.15s/it, tok/s=14440.6]
Inference: 34%|███▍ | 1998/5920 [30:35<1:15:27, 1.15s/it, tok/s=14440.8]
Inference: 34%|███▍ | 1999/5920 [30:36<1:10:40, 1.08s/it, tok/s=14440.8]
Inference: 34%|███▍ | 1999/5920 [30:36<1:10:40, 1.08s/it, tok/s=14435.8]
Inference: 34%|███▍ | 2000/5920 [30:37<1:05:11, 1.00it/s, tok/s=14435.8]
Inference: 34%|███▍ | 2000/5920 [30:37<1:05:11, 1.00it/s, tok/s=14431.8]
Inference: 34%|███▍ | 2001/5920 [30:38<1:08:23, 1.05s/it, tok/s=14431.8]
Inference: 34%|███▍ | 2001/5920 [30:38<1:08:23, 1.05s/it, tok/s=14424.9]
Inference: 34%|███▍ | 2002/5920 [30:38<1:08:22, 1.05s/it, tok/s=14435.9]
Inference: 34%|███▍ | 2003/5920 [30:38<40:37, 1.61it/s, tok/s=14435.9]
Inference: 34%|███▍ | 2003/5920 [30:38<40:37, 1.61it/s, tok/s=14437.0]
Inference: 34%|███▍ | 2004/5920 [30:40<56:03, 1.16it/s, tok/s=14437.0]
Inference: 34%|███▍ | 2004/5920 [30:40<56:03, 1.16it/s, tok/s=14440.6]
Inference: 34%|███▍ | 2005/5920 [30:45<2:05:27, 1.92s/it, tok/s=14440.6]
Inference: 34%|███▍ | 2005/5920 [30:45<2:05:27, 1.92s/it, tok/s=14409.9]
Inference: 34%|███▍ | 2006/5920 [30:46<1:48:12, 1.66s/it, tok/s=14409.9]
Inference: 34%|███▍ | 2006/5920 [30:46<1:48:12, 1.66s/it, tok/s=14410.9]
Inference: 34%|███▍ | 2007/5920 [30:51<2:52:28, 2.64s/it, tok/s=14410.9]
Inference: 34%|███▍ | 2007/5920 [30:51<2:52:28, 2.64s/it, tok/s=14372.8]
Inference: 34%|███▍ | 2008/5920 [30:51<2:52:25, 2.64s/it, tok/s=14373.8]
Inference: 34%|███▍ | 2009/5920 [30:56<2:44:18, 2.52s/it, tok/s=14373.8]
Inference: 34%|███▍ | 2009/5920 [30:56<2:44:18, 2.52s/it, tok/s=14346.5]
Inference: 34%|███▍ | 2010/5920 [30:56<2:10:02, 2.00s/it, tok/s=14346.5]
Inference: 34%|███▍ | 2010/5920 [30:56<2:10:02, 2.00s/it, tok/s=14360.2]
Inference: 34%|███▍ | 2011/5920 [30:59<2:24:18, 2.22s/it, tok/s=14360.2]
Inference: 34%|███▍ | 2011/5920 [30:59<2:24:18, 2.22s/it, tok/s=14354.4]
Inference: 34%|███▍ | 2012/5920 [30:59<1:55:02, 1.77s/it, tok/s=14354.4]
Inference: 34%|███▍ | 2012/5920 [30:59<1:55:02, 1.77s/it, tok/s=14358.0]
Inference: 34%|███▍ | 2013/5920 [31:00<1:30:23, 1.39s/it, tok/s=14358.0]
Inference: 34%|███▍ | 2013/5920 [31:00<1:30:23, 1.39s/it, tok/s=14357.0]
Inference: 34%|███▍ | 2014/5920 [31:00<1:12:37, 1.12s/it, tok/s=14357.0]
Inference: 34%|███▍ | 2014/5920 [31:00<1:12:37, 1.12s/it, tok/s=14362.2]
Inference: 34%|███▍ | 2015/5920 [31:02<1:29:30, 1.38s/it, tok/s=14362.2]
Inference: 34%|███▍ | 2015/5920 [31:02<1:29:30, 1.38s/it, tok/s=14349.4]
Inference: 34%|███▍ | 2016/5920 [31:04<1:38:29, 1.51s/it, tok/s=14349.4]
Inference: 34%|███▍ | 2016/5920 [31:04<1:38:29, 1.51s/it, tok/s=14351.4]
Inference: 34%|███▍ | 2017/5920 [31:05<1:20:33, 1.24s/it, tok/s=14351.4]
Inference: 34%|███▍ | 2017/5920 [31:05<1:20:33, 1.24s/it, tok/s=14349.3]
Inference: 34%|███▍ | 2018/5920 [31:05<1:07:53, 1.04s/it, tok/s=14349.3]
Inference: 34%|███▍ | 2018/5920 [31:05<1:07:53, 1.04s/it, tok/s=14347.8]
Inference: 34%|███▍ | 2019/5920 [31:06<1:00:02, 1.08it/s, tok/s=14347.8]
Inference: 34%|███▍ | 2019/5920 [31:06<1:00:02, 1.08it/s, tok/s=14351.2]
Inference: 34%|███▍ | 2020/5920 [31:06<1:00:01, 1.08it/s, tok/s=14359.0]
Inference: 34%|███▍ | 2021/5920 [31:07<45:30, 1.43it/s, tok/s=14359.0]
Inference: 34%|███▍ | 2021/5920 [31:07<45:30, 1.43it/s, tok/s=14355.2]
Inference: 34%|███▍ | 2022/5920 [31:08<50:31, 1.29it/s, tok/s=14355.2]
Inference: 34%|███▍ | 2022/5920 [31:08<50:31, 1.29it/s, tok/s=14363.2]
Inference: 34%|███▍ | 2023/5920 [31:09<52:30, 1.24it/s, tok/s=14363.2]
Inference: 34%|███▍ | 2023/5920 [31:09<52:30, 1.24it/s, tok/s=14372.4]
Inference: 34%|███▍ | 2024/5920 [31:10<54:47, 1.19it/s, tok/s=14372.4]
Inference: 34%|███▍ | 2024/5920 [31:10<54:47, 1.19it/s, tok/s=14379.0]
Inference: 34%|███▍ | 2025/5920 [31:11<1:05:00, 1.00s/it, tok/s=14379.0]
Inference: 34%|███▍ | 2025/5920 [31:11<1:05:00, 1.00s/it, tok/s=14381.3]
Inference: 34%|███▍ | 2026/5920 [31:11<49:39, 1.31it/s, tok/s=14381.3]
Inference: 34%|███▍ | 2026/5920 [31:11<49:39, 1.31it/s, tok/s=14388.7]
Inference: 34%|███▍ | 2027/5920 [31:12<43:12, 1.50it/s, tok/s=14388.7]
Inference: 34%|███▍ | 2027/5920 [31:12<43:12, 1.50it/s, tok/s=14388.3]
Inference: 34%|███▍ | 2028/5920 [31:16<1:54:49, 1.77s/it, tok/s=14388.3]
Inference: 34%|███▍ | 2028/5920 [31:16<1:54:49, 1.77s/it, tok/s=14363.7]
Inference: 34%|███▍ | 2029/5920 [31:20<2:30:24, 2.32s/it, tok/s=14363.7]
Inference: 34%|███▍ | 2029/5920 [31:20<2:30:24, 2.32s/it, tok/s=14345.2]
Inference: 34%|███▍ | 2030/5920 [31:20<2:30:21, 2.32s/it, tok/s=14355.3]
Inference: 34%|███▍ | 2031/5920 [31:21<1:43:43, 1.60s/it, tok/s=14355.3]
Inference: 34%|███▍ | 2031/5920 [31:21<1:43:43, 1.60s/it, tok/s=14356.3]
Inference: 34%|███▍ | 2032/5920 [31:23<1:44:42, 1.62s/it, tok/s=14356.3]
Inference: 34%|███▍ | 2032/5920 [31:23<1:44:42, 1.62s/it, tok/s=14346.4]
Inference: 34%|███▍ | 2033/5920 [31:26<2:15:16, 2.09s/it, tok/s=14346.4]
Inference: 34%|███▍ | 2033/5920 [31:26<2:15:16, 2.09s/it, tok/s=14330.0]
Inference: 34%|███▍ | 2034/5920 [31:27<2:00:00, 1.85s/it, tok/s=14330.0]
Inference: 34%|███▍ | 2034/5920 [31:27<2:00:00, 1.85s/it, tok/s=14336.7]
Inference: 34%|███▍ | 2035/5920 [31:30<2:16:51, 2.11s/it, tok/s=14336.7]
Inference: 34%|███▍ | 2035/5920 [31:30<2:16:51, 2.11s/it, tok/s=14323.7]
Inference: 34%|███▍ | 2036/5920 [31:33<2:20:31, 2.17s/it, tok/s=14323.7]
Inference: 34%|███▍ | 2036/5920 [31:33<2:20:31, 2.17s/it, tok/s=14321.7]
Inference: 34%|███▍ | 2037/5920 [31:33<1:44:00, 1.61s/it, tok/s=14321.7]
Inference: 34%|███▍ | 2037/5920 [31:33<1:44:00, 1.61s/it, tok/s=14334.8]
Inference: 34%|███▍ | 2038/5920 [31:34<1:34:53, 1.47s/it, tok/s=14334.8]
Inference: 34%|███▍ | 2038/5920 [31:34<1:34:53, 1.47s/it, tok/s=14329.2]
Inference: 34%|███▍ | 2039/5920 [31:34<1:11:52, 1.11s/it, tok/s=14329.2]
Inference: 34%|███▍ | 2039/5920 [31:34<1:11:52, 1.11s/it, tok/s=14329.8]
Inference: 34%|███▍ | 2040/5920 [31:35<1:07:29, 1.04s/it, tok/s=14329.8]
Inference: 34%|███▍ | 2040/5920 [31:35<1:07:29, 1.04s/it, tok/s=14325.9]
Inference: 34%|███▍ | 2041/5920 [31:35<1:07:28, 1.04s/it, tok/s=14334.8]
Inference: 34%|███▍ | 2042/5920 [31:37<1:02:48, 1.03it/s, tok/s=14334.8]
Inference: 34%|███▍ | 2042/5920 [31:37<1:02:48, 1.03it/s, tok/s=14330.0]
Inference: 35%|███▍ | 2043/5920 [31:38<1:01:53, 1.04it/s, tok/s=14330.0]
Inference: 35%|███▍ | 2043/5920 [31:38<1:01:53, 1.04it/s, tok/s=14331.8]
Inference: 35%|███▍ | 2044/5920 [31:40<1:27:25, 1.35s/it, tok/s=14331.8]
Inference: 35%|███▍ | 2044/5920 [31:40<1:27:25, 1.35s/it, tok/s=14316.4]
Inference: 35%|███▍ | 2045/5920 [31:41<1:12:37, 1.12s/it, tok/s=14316.4]
Inference: 35%|███▍ | 2045/5920 [31:41<1:12:37, 1.12s/it, tok/s=14328.5]
Inference: 35%|███▍ | 2046/5920 [31:41<1:00:52, 1.06it/s, tok/s=14328.5]
Inference: 35%|███▍ | 2046/5920 [31:41<1:00:52, 1.06it/s, tok/s=14327.7]
Inference: 35%|███▍ | 2047/5920 [31:44<1:28:01, 1.36s/it, tok/s=14327.7]
Inference: 35%|███▍ | 2047/5920 [31:44<1:28:01, 1.36s/it, tok/s=14317.7]
Inference: 35%|███▍ | 2048/5920 [31:44<1:08:01, 1.05s/it, tok/s=14317.7]
Inference: 35%|███▍ | 2048/5920 [31:44<1:08:01, 1.05s/it, tok/s=14317.8]
Inference: 35%|███▍ | 2049/5920 [31:48<1:56:44, 1.81s/it, tok/s=14317.8]
Inference: 35%|███▍ | 2049/5920 [31:48<1:56:44, 1.81s/it, tok/s=14293.0]
Inference: 35%|███▍ | 2050/5920 [31:49<1:51:12, 1.72s/it, tok/s=14293.0]
Inference: 35%|███▍ | 2050/5920 [31:49<1:51:12, 1.72s/it, tok/s=14291.8]
Inference: 35%|███▍ | 2051/5920 [31:51<1:54:10, 1.77s/it, tok/s=14291.8]
Inference: 35%|███▍ | 2051/5920 [31:51<1:54:10, 1.77s/it, tok/s=14293.6]
Inference: 35%|███▍ | 2052/5920 [31:51<1:25:42, 1.33s/it, tok/s=14293.6]
Inference: 35%|███▍ | 2052/5920 [31:51<1:25:42, 1.33s/it, tok/s=14294.1]
Inference: 35%|███▍ | 2053/5920 [31:58<3:10:34, 2.96s/it, tok/s=14294.1]
Inference: 35%|███▍ | 2053/5920 [31:58<3:10:34, 2.96s/it, tok/s=14256.3]
Inference: 35%|███▍ | 2054/5920 [31:59<2:35:33, 2.41s/it, tok/s=14256.3]
Inference: 35%|███▍ | 2054/5920 [31:59<2:35:33, 2.41s/it, tok/s=14263.6]
Inference: 35%|███▍ | 2055/5920 [31:59<2:35:31, 2.41s/it, tok/s=14267.3]
Inference: 35%|███▍ | 2056/5920 [32:00<1:36:25, 1.50s/it, tok/s=14267.3]
Inference: 35%|███▍ | 2056/5920 [32:00<1:36:25, 1.50s/it, tok/s=14276.7]
Inference: 35%|███▍ | 2057/5920 [32:00<1:20:14, 1.25s/it, tok/s=14276.7]
Inference: 35%|███▍ | 2057/5920 [32:00<1:20:14, 1.25s/it, tok/s=14284.6]
Inference: 35%|███▍ | 2058/5920 [32:02<1:25:31, 1.33s/it, tok/s=14284.6]
Inference: 35%|███▍ | 2058/5920 [32:02<1:25:31, 1.33s/it, tok/s=14287.5]
Inference: 35%|███▍ | 2059/5920 [32:03<1:14:41, 1.16s/it, tok/s=14287.5]
Inference: 35%|███▍ | 2059/5920 [32:03<1:14:41, 1.16s/it, tok/s=14291.2]
Inference: 35%|███▍ | 2060/5920 [32:03<57:41, 1.11it/s, tok/s=14291.2]
Inference: 35%|███▍ | 2060/5920 [32:03<57:41, 1.11it/s, tok/s=14298.7]
Inference: 35%|███▍ | 2061/5920 [32:03<57:41, 1.11it/s, tok/s=14306.5]
Inference: 35%|███▍ | 2062/5920 [32:04<49:40, 1.29it/s, tok/s=14306.5]
Inference: 35%|███▍ | 2062/5920 [32:04<49:40, 1.29it/s, tok/s=14312.5]
Inference: 35%|███▍ | 2063/5920 [32:05<57:01, 1.13it/s, tok/s=14312.5]
Inference: 35%|███▍ | 2063/5920 [32:05<57:01, 1.13it/s, tok/s=14318.9]
Inference: 35%|███▍ | 2064/5920 [32:06<47:29, 1.35it/s, tok/s=14318.9]
Inference: 35%|███▍ | 2064/5920 [32:06<47:29, 1.35it/s, tok/s=14329.1]
Inference: 35%|███▍ | 2065/5920 [32:06<37:04, 1.73it/s, tok/s=14329.1]
Inference: 35%|███▍ | 2065/5920 [32:06<37:04, 1.73it/s, tok/s=14341.0]
Inference: 35%|███▍ | 2066/5920 [32:07<45:15, 1.42it/s, tok/s=14341.0]
Inference: 35%|███▍ | 2066/5920 [32:07<45:15, 1.42it/s, tok/s=14348.4]
Inference: 35%|███▍ | 2067/5920 [32:08<51:47, 1.24it/s, tok/s=14348.4]
Inference: 35%|███▍ | 2067/5920 [32:08<51:47, 1.24it/s, tok/s=14352.2]
Inference: 35%|███▍ | 2068/5920 [32:09<59:19, 1.08it/s, tok/s=14352.2]
Inference: 35%|███▍ | 2068/5920 [32:09<59:19, 1.08it/s, tok/s=14358.8]
Inference: 35%|███▍ | 2069/5920 [32:13<1:46:52, 1.67s/it, tok/s=14358.8]
Inference: 35%|███▍ | 2069/5920 [32:13<1:46:52, 1.67s/it, tok/s=14348.0]
Inference: 35%|███▍ | 2070/5920 [32:14<1:31:52, 1.43s/it, tok/s=14348.0]
Inference: 35%|███▍ | 2070/5920 [32:14<1:31:52, 1.43s/it, tok/s=14354.0]
Inference: 35%|███▍ | 2071/5920 [32:20<3:04:22, 2.87s/it, tok/s=14354.0]
Inference: 35%|███▍ | 2071/5920 [32:20<3:04:22, 2.87s/it, tok/s=14316.9]
Inference: 35%|███▌ | 2072/5920 [32:20<2:16:02, 2.12s/it, tok/s=14316.9]
Inference: 35%|███▌ | 2072/5920 [32:20<2:16:02, 2.12s/it, tok/s=14329.9]
Inference: 35%|███▌ | 2073/5920 [32:25<2:57:51, 2.77s/it, tok/s=14329.9]
Inference: 35%|███▌ | 2073/5920 [32:25<2:57:51, 2.77s/it, tok/s=14313.6]
Inference: 35%|███▌ | 2074/5920 [32:25<2:57:48, 2.77s/it, tok/s=14327.6]
Inference: 35%|███▌ | 2075/5920 [32:27<2:15:53, 2.12s/it, tok/s=14327.6]
Inference: 35%|███▌ | 2075/5920 [32:27<2:15:53, 2.12s/it, tok/s=14321.4]
Inference: 35%|███▌ | 2076/5920 [32:30<2:32:42, 2.38s/it, tok/s=14321.4]
Inference: 35%|███▌ | 2076/5920 [32:30<2:32:42, 2.38s/it, tok/s=14301.8]
Inference: 35%|███▌ | 2077/5920 [32:30<2:32:40, 2.38s/it, tok/s=14316.7]
Inference: 35%|███▌ | 2078/5920 [32:31<1:36:18, 1.50s/it, tok/s=14316.7]
Inference: 35%|███▌ | 2078/5920 [32:31<1:36:18, 1.50s/it, tok/s=14320.6]
Inference: 35%|███▌ | 2079/5920 [32:33<1:38:02, 1.53s/it, tok/s=14320.6]
Inference: 35%|███▌ | 2079/5920 [32:33<1:38:02, 1.53s/it, tok/s=14317.7]
Inference: 35%|███▌ | 2080/5920 [32:33<1:16:32, 1.20s/it, tok/s=14317.7]
Inference: 35%|███▌ | 2080/5920 [32:33<1:16:32, 1.20s/it, tok/s=14331.9]
Inference: 35%|███▌ | 2081/5920 [32:34<1:19:35, 1.24s/it, tok/s=14331.9]
Inference: 35%|███▌ | 2081/5920 [32:34<1:19:35, 1.24s/it, tok/s=14335.1]
Inference: 35%|███▌ | 2082/5920 [32:35<1:07:00, 1.05s/it, tok/s=14335.1]
Inference: 35%|███▌ | 2082/5920 [32:35<1:07:00, 1.05s/it, tok/s=14344.9]
Inference: 35%|███▌ | 2083/5920 [32:37<1:24:19, 1.32s/it, tok/s=14344.9]
Inference: 35%|███▌ | 2083/5920 [32:37<1:24:19, 1.32s/it, tok/s=14339.3]
Inference: 35%|███▌ | 2084/5920 [32:37<1:06:33, 1.04s/it, tok/s=14339.3]
Inference: 35%|███▌ | 2084/5920 [32:37<1:06:33, 1.04s/it, tok/s=14352.2]
Inference: 35%|███▌ | 2085/5920 [32:38<1:01:39, 1.04it/s, tok/s=14352.2]
Inference: 35%|███▌ | 2085/5920 [32:38<1:01:39, 1.04it/s, tok/s=14350.4]
Inference: 35%|███▌ | 2086/5920 [32:38<50:47, 1.26it/s, tok/s=14350.4]
Inference: 35%|███▌ | 2086/5920 [32:38<50:47, 1.26it/s, tok/s=14350.7]
Inference: 35%|███▌ | 2087/5920 [32:43<2:16:02, 2.13s/it, tok/s=14350.7]
Inference: 35%|███▌ | 2087/5920 [32:43<2:16:02, 2.13s/it, tok/s=14327.1]
Inference: 35%|███▌ | 2088/5920 [32:48<3:05:16, 2.90s/it, tok/s=14327.1]
Inference: 35%|███▌ | 2088/5920 [32:48<3:05:16, 2.90s/it, tok/s=14303.4]
Inference: 35%|███▌ | 2089/5920 [32:49<2:33:24, 2.40s/it, tok/s=14303.4]
Inference: 35%|███▌ | 2089/5920 [32:49<2:33:24, 2.40s/it, tok/s=14309.8]
Inference: 35%|███▌ | 2090/5920 [32:50<2:00:56, 1.89s/it, tok/s=14309.8]
Inference: 35%|███▌ | 2090/5920 [32:50<2:00:56, 1.89s/it, tok/s=14315.5]
Inference: 35%|███▌ | 2091/5920 [32:51<1:34:36, 1.48s/it, tok/s=14315.5]
Inference: 35%|███▌ | 2091/5920 [32:51<1:34:36, 1.48s/it, tok/s=14320.6]
Inference: 35%|███▌ | 2092/5920 [32:53<1:51:45, 1.75s/it, tok/s=14320.6]
Inference: 35%|███▌ | 2092/5920 [32:53<1:51:45, 1.75s/it, tok/s=14313.2]
Inference: 35%|███▌ | 2093/5920 [32:54<1:40:55, 1.58s/it, tok/s=14313.2]
Inference: 35%|███▌ | 2093/5920 [32:54<1:40:55, 1.58s/it, tok/s=14319.9]
Inference: 35%|███▌ | 2094/5920 [32:55<1:20:53, 1.27s/it, tok/s=14319.9]
Inference: 35%|███▌ | 2094/5920 [32:55<1:20:53, 1.27s/it, tok/s=14326.8]
Inference: 35%|███▌ | 2095/5920 [32:57<1:31:40, 1.44s/it, tok/s=14326.8]
Inference: 35%|███▌ | 2095/5920 [32:57<1:31:40, 1.44s/it, tok/s=14323.8]
Inference: 35%|███▌ | 2096/5920 [33:00<2:00:23, 1.89s/it, tok/s=14323.8]
Inference: 35%|███▌ | 2096/5920 [33:00<2:00:23, 1.89s/it, tok/s=14305.7]
Inference: 35%|███▌ | 2097/5920 [33:00<1:30:24, 1.42s/it, tok/s=14305.7]
Inference: 35%|███▌ | 2097/5920 [33:00<1:30:24, 1.42s/it, tok/s=14316.6]
Inference: 35%|███▌ | 2098/5920 [33:00<1:13:12, 1.15s/it, tok/s=14316.6]
Inference: 35%|███▌ | 2098/5920 [33:00<1:13:12, 1.15s/it, tok/s=14328.0]
Inference: 35%|███▌ | 2099/5920 [33:01<1:01:42, 1.03it/s, tok/s=14328.0]
Inference: 35%|███▌ | 2099/5920 [33:01<1:01:42, 1.03it/s, tok/s=14332.9]
Inference: 35%|███▌ | 2100/5920 [33:01<48:35, 1.31it/s, tok/s=14332.9]
Inference: 35%|███▌ | 2100/5920 [33:01<48:35, 1.31it/s, tok/s=14346.0]
Inference: 35%|███▌ | 2101/5920 [33:02<46:46, 1.36it/s, tok/s=14346.0]
Inference: 35%|███▌ | 2101/5920 [33:02<46:46, 1.36it/s, tok/s=14344.4]
Inference: 36%|███▌ | 2102/5920 [33:02<38:41, 1.64it/s, tok/s=14344.4]
Inference: 36%|███▌ | 2102/5920 [33:02<38:41, 1.64it/s, tok/s=14344.9]
Inference: 36%|███▌ | 2103/5920 [33:03<39:48, 1.60it/s, tok/s=14344.9]
Inference: 36%|███▌ | 2103/5920 [33:03<39:48, 1.60it/s, tok/s=14347.7]
Inference: 36%|███▌ | 2104/5920 [33:03<31:20, 2.03it/s, tok/s=14347.7]
Inference: 36%|███▌ | 2104/5920 [33:03<31:20, 2.03it/s, tok/s=14350.2]
Inference: 36%|███▌ | 2105/5920 [33:04<45:17, 1.40it/s, tok/s=14350.2]
Inference: 36%|███▌ | 2105/5920 [33:04<45:17, 1.40it/s, tok/s=14352.6]
Inference: 36%|███▌ | 2106/5920 [33:05<51:05, 1.24it/s, tok/s=14352.6]
Inference: 36%|███▌ | 2106/5920 [33:05<51:05, 1.24it/s, tok/s=14355.2]
Inference: 36%|███▌ | 2107/5920 [33:08<1:35:54, 1.51s/it, tok/s=14355.2]
Inference: 36%|███▌ | 2107/5920 [33:08<1:35:54, 1.51s/it, tok/s=14346.8]
Inference: 36%|███▌ | 2108/5920 [33:12<2:07:02, 2.00s/it, tok/s=14346.8]
Inference: 36%|███▌ | 2108/5920 [33:12<2:07:02, 2.00s/it, tok/s=14336.1]
Inference: 36%|███▌ | 2109/5920 [33:12<1:39:03, 1.56s/it, tok/s=14336.1]
Inference: 36%|███▌ | 2109/5920 [33:12<1:39:03, 1.56s/it, tok/s=14344.4]
Inference: 36%|███▌ | 2110/5920 [33:17<2:36:20, 2.46s/it, tok/s=14344.4]
Inference: 36%|███▌ | 2110/5920 [33:17<2:36:20, 2.46s/it, tok/s=14324.2]
Inference: 36%|███▌ | 2111/5920 [33:17<1:57:12, 1.85s/it, tok/s=14324.2]
Inference: 36%|███▌ | 2111/5920 [33:17<1:57:12, 1.85s/it, tok/s=14324.6]
Inference: 36%|███▌ | 2112/5920 [33:19<2:02:55, 1.94s/it, tok/s=14324.6]
Inference: 36%|███▌ | 2112/5920 [33:19<2:02:55, 1.94s/it, tok/s=14319.2]
Inference: 36%|███▌ | 2113/5920 [33:19<1:28:11, 1.39s/it, tok/s=14319.2]
Inference: 36%|███▌ | 2113/5920 [33:19<1:28:11, 1.39s/it, tok/s=14333.5]
Inference: 36%|███▌ | 2114/5920 [33:24<2:35:04, 2.44s/it, tok/s=14333.5]
Inference: 36%|███▌ | 2114/5920 [33:24<2:35:04, 2.44s/it, tok/s=14312.7]
Inference: 36%|███▌ | 2115/5920 [33:25<2:00:29, 1.90s/it, tok/s=14312.7]
Inference: 36%|███▌ | 2115/5920 [33:25<2:00:29, 1.90s/it, tok/s=14318.6]
Inference: 36%|███▌ | 2116/5920 [33:26<1:39:39, 1.57s/it, tok/s=14318.6]
Inference: 36%|███▌ | 2116/5920 [33:26<1:39:39, 1.57s/it, tok/s=14316.0]
Inference: 36%|███▌ | 2117/5920 [33:26<1:15:30, 1.19s/it, tok/s=14316.0]
Inference: 36%|███▌ | 2117/5920 [33:26<1:15:30, 1.19s/it, tok/s=14323.7]
Inference: 36%|███▌ | 2118/5920 [33:26<1:02:15, 1.02it/s, tok/s=14323.7]
Inference: 36%|███▌ | 2118/5920 [33:26<1:02:15, 1.02it/s, tok/s=14334.7]
Inference: 36%|███▌ | 2119/5920 [33:28<1:18:55, 1.25s/it, tok/s=14334.7]
Inference: 36%|███▌ | 2119/5920 [33:28<1:18:55, 1.25s/it, tok/s=14324.6]
Inference: 36%|███▌ | 2120/5920 [33:29<1:15:02, 1.18s/it, tok/s=14324.6]
Inference: 36%|███▌ | 2120/5920 [33:29<1:15:02, 1.18s/it, tok/s=14329.6]
Inference: 36%|███▌ | 2121/5920 [33:31<1:14:57, 1.18s/it, tok/s=14329.6]
Inference: 36%|███▌ | 2121/5920 [33:31<1:14:57, 1.18s/it, tok/s=14330.1]
Inference: 36%|███▌ | 2122/5920 [33:31<1:04:03, 1.01s/it, tok/s=14330.1]
Inference: 36%|███▌ | 2122/5920 [33:31<1:04:03, 1.01s/it, tok/s=14338.7]
Inference: 36%|███▌ | 2123/5920 [33:32<1:05:51, 1.04s/it, tok/s=14338.7]
Inference: 36%|███▌ | 2123/5920 [33:32<1:05:51, 1.04s/it, tok/s=14341.3]
Inference: 36%|███▌ | 2124/5920 [33:34<1:11:05, 1.12s/it, tok/s=14341.3]
Inference: 36%|███▌ | 2124/5920 [33:34<1:11:05, 1.12s/it, tok/s=14342.4]
Inference: 36%|███▌ | 2125/5920 [33:35<1:21:21, 1.29s/it, tok/s=14342.4]
Inference: 36%|███▌ | 2125/5920 [33:35<1:21:21, 1.29s/it, tok/s=14341.5]
Inference: 36%|███▌ | 2126/5920 [33:35<1:01:31, 1.03it/s, tok/s=14341.5]
Inference: 36%|███▌ | 2126/5920 [33:35<1:01:31, 1.03it/s, tok/s=14341.6]
Inference: 36%|███▌ | 2127/5920 [33:36<47:16, 1.34it/s, tok/s=14341.6]
Inference: 36%|███▌ | 2127/5920 [33:36<47:16, 1.34it/s, tok/s=14348.0]
Inference: 36%|███▌ | 2128/5920 [33:37<52:48, 1.20it/s, tok/s=14348.0]
Inference: 36%|███▌ | 2128/5920 [33:37<52:48, 1.20it/s, tok/s=14353.9]
Inference: 36%|███▌ | 2129/5920 [33:37<44:11, 1.43it/s, tok/s=14353.9]
Inference: 36%|███▌ | 2129/5920 [33:37<44:11, 1.43it/s, tok/s=14353.2]
Inference: 36%|███▌ | 2130/5920 [33:39<1:00:45, 1.04it/s, tok/s=14353.2]
Inference: 36%|███▌ | 2130/5920 [33:39<1:00:45, 1.04it/s, tok/s=14354.4]
Inference: 36%|███▌ | 2131/5920 [33:39<56:04, 1.13it/s, tok/s=14354.4]
Inference: 36%|███▌ | 2131/5920 [33:39<56:04, 1.13it/s, tok/s=14362.2]
Inference: 36%|███▌ | 2132/5920 [33:42<1:25:44, 1.36s/it, tok/s=14362.2]
Inference: 36%|███▌ | 2132/5920 [33:42<1:25:44, 1.36s/it, tok/s=14357.9]
Inference: 36%|███▌ | 2133/5920 [33:45<2:03:48, 1.96s/it, tok/s=14357.9]
Inference: 36%|███▌ | 2133/5920 [33:45<2:03:48, 1.96s/it, tok/s=14337.6]
Inference: 36%|███▌ | 2134/5920 [33:47<1:50:52, 1.76s/it, tok/s=14337.6]
Inference: 36%|███▌ | 2134/5920 [33:47<1:50:52, 1.76s/it, tok/s=14343.4]
Inference: 36%|███▌ | 2135/5920 [33:54<3:37:54, 3.45s/it, tok/s=14343.4]
Inference: 36%|███▌ | 2135/5920 [33:54<3:37:54, 3.45s/it, tok/s=14293.1]
Inference: 36%|███▌ | 2136/5920 [33:56<3:08:53, 3.00s/it, tok/s=14293.1]
Inference: 36%|███▌ | 2136/5920 [33:56<3:08:53, 3.00s/it, tok/s=14282.8]
Inference: 36%|███▌ | 2137/5920 [34:00<3:34:28, 3.40s/it, tok/s=14282.8]
Inference: 36%|███▌ | 2137/5920 [34:00<3:34:28, 3.40s/it, tok/s=14260.4]
Inference: 36%|███▌ | 2138/5920 [34:01<2:46:22, 2.64s/it, tok/s=14260.4]
Inference: 36%|███▌ | 2138/5920 [34:01<2:46:22, 2.64s/it, tok/s=14263.6]
Inference: 36%|███▌ | 2139/5920 [34:01<2:04:02, 1.97s/it, tok/s=14263.6]
Inference: 36%|███▌ | 2139/5920 [34:01<2:04:02, 1.97s/it, tok/s=14264.6]
Inference: 36%|███▌ | 2140/5920 [34:05<2:41:42, 2.57s/it, tok/s=14264.6]
Inference: 36%|███▌ | 2140/5920 [34:05<2:41:42, 2.57s/it, tok/s=14241.1]
Inference: 36%|███▌ | 2141/5920 [34:08<2:37:33, 2.50s/it, tok/s=14241.1]
Inference: 36%|███▌ | 2141/5920 [34:08<2:37:33, 2.50s/it, tok/s=14228.6]
Inference: 36%|███▌ | 2142/5920 [34:08<1:53:02, 1.80s/it, tok/s=14228.6]
Inference: 36%|███▌ | 2142/5920 [34:08<1:53:02, 1.80s/it, tok/s=14237.1]
Inference: 36%|███▌ | 2143/5920 [34:08<1:28:18, 1.40s/it, tok/s=14237.1]
Inference: 36%|███▌ | 2143/5920 [34:08<1:28:18, 1.40s/it, tok/s=14236.0]
Inference: 36%|███▌ | 2144/5920 [34:10<1:30:26, 1.44s/it, tok/s=14236.0]
Inference: 36%|███▌ | 2144/5920 [34:10<1:30:26, 1.44s/it, tok/s=14237.2]
Inference: 36%|███▌ | 2145/5920 [34:13<2:00:14, 1.91s/it, tok/s=14237.2]
Inference: 36%|███▌ | 2145/5920 [34:13<2:00:14, 1.91s/it, tok/s=14227.9]
Inference: 36%|███▋ | 2146/5920 [34:13<1:28:35, 1.41s/it, tok/s=14227.9]
Inference: 36%|███▋ | 2146/5920 [34:13<1:28:35, 1.41s/it, tok/s=14229.0]
Inference: 36%|███▋ | 2147/5920 [34:13<1:07:04, 1.07s/it, tok/s=14229.0]
Inference: 36%|███▋ | 2147/5920 [34:13<1:07:04, 1.07s/it, tok/s=14241.7]
Inference: 36%|███▋ | 2148/5920 [34:15<1:08:14, 1.09s/it, tok/s=14241.7]
Inference: 36%|███▋ | 2148/5920 [34:15<1:08:14, 1.09s/it, tok/s=14236.9]
Inference: 36%|███▋ | 2149/5920 [34:20<2:21:23, 2.25s/it, tok/s=14236.9]
Inference: 36%|███▋ | 2149/5920 [34:20<2:21:23, 2.25s/it, tok/s=14216.8]
Inference: 36%|███▋ | 2150/5920 [34:22<2:34:15, 2.45s/it, tok/s=14216.8]
Inference: 36%|███▋ | 2150/5920 [34:22<2:34:15, 2.45s/it, tok/s=14210.5]
Inference: 36%|███▋ | 2151/5920 [34:25<2:38:46, 2.53s/it, tok/s=14210.5]
Inference: 36%|███▋ | 2151/5920 [34:25<2:38:46, 2.53s/it, tok/s=14206.5]
Inference: 36%|███▋ | 2152/5920 [34:25<2:38:43, 2.53s/it, tok/s=14210.3]
Inference: 36%|███▋ | 2153/5920 [34:25<2:38:41, 2.53s/it, tok/s=14212.4]
Inference: 36%|███▋ | 2154/5920 [34:28<1:47:38, 1.71s/it, tok/s=14212.4]
Inference: 36%|███▋ | 2154/5920 [34:28<1:47:38, 1.71s/it, tok/s=14195.0]
Inference: 36%|███▋ | 2155/5920 [34:30<1:44:08, 1.66s/it, tok/s=14195.0]
Inference: 36%|███▋ | 2155/5920 [34:30<1:44:08, 1.66s/it, tok/s=14199.5]
Inference: 36%|███▋ | 2156/5920 [34:31<1:42:43, 1.64s/it, tok/s=14199.5]
Inference: 36%|███▋ | 2156/5920 [34:31<1:42:43, 1.64s/it, tok/s=14203.5]
Inference: 36%|███▋ | 2157/5920 [34:34<1:59:15, 1.90s/it, tok/s=14203.5]
Inference: 36%|███▋ | 2157/5920 [34:34<1:59:15, 1.90s/it, tok/s=14197.5]
Inference: 36%|███▋ | 2158/5920 [34:34<1:33:14, 1.49s/it, tok/s=14197.5]
Inference: 36%|███▋ | 2158/5920 [34:34<1:33:14, 1.49s/it, tok/s=14209.0]
Inference: 36%|███▋ | 2159/5920 [34:38<2:16:16, 2.17s/it, tok/s=14209.0]
Inference: 36%|███▋ | 2159/5920 [34:38<2:16:16, 2.17s/it, tok/s=14193.3]
Inference: 36%|███▋ | 2160/5920 [34:39<1:45:42, 1.69s/it, tok/s=14193.3]
Inference: 36%|███▋ | 2160/5920 [34:39<1:45:42, 1.69s/it, tok/s=14192.8]
Inference: 37%|███▋ | 2161/5920 [34:39<1:25:41, 1.37s/it, tok/s=14192.8]
Inference: 37%|███▋ | 2161/5920 [34:39<1:25:41, 1.37s/it, tok/s=14198.3]
Inference: 37%|███▋ | 2162/5920 [34:41<1:31:24, 1.46s/it, tok/s=14198.3]
Inference: 37%|███▋ | 2162/5920 [34:41<1:31:24, 1.46s/it, tok/s=14201.5]
Inference: 37%|███▋ | 2163/5920 [34:45<2:15:02, 2.16s/it, tok/s=14201.5]
Inference: 37%|███▋ | 2163/5920 [34:45<2:15:02, 2.16s/it, tok/s=14189.2]
Inference: 37%|███▋ | 2164/5920 [34:45<1:40:28, 1.61s/it, tok/s=14189.2]
Inference: 37%|███▋ | 2164/5920 [34:45<1:40:28, 1.61s/it, tok/s=14201.7]
Inference: 37%|███▋ | 2165/5920 [34:48<2:05:41, 2.01s/it, tok/s=14201.7]
Inference: 37%|███▋ | 2165/5920 [34:48<2:05:41, 2.01s/it, tok/s=14185.9]
Inference: 37%|███▋ | 2166/5920 [34:52<2:37:24, 2.52s/it, tok/s=14185.9]
Inference: 37%|███▋ | 2166/5920 [34:52<2:37:24, 2.52s/it, tok/s=14165.6]
Inference: 37%|███▋ | 2167/5920 [34:53<2:15:51, 2.17s/it, tok/s=14165.6]
Inference: 37%|███▋ | 2167/5920 [34:53<2:15:51, 2.17s/it, tok/s=14168.3]
Inference: 37%|███▋ | 2168/5920 [34:58<2:58:11, 2.85s/it, tok/s=14168.3]
Inference: 37%|███▋ | 2168/5920 [34:58<2:58:11, 2.85s/it, tok/s=14142.8]
Inference: 37%|███▋ | 2169/5920 [35:02<3:16:03, 3.14s/it, tok/s=14142.8]
Inference: 37%|███▋ | 2169/5920 [35:02<3:16:03, 3.14s/it, tok/s=14130.6]
Inference: 37%|███▋ | 2170/5920 [35:02<2:30:26, 2.41s/it, tok/s=14130.6]
Inference: 37%|███▋ | 2170/5920 [35:02<2:30:26, 2.41s/it, tok/s=14137.1]
Inference: 37%|███▋ | 2171/5920 [35:04<2:08:41, 2.06s/it, tok/s=14137.1]
Inference: 37%|███▋ | 2171/5920 [35:04<2:08:41, 2.06s/it, tok/s=14141.9]
Inference: 37%|███▋ | 2172/5920 [35:04<1:42:51, 1.65s/it, tok/s=14141.9]
Inference: 37%|███▋ | 2172/5920 [35:04<1:42:51, 1.65s/it, tok/s=14150.1]
Inference: 37%|███▋ | 2173/5920 [35:04<1:42:49, 1.65s/it, tok/s=14163.8]
Inference: 37%|███▋ | 2174/5920 [35:06<1:27:28, 1.40s/it, tok/s=14163.8]
Inference: 37%|███▋ | 2174/5920 [35:06<1:27:28, 1.40s/it, tok/s=14162.0]
Inference: 37%|███▋ | 2175/5920 [35:08<1:29:01, 1.43s/it, tok/s=14162.0]
Inference: 37%|███▋ | 2175/5920 [35:08<1:29:01, 1.43s/it, tok/s=14163.3]
Inference: 37%|███▋ | 2176/5920 [35:10<1:36:43, 1.55s/it, tok/s=14163.3]
Inference: 37%|███▋ | 2176/5920 [35:10<1:36:43, 1.55s/it, tok/s=14164.8]
Inference: 37%|███▋ | 2177/5920 [35:10<1:36:42, 1.55s/it, tok/s=14179.0]
Inference: 37%|███▋ | 2178/5920 [35:14<1:49:58, 1.76s/it, tok/s=14179.0]
Inference: 37%|███▋ | 2178/5920 [35:14<1:49:58, 1.76s/it, tok/s=14160.8]
Inference: 37%|███▋ | 2179/5920 [35:14<1:32:05, 1.48s/it, tok/s=14160.8]
Inference: 37%|███▋ | 2179/5920 [35:14<1:32:05, 1.48s/it, tok/s=14170.1]
Inference: 37%|███▋ | 2180/5920 [35:17<1:47:12, 1.72s/it, tok/s=14170.1]
Inference: 37%|███▋ | 2180/5920 [35:17<1:47:12, 1.72s/it, tok/s=14162.2]
Inference: 37%|███▋ | 2181/5920 [35:17<1:47:10, 1.72s/it, tok/s=14170.6]
Inference: 37%|███▋ | 2182/5920 [35:17<1:47:08, 1.72s/it, tok/s=14181.8]
Inference: 37%|███▋ | 2183/5920 [35:19<1:11:09, 1.14s/it, tok/s=14181.8]
Inference: 37%|███▋ | 2183/5920 [35:19<1:11:09, 1.14s/it, tok/s=14181.8]
Inference: 37%|███▋ | 2184/5920 [35:20<1:17:06, 1.24s/it, tok/s=14181.8]
Inference: 37%|███▋ | 2184/5920 [35:20<1:17:06, 1.24s/it, tok/s=14180.5]
Inference: 37%|███▋ | 2185/5920 [35:26<2:23:36, 2.31s/it, tok/s=14180.5]
Inference: 37%|███▋ | 2185/5920 [35:26<2:23:36, 2.31s/it, tok/s=14143.8]
Inference: 37%|███▋ | 2186/5920 [35:32<3:06:41, 3.00s/it, tok/s=14143.8]
Inference: 37%|███▋ | 2186/5920 [35:32<3:06:41, 3.00s/it, tok/s=14112.0]
Inference: 37%|███▋ | 2187/5920 [35:33<2:47:13, 2.69s/it, tok/s=14112.0]
Inference: 37%|███▋ | 2187/5920 [35:33<2:47:13, 2.69s/it, tok/s=14110.3]
Inference: 37%|███▋ | 2188/5920 [35:34<2:08:42, 2.07s/it, tok/s=14110.3]
Inference: 37%|███▋ | 2188/5920 [35:34<2:08:42, 2.07s/it, tok/s=14117.4]
Inference: 37%|███▋ | 2189/5920 [35:36<2:16:25, 2.19s/it, tok/s=14117.4]
Inference: 37%|███▋ | 2189/5920 [35:36<2:16:25, 2.19s/it, tok/s=14112.8]
Inference: 37%|███▋ | 2190/5920 [35:40<2:41:25, 2.60s/it, tok/s=14112.8]
Inference: 37%|███▋ | 2190/5920 [35:40<2:41:25, 2.60s/it, tok/s=14091.6]
Inference: 37%|███▋ | 2191/5920 [35:41<2:04:42, 2.01s/it, tok/s=14091.6]
Inference: 37%|███▋ | 2191/5920 [35:41<2:04:42, 2.01s/it, tok/s=14092.3]
Inference: 37%|███▋ | 2192/5920 [35:44<2:32:41, 2.46s/it, tok/s=14092.3]
Inference: 37%|███▋ | 2192/5920 [35:44<2:32:41, 2.46s/it, tok/s=14083.0]
Inference: 37%|███▋ | 2193/5920 [35:44<2:32:39, 2.46s/it, tok/s=14092.6]
Inference: 37%|███▋ | 2194/5920 [35:46<1:52:49, 1.82s/it, tok/s=14092.6]
Inference: 37%|███▋ | 2194/5920 [35:46<1:52:49, 1.82s/it, tok/s=14091.1]
Inference: 37%|███▋ | 2195/5920 [35:46<1:27:41, 1.41s/it, tok/s=14091.1]
Inference: 37%|███▋ | 2195/5920 [35:46<1:27:41, 1.41s/it, tok/s=14100.3]
Inference: 37%|███▋ | 2196/5920 [35:51<2:19:54, 2.25s/it, tok/s=14100.3]
Inference: 37%|███▋ | 2196/5920 [35:51<2:19:54, 2.25s/it, tok/s=14072.6]
Inference: 37%|███▋ | 2197/5920 [35:51<1:49:32, 1.77s/it, tok/s=14072.6]
Inference: 37%|███▋ | 2197/5920 [35:51<1:49:32, 1.77s/it, tok/s=14080.5]
Inference: 37%|███▋ | 2198/5920 [35:51<1:49:30, 1.77s/it, tok/s=14085.1]
Inference: 37%|███▋ | 2199/5920 [35:52<1:49:28, 1.77s/it, tok/s=14087.4]
Inference: 37%|███▋ | 2200/5920 [35:53<1:10:20, 1.13s/it, tok/s=14087.4]
Inference: 37%|███▋ | 2200/5920 [35:53<1:10:20, 1.13s/it, tok/s=14089.5]
Inference: 37%|███▋ | 2201/5920 [35:55<1:13:45, 1.19s/it, tok/s=14089.5]
Inference: 37%|███▋ | 2201/5920 [35:55<1:13:45, 1.19s/it, tok/s=14089.4]
Inference: 37%|███▋ | 2202/5920 [35:59<1:59:42, 1.93s/it, tok/s=14089.4]
Inference: 37%|███▋ | 2202/5920 [35:59<1:59:42, 1.93s/it, tok/s=14065.3]
Inference: 37%|███▋ | 2203/5920 [36:00<1:51:43, 1.80s/it, tok/s=14065.3]
Inference: 37%|███▋ | 2203/5920 [36:00<1:51:43, 1.80s/it, tok/s=14060.7]
Inference: 37%|███▋ | 2204/5920 [36:04<2:13:04, 2.15s/it, tok/s=14060.7]
Inference: 37%|███▋ | 2204/5920 [36:04<2:13:04, 2.15s/it, tok/s=14051.1]
Inference: 37%|███▋ | 2205/5920 [36:06<2:13:22, 2.15s/it, tok/s=14051.1]
Inference: 37%|███▋ | 2205/5920 [36:06<2:13:22, 2.15s/it, tok/s=14040.8]
Inference: 37%|███▋ | 2206/5920 [36:06<2:13:20, 2.15s/it, tok/s=14053.5]
Inference: 37%|███▋ | 2207/5920 [36:07<1:35:34, 1.54s/it, tok/s=14053.5]
Inference: 37%|███▋ | 2207/5920 [36:07<1:35:34, 1.54s/it, tok/s=14053.6]
Inference: 37%|███▋ | 2208/5920 [36:11<2:10:27, 2.11s/it, tok/s=14053.6]
Inference: 37%|███▋ | 2208/5920 [36:11<2:10:27, 2.11s/it, tok/s=14042.3]
Inference: 37%|███▋ | 2209/5920 [36:12<1:42:31, 1.66s/it, tok/s=14042.3]
Inference: 37%|███▋ | 2209/5920 [36:12<1:42:31, 1.66s/it, tok/s=14050.1]
Inference: 37%|███▋ | 2210/5920 [36:15<2:04:02, 2.01s/it, tok/s=14050.1]
Inference: 37%|███▋ | 2210/5920 [36:15<2:04:02, 2.01s/it, tok/s=14034.5]
Inference: 37%|███▋ | 2211/5920 [36:15<1:41:27, 1.64s/it, tok/s=14034.5]
Inference: 37%|███▋ | 2211/5920 [36:15<1:41:27, 1.64s/it, tok/s=14043.0]
Inference: 37%|███▋ | 2212/5920 [36:16<1:18:14, 1.27s/it, tok/s=14043.0]
Inference: 37%|███▋ | 2212/5920 [36:16<1:18:14, 1.27s/it, tok/s=14054.8]
Inference: 37%|███▋ | 2213/5920 [36:17<1:16:27, 1.24s/it, tok/s=14054.8]
Inference: 37%|███▋ | 2213/5920 [36:17<1:16:27, 1.24s/it, tok/s=14051.0]
Inference: 37%|███▋ | 2214/5920 [36:17<57:35, 1.07it/s, tok/s=14051.0]
Inference: 37%|███▋ | 2214/5920 [36:17<57:35, 1.07it/s, tok/s=14053.7]
Inference: 37%|███▋ | 2215/5920 [36:17<45:32, 1.36it/s, tok/s=14053.7]
Inference: 37%|███▋ | 2215/5920 [36:17<45:32, 1.36it/s, tok/s=14062.1]
Inference: 37%|███▋ | 2216/5920 [36:18<43:07, 1.43it/s, tok/s=14062.1]
Inference: 37%|███▋ | 2216/5920 [36:18<43:07, 1.43it/s, tok/s=14063.2]
Inference: 37%|███▋ | 2217/5920 [36:18<37:16, 1.66it/s, tok/s=14063.2]
Inference: 37%|███▋ | 2217/5920 [36:18<37:16, 1.66it/s, tok/s=14065.4]
Inference: 37%|███▋ | 2218/5920 [36:22<1:42:31, 1.66s/it, tok/s=14065.4]
Inference: 37%|███▋ | 2218/5920 [36:22<1:42:31, 1.66s/it, tok/s=14045.7]
Inference: 37%|███▋ | 2219/5920 [36:28<3:01:34, 2.94s/it, tok/s=14045.7]
Inference: 37%|███▋ | 2219/5920 [36:28<3:01:34, 2.94s/it, tok/s=14018.5]
Inference: 38%|███▊ | 2220/5920 [36:29<2:12:29, 2.15s/it, tok/s=14018.5]
Inference: 38%|███▊ | 2220/5920 [36:29<2:12:29, 2.15s/it, tok/s=14027.1]
Inference: 38%|███▊ | 2221/5920 [36:31<2:11:42, 2.14s/it, tok/s=14027.1]
Inference: 38%|███▊ | 2221/5920 [36:31<2:11:42, 2.14s/it, tok/s=14024.6]
Inference: 38%|███▊ | 2222/5920 [36:31<1:34:39, 1.54s/it, tok/s=14024.6]
Inference: 38%|███▊ | 2222/5920 [36:31<1:34:39, 1.54s/it, tok/s=14032.1]
Inference: 38%|███▊ | 2223/5920 [36:32<1:28:20, 1.43s/it, tok/s=14032.1]
Inference: 38%|███▊ | 2223/5920 [36:32<1:28:20, 1.43s/it, tok/s=14027.8]
Inference: 38%|███▊ | 2224/5920 [36:35<1:58:51, 1.93s/it, tok/s=14027.8]
Inference: 38%|███▊ | 2224/5920 [36:35<1:58:51, 1.93s/it, tok/s=14021.8]
Inference: 38%|███▊ | 2225/5920 [36:36<1:36:05, 1.56s/it, tok/s=14021.8]
Inference: 38%|███▊ | 2225/5920 [36:36<1:36:05, 1.56s/it, tok/s=14022.2]
Inference: 38%|███▊ | 2226/5920 [36:36<1:36:03, 1.56s/it, tok/s=14035.2]
Inference: 38%|███▊ | 2227/5920 [36:37<1:02:58, 1.02s/it, tok/s=14035.2]
Inference: 38%|███▊ | 2227/5920 [36:37<1:02:58, 1.02s/it, tok/s=14044.0]
Inference: 38%|███▊ | 2228/5920 [36:37<52:36, 1.17it/s, tok/s=14044.0]
Inference: 38%|███▊ | 2228/5920 [36:37<52:36, 1.17it/s, tok/s=14046.3]
Inference: 38%|███▊ | 2229/5920 [36:37<48:23, 1.27it/s, tok/s=14046.3]
Inference: 38%|███▊ | 2229/5920 [36:37<48:23, 1.27it/s, tok/s=14045.8]
Inference: 38%|███▊ | 2230/5920 [36:41<1:32:26, 1.50s/it, tok/s=14045.8]
Inference: 38%|███▊ | 2230/5920 [36:41<1:32:26, 1.50s/it, tok/s=14033.9]
Inference: 38%|███▊ | 2231/5920 [36:41<1:32:25, 1.50s/it, tok/s=14043.7]
Inference: 38%|███▊ | 2232/5920 [36:41<57:00, 1.08it/s, tok/s=14043.7]
Inference: 38%|███▊ | 2232/5920 [36:41<57:00, 1.08it/s, tok/s=14055.5]
Inference: 38%|███▊ | 2233/5920 [36:41<44:57, 1.37it/s, tok/s=14055.5]
Inference: 38%|███▊ | 2233/5920 [36:41<44:57, 1.37it/s, tok/s=14058.1]
Inference: 38%|███▊ | 2234/5920 [36:44<1:18:47, 1.28s/it, tok/s=14058.1]
Inference: 38%|███▊ | 2234/5920 [36:44<1:18:47, 1.28s/it, tok/s=14044.9]
Inference: 38%|███▊ | 2235/5920 [36:46<1:31:55, 1.50s/it, tok/s=14044.9]
Inference: 38%|███▊ | 2235/5920 [36:46<1:31:55, 1.50s/it, tok/s=14045.2]
Inference: 38%|███▊ | 2236/5920 [36:48<1:26:38, 1.41s/it, tok/s=14045.2]
Inference: 38%|███▊ | 2236/5920 [36:48<1:26:38, 1.41s/it, tok/s=14047.9]
Inference: 38%|███▊ | 2237/5920 [36:48<1:09:13, 1.13s/it, tok/s=14047.9]
Inference: 38%|███▊ | 2237/5920 [36:48<1:09:13, 1.13s/it, tok/s=14053.7]
Inference: 38%|███▊ | 2238/5920 [36:50<1:28:01, 1.43s/it, tok/s=14053.7]
Inference: 38%|███▊ | 2238/5920 [36:50<1:28:01, 1.43s/it, tok/s=14053.6]
Inference: 38%|███▊ | 2239/5920 [36:54<2:05:25, 2.04s/it, tok/s=14053.6]
Inference: 38%|███▊ | 2239/5920 [36:54<2:05:25, 2.04s/it, tok/s=14041.4]
Inference: 38%|███▊ | 2240/5920 [36:57<2:24:59, 2.36s/it, tok/s=14041.4]
Inference: 38%|███▊ | 2240/5920 [36:57<2:24:59, 2.36s/it, tok/s=14035.1]
Inference: 38%|███▊ | 2241/5920 [37:04<3:50:11, 3.75s/it, tok/s=14035.1]
Inference: 38%|███▊ | 2241/5920 [37:04<3:50:11, 3.75s/it, tok/s=14001.4]
Inference: 38%|███▊ | 2242/5920 [37:07<3:30:42, 3.44s/it, tok/s=14001.4]
Inference: 38%|███▊ | 2242/5920 [37:07<3:30:42, 3.44s/it, tok/s=13993.4]
Inference: 38%|███▊ | 2243/5920 [37:07<2:31:53, 2.48s/it, tok/s=13993.4]
Inference: 38%|███▊ | 2243/5920 [37:07<2:31:53, 2.48s/it, tok/s=13996.2]
Inference: 38%|███▊ | 2244/5920 [37:07<1:59:24, 1.95s/it, tok/s=13996.2]
Inference: 38%|███▊ | 2244/5920 [37:07<1:59:24, 1.95s/it, tok/s=13995.5]
Inference: 38%|███▊ | 2245/5920 [37:08<1:36:32, 1.58s/it, tok/s=13995.5]
Inference: 38%|███▊ | 2245/5920 [37:08<1:36:32, 1.58s/it, tok/s=13995.4]
Inference: 38%|███▊ | 2246/5920 [37:08<1:12:27, 1.18s/it, tok/s=13995.4]
Inference: 38%|███▊ | 2246/5920 [37:08<1:12:27, 1.18s/it, tok/s=14007.2]
Inference: 38%|███▊ | 2247/5920 [37:09<57:44, 1.06it/s, tok/s=14007.2]
Inference: 38%|███▊ | 2247/5920 [37:09<57:44, 1.06it/s, tok/s=14013.4]
Inference: 38%|███▊ | 2248/5920 [37:10<57:01, 1.07it/s, tok/s=14013.4]
Inference: 38%|███▊ | 2248/5920 [37:10<57:01, 1.07it/s, tok/s=14017.0]
Inference: 38%|███▊ | 2249/5920 [37:12<1:17:46, 1.27s/it, tok/s=14017.0]
Inference: 38%|███▊ | 2249/5920 [37:12<1:17:46, 1.27s/it, tok/s=14015.8]
Inference: 38%|███▊ | 2250/5920 [37:12<58:58, 1.04it/s, tok/s=14015.8]
Inference: 38%|███▊ | 2250/5920 [37:12<58:58, 1.04it/s, tok/s=14023.0]
Inference: 38%|███▊ | 2251/5920 [37:17<2:11:52, 2.16s/it, tok/s=14023.0]
Inference: 38%|███▊ | 2251/5920 [37:17<2:11:52, 2.16s/it, tok/s=14003.7]
Inference: 38%|███▊ | 2252/5920 [37:19<2:11:22, 2.15s/it, tok/s=14003.7]
Inference: 38%|███▊ | 2252/5920 [37:19<2:11:22, 2.15s/it, tok/s=14000.1]
Inference: 38%|███▊ | 2253/5920 [37:22<2:15:57, 2.22s/it, tok/s=14000.1]
Inference: 38%|███▊ | 2253/5920 [37:22<2:15:57, 2.22s/it, tok/s=13989.3]
Inference: 38%|███▊ | 2254/5920 [37:23<2:00:39, 1.97s/it, tok/s=13989.3]
Inference: 38%|███▊ | 2254/5920 [37:23<2:00:39, 1.97s/it, tok/s=13986.7]
Inference: 38%|███▊ | 2255/5920 [37:26<2:26:07, 2.39s/it, tok/s=13986.7]
Inference: 38%|███▊ | 2255/5920 [37:26<2:26:07, 2.39s/it, tok/s=13976.1]
Inference: 38%|███▊ | 2256/5920 [37:27<1:57:32, 1.92s/it, tok/s=13976.1]
Inference: 38%|███▊ | 2256/5920 [37:27<1:57:32, 1.92s/it, tok/s=13982.3]
Inference: 38%|███▊ | 2257/5920 [37:27<1:29:03, 1.46s/it, tok/s=13982.3]
Inference: 38%|███▊ | 2257/5920 [37:27<1:29:03, 1.46s/it, tok/s=13991.1]
Inference: 38%|███▊ | 2258/5920 [37:28<1:29:02, 1.46s/it, tok/s=14001.8]
Inference: 38%|███▊ | 2259/5920 [37:28<51:19, 1.19it/s, tok/s=14001.8]
Inference: 38%|███▊ | 2259/5920 [37:28<51:19, 1.19it/s, tok/s=14009.8]
Inference: 38%|███▊ | 2260/5920 [37:28<51:18, 1.19it/s, tok/s=14022.6]
Inference: 38%|███▊ | 2261/5920 [37:28<38:40, 1.58it/s, tok/s=14022.6]
Inference: 38%|███▊ | 2261/5920 [37:28<38:40, 1.58it/s, tok/s=14023.7]
Inference: 38%|███▊ | 2262/5920 [37:30<58:38, 1.04it/s, tok/s=14023.7]
Inference: 38%|███▊ | 2262/5920 [37:30<58:38, 1.04it/s, tok/s=14019.3]
Inference: 38%|███▊ | 2263/5920 [37:32<1:12:39, 1.19s/it, tok/s=14019.3]
Inference: 38%|███▊ | 2263/5920 [37:32<1:12:39, 1.19s/it, tok/s=14020.8]
Inference: 38%|███▊ | 2264/5920 [37:33<1:02:03, 1.02s/it, tok/s=14020.8]
Inference: 38%|███▊ | 2264/5920 [37:33<1:02:03, 1.02s/it, tok/s=14026.1]
Inference: 38%|███▊ | 2265/5920 [37:35<1:13:07, 1.20s/it, tok/s=14026.1]
Inference: 38%|███▊ | 2265/5920 [37:35<1:13:07, 1.20s/it, tok/s=14025.8]
Inference: 38%|███▊ | 2266/5920 [37:35<59:22, 1.03it/s, tok/s=14025.8]
Inference: 38%|███▊ | 2266/5920 [37:35<59:22, 1.03it/s, tok/s=14034.2]
Inference: 38%|███▊ | 2267/5920 [37:37<1:13:53, 1.21s/it, tok/s=14034.2]
Inference: 38%|███▊ | 2267/5920 [37:37<1:13:53, 1.21s/it, tok/s=14031.4]
Inference: 38%|███▊ | 2268/5920 [37:38<1:04:56, 1.07s/it, tok/s=14031.4]
Inference: 38%|███▊ | 2268/5920 [37:38<1:04:56, 1.07s/it, tok/s=14036.4]
Inference: 38%|███▊ | 2269/5920 [37:38<1:00:47, 1.00it/s, tok/s=14036.4]
Inference: 38%|███▊ | 2269/5920 [37:38<1:00:47, 1.00it/s, tok/s=14041.3]
Inference: 38%|███▊ | 2270/5920 [37:39<45:54, 1.33it/s, tok/s=14041.3]
Inference: 38%|███▊ | 2270/5920 [37:39<45:54, 1.33it/s, tok/s=14051.6]
Inference: 38%|███▊ | 2271/5920 [37:43<1:58:15, 1.94s/it, tok/s=14051.6]
Inference: 38%|███▊ | 2271/5920 [37:43<1:58:15, 1.94s/it, tok/s=14035.2]
Inference: 38%|███▊ | 2272/5920 [37:43<1:58:13, 1.94s/it, tok/s=14045.7]
Inference: 38%|███▊ | 2273/5920 [37:45<1:34:34, 1.56s/it, tok/s=14045.7]
Inference: 38%|███▊ | 2273/5920 [37:45<1:34:34, 1.56s/it, tok/s=14037.1]
Inference: 38%|███▊ | 2274/5920 [37:47<1:41:03, 1.66s/it, tok/s=14037.1]
Inference: 38%|███▊ | 2274/5920 [37:47<1:41:03, 1.66s/it, tok/s=14035.9]
Inference: 38%|███▊ | 2275/5920 [37:48<1:16:42, 1.26s/it, tok/s=14035.9]
Inference: 38%|███▊ | 2275/5920 [37:48<1:16:42, 1.26s/it, tok/s=14044.1]
Inference: 38%|███▊ | 2276/5920 [37:48<1:16:40, 1.26s/it, tok/s=14055.1]
Inference: 38%|███▊ | 2277/5920 [37:49<1:07:08, 1.11s/it, tok/s=14055.1]
Inference: 38%|███▊ | 2277/5920 [37:49<1:07:08, 1.11s/it, tok/s=14053.4]
Inference: 38%|███▊ | 2278/5920 [37:52<1:34:35, 1.56s/it, tok/s=14053.4]
Inference: 38%|███▊ | 2278/5920 [37:52<1:34:35, 1.56s/it, tok/s=14044.8]
Inference: 38%|███▊ | 2279/5920 [37:55<1:49:28, 1.80s/it, tok/s=14044.8]
Inference: 38%|███▊ | 2279/5920 [37:55<1:49:28, 1.80s/it, tok/s=14033.6]
Inference: 39%|███▊ | 2280/5920 [37:57<1:53:31, 1.87s/it, tok/s=14033.6]
Inference: 39%|███▊ | 2280/5920 [37:57<1:53:31, 1.87s/it, tok/s=14032.8]
Inference: 39%|███▊ | 2281/5920 [37:59<1:52:10, 1.85s/it, tok/s=14032.8]
Inference: 39%|███▊ | 2281/5920 [37:59<1:52:10, 1.85s/it, tok/s=14031.5]
Inference: 39%|███▊ | 2282/5920 [37:59<1:23:43, 1.38s/it, tok/s=14031.5]
Inference: 39%|███▊ | 2282/5920 [37:59<1:23:43, 1.38s/it, tok/s=14043.6]
Inference: 39%|███▊ | 2283/5920 [37:59<1:02:26, 1.03s/it, tok/s=14043.6]
Inference: 39%|███▊ | 2283/5920 [37:59<1:02:26, 1.03s/it, tok/s=14048.5]
Inference: 39%|███▊ | 2284/5920 [38:00<1:07:06, 1.11s/it, tok/s=14048.5]
Inference: 39%|███▊ | 2284/5920 [38:00<1:07:06, 1.11s/it, tok/s=14050.2]
Inference: 39%|███▊ | 2285/5920 [38:05<2:16:46, 2.26s/it, tok/s=14050.2]
Inference: 39%|███▊ | 2285/5920 [38:05<2:16:46, 2.26s/it, tok/s=14029.2]
Inference: 39%|███▊ | 2286/5920 [38:06<1:41:59, 1.68s/it, tok/s=14029.2]
Inference: 39%|███▊ | 2286/5920 [38:06<1:41:59, 1.68s/it, tok/s=14038.4]
Inference: 39%|███▊ | 2287/5920 [38:07<1:33:34, 1.55s/it, tok/s=14038.4]
Inference: 39%|███▊ | 2287/5920 [38:07<1:33:34, 1.55s/it, tok/s=14039.5]
Inference: 39%|███▊ | 2288/5920 [38:08<1:30:45, 1.50s/it, tok/s=14039.5]
Inference: 39%|███▊ | 2288/5920 [38:08<1:30:45, 1.50s/it, tok/s=14040.5]
Inference: 39%|███▊ | 2289/5920 [38:09<1:11:32, 1.18s/it, tok/s=14040.5]
Inference: 39%|███▊ | 2289/5920 [38:09<1:11:32, 1.18s/it, tok/s=14048.8]
Inference: 39%|███▊ | 2290/5920 [38:14<2:22:51, 2.36s/it, tok/s=14048.8]
Inference: 39%|███▊ | 2290/5920 [38:14<2:22:51, 2.36s/it, tok/s=14027.4]
Inference: 39%|███▊ | 2291/5920 [38:14<1:46:23, 1.76s/it, tok/s=14027.4]
Inference: 39%|███▊ | 2291/5920 [38:14<1:46:23, 1.76s/it, tok/s=14034.9]
Inference: 39%|███▊ | 2292/5920 [38:16<1:44:51, 1.73s/it, tok/s=14034.9]
Inference: 39%|███▊ | 2292/5920 [38:16<1:44:51, 1.73s/it, tok/s=14033.1]
Inference: 39%|███▊ | 2293/5920 [38:17<1:35:33, 1.58s/it, tok/s=14033.1]
Inference: 39%|███▊ | 2293/5920 [38:17<1:35:33, 1.58s/it, tok/s=14035.0]
Inference: 39%|███▉ | 2294/5920 [38:19<1:33:08, 1.54s/it, tok/s=14035.0]
Inference: 39%|███▉ | 2294/5920 [38:19<1:33:08, 1.54s/it, tok/s=14038.8]
Inference: 39%|███▉ | 2295/5920 [38:19<1:13:32, 1.22s/it, tok/s=14038.8]
Inference: 39%|███▉ | 2295/5920 [38:19<1:13:32, 1.22s/it, tok/s=14045.9]
Inference: 39%|███▉ | 2296/5920 [38:20<1:03:11, 1.05s/it, tok/s=14045.9]
Inference: 39%|███▉ | 2296/5920 [38:20<1:03:11, 1.05s/it, tok/s=14051.6]
Inference: 39%|███▉ | 2297/5920 [38:20<53:34, 1.13it/s, tok/s=14051.6]
Inference: 39%|███▉ | 2297/5920 [38:20<53:34, 1.13it/s, tok/s=14057.8]
Inference: 39%|███▉ | 2298/5920 [38:21<45:08, 1.34it/s, tok/s=14057.8]
Inference: 39%|███▉ | 2298/5920 [38:21<45:08, 1.34it/s, tok/s=14059.3]
Inference: 39%|███▉ | 2299/5920 [38:21<39:36, 1.52it/s, tok/s=14059.3]
Inference: 39%|███▉ | 2299/5920 [38:21<39:36, 1.52it/s, tok/s=14061.3]
Inference: 39%|███▉ | 2300/5920 [38:21<31:12, 1.93it/s, tok/s=14061.3]
Inference: 39%|███▉ | 2300/5920 [38:21<31:12, 1.93it/s, tok/s=14064.7]
Inference: 39%|███▉ | 2301/5920 [38:23<52:39, 1.15it/s, tok/s=14064.7]
Inference: 39%|███▉ | 2301/5920 [38:23<52:39, 1.15it/s, tok/s=14065.8]
Inference: 39%|███▉ | 2302/5920 [38:25<1:14:25, 1.23s/it, tok/s=14065.8]
Inference: 39%|███▉ | 2302/5920 [38:25<1:14:25, 1.23s/it, tok/s=14063.9]
Inference: 39%|███▉ | 2303/5920 [38:27<1:21:46, 1.36s/it, tok/s=14063.9]
Inference: 39%|███▉ | 2303/5920 [38:27<1:21:46, 1.36s/it, tok/s=14063.0]
Inference: 39%|███▉ | 2304/5920 [38:28<1:24:30, 1.40s/it, tok/s=14063.0]
Inference: 39%|███▉ | 2304/5920 [38:28<1:24:30, 1.40s/it, tok/s=14061.8]
Inference: 39%|███▉ | 2305/5920 [38:28<1:01:26, 1.02s/it, tok/s=14061.8]
Inference: 39%|███▉ | 2305/5920 [38:28<1:01:26, 1.02s/it, tok/s=14065.6]
Inference: 39%|███▉ | 2306/5920 [38:31<1:26:11, 1.43s/it, tok/s=14065.6]
Inference: 39%|███▉ | 2306/5920 [38:31<1:26:11, 1.43s/it, tok/s=14061.4]
Inference: 39%|███▉ | 2307/5920 [38:31<1:06:08, 1.10s/it, tok/s=14061.4]
Inference: 39%|███▉ | 2307/5920 [38:31<1:06:08, 1.10s/it, tok/s=14069.3]
Inference: 39%|███▉ | 2308/5920 [38:31<1:06:07, 1.10s/it, tok/s=14081.9]
Inference: 39%|███▉ | 2309/5920 [38:31<1:06:05, 1.10s/it, tok/s=14095.1]
Inference: 39%|███▉ | 2310/5920 [38:32<39:08, 1.54it/s, tok/s=14095.1]
Inference: 39%|███▉ | 2310/5920 [38:32<39:08, 1.54it/s, tok/s=14095.9]
Inference: 39%|███▉ | 2311/5920 [38:32<33:19, 1.80it/s, tok/s=14095.9]
Inference: 39%|███▉ | 2311/5920 [38:32<33:19, 1.80it/s, tok/s=14100.2]
Inference: 39%|███▉ | 2312/5920 [38:35<1:10:56, 1.18s/it, tok/s=14100.2]
Inference: 39%|███▉ | 2312/5920 [38:35<1:10:56, 1.18s/it, tok/s=14085.1]
Inference: 39%|███▉ | 2313/5920 [38:36<1:10:55, 1.18s/it, tok/s=14094.7]
Inference: 39%|███▉ | 2314/5920 [38:36<44:48, 1.34it/s, tok/s=14094.7]
Inference: 39%|███▉ | 2314/5920 [38:36<44:48, 1.34it/s, tok/s=14098.8]
Inference: 39%|███▉ | 2315/5920 [38:37<57:44, 1.04it/s, tok/s=14098.8]
Inference: 39%|███▉ | 2315/5920 [38:37<57:44, 1.04it/s, tok/s=14097.7]
Inference: 39%|███▉ | 2316/5920 [38:38<55:23, 1.08it/s, tok/s=14097.7]
Inference: 39%|███▉ | 2316/5920 [38:38<55:23, 1.08it/s, tok/s=14100.0]
Inference: 39%|███▉ | 2317/5920 [38:39<48:10, 1.25it/s, tok/s=14100.0]
Inference: 39%|███▉ | 2317/5920 [38:39<48:10, 1.25it/s, tok/s=14101.2]
Inference: 39%|███▉ | 2318/5920 [38:40<1:02:27, 1.04s/it, tok/s=14101.2]
Inference: 39%|███▉ | 2318/5920 [38:40<1:02:27, 1.04s/it, tok/s=14095.2]
Inference: 39%|███▉ | 2319/5920 [38:40<1:02:26, 1.04s/it, tok/s=14104.3]
Inference: 39%|███▉ | 2320/5920 [38:42<59:45, 1.00it/s, tok/s=14104.3]
Inference: 39%|███▉ | 2320/5920 [38:42<59:45, 1.00it/s, tok/s=14103.0]
Inference: 39%|███▉ | 2321/5920 [38:43<53:01, 1.13it/s, tok/s=14103.0]
Inference: 39%|███▉ | 2321/5920 [38:43<53:01, 1.13it/s, tok/s=14106.4]
Inference: 39%|███▉ | 2322/5920 [38:43<41:14, 1.45it/s, tok/s=14106.4]
Inference: 39%|███▉ | 2322/5920 [38:43<41:14, 1.45it/s, tok/s=14117.4]
Inference: 39%|███▉ | 2323/5920 [38:44<47:01, 1.27it/s, tok/s=14117.4]
Inference: 39%|███▉ | 2323/5920 [38:44<47:01, 1.27it/s, tok/s=14120.6]
Inference: 39%|███▉ | 2324/5920 [38:44<43:20, 1.38it/s, tok/s=14120.6]
Inference: 39%|███▉ | 2324/5920 [38:44<43:20, 1.38it/s, tok/s=14123.2]
Inference: 39%|███▉ | 2325/5920 [38:45<37:54, 1.58it/s, tok/s=14123.2]
Inference: 39%|███▉ | 2325/5920 [38:45<37:54, 1.58it/s, tok/s=14132.0]
Inference: 39%|███▉ | 2326/5920 [38:45<31:16, 1.92it/s, tok/s=14132.0]
Inference: 39%|███▉ | 2326/5920 [38:45<31:16, 1.92it/s, tok/s=14141.0]
Inference: 39%|███▉ | 2327/5920 [38:45<25:43, 2.33it/s, tok/s=14141.0]
Inference: 39%|███▉ | 2327/5920 [38:45<25:43, 2.33it/s, tok/s=14149.9]
Inference: 39%|███▉ | 2328/5920 [38:46<22:15, 2.69it/s, tok/s=14149.9]
Inference: 39%|███▉ | 2328/5920 [38:46<22:15, 2.69it/s, tok/s=14154.9]
Inference: 39%|███▉ | 2329/5920 [38:46<19:19, 3.10it/s, tok/s=14154.9]
Inference: 39%|███▉ | 2329/5920 [38:46<19:19, 3.10it/s, tok/s=14162.2]
Inference: 39%|███▉ | 2330/5920 [38:48<55:34, 1.08it/s, tok/s=14162.2]
Inference: 39%|███▉ | 2330/5920 [38:48<55:34, 1.08it/s, tok/s=14151.7]
Inference: 39%|███▉ | 2331/5920 [38:51<1:37:28, 1.63s/it, tok/s=14151.7]
Inference: 39%|███▉ | 2331/5920 [38:51<1:37:28, 1.63s/it, tok/s=14141.6]
Inference: 39%|███▉ | 2332/5920 [38:52<1:17:11, 1.29s/it, tok/s=14141.6]
Inference: 39%|███▉ | 2332/5920 [38:52<1:17:11, 1.29s/it, tok/s=14143.8]
Inference: 39%|███▉ | 2333/5920 [38:55<1:41:58, 1.71s/it, tok/s=14143.8]
Inference: 39%|███▉ | 2333/5920 [38:55<1:41:58, 1.71s/it, tok/s=14140.5]
Inference: 39%|███▉ | 2334/5920 [38:58<2:07:32, 2.13s/it, tok/s=14140.5]
Inference: 39%|███▉ | 2334/5920 [38:58<2:07:32, 2.13s/it, tok/s=14125.4]
Inference: 39%|███▉ | 2335/5920 [38:58<1:32:57, 1.56s/it, tok/s=14125.4]
Inference: 39%|███▉ | 2335/5920 [38:58<1:32:57, 1.56s/it, tok/s=14134.0]
Inference: 39%|███▉ | 2336/5920 [38:59<1:18:33, 1.32s/it, tok/s=14134.0]
Inference: 39%|███▉ | 2336/5920 [38:59<1:18:33, 1.32s/it, tok/s=14134.5]
Inference: 39%|███▉ | 2337/5920 [39:00<1:17:22, 1.30s/it, tok/s=14134.5]
Inference: 39%|███▉ | 2337/5920 [39:00<1:17:22, 1.30s/it, tok/s=14134.1]
Inference: 39%|███▉ | 2338/5920 [39:01<1:06:27, 1.11s/it, tok/s=14134.1]
Inference: 39%|███▉ | 2338/5920 [39:01<1:06:27, 1.11s/it, tok/s=14134.0]
Inference: 40%|███▉ | 2339/5920 [39:01<1:06:26, 1.11s/it, tok/s=14143.9]
Inference: 40%|███▉ | 2340/5920 [39:01<41:47, 1.43it/s, tok/s=14143.9]
Inference: 40%|███▉ | 2340/5920 [39:01<41:47, 1.43it/s, tok/s=14152.1]
Inference: 40%|███▉ | 2341/5920 [39:02<41:48, 1.43it/s, tok/s=14152.1]
Inference: 40%|███▉ | 2341/5920 [39:02<41:48, 1.43it/s, tok/s=14152.3]
Inference: 40%|███▉ | 2342/5920 [39:03<56:41, 1.05it/s, tok/s=14152.3]
Inference: 40%|███▉ | 2342/5920 [39:03<56:41, 1.05it/s, tok/s=14151.4]
Inference: 40%|███▉ | 2343/5920 [39:05<59:21, 1.00it/s, tok/s=14151.4]
Inference: 40%|███▉ | 2343/5920 [39:05<59:21, 1.00it/s, tok/s=14154.3]
Inference: 40%|███▉ | 2344/5920 [39:05<49:40, 1.20it/s, tok/s=14154.3]
Inference: 40%|███▉ | 2344/5920 [39:05<49:40, 1.20it/s, tok/s=14158.5]
Inference: 40%|███▉ | 2345/5920 [39:07<1:05:09, 1.09s/it, tok/s=14158.5]
Inference: 40%|███▉ | 2345/5920 [39:07<1:05:09, 1.09s/it, tok/s=14158.1]
Inference: 40%|███▉ | 2346/5920 [39:09<1:21:28, 1.37s/it, tok/s=14158.1]
Inference: 40%|███▉ | 2346/5920 [39:09<1:21:28, 1.37s/it, tok/s=14156.6]
Inference: 40%|███▉ | 2347/5920 [39:18<3:39:24, 3.68s/it, tok/s=14156.6]
Inference: 40%|███▉ | 2347/5920 [39:18<3:39:24, 3.68s/it, tok/s=14106.8]
Inference: 40%|███▉ | 2348/5920 [39:19<2:53:20, 2.91s/it, tok/s=14106.8]
Inference: 40%|███▉ | 2348/5920 [39:19<2:53:20, 2.91s/it, tok/s=14108.6]
Inference: 40%|███▉ | 2349/5920 [39:21<2:29:31, 2.51s/it, tok/s=14108.6]
Inference: 40%|███▉ | 2349/5920 [39:21<2:29:31, 2.51s/it, tok/s=14105.5]
Inference: 40%|███▉ | 2350/5920 [39:23<2:29:46, 2.52s/it, tok/s=14105.5]
Inference: 40%|███▉ | 2350/5920 [39:23<2:29:46, 2.52s/it, tok/s=14101.1]
Inference: 40%|███▉ | 2351/5920 [39:23<2:29:44, 2.52s/it, tok/s=14101.2]
Inference: 40%|███▉ | 2352/5920 [39:26<2:03:56, 2.08s/it, tok/s=14101.2]
Inference: 40%|███▉ | 2352/5920 [39:26<2:03:56, 2.08s/it, tok/s=14083.5]
Inference: 40%|███▉ | 2353/5920 [39:27<1:44:58, 1.77s/it, tok/s=14083.5]
Inference: 40%|███▉ | 2353/5920 [39:27<1:44:58, 1.77s/it, tok/s=14088.6]
Inference: 40%|███▉ | 2354/5920 [39:29<1:52:42, 1.90s/it, tok/s=14088.6]
Inference: 40%|███▉ | 2354/5920 [39:29<1:52:42, 1.90s/it, tok/s=14087.0]
Inference: 40%|███▉ | 2355/5920 [39:32<2:01:07, 2.04s/it, tok/s=14087.0]
Inference: 40%|███▉ | 2355/5920 [39:32<2:01:07, 2.04s/it, tok/s=14073.2]
Inference: 40%|███▉ | 2356/5920 [39:33<1:48:22, 1.82s/it, tok/s=14073.2]
Inference: 40%|███▉ | 2356/5920 [39:33<1:48:22, 1.82s/it, tok/s=14071.0]
Inference: 40%|███▉ | 2357/5920 [39:34<1:37:53, 1.65s/it, tok/s=14071.0]
Inference: 40%|███▉ | 2357/5920 [39:34<1:37:53, 1.65s/it, tok/s=14068.2]
Inference: 40%|███▉ | 2358/5920 [39:36<1:40:20, 1.69s/it, tok/s=14068.2]
Inference: 40%|███▉ | 2358/5920 [39:36<1:40:20, 1.69s/it, tok/s=14058.2]
Inference: 40%|███▉ | 2359/5920 [39:38<1:37:02, 1.64s/it, tok/s=14058.2]
Inference: 40%|███▉ | 2359/5920 [39:38<1:37:02, 1.64s/it, tok/s=14058.7]
Inference: 40%|███▉ | 2360/5920 [39:38<1:37:00, 1.64s/it, tok/s=14061.9]
Inference: 40%|███▉ | 2361/5920 [39:39<1:09:40, 1.17s/it, tok/s=14061.9]
Inference: 40%|███▉ | 2361/5920 [39:39<1:09:40, 1.17s/it, tok/s=14059.5]
Inference: 40%|███▉ | 2362/5920 [39:39<55:15, 1.07it/s, tok/s=14059.5]
Inference: 40%|███▉ | 2362/5920 [39:39<55:15, 1.07it/s, tok/s=14069.2]
Inference: 40%|███▉ | 2363/5920 [39:39<55:14, 1.07it/s, tok/s=14077.3]
Inference: 40%|███▉ | 2364/5920 [39:41<58:10, 1.02it/s, tok/s=14077.3]
Inference: 40%|███▉ | 2364/5920 [39:41<58:10, 1.02it/s, tok/s=14075.2]
Inference: 40%|███▉ | 2365/5920 [39:42<56:03, 1.06it/s, tok/s=14075.2]
Inference: 40%|███▉ | 2365/5920 [39:42<56:03, 1.06it/s, tok/s=14077.8]
Inference: 40%|███▉ | 2366/5920 [39:44<1:09:54, 1.18s/it, tok/s=14077.8]
Inference: 40%|███▉ | 2366/5920 [39:44<1:09:54, 1.18s/it, tok/s=14067.2]
Inference: 40%|███▉ | 2367/5920 [39:45<1:15:45, 1.28s/it, tok/s=14067.2]
Inference: 40%|███▉ | 2367/5920 [39:45<1:15:45, 1.28s/it, tok/s=14066.9]
Inference: 40%|████ | 2368/5920 [39:46<59:42, 1.01s/it, tok/s=14066.9]
Inference: 40%|████ | 2368/5920 [39:46<59:42, 1.01s/it, tok/s=14066.1]
Inference: 40%|████ | 2369/5920 [39:46<56:36, 1.05it/s, tok/s=14066.1]
Inference: 40%|████ | 2369/5920 [39:46<56:36, 1.05it/s, tok/s=14065.8]
Inference: 40%|████ | 2370/5920 [39:47<56:35, 1.05it/s, tok/s=14065.9]
Inference: 40%|████ | 2371/5920 [39:47<34:40, 1.71it/s, tok/s=14065.9]
Inference: 40%|████ | 2371/5920 [39:47<34:40, 1.71it/s, tok/s=14074.7]
Inference: 40%|████ | 2372/5920 [39:47<30:36, 1.93it/s, tok/s=14074.7]
Inference: 40%|████ | 2372/5920 [39:47<30:36, 1.93it/s, tok/s=14073.6]
Inference: 40%|████ | 2373/5920 [39:48<31:15, 1.89it/s, tok/s=14073.6]
Inference: 40%|████ | 2373/5920 [39:48<31:15, 1.89it/s, tok/s=14077.5]
Inference: 40%|████ | 2374/5920 [39:50<1:02:17, 1.05s/it, tok/s=14077.5]
Inference: 40%|████ | 2374/5920 [39:50<1:02:17, 1.05s/it, tok/s=14073.1]
Inference: 40%|████ | 2375/5920 [39:51<1:06:38, 1.13s/it, tok/s=14073.1]
Inference: 40%|████ | 2375/5920 [39:51<1:06:38, 1.13s/it, tok/s=14070.3]
Inference: 40%|████ | 2376/5920 [39:51<1:06:37, 1.13s/it, tok/s=14075.3]
Inference: 40%|████ | 2377/5920 [39:52<39:41, 1.49it/s, tok/s=14075.3]
Inference: 40%|████ | 2377/5920 [39:52<39:41, 1.49it/s, tok/s=14081.0]
Inference: 40%|████ | 2378/5920 [39:55<1:22:24, 1.40s/it, tok/s=14081.0]
Inference: 40%|████ | 2378/5920 [39:55<1:22:24, 1.40s/it, tok/s=14069.7]
Inference: 40%|████ | 2379/5920 [39:55<1:04:14, 1.09s/it, tok/s=14069.7]
Inference: 40%|████ | 2379/5920 [39:55<1:04:14, 1.09s/it, tok/s=14075.2]
Inference: 40%|████ | 2380/5920 [39:56<49:01, 1.20it/s, tok/s=14075.2]
Inference: 40%|████ | 2380/5920 [39:56<49:01, 1.20it/s, tok/s=14078.7]
Inference: 40%|████ | 2381/5920 [39:56<37:52, 1.56it/s, tok/s=14078.7]
Inference: 40%|████ | 2381/5920 [39:56<37:52, 1.56it/s, tok/s=14078.7]
Inference: 40%|████ | 2382/5920 [39:56<37:52, 1.56it/s, tok/s=14079.3]
Inference: 40%|████ | 2383/5920 [39:56<37:51, 1.56it/s, tok/s=14079.9]
Inference: 40%|████ | 2384/5920 [39:56<19:24, 3.04it/s, tok/s=14079.9]
Inference: 40%|████ | 2384/5920 [39:56<19:24, 3.04it/s, tok/s=14089.5]
Inference: 40%|████ | 2385/5920 [39:57<22:34, 2.61it/s, tok/s=14089.5]
Inference: 40%|████ | 2385/5920 [39:57<22:34, 2.61it/s, tok/s=14095.9]
Inference: 40%|████ | 2386/5920 [39:58<32:54, 1.79it/s, tok/s=14095.9]
Inference: 40%|████ | 2386/5920 [39:58<32:54, 1.79it/s, tok/s=14094.2]
Inference: 40%|████ | 2387/5920 [39:58<35:02, 1.68it/s, tok/s=14094.2]
Inference: 40%|████ | 2387/5920 [39:58<35:02, 1.68it/s, tok/s=14101.2]
Inference: 40%|████ | 2388/5920 [39:59<40:26, 1.46it/s, tok/s=14101.2]
Inference: 40%|████ | 2388/5920 [39:59<40:26, 1.46it/s, tok/s=14108.2]
Inference: 40%|████ | 2389/5920 [40:01<49:01, 1.20it/s, tok/s=14108.2]
Inference: 40%|████ | 2389/5920 [40:01<49:01, 1.20it/s, tok/s=14111.4]
Inference: 40%|████ | 2390/5920 [40:01<41:49, 1.41it/s, tok/s=14111.4]
Inference: 40%|████ | 2390/5920 [40:01<41:49, 1.41it/s, tok/s=14120.9]
Inference: 40%|████ | 2391/5920 [40:01<41:48, 1.41it/s, tok/s=14130.4]
Inference: 40%|████ | 2392/5920 [40:01<27:57, 2.10it/s, tok/s=14130.4]
Inference: 40%|████ | 2392/5920 [40:01<27:57, 2.10it/s, tok/s=14129.5]
Inference: 40%|████ | 2393/5920 [40:01<27:57, 2.10it/s, tok/s=14139.9]
Inference: 40%|████ | 2394/5920 [40:02<20:27, 2.87it/s, tok/s=14139.9]
Inference: 40%|████ | 2394/5920 [40:02<20:27, 2.87it/s, tok/s=14144.6]
Inference: 40%|████ | 2395/5920 [40:03<30:54, 1.90it/s, tok/s=14144.6]
Inference: 40%|████ | 2395/5920 [40:03<30:54, 1.90it/s, tok/s=14144.2]
Inference: 40%|████ | 2396/5920 [40:03<29:54, 1.96it/s, tok/s=14144.2]
Inference: 40%|████ | 2396/5920 [40:03<29:54, 1.96it/s, tok/s=14151.8]
Inference: 40%|████ | 2397/5920 [40:04<36:52, 1.59it/s, tok/s=14151.8]
Inference: 40%|████ | 2397/5920 [40:04<36:52, 1.59it/s, tok/s=14146.6]
Inference: 41%|████ | 2398/5920 [40:04<36:51, 1.59it/s, tok/s=14155.8]
Inference: 41%|████ | 2399/5920 [40:06<43:20, 1.35it/s, tok/s=14155.8]
Inference: 41%|████ | 2399/5920 [40:06<43:20, 1.35it/s, tok/s=14151.0]
Inference: 41%|████ | 2400/5920 [40:06<43:19, 1.35it/s, tok/s=14161.1]
Inference: 41%|████ | 2401/5920 [40:07<40:39, 1.44it/s, tok/s=14161.1]
Inference: 41%|████ | 2401/5920 [40:07<40:39, 1.44it/s, tok/s=14165.3]
Inference: 41%|████ | 2402/5920 [40:07<40:38, 1.44it/s, tok/s=14165.6]
Inference: 41%|████ | 2403/5920 [40:11<1:09:18, 1.18s/it, tok/s=14165.6]
Inference: 41%|████ | 2403/5920 [40:11<1:09:18, 1.18s/it, tok/s=14151.4]
Inference: 41%|████ | 2404/5920 [40:14<1:23:14, 1.42s/it, tok/s=14151.4]
Inference: 41%|████ | 2404/5920 [40:14<1:23:14, 1.42s/it, tok/s=14144.6]
Inference: 41%|████ | 2405/5920 [40:14<1:13:19, 1.25s/it, tok/s=14144.6]
Inference: 41%|████ | 2405/5920 [40:14<1:13:19, 1.25s/it, tok/s=14151.4]
Inference: 41%|████ | 2406/5920 [40:17<1:26:16, 1.47s/it, tok/s=14151.4]
Inference: 41%|████ | 2406/5920 [40:17<1:26:16, 1.47s/it, tok/s=14151.2]
Inference: 41%|████ | 2407/5920 [40:17<1:07:08, 1.15s/it, tok/s=14151.2]
Inference: 41%|████ | 2407/5920 [40:17<1:07:08, 1.15s/it, tok/s=14150.9]
Inference: 41%|████ | 2408/5920 [40:17<56:18, 1.04it/s, tok/s=14150.9]
Inference: 41%|████ | 2408/5920 [40:17<56:18, 1.04it/s, tok/s=14159.2]
Inference: 41%|████ | 2409/5920 [40:17<44:03, 1.33it/s, tok/s=14159.2]
Inference: 41%|████ | 2409/5920 [40:17<44:03, 1.33it/s, tok/s=14162.5]
Inference: 41%|████ | 2410/5920 [40:18<44:49, 1.31it/s, tok/s=14162.5]
Inference: 41%|████ | 2410/5920 [40:18<44:49, 1.31it/s, tok/s=14158.3]
Inference: 41%|████ | 2411/5920 [40:20<56:22, 1.04it/s, tok/s=14158.3]
Inference: 41%|████ | 2411/5920 [40:20<56:22, 1.04it/s, tok/s=14158.2]
Inference: 41%|████ | 2412/5920 [40:21<52:58, 1.10it/s, tok/s=14158.2]
Inference: 41%|████ | 2412/5920 [40:21<52:58, 1.10it/s, tok/s=14165.0]
Inference: 41%|████ | 2413/5920 [40:21<51:33, 1.13it/s, tok/s=14165.0]
Inference: 41%|████ | 2413/5920 [40:21<51:33, 1.13it/s, tok/s=14166.5]
Inference: 41%|████ | 2414/5920 [40:23<1:13:34, 1.26s/it, tok/s=14166.5]
Inference: 41%|████ | 2414/5920 [40:23<1:13:34, 1.26s/it, tok/s=14154.6]
Inference: 41%|████ | 2415/5920 [40:24<1:04:50, 1.11s/it, tok/s=14154.6]
Inference: 41%|████ | 2415/5920 [40:24<1:04:50, 1.11s/it, tok/s=14159.9]
Inference: 41%|████ | 2416/5920 [40:25<52:09, 1.12it/s, tok/s=14159.9]
Inference: 41%|████ | 2416/5920 [40:25<52:09, 1.12it/s, tok/s=14161.9]
Inference: 41%|████ | 2417/5920 [40:25<48:52, 1.19it/s, tok/s=14161.9]
Inference: 41%|████ | 2417/5920 [40:25<48:52, 1.19it/s, tok/s=14162.3]
Inference: 41%|████ | 2418/5920 [40:26<44:44, 1.30it/s, tok/s=14162.3]
Inference: 41%|████ | 2418/5920 [40:26<44:44, 1.30it/s, tok/s=14159.4]
Inference: 41%|████ | 2419/5920 [40:26<36:25, 1.60it/s, tok/s=14159.4]
Inference: 41%|████ | 2419/5920 [40:26<36:25, 1.60it/s, tok/s=14158.2]
Inference: 41%|████ | 2420/5920 [40:28<57:47, 1.01it/s, tok/s=14158.2]
Inference: 41%|████ | 2420/5920 [40:28<57:47, 1.01it/s, tok/s=14158.3]
Inference: 41%|████ | 2421/5920 [40:28<42:42, 1.37it/s, tok/s=14158.3]
Inference: 41%|████ | 2421/5920 [40:28<42:42, 1.37it/s, tok/s=14157.9]
Inference: 41%|████ | 2422/5920 [40:28<33:44, 1.73it/s, tok/s=14157.9]
Inference: 41%|████ | 2422/5920 [40:28<33:44, 1.73it/s, tok/s=14162.6]
Inference: 41%|████ | 2423/5920 [40:29<34:15, 1.70it/s, tok/s=14162.6]
Inference: 41%|████ | 2423/5920 [40:29<34:15, 1.70it/s, tok/s=14159.6]
Inference: 41%|████ | 2424/5920 [40:30<36:03, 1.62it/s, tok/s=14159.6]
Inference: 41%|████ | 2424/5920 [40:30<36:03, 1.62it/s, tok/s=14156.3]
Inference: 41%|████ | 2425/5920 [40:30<34:40, 1.68it/s, tok/s=14156.3]
Inference: 41%|████ | 2425/5920 [40:30<34:40, 1.68it/s, tok/s=14157.0]
Inference: 41%|████ | 2426/5920 [40:31<29:54, 1.95it/s, tok/s=14157.0]
Inference: 41%|████ | 2426/5920 [40:31<29:54, 1.95it/s, tok/s=14156.3]
Inference: 41%|████ | 2427/5920 [40:31<29:53, 1.95it/s, tok/s=14161.7]
Inference: 41%|████ | 2428/5920 [40:32<31:25, 1.85it/s, tok/s=14161.7]
Inference: 41%|████ | 2428/5920 [40:32<31:25, 1.85it/s, tok/s=14166.6]
Inference: 41%|████ | 2429/5920 [40:33<48:34, 1.20it/s, tok/s=14166.6]
Inference: 41%|████ | 2429/5920 [40:33<48:34, 1.20it/s, tok/s=14167.6]
Inference: 41%|████ | 2430/5920 [40:34<50:38, 1.15it/s, tok/s=14167.6]
Inference: 41%|████ | 2430/5920 [40:34<50:38, 1.15it/s, tok/s=14167.2]
Inference: 41%|████ | 2431/5920 [40:35<39:43, 1.46it/s, tok/s=14167.2]
Inference: 41%|████ | 2431/5920 [40:35<39:43, 1.46it/s, tok/s=14171.5]
Inference: 41%|████ | 2432/5920 [40:35<32:54, 1.77it/s, tok/s=14171.5]
Inference: 41%|████ | 2432/5920 [40:35<32:54, 1.77it/s, tok/s=14175.0]
Inference: 41%|████ | 2433/5920 [40:35<26:22, 2.20it/s, tok/s=14175.0]
Inference: 41%|████ | 2433/5920 [40:35<26:22, 2.20it/s, tok/s=14181.0]
Inference: 41%|████ | 2434/5920 [40:36<37:12, 1.56it/s, tok/s=14181.0]
Inference: 41%|████ | 2434/5920 [40:36<37:12, 1.56it/s, tok/s=14186.7]
Inference: 41%|████ | 2435/5920 [40:37<39:03, 1.49it/s, tok/s=14186.7]
Inference: 41%|████ | 2435/5920 [40:37<39:03, 1.49it/s, tok/s=14192.2]
Inference: 41%|████ | 2436/5920 [40:38<43:55, 1.32it/s, tok/s=14192.2]
Inference: 41%|████ | 2436/5920 [40:38<43:55, 1.32it/s, tok/s=14196.8]
Inference: 41%|████ | 2437/5920 [40:39<56:34, 1.03it/s, tok/s=14196.8]
Inference: 41%|████ | 2437/5920 [40:39<56:34, 1.03it/s, tok/s=14194.8]
Inference: 41%|████ | 2438/5920 [40:40<46:12, 1.26it/s, tok/s=14194.8]
Inference: 41%|████ | 2438/5920 [40:40<46:12, 1.26it/s, tok/s=14203.9]
Inference: 41%|████ | 2439/5920 [40:40<45:14, 1.28it/s, tok/s=14203.9]
Inference: 41%|████ | 2439/5920 [40:40<45:14, 1.28it/s, tok/s=14208.0]
Inference: 41%|████ | 2440/5920 [40:41<36:16, 1.60it/s, tok/s=14208.0]
Inference: 41%|████ | 2440/5920 [40:41<36:16, 1.60it/s, tok/s=14214.8]
Inference: 41%|████ | 2441/5920 [40:41<36:16, 1.60it/s, tok/s=14215.3]
Inference: 41%|████▏ | 2442/5920 [40:41<22:40, 2.56it/s, tok/s=14215.3]
Inference: 41%|████▏ | 2442/5920 [40:41<22:40, 2.56it/s, tok/s=14225.5]
Inference: 41%|████▏ | 2443/5920 [40:44<1:02:27, 1.08s/it, tok/s=14225.5]
Inference: 41%|████▏ | 2443/5920 [40:44<1:02:27, 1.08s/it, tok/s=14212.3]
Inference: 41%|████▏ | 2444/5920 [40:47<1:24:14, 1.45s/it, tok/s=14212.3]
Inference: 41%|████▏ | 2444/5920 [40:47<1:24:14, 1.45s/it, tok/s=14198.1]
Inference: 41%|████▏ | 2445/5920 [40:52<2:32:01, 2.62s/it, tok/s=14198.1]
Inference: 41%|████▏ | 2445/5920 [40:52<2:32:01, 2.62s/it, tok/s=14175.9]
Inference: 41%|████▏ | 2446/5920 [40:53<2:03:26, 2.13s/it, tok/s=14175.9]
Inference: 41%|████▏ | 2446/5920 [40:53<2:03:26, 2.13s/it, tok/s=14181.7]
Inference: 41%|████▏ | 2447/5920 [40:54<1:42:37, 1.77s/it, tok/s=14181.7]
Inference: 41%|████▏ | 2447/5920 [40:54<1:42:37, 1.77s/it, tok/s=14177.2]
Inference: 41%|████▏ | 2448/5920 [40:59<2:27:07, 2.54s/it, tok/s=14177.2]
Inference: 41%|████▏ | 2448/5920 [40:59<2:27:07, 2.54s/it, tok/s=14163.0]
Inference: 41%|████▏ | 2449/5920 [41:00<2:15:09, 2.34s/it, tok/s=14163.0]
Inference: 41%|████▏ | 2449/5920 [41:00<2:15:09, 2.34s/it, tok/s=14158.2]
Inference: 41%|████▏ | 2450/5920 [41:02<1:57:34, 2.03s/it, tok/s=14158.2]
Inference: 41%|████▏ | 2450/5920 [41:02<1:57:34, 2.03s/it, tok/s=14159.3]
Inference: 41%|████▏ | 2451/5920 [41:04<1:53:28, 1.96s/it, tok/s=14159.3]
Inference: 41%|████▏ | 2451/5920 [41:04<1:53:28, 1.96s/it, tok/s=14159.2]
Inference: 41%|████▏ | 2452/5920 [41:04<1:31:49, 1.59s/it, tok/s=14159.2]
Inference: 41%|████▏ | 2452/5920 [41:04<1:31:49, 1.59s/it, tok/s=14167.4]
Inference: 41%|████▏ | 2453/5920 [41:06<1:43:19, 1.79s/it, tok/s=14167.4]
Inference: 41%|████▏ | 2453/5920 [41:06<1:43:19, 1.79s/it, tok/s=14156.0]
Inference: 41%|████▏ | 2454/5920 [41:07<1:23:57, 1.45s/it, tok/s=14156.0]
Inference: 41%|████▏ | 2454/5920 [41:07<1:23:57, 1.45s/it, tok/s=14157.8]
Inference: 41%|████▏ | 2455/5920 [41:07<1:23:55, 1.45s/it, tok/s=14166.2]
Inference: 41%|████▏ | 2456/5920 [41:08<1:01:46, 1.07s/it, tok/s=14166.2]
Inference: 41%|████▏ | 2456/5920 [41:08<1:01:46, 1.07s/it, tok/s=14160.2]
Inference: 42%|████▏ | 2457/5920 [41:08<1:01:45, 1.07s/it, tok/s=14160.4]
Inference: 42%|████▏ | 2458/5920 [41:11<1:05:43, 1.14s/it, tok/s=14160.4]
Inference: 42%|████▏ | 2458/5920 [41:11<1:05:43, 1.14s/it, tok/s=14156.9]
Inference: 42%|████▏ | 2459/5920 [41:11<1:05:42, 1.14s/it, tok/s=14168.8]
Inference: 42%|████▏ | 2460/5920 [41:11<45:39, 1.26it/s, tok/s=14168.8]
Inference: 42%|████▏ | 2460/5920 [41:11<45:39, 1.26it/s, tok/s=14179.3]
Inference: 42%|████▏ | 2461/5920 [41:11<38:46, 1.49it/s, tok/s=14179.3]
Inference: 42%|████▏ | 2461/5920 [41:11<38:46, 1.49it/s, tok/s=14178.7]
Inference: 42%|████▏ | 2462/5920 [41:12<36:34, 1.58it/s, tok/s=14178.7]
Inference: 42%|████▏ | 2462/5920 [41:12<36:34, 1.58it/s, tok/s=14177.6]
Inference: 42%|████▏ | 2463/5920 [41:12<36:33, 1.58it/s, tok/s=14187.8]
Inference: 42%|████▏ | 2464/5920 [41:13<30:47, 1.87it/s, tok/s=14187.8]
Inference: 42%|████▏ | 2464/5920 [41:13<30:47, 1.87it/s, tok/s=14194.6]
Inference: 42%|████▏ | 2465/5920 [41:14<34:30, 1.67it/s, tok/s=14194.6]
Inference: 42%|████▏ | 2465/5920 [41:14<34:30, 1.67it/s, tok/s=14190.3]
Inference: 42%|████▏ | 2466/5920 [41:14<28:50, 2.00it/s, tok/s=14190.3]
Inference: 42%|████▏ | 2466/5920 [41:14<28:50, 2.00it/s, tok/s=14190.0]
Inference: 42%|████▏ | 2467/5920 [41:14<30:41, 1.87it/s, tok/s=14190.0]
Inference: 42%|████▏ | 2467/5920 [41:14<30:41, 1.87it/s, tok/s=14186.9]
Inference: 42%|████▏ | 2468/5920 [41:15<27:44, 2.07it/s, tok/s=14186.9]
Inference: 42%|████▏ | 2468/5920 [41:15<27:44, 2.07it/s, tok/s=14196.6]
Inference: 42%|████▏ | 2469/5920 [41:15<27:44, 2.07it/s, tok/s=14207.8]
Inference: 42%|████▏ | 2470/5920 [41:15<27:43, 2.07it/s, tok/s=14208.9]
Inference: 42%|████▏ | 2471/5920 [41:15<15:22, 3.74it/s, tok/s=14208.9]
Inference: 42%|████▏ | 2471/5920 [41:15<15:22, 3.74it/s, tok/s=14216.4]
Inference: 42%|████▏ | 2472/5920 [41:16<26:52, 2.14it/s, tok/s=14216.4]
Inference: 42%|████▏ | 2472/5920 [41:16<26:52, 2.14it/s, tok/s=14209.6]
Inference: 42%|████▏ | 2473/5920 [41:16<22:35, 2.54it/s, tok/s=14209.6]
Inference: 42%|████▏ | 2473/5920 [41:16<22:35, 2.54it/s, tok/s=14209.4]
Inference: 42%|████▏ | 2474/5920 [41:17<31:24, 1.83it/s, tok/s=14209.4]
Inference: 42%|████▏ | 2474/5920 [41:17<31:24, 1.83it/s, tok/s=14205.0]
Inference: 42%|████▏ | 2475/5920 [41:19<43:47, 1.31it/s, tok/s=14205.0]
Inference: 42%|████▏ | 2475/5920 [41:19<43:47, 1.31it/s, tok/s=14209.3]
Inference: 42%|████▏ | 2476/5920 [41:19<34:34, 1.66it/s, tok/s=14209.3]
Inference: 42%|████▏ | 2476/5920 [41:19<34:34, 1.66it/s, tok/s=14209.1]
Inference: 42%|████▏ | 2477/5920 [41:20<40:26, 1.42it/s, tok/s=14209.1]
Inference: 42%|████▏ | 2477/5920 [41:20<40:26, 1.42it/s, tok/s=14204.7]
Inference: 42%|████▏ | 2478/5920 [41:20<34:00, 1.69it/s, tok/s=14204.7]
Inference: 42%|████▏ | 2478/5920 [41:20<34:00, 1.69it/s, tok/s=14203.8]
Inference: 42%|████▏ | 2479/5920 [41:20<27:07, 2.11it/s, tok/s=14203.8]
Inference: 42%|████▏ | 2479/5920 [41:20<27:07, 2.11it/s, tok/s=14214.8]
Inference: 42%|████▏ | 2480/5920 [41:20<27:06, 2.11it/s, tok/s=14218.9]
Inference: 42%|████▏ | 2481/5920 [41:21<18:32, 3.09it/s, tok/s=14218.9]
Inference: 42%|████▏ | 2481/5920 [41:21<18:32, 3.09it/s, tok/s=14227.0]
Inference: 42%|████▏ | 2482/5920 [41:21<23:38, 2.42it/s, tok/s=14227.0]
Inference: 42%|████▏ | 2482/5920 [41:21<23:38, 2.42it/s, tok/s=14223.7]
Inference: 42%|████▏ | 2483/5920 [41:22<20:49, 2.75it/s, tok/s=14223.7]
Inference: 42%|████▏ | 2483/5920 [41:22<20:49, 2.75it/s, tok/s=14233.7]
Inference: 42%|████▏ | 2484/5920 [41:22<19:49, 2.89it/s, tok/s=14233.7]
Inference: 42%|████▏ | 2484/5920 [41:22<19:49, 2.89it/s, tok/s=14237.7]
Inference: 42%|████▏ | 2485/5920 [41:23<37:43, 1.52it/s, tok/s=14237.7]
Inference: 42%|████▏ | 2485/5920 [41:23<37:43, 1.52it/s, tok/s=14237.4]
Inference: 42%|████▏ | 2486/5920 [41:24<42:13, 1.36it/s, tok/s=14237.4]
Inference: 42%|████▏ | 2486/5920 [41:24<42:13, 1.36it/s, tok/s=14232.7]
Inference: 42%|████▏ | 2487/5920 [41:24<42:12, 1.36it/s, tok/s=14244.3]
Inference: 42%|████▏ | 2488/5920 [41:25<26:34, 2.15it/s, tok/s=14244.3]
Inference: 42%|████▏ | 2488/5920 [41:25<26:34, 2.15it/s, tok/s=14249.9]
Inference: 42%|████▏ | 2489/5920 [41:26<36:17, 1.58it/s, tok/s=14249.9]
Inference: 42%|████▏ | 2489/5920 [41:26<36:17, 1.58it/s, tok/s=14244.1]
Inference: 42%|████▏ | 2490/5920 [41:26<36:17, 1.58it/s, tok/s=14254.0]
Inference: 42%|████▏ | 2491/5920 [41:26<25:00, 2.29it/s, tok/s=14254.0]
Inference: 42%|████▏ | 2491/5920 [41:26<25:00, 2.29it/s, tok/s=14264.8]
Inference: 42%|████▏ | 2492/5920 [41:27<26:30, 2.16it/s, tok/s=14264.8]
Inference: 42%|████▏ | 2492/5920 [41:27<26:30, 2.16it/s, tok/s=14271.2]
Inference: 42%|████▏ | 2493/5920 [41:27<32:49, 1.74it/s, tok/s=14271.2]
Inference: 42%|████▏ | 2493/5920 [41:27<32:49, 1.74it/s, tok/s=14275.3]
Inference: 42%|████▏ | 2494/5920 [41:27<32:48, 1.74it/s, tok/s=14276.0]
Inference: 42%|████▏ | 2495/5920 [41:28<21:55, 2.60it/s, tok/s=14276.0]
Inference: 42%|████▏ | 2495/5920 [41:28<21:55, 2.60it/s, tok/s=14280.6]
Inference: 42%|████▏ | 2496/5920 [41:29<29:50, 1.91it/s, tok/s=14280.6]
Inference: 42%|████▏ | 2496/5920 [41:29<29:50, 1.91it/s, tok/s=14282.5]
Inference: 42%|████▏ | 2497/5920 [41:29<32:49, 1.74it/s, tok/s=14282.5]
Inference: 42%|████▏ | 2497/5920 [41:29<32:49, 1.74it/s, tok/s=14278.9]
Inference: 42%|████▏ | 2498/5920 [41:31<42:48, 1.33it/s, tok/s=14278.9]
Inference: 42%|████▏ | 2498/5920 [41:31<42:48, 1.33it/s, tok/s=14283.8]
Inference: 42%|████▏ | 2499/5920 [41:32<44:04, 1.29it/s, tok/s=14283.8]
Inference: 42%|████▏ | 2499/5920 [41:32<44:04, 1.29it/s, tok/s=14290.1]
Inference: 42%|████▏ | 2500/5920 [41:32<43:37, 1.31it/s, tok/s=14290.1]
Inference: 42%|████▏ | 2500/5920 [41:32<43:37, 1.31it/s, tok/s=14289.9]
Inference: 42%|████▏ | 2501/5920 [41:33<38:07, 1.49it/s, tok/s=14289.9]
Inference: 42%|████▏ | 2501/5920 [41:33<38:07, 1.49it/s, tok/s=14295.6]
Inference: 42%|████▏ | 2502/5920 [41:33<30:50, 1.85it/s, tok/s=14295.6]
Inference: 42%|████▏ | 2502/5920 [41:33<30:50, 1.85it/s, tok/s=14306.4]
Inference: 42%|████▏ | 2503/5920 [41:33<27:16, 2.09it/s, tok/s=14306.4]
Inference: 42%|████▏ | 2503/5920 [41:33<27:16, 2.09it/s, tok/s=14313.6]
Inference: 42%|████▏ | 2504/5920 [41:35<47:10, 1.21it/s, tok/s=14313.6]
Inference: 42%|████▏ | 2504/5920 [41:35<47:10, 1.21it/s, tok/s=14305.1]
Inference: 42%|████▏ | 2505/5920 [41:35<36:29, 1.56it/s, tok/s=14305.1]
Inference: 42%|████▏ | 2505/5920 [41:35<36:29, 1.56it/s, tok/s=14316.1]
Inference: 42%|████▏ | 2506/5920 [41:36<35:16, 1.61it/s, tok/s=14316.1]
Inference: 42%|████▏ | 2506/5920 [41:36<35:16, 1.61it/s, tok/s=14324.9]
Inference: 42%|████▏ | 2507/5920 [41:36<33:55, 1.68it/s, tok/s=14324.9]
Inference: 42%|████▏ | 2507/5920 [41:36<33:55, 1.68it/s, tok/s=14333.6]
Inference: 42%|████▏ | 2508/5920 [41:37<33:04, 1.72it/s, tok/s=14333.6]
Inference: 42%|████▏ | 2508/5920 [41:37<33:04, 1.72it/s, tok/s=14341.8]
Inference: 42%|████▏ | 2509/5920 [41:40<1:15:29, 1.33s/it, tok/s=14341.8]
Inference: 42%|████▏ | 2509/5920 [41:40<1:15:29, 1.33s/it, tok/s=14330.0]
Inference: 42%|████▏ | 2510/5920 [41:40<56:04, 1.01it/s, tok/s=14330.0]
Inference: 42%|████▏ | 2510/5920 [41:40<56:04, 1.01it/s, tok/s=14337.0]
Inference: 42%|████▏ | 2511/5920 [41:40<44:17, 1.28it/s, tok/s=14337.0]
Inference: 42%|████▏ | 2511/5920 [41:40<44:17, 1.28it/s, tok/s=14344.2]
Inference: 42%|████▏ | 2512/5920 [41:41<45:32, 1.25it/s, tok/s=14344.2]
Inference: 42%|████▏ | 2512/5920 [41:41<45:32, 1.25it/s, tok/s=14351.3]
Inference: 42%|████▏ | 2513/5920 [41:41<45:31, 1.25it/s, tok/s=14351.9]
Inference: 42%|████▏ | 2514/5920 [41:42<38:29, 1.47it/s, tok/s=14351.9]
Inference: 42%|████▏ | 2514/5920 [41:42<38:29, 1.47it/s, tok/s=14356.5]
Inference: 42%|████▏ | 2515/5920 [41:42<32:36, 1.74it/s, tok/s=14356.5]
Inference: 42%|████▏ | 2515/5920 [41:42<32:36, 1.74it/s, tok/s=14360.0]
Inference: 42%|████▎ | 2516/5920 [41:43<32:36, 1.74it/s, tok/s=14360.8]
Inference: 43%|████▎ | 2517/5920 [41:43<25:27, 2.23it/s, tok/s=14360.8]
Inference: 43%|████▎ | 2517/5920 [41:43<25:27, 2.23it/s, tok/s=14359.0]
Inference: 43%|████▎ | 2518/5920 [41:43<23:21, 2.43it/s, tok/s=14359.0]
Inference: 43%|████▎ | 2518/5920 [41:43<23:21, 2.43it/s, tok/s=14363.1]
Inference: 43%|████▎ | 2519/5920 [41:45<45:09, 1.26it/s, tok/s=14363.1]
Inference: 43%|████▎ | 2519/5920 [41:45<45:09, 1.26it/s, tok/s=14353.7]
Inference: 43%|████▎ | 2520/5920 [41:45<45:08, 1.26it/s, tok/s=14357.6]
Inference: 43%|████▎ | 2521/5920 [41:47<46:37, 1.22it/s, tok/s=14357.6]
Inference: 43%|████▎ | 2521/5920 [41:47<46:37, 1.22it/s, tok/s=14349.5]
Inference: 43%|████▎ | 2522/5920 [41:47<41:38, 1.36it/s, tok/s=14349.5]
Inference: 43%|████▎ | 2522/5920 [41:47<41:38, 1.36it/s, tok/s=14347.4]
Inference: 43%|████▎ | 2523/5920 [41:48<40:01, 1.41it/s, tok/s=14347.4]
Inference: 43%|████▎ | 2523/5920 [41:48<40:01, 1.41it/s, tok/s=14344.6]
Inference: 43%|████▎ | 2524/5920 [41:48<32:50, 1.72it/s, tok/s=14344.6]
Inference: 43%|████▎ | 2524/5920 [41:48<32:50, 1.72it/s, tok/s=14355.0]
Inference: 43%|████▎ | 2525/5920 [41:50<42:21, 1.34it/s, tok/s=14355.0]
Inference: 43%|████▎ | 2525/5920 [41:50<42:21, 1.34it/s, tok/s=14348.7]
Inference: 43%|████▎ | 2526/5920 [41:50<42:20, 1.34it/s, tok/s=14360.6]
Inference: 43%|████▎ | 2527/5920 [41:50<27:45, 2.04it/s, tok/s=14360.6]
Inference: 43%|████▎ | 2527/5920 [41:50<27:45, 2.04it/s, tok/s=14359.4]
Inference: 43%|████▎ | 2528/5920 [41:50<27:44, 2.04it/s, tok/s=14370.9]
Inference: 43%|████▎ | 2529/5920 [41:50<23:27, 2.41it/s, tok/s=14370.9]
Inference: 43%|████▎ | 2529/5920 [41:50<23:27, 2.41it/s, tok/s=14368.4]
Inference: 43%|████▎ | 2530/5920 [41:51<23:43, 2.38it/s, tok/s=14368.4]
Inference: 43%|████▎ | 2530/5920 [41:51<23:43, 2.38it/s, tok/s=14371.9]
Inference: 43%|████▎ | 2531/5920 [41:51<24:49, 2.28it/s, tok/s=14371.9]
Inference: 43%|████▎ | 2531/5920 [41:51<24:49, 2.28it/s, tok/s=14376.3]
Inference: 43%|████▎ | 2532/5920 [41:53<43:41, 1.29it/s, tok/s=14376.3]
Inference: 43%|████▎ | 2532/5920 [41:53<43:41, 1.29it/s, tok/s=14367.1]
Inference: 43%|████▎ | 2533/5920 [41:54<40:03, 1.41it/s, tok/s=14367.1]
Inference: 43%|████▎ | 2533/5920 [41:54<40:03, 1.41it/s, tok/s=14375.7]
Inference: 43%|████▎ | 2534/5920 [41:54<37:44, 1.50it/s, tok/s=14375.7]
Inference: 43%|████▎ | 2534/5920 [41:54<37:44, 1.50it/s, tok/s=14373.8]
Inference: 43%|████▎ | 2535/5920 [41:54<29:50, 1.89it/s, tok/s=14373.8]
Inference: 43%|████▎ | 2535/5920 [41:54<29:50, 1.89it/s, tok/s=14378.8]
Inference: 43%|████▎ | 2536/5920 [41:55<23:22, 2.41it/s, tok/s=14378.8]
Inference: 43%|████▎ | 2536/5920 [41:55<23:22, 2.41it/s, tok/s=14390.1]
Inference: 43%|████▎ | 2537/5920 [41:55<22:30, 2.50it/s, tok/s=14390.1]
Inference: 43%|████▎ | 2537/5920 [41:55<22:30, 2.50it/s, tok/s=14388.5]
Inference: 43%|████▎ | 2538/5920 [41:55<21:07, 2.67it/s, tok/s=14388.5]
Inference: 43%|████▎ | 2538/5920 [41:55<21:07, 2.67it/s, tok/s=14387.3]
Inference: 43%|████▎ | 2539/5920 [41:56<27:27, 2.05it/s, tok/s=14387.3]
Inference: 43%|████▎ | 2539/5920 [41:56<27:27, 2.05it/s, tok/s=14395.0]
Inference: 43%|████▎ | 2540/5920 [41:56<27:26, 2.05it/s, tok/s=14395.4]
Inference: 43%|████▎ | 2541/5920 [41:58<35:30, 1.59it/s, tok/s=14395.4]
Inference: 43%|████▎ | 2541/5920 [41:58<35:30, 1.59it/s, tok/s=14387.3]
Inference: 43%|████▎ | 2542/5920 [41:58<32:22, 1.74it/s, tok/s=14387.3]
Inference: 43%|████▎ | 2542/5920 [41:58<32:22, 1.74it/s, tok/s=14394.4]
Inference: 43%|████▎ | 2543/5920 [41:58<32:22, 1.74it/s, tok/s=14395.0]
Inference: 43%|████▎ | 2544/5920 [42:01<52:24, 1.07it/s, tok/s=14395.0]
Inference: 43%|████▎ | 2544/5920 [42:01<52:24, 1.07it/s, tok/s=14380.3]
Inference: 43%|████▎ | 2545/5920 [42:01<52:23, 1.07it/s, tok/s=14381.9]
Inference: 43%|████▎ | 2546/5920 [42:01<36:38, 1.53it/s, tok/s=14381.9]
Inference: 43%|████▎ | 2546/5920 [42:01<36:38, 1.53it/s, tok/s=14388.4]
Inference: 43%|████▎ | 2547/5920 [42:02<35:26, 1.59it/s, tok/s=14388.4]
Inference: 43%|████▎ | 2547/5920 [42:02<35:26, 1.59it/s, tok/s=14385.7]
Inference: 43%|████▎ | 2548/5920 [42:02<35:15, 1.59it/s, tok/s=14385.7]
Inference: 43%|████▎ | 2548/5920 [42:02<35:15, 1.59it/s, tok/s=14385.5]
Inference: 43%|████▎ | 2549/5920 [42:04<51:37, 1.09it/s, tok/s=14385.5]
Inference: 43%|████▎ | 2549/5920 [42:04<51:37, 1.09it/s, tok/s=14377.7]
Inference: 43%|████▎ | 2550/5920 [42:05<44:19, 1.27it/s, tok/s=14377.7]
Inference: 43%|████▎ | 2550/5920 [42:05<44:19, 1.27it/s, tok/s=14387.3]
Inference: 43%|████▎ | 2551/5920 [42:05<44:55, 1.25it/s, tok/s=14387.3]
Inference: 43%|████▎ | 2551/5920 [42:05<44:55, 1.25it/s, tok/s=14391.4]
Inference: 43%|████▎ | 2552/5920 [42:06<39:51, 1.41it/s, tok/s=14391.4]
Inference: 43%|████▎ | 2552/5920 [42:06<39:51, 1.41it/s, tok/s=14394.3]
Inference: 43%|████▎ | 2553/5920 [42:06<30:28, 1.84it/s, tok/s=14394.3]
Inference: 43%|████▎ | 2553/5920 [42:06<30:28, 1.84it/s, tok/s=14405.5]
Inference: 43%|████▎ | 2554/5920 [42:06<23:33, 2.38it/s, tok/s=14405.5]
Inference: 43%|████▎ | 2554/5920 [42:06<23:33, 2.38it/s, tok/s=14405.4]
Inference: 43%|████▎ | 2555/5920 [42:06<18:46, 2.99it/s, tok/s=14405.4]
Inference: 43%|████▎ | 2555/5920 [42:06<18:46, 2.99it/s, tok/s=14415.7]
Inference: 43%|████▎ | 2556/5920 [42:08<34:08, 1.64it/s, tok/s=14415.7]
Inference: 43%|████▎ | 2556/5920 [42:08<34:08, 1.64it/s, tok/s=14409.1]
Inference: 43%|████▎ | 2557/5920 [42:09<44:16, 1.27it/s, tok/s=14409.1]
Inference: 43%|████▎ | 2557/5920 [42:09<44:16, 1.27it/s, tok/s=14403.1]
Inference: 43%|████▎ | 2558/5920 [42:09<33:07, 1.69it/s, tok/s=14403.1]
Inference: 43%|████▎ | 2558/5920 [42:09<33:07, 1.69it/s, tok/s=14411.5]
Inference: 43%|████▎ | 2559/5920 [42:09<33:06, 1.69it/s, tok/s=14422.9]
Inference: 43%|████▎ | 2560/5920 [42:09<21:02, 2.66it/s, tok/s=14422.9]
Inference: 43%|████▎ | 2560/5920 [42:09<21:02, 2.66it/s, tok/s=14430.4]
Inference: 43%|████▎ | 2561/5920 [42:11<36:15, 1.54it/s, tok/s=14430.4]
Inference: 43%|████▎ | 2561/5920 [42:11<36:15, 1.54it/s, tok/s=14422.6]
Inference: 43%|████▎ | 2562/5920 [42:11<28:42, 1.95it/s, tok/s=14422.6]
Inference: 43%|████▎ | 2562/5920 [42:11<28:42, 1.95it/s, tok/s=14422.7]
Inference: 43%|████▎ | 2563/5920 [42:11<31:28, 1.78it/s, tok/s=14422.7]
Inference: 43%|████▎ | 2563/5920 [42:11<31:28, 1.78it/s, tok/s=14419.2]
Inference: 43%|████▎ | 2564/5920 [42:12<35:12, 1.59it/s, tok/s=14419.2]
Inference: 43%|████▎ | 2564/5920 [42:12<35:12, 1.59it/s, tok/s=14415.6]
Inference: 43%|████▎ | 2565/5920 [42:13<35:33, 1.57it/s, tok/s=14415.6]
Inference: 43%|████▎ | 2565/5920 [42:13<35:33, 1.57it/s, tok/s=14412.5]
Inference: 43%|████▎ | 2566/5920 [42:14<36:54, 1.51it/s, tok/s=14412.5]
Inference: 43%|████▎ | 2566/5920 [42:14<36:54, 1.51it/s, tok/s=14409.3]
Inference: 43%|████▎ | 2567/5920 [42:15<49:29, 1.13it/s, tok/s=14409.3]
Inference: 43%|████▎ | 2567/5920 [42:15<49:29, 1.13it/s, tok/s=14401.6]
Inference: 43%|████▎ | 2568/5920 [42:15<41:18, 1.35it/s, tok/s=14401.6]
Inference: 43%|████▎ | 2568/5920 [42:15<41:18, 1.35it/s, tok/s=14400.1]
Inference: 43%|████▎ | 2569/5920 [42:16<37:03, 1.51it/s, tok/s=14400.1]
Inference: 43%|████▎ | 2569/5920 [42:16<37:03, 1.51it/s, tok/s=14397.8]
Inference: 43%|████▎ | 2570/5920 [42:16<29:25, 1.90it/s, tok/s=14397.8]
Inference: 43%|████▎ | 2570/5920 [42:16<29:25, 1.90it/s, tok/s=14398.3]
Inference: 43%|████▎ | 2571/5920 [42:17<36:07, 1.55it/s, tok/s=14398.3]
Inference: 43%|████▎ | 2571/5920 [42:17<36:07, 1.55it/s, tok/s=14404.9]
Inference: 43%|████▎ | 2572/5920 [42:17<31:49, 1.75it/s, tok/s=14404.9]
Inference: 43%|████▎ | 2572/5920 [42:17<31:49, 1.75it/s, tok/s=14413.5]
Inference: 43%|████▎ | 2573/5920 [42:18<32:20, 1.72it/s, tok/s=14413.5]
Inference: 43%|████▎ | 2573/5920 [42:18<32:20, 1.72it/s, tok/s=14411.4]
Inference: 43%|████▎ | 2574/5920 [42:19<35:22, 1.58it/s, tok/s=14411.4]
Inference: 43%|████▎ | 2574/5920 [42:19<35:22, 1.58it/s, tok/s=14407.7]
Inference: 43%|████▎ | 2575/5920 [42:19<28:51, 1.93it/s, tok/s=14407.7]
Inference: 43%|████▎ | 2575/5920 [42:19<28:51, 1.93it/s, tok/s=14418.2]
Inference: 44%|████▎ | 2576/5920 [42:19<27:28, 2.03it/s, tok/s=14418.2]
Inference: 44%|████▎ | 2576/5920 [42:19<27:28, 2.03it/s, tok/s=14420.8]
Inference: 44%|████▎ | 2577/5920 [42:19<27:28, 2.03it/s, tok/s=14431.7]
Inference: 44%|████▎ | 2578/5920 [42:20<17:12, 3.24it/s, tok/s=14431.7]
Inference: 44%|████▎ | 2578/5920 [42:20<17:12, 3.24it/s, tok/s=14437.4]
Inference: 44%|████▎ | 2579/5920 [42:20<16:10, 3.44it/s, tok/s=14437.4]
Inference: 44%|████▎ | 2579/5920 [42:20<16:10, 3.44it/s, tok/s=14445.5]
Inference: 44%|████▎ | 2580/5920 [42:20<18:13, 3.05it/s, tok/s=14445.5]
Inference: 44%|████▎ | 2580/5920 [42:20<18:13, 3.05it/s, tok/s=14443.7]
Inference: 44%|████▎ | 2581/5920 [42:21<26:58, 2.06it/s, tok/s=14443.7]
Inference: 44%|████▎ | 2581/5920 [42:21<26:58, 2.06it/s, tok/s=14445.4]
Inference: 44%|████▎ | 2582/5920 [42:22<25:56, 2.14it/s, tok/s=14445.4]
Inference: 44%|████▎ | 2582/5920 [42:22<25:56, 2.14it/s, tok/s=14445.3]
Inference: 44%|████▎ | 2583/5920 [42:22<29:23, 1.89it/s, tok/s=14445.3]
Inference: 44%|████▎ | 2583/5920 [42:22<29:23, 1.89it/s, tok/s=14442.4]
Inference: 44%|████▎ | 2584/5920 [42:23<33:50, 1.64it/s, tok/s=14442.4]
Inference: 44%|████▎ | 2584/5920 [42:23<33:50, 1.64it/s, tok/s=14438.3]
Inference: 44%|████▎ | 2585/5920 [42:23<27:54, 1.99it/s, tok/s=14438.3]
Inference: 44%|████▎ | 2585/5920 [42:23<27:54, 1.99it/s, tok/s=14443.3]
Inference: 44%|████▎ | 2586/5920 [42:23<27:53, 1.99it/s, tok/s=14443.9]
Inference: 44%|████▎ | 2587/5920 [42:24<17:49, 3.12it/s, tok/s=14443.9]
Inference: 44%|████▎ | 2587/5920 [42:24<17:49, 3.12it/s, tok/s=14443.9]
Inference: 44%|████▎ | 2588/5920 [42:24<17:49, 3.12it/s, tok/s=14455.2]
Inference: 44%|████▎ | 2589/5920 [42:24<17:05, 3.25it/s, tok/s=14455.2]
Inference: 44%|████▎ | 2589/5920 [42:24<17:05, 3.25it/s, tok/s=14453.5]
Inference: 44%|████▍ | 2590/5920 [42:25<19:13, 2.89it/s, tok/s=14453.5]
Inference: 44%|████▍ | 2590/5920 [42:25<19:13, 2.89it/s, tok/s=14451.3]
Inference: 44%|████▍ | 2591/5920 [42:25<19:13, 2.89it/s, tok/s=14461.4]
Inference: 44%|████▍ | 2592/5920 [42:25<19:13, 2.89it/s, tok/s=14466.1]
Inference: 44%|████▍ | 2593/5920 [42:25<11:36, 4.78it/s, tok/s=14466.1]
Inference: 44%|████▍ | 2593/5920 [42:25<11:36, 4.78it/s, tok/s=14466.5]
Inference: 44%|████▍ | 2594/5920 [42:25<10:49, 5.12it/s, tok/s=14466.5]
Inference: 44%|████▍ | 2594/5920 [42:25<10:49, 5.12it/s, tok/s=14466.4]
Inference: 44%|████▍ | 2595/5920 [42:25<10:49, 5.12it/s, tok/s=14466.9]
Inference: 44%|████▍ | 2596/5920 [42:25<10:49, 5.12it/s, tok/s=14469.2]
Inference: 44%|████▍ | 2597/5920 [42:25<07:18, 7.57it/s, tok/s=14469.2]
Inference: 44%|████▍ | 2597/5920 [42:25<07:18, 7.57it/s, tok/s=14471.2]
Inference: 44%|████▍ | 2598/5920 [42:25<07:18, 7.57it/s, tok/s=14472.8]
Inference: 44%|████▍ | 2599/5920 [42:26<08:30, 6.50it/s, tok/s=14472.8]
Inference: 44%|████▍ | 2599/5920 [42:26<08:30, 6.50it/s, tok/s=14476.7]
Inference: 44%|████▍ | 2600/5920 [42:26<10:42, 5.16it/s, tok/s=14476.7]
Inference: 44%|████▍ | 2600/5920 [42:26<10:42, 5.16it/s, tok/s=14475.9]
Inference: 44%|████▍ | 2601/5920 [42:26<09:41, 5.71it/s, tok/s=14475.9]
Inference: 44%|████▍ | 2601/5920 [42:26<09:41, 5.71it/s, tok/s=14476.3]
Inference: 44%|████▍ | 2602/5920 [42:27<20:03, 2.76it/s, tok/s=14476.3]
Inference: 44%|████▍ | 2602/5920 [42:27<20:03, 2.76it/s, tok/s=14482.1]
Inference: 44%|████▍ | 2603/5920 [42:27<20:02, 2.76it/s, tok/s=14482.8]
Inference: 44%|████▍ | 2604/5920 [42:28<18:19, 3.02it/s, tok/s=14482.8]
Inference: 44%|████▍ | 2604/5920 [42:28<18:19, 3.02it/s, tok/s=14480.8]
Inference: 44%|████▍ | 2605/5920 [42:28<21:00, 2.63it/s, tok/s=14480.8]
Inference: 44%|████▍ | 2605/5920 [42:28<21:00, 2.63it/s, tok/s=14478.3]
Inference: 44%|████▍ | 2606/5920 [42:28<19:49, 2.79it/s, tok/s=14478.3]
Inference: 44%|████▍ | 2606/5920 [42:28<19:49, 2.79it/s, tok/s=14477.5]
Inference: 44%|████▍ | 2607/5920 [42:29<17:02, 3.24it/s, tok/s=14477.5]
Inference: 44%|████▍ | 2607/5920 [42:29<17:02, 3.24it/s, tok/s=14477.3]
Inference: 44%|████▍ | 2608/5920 [42:29<14:09, 3.90it/s, tok/s=14477.3]
Inference: 44%|████▍ | 2608/5920 [42:29<14:09, 3.90it/s, tok/s=14486.1]
Inference: 44%|████▍ | 2609/5920 [42:29<18:26, 2.99it/s, tok/s=14486.1]
Inference: 44%|████▍ | 2609/5920 [42:29<18:26, 2.99it/s, tok/s=14494.8]
Inference: 44%|████▍ | 2610/5920 [42:29<18:26, 2.99it/s, tok/s=14498.9]
Inference: 44%|████▍ | 2611/5920 [42:31<31:20, 1.76it/s, tok/s=14498.9]
Inference: 44%|████▍ | 2611/5920 [42:31<31:20, 1.76it/s, tok/s=14489.8]
Inference: 44%|████▍ | 2612/5920 [42:32<31:03, 1.78it/s, tok/s=14489.8]
Inference: 44%|████▍ | 2612/5920 [42:32<31:03, 1.78it/s, tok/s=14498.6]
Inference: 44%|████▍ | 2613/5920 [42:32<24:36, 2.24it/s, tok/s=14498.6]
Inference: 44%|████▍ | 2613/5920 [42:32<24:36, 2.24it/s, tok/s=14498.4]
Inference: 44%|████▍ | 2614/5920 [42:33<34:49, 1.58it/s, tok/s=14498.4]
Inference: 44%|████▍ | 2614/5920 [42:33<34:49, 1.58it/s, tok/s=14501.7]
Inference: 44%|████▍ | 2615/5920 [42:33<27:40, 1.99it/s, tok/s=14501.7]
Inference: 44%|████▍ | 2615/5920 [42:33<27:40, 1.99it/s, tok/s=14501.3]
Inference: 44%|████▍ | 2616/5920 [42:33<24:55, 2.21it/s, tok/s=14501.3]
Inference: 44%|████▍ | 2616/5920 [42:33<24:55, 2.21it/s, tok/s=14501.7]
Inference: 44%|████▍ | 2617/5920 [42:34<21:30, 2.56it/s, tok/s=14501.7]
Inference: 44%|████▍ | 2617/5920 [42:34<21:30, 2.56it/s, tok/s=14500.8]
Inference: 44%|████▍ | 2618/5920 [42:34<21:29, 2.56it/s, tok/s=14501.6]
Inference: 44%|████▍ | 2619/5920 [42:34<15:10, 3.62it/s, tok/s=14501.6]
Inference: 44%|████▍ | 2619/5920 [42:34<15:10, 3.62it/s, tok/s=14505.9]
Inference: 44%|████▍ | 2620/5920 [42:34<14:05, 3.90it/s, tok/s=14505.9]
Inference: 44%|████▍ | 2620/5920 [42:34<14:05, 3.90it/s, tok/s=14506.5]
Inference: 44%|████▍ | 2621/5920 [42:34<16:26, 3.34it/s, tok/s=14506.5]
Inference: 44%|████▍ | 2621/5920 [42:34<16:26, 3.34it/s, tok/s=14515.9]
Inference: 44%|████▍ | 2622/5920 [42:35<21:31, 2.55it/s, tok/s=14515.9]
Inference: 44%|████▍ | 2622/5920 [42:35<21:31, 2.55it/s, tok/s=14512.7]
Inference: 44%|████▍ | 2623/5920 [42:35<21:31, 2.55it/s, tok/s=14512.9]
Inference: 44%|████▍ | 2624/5920 [42:36<18:39, 2.94it/s, tok/s=14512.9]
Inference: 44%|████▍ | 2624/5920 [42:36<18:39, 2.94it/s, tok/s=14510.9]
Inference: 44%|████▍ | 2625/5920 [42:36<19:44, 2.78it/s, tok/s=14510.9]
Inference: 44%|████▍ | 2625/5920 [42:36<19:44, 2.78it/s, tok/s=14509.1]
Inference: 44%|████▍ | 2626/5920 [42:37<22:11, 2.47it/s, tok/s=14509.1]
Inference: 44%|████▍ | 2626/5920 [42:37<22:11, 2.47it/s, tok/s=14506.5]
Inference: 44%|████▍ | 2627/5920 [42:37<18:59, 2.89it/s, tok/s=14506.5]
Inference: 44%|████▍ | 2627/5920 [42:37<18:59, 2.89it/s, tok/s=14507.3]
Inference: 44%|████▍ | 2628/5920 [42:37<18:32, 2.96it/s, tok/s=14507.3]
Inference: 44%|████▍ | 2628/5920 [42:37<18:32, 2.96it/s, tok/s=14512.4]
Inference: 44%|████▍ | 2629/5920 [42:38<21:26, 2.56it/s, tok/s=14512.4]
Inference: 44%|████▍ | 2629/5920 [42:38<21:26, 2.56it/s, tok/s=14521.3]
Inference: 44%|████▍ | 2630/5920 [42:38<24:04, 2.28it/s, tok/s=14521.3]
Inference: 44%|████▍ | 2630/5920 [42:38<24:04, 2.28it/s, tok/s=14520.2]
Inference: 44%|████▍ | 2631/5920 [42:38<21:14, 2.58it/s, tok/s=14520.2]
Inference: 44%|████▍ | 2631/5920 [42:38<21:14, 2.58it/s, tok/s=14526.0]
Inference: 44%|████▍ | 2632/5920 [42:39<17:37, 3.11it/s, tok/s=14526.0]
Inference: 44%|████▍ | 2632/5920 [42:39<17:37, 3.11it/s, tok/s=14527.0]
Inference: 44%|████▍ | 2633/5920 [42:39<17:36, 3.11it/s, tok/s=14538.6]
Inference: 44%|████▍ | 2634/5920 [42:39<11:45, 4.66it/s, tok/s=14538.6]
Inference: 44%|████▍ | 2634/5920 [42:39<11:45, 4.66it/s, tok/s=14547.4]
Inference: 45%|████▍ | 2635/5920 [42:40<28:57, 1.89it/s, tok/s=14547.4]
Inference: 45%|████▍ | 2635/5920 [42:40<28:57, 1.89it/s, tok/s=14540.4]
Inference: 45%|████▍ | 2636/5920 [42:40<28:57, 1.89it/s, tok/s=14541.9]
Inference: 45%|████▍ | 2637/5920 [42:41<23:55, 2.29it/s, tok/s=14541.9]
Inference: 45%|████▍ | 2637/5920 [42:41<23:55, 2.29it/s, tok/s=14539.5]
Inference: 45%|████▍ | 2638/5920 [42:42<26:54, 2.03it/s, tok/s=14539.5]
Inference: 45%|████▍ | 2638/5920 [42:42<26:54, 2.03it/s, tok/s=14544.1]
Inference: 45%|████▍ | 2639/5920 [42:43<39:18, 1.39it/s, tok/s=14544.1]
Inference: 45%|████▍ | 2639/5920 [42:43<39:18, 1.39it/s, tok/s=14536.6]
Inference: 45%|████▍ | 2640/5920 [42:44<45:27, 1.20it/s, tok/s=14536.6]
Inference: 45%|████▍ | 2640/5920 [42:44<45:27, 1.20it/s, tok/s=14531.3]
Inference: 45%|████▍ | 2641/5920 [42:44<45:26, 1.20it/s, tok/s=14542.9]
Inference: 45%|████▍ | 2642/5920 [42:45<31:23, 1.74it/s, tok/s=14542.9]
Inference: 45%|████▍ | 2642/5920 [42:45<31:23, 1.74it/s, tok/s=14541.2]
Inference: 45%|████▍ | 2643/5920 [42:45<28:53, 1.89it/s, tok/s=14541.2]
Inference: 45%|████▍ | 2643/5920 [42:45<28:53, 1.89it/s, tok/s=14539.6]
Inference: 45%|████▍ | 2644/5920 [42:45<23:45, 2.30it/s, tok/s=14539.6]
Inference: 45%|████▍ | 2644/5920 [42:45<23:45, 2.30it/s, tok/s=14539.4]
Inference: 45%|████▍ | 2645/5920 [42:46<29:13, 1.87it/s, tok/s=14539.4]
Inference: 45%|████▍ | 2645/5920 [42:46<29:13, 1.87it/s, tok/s=14535.5]
Inference: 45%|████▍ | 2646/5920 [42:46<29:12, 1.87it/s, tok/s=14542.3]
Inference: 45%|████▍ | 2647/5920 [42:47<31:36, 1.73it/s, tok/s=14542.3]
Inference: 45%|████▍ | 2647/5920 [42:47<31:36, 1.73it/s, tok/s=14536.9]
Inference: 45%|████▍ | 2648/5920 [42:48<31:21, 1.74it/s, tok/s=14536.9]
Inference: 45%|████▍ | 2648/5920 [42:48<31:21, 1.74it/s, tok/s=14539.8]
Inference: 45%|████▍ | 2649/5920 [42:48<27:01, 2.02it/s, tok/s=14539.8]
Inference: 45%|████▍ | 2649/5920 [42:48<27:01, 2.02it/s, tok/s=14539.1]
Inference: 45%|████▍ | 2650/5920 [42:49<40:03, 1.36it/s, tok/s=14539.1]
Inference: 45%|████▍ | 2650/5920 [42:49<40:03, 1.36it/s, tok/s=14542.5]
Inference: 45%|████▍ | 2651/5920 [42:50<35:11, 1.55it/s, tok/s=14542.5]
Inference: 45%|████▍ | 2651/5920 [42:50<35:11, 1.55it/s, tok/s=14540.8]
Inference: 45%|████▍ | 2652/5920 [42:50<28:39, 1.90it/s, tok/s=14540.8]
Inference: 45%|████▍ | 2652/5920 [42:50<28:39, 1.90it/s, tok/s=14548.3]
Inference: 45%|████▍ | 2653/5920 [42:51<32:20, 1.68it/s, tok/s=14548.3]
Inference: 45%|████▍ | 2653/5920 [42:51<32:20, 1.68it/s, tok/s=14544.5]
Inference: 45%|████▍ | 2654/5920 [42:51<32:19, 1.68it/s, tok/s=14544.4]
Inference: 45%|████▍ | 2655/5920 [42:51<23:51, 2.28it/s, tok/s=14544.4]
Inference: 45%|████▍ | 2655/5920 [42:51<23:51, 2.28it/s, tok/s=14543.4]
Inference: 45%|████▍ | 2656/5920 [42:52<31:59, 1.70it/s, tok/s=14543.4]
Inference: 45%|████▍ | 2656/5920 [42:52<31:59, 1.70it/s, tok/s=14537.7]
Inference: 45%|████▍ | 2657/5920 [42:53<26:19, 2.07it/s, tok/s=14537.7]
Inference: 45%|████▍ | 2657/5920 [42:53<26:19, 2.07it/s, tok/s=14546.7]
Inference: 45%|████▍ | 2658/5920 [42:53<23:51, 2.28it/s, tok/s=14546.7]
Inference: 45%|████▍ | 2658/5920 [42:53<23:51, 2.28it/s, tok/s=14545.8]
Inference: 45%|████▍ | 2659/5920 [42:54<35:40, 1.52it/s, tok/s=14545.8]
Inference: 45%|████▍ | 2659/5920 [42:54<35:40, 1.52it/s, tok/s=14539.7]
Inference: 45%|████▍ | 2660/5920 [42:54<31:40, 1.72it/s, tok/s=14539.7]
Inference: 45%|████▍ | 2660/5920 [42:54<31:40, 1.72it/s, tok/s=14537.9]
Inference: 45%|████▍ | 2661/5920 [42:55<34:14, 1.59it/s, tok/s=14537.9]
Inference: 45%|████▍ | 2661/5920 [42:55<34:14, 1.59it/s, tok/s=14534.9]
Inference: 45%|████▍ | 2662/5920 [42:57<45:43, 1.19it/s, tok/s=14534.9]
Inference: 45%|████▍ | 2662/5920 [42:57<45:43, 1.19it/s, tok/s=14528.2]
Inference: 45%|████▍ | 2663/5920 [42:58<53:24, 1.02it/s, tok/s=14528.2]
Inference: 45%|████▍ | 2663/5920 [42:58<53:24, 1.02it/s, tok/s=14530.1]
Inference: 45%|████▌ | 2664/5920 [42:58<43:13, 1.26it/s, tok/s=14530.1]
Inference: 45%|████▌ | 2664/5920 [42:58<43:13, 1.26it/s, tok/s=14538.1]
Inference: 45%|████▌ | 2665/5920 [42:59<41:53, 1.30it/s, tok/s=14538.1]
Inference: 45%|████▌ | 2665/5920 [42:59<41:53, 1.30it/s, tok/s=14545.8]
Inference: 45%|████▌ | 2666/5920 [43:00<40:09, 1.35it/s, tok/s=14545.8]
Inference: 45%|████▌ | 2666/5920 [43:00<40:09, 1.35it/s, tok/s=14542.9]
Inference: 45%|████▌ | 2667/5920 [43:00<29:54, 1.81it/s, tok/s=14542.9]
Inference: 45%|████▌ | 2667/5920 [43:00<29:54, 1.81it/s, tok/s=14553.2]
Inference: 45%|████▌ | 2668/5920 [43:00<30:13, 1.79it/s, tok/s=14553.2]
Inference: 45%|████▌ | 2668/5920 [43:00<30:13, 1.79it/s, tok/s=14551.4]
Inference: 45%|████▌ | 2669/5920 [43:00<23:37, 2.29it/s, tok/s=14551.4]
Inference: 45%|████▌ | 2669/5920 [43:00<23:37, 2.29it/s, tok/s=14551.2]
Inference: 45%|████▌ | 2670/5920 [43:00<23:37, 2.29it/s, tok/s=14558.0]
Inference: 45%|████▌ | 2671/5920 [43:01<19:20, 2.80it/s, tok/s=14558.0]
Inference: 45%|████▌ | 2671/5920 [43:01<19:20, 2.80it/s, tok/s=14557.4]
Inference: 45%|████▌ | 2672/5920 [43:01<19:20, 2.80it/s, tok/s=14558.1]
Inference: 45%|████▌ | 2673/5920 [43:01<16:36, 3.26it/s, tok/s=14558.1]
Inference: 45%|████▌ | 2673/5920 [43:01<16:36, 3.26it/s, tok/s=14557.1]
Inference: 45%|████▌ | 2674/5920 [43:02<16:44, 3.23it/s, tok/s=14557.1]
Inference: 45%|████▌ | 2674/5920 [43:02<16:44, 3.23it/s, tok/s=14555.9]
Inference: 45%|████▌ | 2675/5920 [43:02<18:05, 2.99it/s, tok/s=14555.9]
Inference: 45%|████▌ | 2675/5920 [43:02<18:05, 2.99it/s, tok/s=14558.1]
Inference: 45%|████▌ | 2676/5920 [43:03<23:33, 2.30it/s, tok/s=14558.1]
Inference: 45%|████▌ | 2676/5920 [43:03<23:33, 2.30it/s, tok/s=14554.7]
Inference: 45%|████▌ | 2677/5920 [43:03<19:16, 2.81it/s, tok/s=14554.7]
Inference: 45%|████▌ | 2677/5920 [43:03<19:16, 2.81it/s, tok/s=14554.4]
Inference: 45%|████▌ | 2678/5920 [43:03<16:18, 3.31it/s, tok/s=14554.4]
Inference: 45%|████▌ | 2678/5920 [43:03<16:18, 3.31it/s, tok/s=14555.9]
Inference: 45%|████▌ | 2679/5920 [43:04<20:28, 2.64it/s, tok/s=14555.9]
Inference: 45%|████▌ | 2679/5920 [43:04<20:28, 2.64it/s, tok/s=14553.3]
Inference: 45%|████▌ | 2680/5920 [43:04<18:51, 2.86it/s, tok/s=14553.3]
Inference: 45%|████▌ | 2680/5920 [43:04<18:51, 2.86it/s, tok/s=14552.2]
Inference: 45%|████▌ | 2681/5920 [43:04<19:34, 2.76it/s, tok/s=14552.2]
Inference: 45%|████▌ | 2681/5920 [43:04<19:34, 2.76it/s, tok/s=14561.7]
Inference: 45%|████▌ | 2682/5920 [43:05<17:34, 3.07it/s, tok/s=14561.7]
Inference: 45%|████▌ | 2682/5920 [43:05<17:34, 3.07it/s, tok/s=14560.9]
Inference: 45%|████▌ | 2683/5920 [43:05<15:10, 3.55it/s, tok/s=14560.9]
Inference: 45%|████▌ | 2683/5920 [43:05<15:10, 3.55it/s, tok/s=14562.5]
Inference: 45%|████▌ | 2684/5920 [43:05<13:03, 4.13it/s, tok/s=14562.5]
Inference: 45%|████▌ | 2684/5920 [43:05<13:03, 4.13it/s, tok/s=14571.2]
Inference: 45%|████▌ | 2685/5920 [43:05<17:07, 3.15it/s, tok/s=14571.2]
Inference: 45%|████▌ | 2685/5920 [43:05<17:07, 3.15it/s, tok/s=14568.7]
Inference: 45%|████▌ | 2686/5920 [43:06<16:44, 3.22it/s, tok/s=14568.7]
Inference: 45%|████▌ | 2686/5920 [43:06<16:44, 3.22it/s, tok/s=14567.6]
Inference: 45%|████▌ | 2687/5920 [43:06<14:16, 3.78it/s, tok/s=14567.6]
Inference: 45%|████▌ | 2687/5920 [43:06<14:16, 3.78it/s, tok/s=14567.1]
Inference: 45%|████▌ | 2688/5920 [43:06<12:45, 4.22it/s, tok/s=14567.1]
Inference: 45%|████▌ | 2688/5920 [43:06<12:45, 4.22it/s, tok/s=14572.9]
Inference: 45%|████▌ | 2689/5920 [43:06<12:53, 4.18it/s, tok/s=14572.9]
Inference: 45%|████▌ | 2689/5920 [43:06<12:53, 4.18it/s, tok/s=14572.1]
Inference: 45%|████▌ | 2690/5920 [43:07<14:02, 3.84it/s, tok/s=14572.1]
Inference: 45%|████▌ | 2690/5920 [43:07<14:02, 3.84it/s, tok/s=14571.5]
Inference: 45%|████▌ | 2691/5920 [43:07<12:58, 4.15it/s, tok/s=14571.5]
Inference: 45%|████▌ | 2691/5920 [43:07<12:58, 4.15it/s, tok/s=14570.9]
Inference: 45%|████▌ | 2692/5920 [43:07<13:15, 4.06it/s, tok/s=14570.9]
Inference: 45%|████▌ | 2692/5920 [43:07<13:15, 4.06it/s, tok/s=14570.0]
Inference: 45%|████▌ | 2693/5920 [43:07<13:15, 4.06it/s, tok/s=14570.1]
Inference: 46%|████▌ | 2694/5920 [43:07<08:58, 5.99it/s, tok/s=14570.1]
Inference: 46%|████▌ | 2694/5920 [43:07<08:58, 5.99it/s, tok/s=14570.1]
Inference: 46%|████▌ | 2695/5920 [43:07<08:58, 5.99it/s, tok/s=14570.1]
Inference: 46%|████▌ | 2696/5920 [43:08<17:15, 3.11it/s, tok/s=14570.1]
Inference: 46%|████▌ | 2696/5920 [43:08<17:15, 3.11it/s, tok/s=14564.7]
Inference: 46%|████▌ | 2697/5920 [43:09<15:15, 3.52it/s, tok/s=14564.7]
Inference: 46%|████▌ | 2697/5920 [43:09<15:15, 3.52it/s, tok/s=14575.5]
Inference: 46%|████▌ | 2698/5920 [43:09<13:57, 3.85it/s, tok/s=14575.5]
Inference: 46%|████▌ | 2698/5920 [43:09<13:57, 3.85it/s, tok/s=14575.0]
Inference: 46%|████▌ | 2699/5920 [43:09<16:31, 3.25it/s, tok/s=14575.0]
Inference: 46%|████▌ | 2699/5920 [43:09<16:31, 3.25it/s, tok/s=14579.9]
Inference: 46%|████▌ | 2700/5920 [43:09<16:30, 3.25it/s, tok/s=14580.9]
Inference: 46%|████▌ | 2701/5920 [43:10<19:15, 2.79it/s, tok/s=14580.9]
Inference: 46%|████▌ | 2701/5920 [43:10<19:15, 2.79it/s, tok/s=14577.2]
Inference: 46%|████▌ | 2702/5920 [43:10<19:14, 2.79it/s, tok/s=14577.2]
Inference: 46%|████▌ | 2702/5920 [43:10<19:14, 2.79it/s, tok/s=14582.8]
Inference: 46%|████▌ | 2703/5920 [43:11<18:38, 2.88it/s, tok/s=14582.8]
Inference: 46%|████▌ | 2703/5920 [43:11<18:38, 2.88it/s, tok/s=14586.1]
Inference: 46%|████▌ | 2704/5920 [43:12<26:14, 2.04it/s, tok/s=14586.1]
Inference: 46%|████▌ | 2704/5920 [43:12<26:14, 2.04it/s, tok/s=14581.5]
Inference: 46%|████▌ | 2705/5920 [43:12<21:30, 2.49it/s, tok/s=14581.5]
Inference: 46%|████▌ | 2705/5920 [43:12<21:30, 2.49it/s, tok/s=14581.8]
Inference: 46%|████▌ | 2706/5920 [43:12<21:29, 2.49it/s, tok/s=14584.1]
Inference: 46%|████▌ | 2707/5920 [43:12<16:27, 3.25it/s, tok/s=14584.1]
Inference: 46%|████▌ | 2707/5920 [43:12<16:27, 3.25it/s, tok/s=14591.1]
Inference: 46%|████▌ | 2708/5920 [43:13<17:40, 3.03it/s, tok/s=14591.1]
Inference: 46%|████▌ | 2708/5920 [43:13<17:40, 3.03it/s, tok/s=14589.4]
Inference: 46%|████▌ | 2709/5920 [43:13<24:34, 2.18it/s, tok/s=14589.4]
Inference: 46%|████▌ | 2709/5920 [43:13<24:34, 2.18it/s, tok/s=14592.3]
Inference: 46%|████▌ | 2710/5920 [43:14<32:01, 1.67it/s, tok/s=14592.3]
Inference: 46%|████▌ | 2710/5920 [43:14<32:01, 1.67it/s, tok/s=14595.0]
Inference: 46%|████▌ | 2711/5920 [43:14<24:55, 2.15it/s, tok/s=14595.0]
Inference: 46%|████▌ | 2711/5920 [43:14<24:55, 2.15it/s, tok/s=14596.5]
Inference: 46%|████▌ | 2712/5920 [43:15<20:07, 2.66it/s, tok/s=14596.5]
Inference: 46%|████▌ | 2712/5920 [43:15<20:07, 2.66it/s, tok/s=14596.7]
Inference: 46%|████▌ | 2713/5920 [43:15<20:06, 2.66it/s, tok/s=14599.1]
Inference: 46%|████▌ | 2714/5920 [43:15<15:16, 3.50it/s, tok/s=14599.1]
Inference: 46%|████▌ | 2714/5920 [43:15<15:16, 3.50it/s, tok/s=14597.9]
Inference: 46%|████▌ | 2715/5920 [43:15<14:47, 3.61it/s, tok/s=14597.9]
Inference: 46%|████▌ | 2715/5920 [43:15<14:47, 3.61it/s, tok/s=14597.2]
Inference: 46%|████▌ | 2716/5920 [43:16<20:49, 2.56it/s, tok/s=14597.2]
Inference: 46%|████▌ | 2716/5920 [43:16<20:49, 2.56it/s, tok/s=14604.8]
Inference: 46%|████▌ | 2717/5920 [43:16<20:49, 2.56it/s, tok/s=14605.2]
Inference: 46%|████▌ | 2718/5920 [43:16<15:28, 3.45it/s, tok/s=14605.2]
Inference: 46%|████▌ | 2718/5920 [43:16<15:28, 3.45it/s, tok/s=14604.2]
Inference: 46%|████▌ | 2719/5920 [43:17<21:20, 2.50it/s, tok/s=14604.2]
Inference: 46%|████▌ | 2719/5920 [43:17<21:20, 2.50it/s, tok/s=14605.0]
Inference: 46%|████▌ | 2720/5920 [43:18<27:20, 1.95it/s, tok/s=14605.0]
Inference: 46%|████▌ | 2720/5920 [43:18<27:20, 1.95it/s, tok/s=14600.8]
Inference: 46%|████▌ | 2721/5920 [43:18<27:42, 1.92it/s, tok/s=14600.8]
Inference: 46%|████▌ | 2721/5920 [43:18<27:42, 1.92it/s, tok/s=14605.5]
Inference: 46%|████▌ | 2722/5920 [43:20<48:15, 1.10it/s, tok/s=14605.5]
Inference: 46%|████▌ | 2722/5920 [43:20<48:15, 1.10it/s, tok/s=14606.3]
Inference: 46%|████▌ | 2723/5920 [43:21<38:00, 1.40it/s, tok/s=14606.3]
Inference: 46%|████▌ | 2723/5920 [43:21<38:00, 1.40it/s, tok/s=14615.0]
Inference: 46%|████▌ | 2724/5920 [43:21<33:36, 1.58it/s, tok/s=14615.0]
Inference: 46%|████▌ | 2724/5920 [43:21<33:36, 1.58it/s, tok/s=14623.4]
Inference: 46%|████▌ | 2725/5920 [43:21<33:35, 1.58it/s, tok/s=14624.0]
Inference: 46%|████▌ | 2726/5920 [43:22<28:59, 1.84it/s, tok/s=14624.0]
Inference: 46%|████▌ | 2726/5920 [43:22<28:59, 1.84it/s, tok/s=14619.7]
Inference: 46%|████▌ | 2727/5920 [43:22<24:19, 2.19it/s, tok/s=14619.7]
Inference: 46%|████▌ | 2727/5920 [43:22<24:19, 2.19it/s, tok/s=14619.5]
Inference: 46%|████▌ | 2728/5920 [43:23<29:25, 1.81it/s, tok/s=14619.5]
Inference: 46%|████▌ | 2728/5920 [43:23<29:25, 1.81it/s, tok/s=14619.3]
Inference: 46%|████▌ | 2729/5920 [43:23<29:24, 1.81it/s, tok/s=14621.2]
Inference: 46%|████▌ | 2730/5920 [43:23<20:51, 2.55it/s, tok/s=14621.2]
Inference: 46%|████▌ | 2730/5920 [43:23<20:51, 2.55it/s, tok/s=14620.1]
Inference: 46%|████▌ | 2731/5920 [43:23<18:35, 2.86it/s, tok/s=14620.1]
Inference: 46%|████▌ | 2731/5920 [43:23<18:35, 2.86it/s, tok/s=14619.6]
Inference: 46%|████▌ | 2732/5920 [43:24<18:39, 2.85it/s, tok/s=14619.6]
Inference: 46%|████▌ | 2732/5920 [43:24<18:39, 2.85it/s, tok/s=14623.0]
Inference: 46%|████▌ | 2733/5920 [43:24<20:35, 2.58it/s, tok/s=14623.0]
Inference: 46%|████▌ | 2733/5920 [43:24<20:35, 2.58it/s, tok/s=14620.9]
Inference: 46%|████▌ | 2734/5920 [43:24<17:44, 2.99it/s, tok/s=14620.9]
Inference: 46%|████▌ | 2734/5920 [43:24<17:44, 2.99it/s, tok/s=14631.4]
Inference: 46%|████▌ | 2735/5920 [43:25<19:54, 2.67it/s, tok/s=14631.4]
Inference: 46%|████▌ | 2735/5920 [43:25<19:54, 2.67it/s, tok/s=14638.5]
Inference: 46%|████▌ | 2736/5920 [43:25<19:54, 2.67it/s, tok/s=14642.0]
Inference: 46%|████▌ | 2737/5920 [43:26<21:05, 2.51it/s, tok/s=14642.0]
Inference: 46%|████▌ | 2737/5920 [43:26<21:05, 2.51it/s, tok/s=14638.7]
Inference: 46%|████▋ | 2738/5920 [43:26<18:26, 2.88it/s, tok/s=14638.7]
Inference: 46%|████▋ | 2738/5920 [43:26<18:26, 2.88it/s, tok/s=14638.3]
Inference: 46%|████▋ | 2739/5920 [43:26<18:21, 2.89it/s, tok/s=14638.3]
Inference: 46%|████▋ | 2739/5920 [43:26<18:21, 2.89it/s, tok/s=14637.1]
Inference: 46%|████▋ | 2740/5920 [43:26<18:21, 2.89it/s, tok/s=14637.6]
Inference: 46%|████▋ | 2741/5920 [43:26<11:48, 4.48it/s, tok/s=14637.6]
Inference: 46%|████▋ | 2741/5920 [43:26<11:48, 4.48it/s, tok/s=14637.9]
Inference: 46%|████▋ | 2742/5920 [43:27<11:07, 4.76it/s, tok/s=14637.9]
Inference: 46%|████▋ | 2742/5920 [43:27<11:07, 4.76it/s, tok/s=14646.9]
Inference: 46%|████▋ | 2743/5920 [43:27<09:52, 5.36it/s, tok/s=14646.9]
Inference: 46%|████▋ | 2743/5920 [43:27<09:52, 5.36it/s, tok/s=14646.9]
Inference: 46%|████▋ | 2744/5920 [43:28<22:48, 2.32it/s, tok/s=14646.9]
Inference: 46%|████▋ | 2744/5920 [43:28<22:48, 2.32it/s, tok/s=14641.6]
Inference: 46%|████▋ | 2745/5920 [43:28<19:11, 2.76it/s, tok/s=14641.6]
Inference: 46%|████▋ | 2745/5920 [43:28<19:11, 2.76it/s, tok/s=14641.0]
Inference: 46%|████▋ | 2746/5920 [43:28<20:46, 2.55it/s, tok/s=14641.0]
Inference: 46%|████▋ | 2746/5920 [43:28<20:46, 2.55it/s, tok/s=14643.7]
Inference: 46%|████▋ | 2747/5920 [43:29<22:09, 2.39it/s, tok/s=14643.7]
Inference: 46%|████▋ | 2747/5920 [43:29<22:09, 2.39it/s, tok/s=14652.5]
Inference: 46%|████▋ | 2748/5920 [43:29<22:09, 2.39it/s, tok/s=14656.6]
Inference: 46%|████▋ | 2749/5920 [43:30<19:43, 2.68it/s, tok/s=14656.6]
Inference: 46%|████▋ | 2749/5920 [43:30<19:43, 2.68it/s, tok/s=14664.4]
Inference: 46%|████▋ | 2750/5920 [43:30<25:37, 2.06it/s, tok/s=14664.4]
Inference: 46%|████▋ | 2750/5920 [43:30<25:37, 2.06it/s, tok/s=14660.2]
Inference: 46%|████▋ | 2751/5920 [43:31<26:33, 1.99it/s, tok/s=14660.2]
Inference: 46%|████▋ | 2751/5920 [43:31<26:33, 1.99it/s, tok/s=14657.5]
Inference: 46%|████▋ | 2752/5920 [43:31<22:15, 2.37it/s, tok/s=14657.5]
Inference: 46%|████▋ | 2752/5920 [43:31<22:15, 2.37it/s, tok/s=14656.8]
Inference: 47%|████▋ | 2753/5920 [43:32<22:53, 2.31it/s, tok/s=14656.8]
Inference: 47%|████▋ | 2753/5920 [43:32<22:53, 2.31it/s, tok/s=14655.2]
Inference: 47%|████▋ | 2754/5920 [43:32<22:53, 2.31it/s, tok/s=14655.1]
Inference: 47%|████▋ | 2755/5920 [43:32<14:53, 3.54it/s, tok/s=14655.1]
Inference: 47%|████▋ | 2755/5920 [43:32<14:53, 3.54it/s, tok/s=14655.3]
Inference: 47%|████▋ | 2756/5920 [43:32<16:20, 3.23it/s, tok/s=14655.3]
Inference: 47%|████▋ | 2756/5920 [43:32<16:20, 3.23it/s, tok/s=14654.0]
Inference: 47%|████▋ | 2757/5920 [43:32<16:20, 3.23it/s, tok/s=14654.2]
Inference: 47%|████▋ | 2758/5920 [43:33<13:09, 4.00it/s, tok/s=14654.2]
Inference: 47%|████▋ | 2758/5920 [43:33<13:09, 4.00it/s, tok/s=14653.5]
Inference: 47%|████▋ | 2759/5920 [43:33<13:36, 3.87it/s, tok/s=14653.5]
Inference: 47%|████▋ | 2759/5920 [43:33<13:36, 3.87it/s, tok/s=14653.1]
Inference: 47%|████▋ | 2760/5920 [43:33<12:36, 4.18it/s, tok/s=14653.1]
Inference: 47%|████▋ | 2760/5920 [43:33<12:36, 4.18it/s, tok/s=14652.6]
Inference: 47%|████▋ | 2761/5920 [43:33<14:51, 3.54it/s, tok/s=14652.6]
Inference: 47%|████▋ | 2761/5920 [43:33<14:51, 3.54it/s, tok/s=14652.9]
Inference: 47%|████▋ | 2762/5920 [43:34<14:58, 3.51it/s, tok/s=14652.9]
Inference: 47%|████▋ | 2762/5920 [43:34<14:58, 3.51it/s, tok/s=14652.5]
Inference: 47%|████▋ | 2763/5920 [43:34<14:58, 3.51it/s, tok/s=14663.6]
Inference: 47%|████▋ | 2764/5920 [43:34<14:20, 3.67it/s, tok/s=14663.6]
Inference: 47%|████▋ | 2764/5920 [43:34<14:20, 3.67it/s, tok/s=14662.4]
Inference: 47%|████▋ | 2765/5920 [43:34<13:31, 3.89it/s, tok/s=14662.4]
Inference: 47%|████▋ | 2765/5920 [43:34<13:31, 3.89it/s, tok/s=14669.9]
Inference: 47%|████▋ | 2766/5920 [43:35<19:54, 2.64it/s, tok/s=14669.9]
Inference: 47%|████▋ | 2766/5920 [43:35<19:54, 2.64it/s, tok/s=14666.8]
Inference: 47%|████▋ | 2767/5920 [43:36<25:18, 2.08it/s, tok/s=14666.8]
Inference: 47%|████▋ | 2767/5920 [43:36<25:18, 2.08it/s, tok/s=14667.0]
Inference: 47%|████▋ | 2768/5920 [43:36<25:17, 2.08it/s, tok/s=14667.6]
Inference: 47%|████▋ | 2769/5920 [43:36<25:17, 2.08it/s, tok/s=14675.2]
Inference: 47%|████▋ | 2770/5920 [43:36<25:16, 2.08it/s, tok/s=14676.3]
Inference: 47%|████▋ | 2771/5920 [43:37<17:32, 2.99it/s, tok/s=14676.3]
Inference: 47%|████▋ | 2771/5920 [43:37<17:32, 2.99it/s, tok/s=14682.9]
Inference: 47%|████▋ | 2772/5920 [43:38<26:50, 1.95it/s, tok/s=14682.9]
Inference: 47%|████▋ | 2772/5920 [43:38<26:50, 1.95it/s, tok/s=14676.1]
Inference: 47%|████▋ | 2773/5920 [43:38<26:50, 1.95it/s, tok/s=14685.6]
Inference: 47%|████▋ | 2774/5920 [43:38<18:51, 2.78it/s, tok/s=14685.6]
Inference: 47%|████▋ | 2774/5920 [43:38<18:51, 2.78it/s, tok/s=14685.5]
Inference: 47%|████▋ | 2775/5920 [43:39<17:20, 3.02it/s, tok/s=14685.5]
Inference: 47%|████▋ | 2775/5920 [43:39<17:20, 3.02it/s, tok/s=14685.2]
Inference: 47%|████▋ | 2776/5920 [43:39<16:29, 3.18it/s, tok/s=14685.2]
Inference: 47%|████▋ | 2776/5920 [43:39<16:29, 3.18it/s, tok/s=14684.8]
Inference: 47%|████▋ | 2777/5920 [43:39<16:29, 3.18it/s, tok/s=14685.7]
Inference: 47%|████▋ | 2778/5920 [43:39<16:28, 3.18it/s, tok/s=14685.9]
Inference: 47%|████▋ | 2779/5920 [43:39<09:39, 5.42it/s, tok/s=14685.9]
Inference: 47%|████▋ | 2779/5920 [43:39<09:39, 5.42it/s, tok/s=14686.0]
Inference: 47%|████▋ | 2780/5920 [43:39<09:39, 5.42it/s, tok/s=14685.3]
Inference: 47%|████▋ | 2781/5920 [43:39<08:56, 5.85it/s, tok/s=14685.3]
Inference: 47%|████▋ | 2781/5920 [43:39<08:56, 5.85it/s, tok/s=14690.0]
Inference: 47%|████▋ | 2782/5920 [43:40<15:32, 3.37it/s, tok/s=14690.0]
Inference: 47%|████▋ | 2782/5920 [43:40<15:32, 3.37it/s, tok/s=14690.7]
Inference: 47%|████▋ | 2783/5920 [43:41<16:49, 3.11it/s, tok/s=14690.7]
Inference: 47%|████▋ | 2783/5920 [43:41<16:49, 3.11it/s, tok/s=14689.3]
Inference: 47%|████▋ | 2784/5920 [43:41<15:10, 3.44it/s, tok/s=14689.3]
Inference: 47%|████▋ | 2784/5920 [43:41<15:10, 3.44it/s, tok/s=14689.0]
Inference: 47%|████▋ | 2785/5920 [43:41<15:10, 3.44it/s, tok/s=14689.7]
Inference: 47%|████▋ | 2786/5920 [43:42<18:10, 2.87it/s, tok/s=14689.7]
Inference: 47%|████▋ | 2786/5920 [43:42<18:10, 2.87it/s, tok/s=14686.6]
Inference: 47%|████▋ | 2787/5920 [43:42<18:10, 2.87it/s, tok/s=14687.0]
Inference: 47%|████▋ | 2788/5920 [43:42<13:53, 3.76it/s, tok/s=14687.0]
Inference: 47%|████▋ | 2788/5920 [43:42<13:53, 3.76it/s, tok/s=14687.0]
Inference: 47%|████▋ | 2789/5920 [43:43<19:05, 2.73it/s, tok/s=14687.0]
Inference: 47%|████▋ | 2789/5920 [43:43<19:05, 2.73it/s, tok/s=14683.2]
Inference: 47%|████▋ | 2790/5920 [43:43<22:15, 2.34it/s, tok/s=14683.2]
Inference: 47%|████▋ | 2790/5920 [43:43<22:15, 2.34it/s, tok/s=14681.2]
Inference: 47%|████▋ | 2791/5920 [43:43<18:36, 2.80it/s, tok/s=14681.2]
Inference: 47%|████▋ | 2791/5920 [43:43<18:36, 2.80it/s, tok/s=14687.4]
Inference: 47%|████▋ | 2792/5920 [43:44<16:47, 3.10it/s, tok/s=14687.4]
Inference: 47%|████▋ | 2792/5920 [43:44<16:47, 3.10it/s, tok/s=14687.2]
Inference: 47%|████▋ | 2793/5920 [43:44<20:53, 2.49it/s, tok/s=14687.2]
Inference: 47%|████▋ | 2793/5920 [43:44<20:53, 2.49it/s, tok/s=14686.6]
Inference: 47%|████▋ | 2794/5920 [43:44<20:53, 2.49it/s, tok/s=14692.9]
Inference: 47%|████▋ | 2795/5920 [43:45<16:40, 3.12it/s, tok/s=14692.9]
Inference: 47%|████▋ | 2795/5920 [43:45<16:40, 3.12it/s, tok/s=14691.3]
Inference: 47%|████▋ | 2796/5920 [43:45<16:08, 3.23it/s, tok/s=14691.3]
Inference: 47%|████▋ | 2796/5920 [43:45<16:08, 3.23it/s, tok/s=14691.3]
Inference: 47%|████▋ | 2797/5920 [43:45<13:41, 3.80it/s, tok/s=14691.3]
Inference: 47%|████▋ | 2797/5920 [43:45<13:41, 3.80it/s, tok/s=14699.5]
Inference: 47%|████▋ | 2798/5920 [43:46<21:53, 2.38it/s, tok/s=14699.5]
Inference: 47%|████▋ | 2798/5920 [43:46<21:53, 2.38it/s, tok/s=14695.3]
Inference: 47%|████▋ | 2799/5920 [43:46<21:24, 2.43it/s, tok/s=14695.3]
Inference: 47%|████▋ | 2799/5920 [43:46<21:24, 2.43it/s, tok/s=14694.2]
Inference: 47%|████▋ | 2800/5920 [43:47<19:15, 2.70it/s, tok/s=14694.2]
Inference: 47%|████▋ | 2800/5920 [43:47<19:15, 2.70it/s, tok/s=14693.6]
Inference: 47%|████▋ | 2801/5920 [43:47<15:28, 3.36it/s, tok/s=14693.6]
Inference: 47%|████▋ | 2801/5920 [43:47<15:28, 3.36it/s, tok/s=14695.1]
Inference: 47%|████▋ | 2802/5920 [43:47<17:07, 3.04it/s, tok/s=14695.1]
Inference: 47%|████▋ | 2802/5920 [43:47<17:07, 3.04it/s, tok/s=14697.9]
Inference: 47%|████▋ | 2803/5920 [43:47<17:06, 3.04it/s, tok/s=14697.9]
Inference: 47%|████▋ | 2804/5920 [43:48<21:21, 2.43it/s, tok/s=14697.9]
Inference: 47%|████▋ | 2804/5920 [43:48<21:21, 2.43it/s, tok/s=14693.7]
Inference: 47%|████▋ | 2805/5920 [43:48<18:12, 2.85it/s, tok/s=14693.7]
Inference: 47%|████▋ | 2805/5920 [43:48<18:12, 2.85it/s, tok/s=14693.6]
Inference: 47%|████▋ | 2806/5920 [43:49<28:21, 1.83it/s, tok/s=14693.6]
Inference: 47%|████▋ | 2806/5920 [43:49<28:21, 1.83it/s, tok/s=14697.8]
Inference: 47%|████▋ | 2807/5920 [43:50<23:06, 2.25it/s, tok/s=14697.8]
Inference: 47%|████▋ | 2807/5920 [43:50<23:06, 2.25it/s, tok/s=14697.7]
Inference: 47%|████▋ | 2808/5920 [43:50<18:40, 2.78it/s, tok/s=14697.7]
Inference: 47%|████▋ | 2808/5920 [43:50<18:40, 2.78it/s, tok/s=14697.8]
Inference: 47%|████▋ | 2809/5920 [43:50<16:45, 3.09it/s, tok/s=14697.8]
Inference: 47%|████▋ | 2809/5920 [43:50<16:45, 3.09it/s, tok/s=14703.4]
Inference: 47%|████▋ | 2810/5920 [43:50<21:17, 2.43it/s, tok/s=14703.4]
Inference: 47%|████▋ | 2810/5920 [43:50<21:17, 2.43it/s, tok/s=14700.5]
Inference: 47%|████▋ | 2811/5920 [43:51<21:17, 2.43it/s, tok/s=14701.7]
Inference: 48%|████▊ | 2812/5920 [43:51<21:17, 2.43it/s, tok/s=14703.2]
Inference: 48%|████▊ | 2813/5920 [43:51<21:16, 2.43it/s, tok/s=14707.3]
Inference: 48%|████▊ | 2814/5920 [43:51<11:40, 4.43it/s, tok/s=14707.3]
Inference: 48%|████▊ | 2814/5920 [43:51<11:40, 4.43it/s, tok/s=14705.9]
Inference: 48%|████▊ | 2815/5920 [43:51<12:06, 4.28it/s, tok/s=14705.9]
Inference: 48%|████▊ | 2815/5920 [43:51<12:06, 4.28it/s, tok/s=14705.6]
Inference: 48%|████▊ | 2816/5920 [43:51<12:05, 4.28it/s, tok/s=14706.3]
Inference: 48%|████▊ | 2817/5920 [43:51<09:52, 5.24it/s, tok/s=14706.3]
Inference: 48%|████▊ | 2817/5920 [43:51<09:52, 5.24it/s, tok/s=14706.0]
Inference: 48%|████▊ | 2818/5920 [43:52<09:52, 5.23it/s, tok/s=14706.0]
Inference: 48%|████▊ | 2818/5920 [43:52<09:52, 5.23it/s, tok/s=14711.0]
Inference: 48%|████▊ | 2819/5920 [43:52<10:05, 5.12it/s, tok/s=14711.0]
Inference: 48%|████▊ | 2819/5920 [43:52<10:05, 5.12it/s, tok/s=14710.3]
Inference: 48%|████▊ | 2820/5920 [43:53<23:37, 2.19it/s, tok/s=14710.3]
Inference: 48%|████▊ | 2820/5920 [43:53<23:37, 2.19it/s, tok/s=14703.5]
Inference: 48%|████▊ | 2821/5920 [43:53<21:16, 2.43it/s, tok/s=14703.5]
Inference: 48%|████▊ | 2821/5920 [43:53<21:16, 2.43it/s, tok/s=14702.5]
Inference: 48%|████▊ | 2822/5920 [43:54<21:19, 2.42it/s, tok/s=14702.5]
Inference: 48%|████▊ | 2822/5920 [43:54<21:19, 2.42it/s, tok/s=14707.9]
Inference: 48%|████▊ | 2823/5920 [43:54<17:27, 2.96it/s, tok/s=14707.9]
Inference: 48%|████▊ | 2823/5920 [43:54<17:27, 2.96it/s, tok/s=14708.0]
Inference: 48%|████▊ | 2824/5920 [43:54<15:05, 3.42it/s, tok/s=14708.0]
Inference: 48%|████▊ | 2824/5920 [43:54<15:05, 3.42it/s, tok/s=14709.2]
Inference: 48%|████▊ | 2825/5920 [43:55<25:25, 2.03it/s, tok/s=14709.2]
Inference: 48%|████▊ | 2825/5920 [43:55<25:25, 2.03it/s, tok/s=14704.4]
Inference: 48%|████▊ | 2826/5920 [43:56<27:12, 1.90it/s, tok/s=14704.4]
Inference: 48%|████▊ | 2826/5920 [43:56<27:12, 1.90it/s, tok/s=14702.8]
Inference: 48%|████▊ | 2827/5920 [43:56<23:13, 2.22it/s, tok/s=14702.8]
Inference: 48%|████▊ | 2827/5920 [43:56<23:13, 2.22it/s, tok/s=14710.3]
Inference: 48%|████▊ | 2828/5920 [43:57<24:17, 2.12it/s, tok/s=14710.3]
Inference: 48%|████▊ | 2828/5920 [43:57<24:17, 2.12it/s, tok/s=14708.3]
Inference: 48%|████▊ | 2829/5920 [43:57<24:16, 2.12it/s, tok/s=14719.3]
Inference: 48%|████▊ | 2830/5920 [43:57<20:49, 2.47it/s, tok/s=14719.3]
Inference: 48%|████▊ | 2830/5920 [43:57<20:49, 2.47it/s, tok/s=14716.9]
Inference: 48%|████▊ | 2831/5920 [43:58<20:07, 2.56it/s, tok/s=14716.9]
Inference: 48%|████▊ | 2831/5920 [43:58<20:07, 2.56it/s, tok/s=14715.7]
Inference: 48%|████▊ | 2832/5920 [43:58<21:14, 2.42it/s, tok/s=14715.7]
Inference: 48%|████▊ | 2832/5920 [43:58<21:14, 2.42it/s, tok/s=14713.7]
Inference: 48%|████▊ | 2833/5920 [43:59<22:21, 2.30it/s, tok/s=14713.7]
Inference: 48%|████▊ | 2833/5920 [43:59<22:21, 2.30it/s, tok/s=14711.7]
Inference: 48%|████▊ | 2834/5920 [43:59<22:20, 2.30it/s, tok/s=14713.0]
Inference: 48%|████▊ | 2835/5920 [43:59<19:10, 2.68it/s, tok/s=14713.0]
Inference: 48%|████▊ | 2835/5920 [43:59<19:10, 2.68it/s, tok/s=14710.4]
Inference: 48%|████▊ | 2836/5920 [44:00<25:52, 1.99it/s, tok/s=14710.4]
Inference: 48%|████▊ | 2836/5920 [44:00<25:52, 1.99it/s, tok/s=14706.2]
Inference: 48%|████▊ | 2837/5920 [44:00<25:52, 1.99it/s, tok/s=14715.5]
Inference: 48%|████▊ | 2838/5920 [44:00<17:21, 2.96it/s, tok/s=14715.5]
Inference: 48%|████▊ | 2838/5920 [44:00<17:21, 2.96it/s, tok/s=14715.4]
Inference: 48%|████▊ | 2839/5920 [44:00<17:20, 2.96it/s, tok/s=14717.2]
Inference: 48%|████▊ | 2840/5920 [44:00<11:56, 4.30it/s, tok/s=14717.2]
Inference: 48%|████▊ | 2840/5920 [44:00<11:56, 4.30it/s, tok/s=14717.5]
Inference: 48%|████▊ | 2841/5920 [44:01<11:42, 4.38it/s, tok/s=14717.5]
Inference: 48%|████▊ | 2841/5920 [44:01<11:42, 4.38it/s, tok/s=14716.8]
Inference: 48%|████▊ | 2842/5920 [44:01<10:33, 4.86it/s, tok/s=14716.8]
Inference: 48%|████▊ | 2842/5920 [44:01<10:33, 4.86it/s, tok/s=14717.0]
Inference: 48%|████▊ | 2843/5920 [44:01<11:30, 4.46it/s, tok/s=14717.0]
Inference: 48%|████▊ | 2843/5920 [44:01<11:30, 4.46it/s, tok/s=14723.3]
Inference: 48%|████▊ | 2844/5920 [44:02<24:28, 2.09it/s, tok/s=14723.3]
Inference: 48%|████▊ | 2844/5920 [44:02<24:28, 2.09it/s, tok/s=14717.5]
Inference: 48%|████▊ | 2845/5920 [44:03<27:57, 1.83it/s, tok/s=14717.5]
Inference: 48%|████▊ | 2845/5920 [44:03<27:57, 1.83it/s, tok/s=14713.9]
Inference: 48%|████▊ | 2846/5920 [44:03<25:32, 2.01it/s, tok/s=14713.9]
Inference: 48%|████▊ | 2846/5920 [44:03<25:32, 2.01it/s, tok/s=14713.1]
Inference: 48%|████▊ | 2847/5920 [44:03<25:32, 2.01it/s, tok/s=14713.5]
Inference: 48%|████▊ | 2848/5920 [44:04<18:04, 2.83it/s, tok/s=14713.5]
Inference: 48%|████▊ | 2848/5920 [44:04<18:04, 2.83it/s, tok/s=14713.2]
Inference: 48%|████▊ | 2849/5920 [44:04<24:29, 2.09it/s, tok/s=14713.2]
Inference: 48%|████▊ | 2849/5920 [44:04<24:29, 2.09it/s, tok/s=14709.4]
Inference: 48%|████▊ | 2850/5920 [44:04<24:28, 2.09it/s, tok/s=14710.3]
Inference: 48%|████▊ | 2851/5920 [44:05<21:48, 2.35it/s, tok/s=14710.3]
Inference: 48%|████▊ | 2851/5920 [44:05<21:48, 2.35it/s, tok/s=14707.3]
Inference: 48%|████▊ | 2852/5920 [44:05<21:47, 2.35it/s, tok/s=14708.0]
Inference: 48%|████▊ | 2853/5920 [44:06<18:22, 2.78it/s, tok/s=14708.0]
Inference: 48%|████▊ | 2853/5920 [44:06<18:22, 2.78it/s, tok/s=14706.6]
Inference: 48%|████▊ | 2854/5920 [44:06<18:21, 2.78it/s, tok/s=14707.3]
Inference: 48%|████▊ | 2855/5920 [44:07<21:22, 2.39it/s, tok/s=14707.3]
Inference: 48%|████▊ | 2855/5920 [44:07<21:22, 2.39it/s, tok/s=14710.0]
Inference: 48%|████▊ | 2856/5920 [44:07<21:46, 2.34it/s, tok/s=14710.0]
Inference: 48%|████▊ | 2856/5920 [44:07<21:46, 2.34it/s, tok/s=14708.5]
Inference: 48%|████▊ | 2857/5920 [44:08<27:23, 1.86it/s, tok/s=14708.5]
Inference: 48%|████▊ | 2857/5920 [44:08<27:23, 1.86it/s, tok/s=14703.9]
Inference: 48%|████▊ | 2858/5920 [44:09<28:05, 1.82it/s, tok/s=14703.9]
Inference: 48%|████▊ | 2858/5920 [44:09<28:05, 1.82it/s, tok/s=14702.2]
Inference: 48%|████▊ | 2859/5920 [44:09<22:34, 2.26it/s, tok/s=14702.2]
Inference: 48%|████▊ | 2859/5920 [44:09<22:34, 2.26it/s, tok/s=14702.2]
Inference: 48%|████▊ | 2860/5920 [44:10<30:07, 1.69it/s, tok/s=14702.2]
Inference: 48%|████▊ | 2860/5920 [44:10<30:07, 1.69it/s, tok/s=14706.9]
Inference: 48%|████▊ | 2861/5920 [44:10<26:15, 1.94it/s, tok/s=14706.9]
Inference: 48%|████▊ | 2861/5920 [44:10<26:15, 1.94it/s, tok/s=14706.3]
Inference: 48%|████▊ | 2862/5920 [44:10<22:45, 2.24it/s, tok/s=14706.3]
Inference: 48%|████▊ | 2862/5920 [44:10<22:45, 2.24it/s, tok/s=14714.5]
Inference: 48%|████▊ | 2863/5920 [44:11<18:56, 2.69it/s, tok/s=14714.5]
Inference: 48%|████▊ | 2863/5920 [44:11<18:56, 2.69it/s, tok/s=14713.9]
Inference: 48%|████▊ | 2864/5920 [44:12<31:28, 1.62it/s, tok/s=14713.9]
Inference: 48%|████▊ | 2864/5920 [44:12<31:28, 1.62it/s, tok/s=14708.0]
Inference: 48%|████▊ | 2865/5920 [44:12<23:50, 2.14it/s, tok/s=14708.0]
Inference: 48%|████▊ | 2865/5920 [44:12<23:50, 2.14it/s, tok/s=14718.7]
Inference: 48%|████▊ | 2866/5920 [44:12<23:09, 2.20it/s, tok/s=14718.7]
Inference: 48%|████▊ | 2866/5920 [44:12<23:09, 2.20it/s, tok/s=14717.5]
Inference: 48%|████▊ | 2867/5920 [44:12<18:20, 2.77it/s, tok/s=14717.5]
Inference: 48%|████▊ | 2867/5920 [44:12<18:20, 2.77it/s, tok/s=14717.2]
Inference: 48%|████▊ | 2868/5920 [44:13<17:12, 2.95it/s, tok/s=14717.2]
Inference: 48%|████▊ | 2868/5920 [44:13<17:12, 2.95it/s, tok/s=14716.3]
Inference: 48%|████▊ | 2869/5920 [44:13<14:15, 3.57it/s, tok/s=14716.3]
Inference: 48%|████▊ | 2869/5920 [44:13<14:15, 3.57it/s, tok/s=14716.5]
Inference: 48%|████▊ | 2870/5920 [44:13<11:39, 4.36it/s, tok/s=14716.5]
Inference: 48%|████▊ | 2870/5920 [44:13<11:39, 4.36it/s, tok/s=14717.0]
Inference: 48%|████▊ | 2871/5920 [44:14<16:02, 3.17it/s, tok/s=14717.0]
Inference: 48%|████▊ | 2871/5920 [44:14<16:02, 3.17it/s, tok/s=14715.6]
Inference: 49%|████▊ | 2872/5920 [44:14<16:02, 3.17it/s, tok/s=14715.8]
Inference: 49%|████▊ | 2873/5920 [44:14<13:54, 3.65it/s, tok/s=14715.8]
Inference: 49%|████▊ | 2873/5920 [44:14<13:54, 3.65it/s, tok/s=14714.2]
Inference: 49%|████▊ | 2874/5920 [44:14<15:53, 3.19it/s, tok/s=14714.2]
Inference: 49%|████▊ | 2874/5920 [44:14<15:53, 3.19it/s, tok/s=14721.4]
Inference: 49%|████▊ | 2875/5920 [44:15<15:59, 3.17it/s, tok/s=14721.4]
Inference: 49%|████▊ | 2875/5920 [44:15<15:59, 3.17it/s, tok/s=14720.6]
Inference: 49%|████▊ | 2876/5920 [44:15<13:29, 3.76it/s, tok/s=14720.6]
Inference: 49%|████▊ | 2876/5920 [44:15<13:29, 3.76it/s, tok/s=14721.3]
Inference: 49%|████▊ | 2877/5920 [44:15<13:28, 3.76it/s, tok/s=14723.5]
Inference: 49%|████▊ | 2878/5920 [44:15<09:11, 5.52it/s, tok/s=14723.5]
Inference: 49%|████▊ | 2878/5920 [44:15<09:11, 5.52it/s, tok/s=14723.3]
Inference: 49%|████▊ | 2879/5920 [44:15<08:54, 5.69it/s, tok/s=14723.3]
Inference: 49%|████▊ | 2879/5920 [44:15<08:54, 5.69it/s, tok/s=14723.2]
Inference: 49%|████▊ | 2880/5920 [44:15<09:57, 5.09it/s, tok/s=14723.2]
Inference: 49%|████▊ | 2880/5920 [44:15<09:57, 5.09it/s, tok/s=14733.2]
Inference: 49%|████▊ | 2881/5920 [44:16<13:50, 3.66it/s, tok/s=14733.2]
Inference: 49%|████▊ | 2881/5920 [44:16<13:50, 3.66it/s, tok/s=14730.9]
Inference: 49%|████▊ | 2882/5920 [44:16<15:40, 3.23it/s, tok/s=14730.9]
Inference: 49%|████▊ | 2882/5920 [44:16<15:40, 3.23it/s, tok/s=14735.6]
Inference: 49%|████▊ | 2883/5920 [44:17<14:16, 3.55it/s, tok/s=14735.6]
Inference: 49%|████▊ | 2883/5920 [44:17<14:16, 3.55it/s, tok/s=14735.0]
Inference: 49%|████▊ | 2884/5920 [44:17<14:16, 3.55it/s, tok/s=14735.5]
Inference: 49%|████▊ | 2885/5920 [44:17<14:16, 3.55it/s, tok/s=14736.0]
Inference: 49%|████▉ | 2886/5920 [44:17<07:29, 6.76it/s, tok/s=14736.0]
Inference: 49%|████▉ | 2886/5920 [44:17<07:29, 6.76it/s, tok/s=14736.8]
Inference: 49%|████▉ | 2887/5920 [44:17<07:28, 6.76it/s, tok/s=14737.3]
Inference: 49%|████▉ | 2888/5920 [44:17<06:16, 8.06it/s, tok/s=14737.3]
Inference: 49%|████▉ | 2888/5920 [44:17<06:16, 8.06it/s, tok/s=14738.0]
Inference: 49%|████▉ | 2889/5920 [44:17<06:16, 8.06it/s, tok/s=14737.6]
Inference: 49%|████▉ | 2890/5920 [44:18<11:01, 4.58it/s, tok/s=14737.6]
Inference: 49%|████▉ | 2890/5920 [44:18<11:01, 4.58it/s, tok/s=14735.5]
Inference: 49%|████▉ | 2891/5920 [44:18<11:01, 4.58it/s, tok/s=14736.6]
Inference: 49%|████▉ | 2892/5920 [44:18<10:05, 5.00it/s, tok/s=14736.6]
Inference: 49%|████▉ | 2892/5920 [44:18<10:05, 5.00it/s, tok/s=14736.8]
Inference: 49%|████▉ | 2893/5920 [44:18<10:04, 5.00it/s, tok/s=14736.9]
Inference: 49%|████▉ | 2894/5920 [44:18<08:45, 5.76it/s, tok/s=14736.9]
Inference: 49%|████▉ | 2894/5920 [44:18<08:45, 5.76it/s, tok/s=14742.4]
Inference: 49%|████▉ | 2895/5920 [44:18<08:45, 5.76it/s, tok/s=14742.4]
Inference: 49%|████▉ | 2895/5920 [44:18<08:45, 5.76it/s, tok/s=14742.0]
Inference: 49%|████▉ | 2896/5920 [44:19<09:29, 5.31it/s, tok/s=14742.0]
Inference: 49%|████▉ | 2896/5920 [44:19<09:29, 5.31it/s, tok/s=14741.7]
Inference: 49%|████▉ | 2897/5920 [44:19<08:32, 5.90it/s, tok/s=14741.7]
Inference: 49%|████▉ | 2897/5920 [44:19<08:32, 5.90it/s, tok/s=14741.5]
Inference: 49%|████▉ | 2898/5920 [44:19<15:49, 3.18it/s, tok/s=14741.5]
Inference: 49%|████▉ | 2898/5920 [44:19<15:49, 3.18it/s, tok/s=14737.8]
Inference: 49%|████▉ | 2899/5920 [44:20<15:21, 3.28it/s, tok/s=14737.8]
Inference: 49%|████▉ | 2899/5920 [44:20<15:21, 3.28it/s, tok/s=14740.9]
Inference: 49%|████▉ | 2900/5920 [44:20<13:04, 3.85it/s, tok/s=14740.9]
Inference: 49%|████▉ | 2900/5920 [44:20<13:04, 3.85it/s, tok/s=14740.9]
Inference: 49%|████▉ | 2901/5920 [44:20<13:04, 3.85it/s, tok/s=14741.9]
Inference: 49%|████▉ | 2902/5920 [44:20<10:22, 4.85it/s, tok/s=14741.9]
Inference: 49%|████▉ | 2902/5920 [44:20<10:22, 4.85it/s, tok/s=14741.4]
Inference: 49%|████▉ | 2903/5920 [44:20<10:21, 4.85it/s, tok/s=14741.9]
Inference: 49%|████▉ | 2904/5920 [44:20<08:14, 6.10it/s, tok/s=14741.9]
Inference: 49%|████▉ | 2904/5920 [44:20<08:14, 6.10it/s, tok/s=14742.1]
Inference: 49%|████▉ | 2905/5920 [44:21<15:46, 3.19it/s, tok/s=14742.1]
Inference: 49%|████▉ | 2905/5920 [44:21<15:46, 3.19it/s, tok/s=14738.6]
Inference: 49%|████▉ | 2906/5920 [44:21<15:45, 3.19it/s, tok/s=14739.0]
Inference: 49%|████▉ | 2907/5920 [44:22<16:27, 3.05it/s, tok/s=14739.0]
Inference: 49%|████▉ | 2907/5920 [44:22<16:27, 3.05it/s, tok/s=14737.1]
Inference: 49%|████▉ | 2908/5920 [44:22<17:18, 2.90it/s, tok/s=14737.1]
Inference: 49%|████▉ | 2908/5920 [44:22<17:18, 2.90it/s, tok/s=14742.0]
Inference: 49%|████▉ | 2909/5920 [44:22<15:05, 3.32it/s, tok/s=14742.0]
Inference: 49%|████▉ | 2909/5920 [44:22<15:05, 3.32it/s, tok/s=14741.7]
Inference: 49%|████▉ | 2910/5920 [44:23<15:34, 3.22it/s, tok/s=14741.7]
Inference: 49%|████▉ | 2910/5920 [44:23<15:34, 3.22it/s, tok/s=14740.5]
Inference: 49%|████▉ | 2911/5920 [44:23<18:26, 2.72it/s, tok/s=14740.5]
Inference: 49%|████▉ | 2911/5920 [44:23<18:26, 2.72it/s, tok/s=14738.0]
Inference: 49%|████▉ | 2912/5920 [44:23<18:25, 2.72it/s, tok/s=14739.1]
Inference: 49%|████▉ | 2913/5920 [44:24<14:18, 3.50it/s, tok/s=14739.1]
Inference: 49%|████▉ | 2913/5920 [44:24<14:18, 3.50it/s, tok/s=14737.9]
Inference: 49%|████▉ | 2914/5920 [44:24<12:22, 4.05it/s, tok/s=14737.9]
Inference: 49%|████▉ | 2914/5920 [44:24<12:22, 4.05it/s, tok/s=14738.0]
Inference: 49%|████▉ | 2915/5920 [44:24<12:21, 4.05it/s, tok/s=14738.7]
Inference: 49%|████▉ | 2916/5920 [44:24<12:54, 3.88it/s, tok/s=14738.7]
Inference: 49%|████▉ | 2916/5920 [44:24<12:54, 3.88it/s, tok/s=14736.9]
Inference: 49%|████▉ | 2917/5920 [44:25<15:07, 3.31it/s, tok/s=14736.9]
Inference: 49%|████▉ | 2917/5920 [44:25<15:07, 3.31it/s, tok/s=14734.9]
Inference: 49%|████▉ | 2918/5920 [44:25<17:22, 2.88it/s, tok/s=14734.9]
Inference: 49%|████▉ | 2918/5920 [44:25<17:22, 2.88it/s, tok/s=14733.9]
Inference: 49%|████▉ | 2919/5920 [44:25<17:21, 2.88it/s, tok/s=14734.4]
Inference: 49%|████▉ | 2920/5920 [44:25<11:47, 4.24it/s, tok/s=14734.4]
Inference: 49%|████▉ | 2920/5920 [44:25<11:47, 4.24it/s, tok/s=14735.9]
Inference: 49%|████▉ | 2921/5920 [44:25<11:47, 4.24it/s, tok/s=14736.1]
Inference: 49%|████▉ | 2922/5920 [44:26<11:55, 4.19it/s, tok/s=14736.1]
Inference: 49%|████▉ | 2922/5920 [44:26<11:55, 4.19it/s, tok/s=14741.4]
Inference: 49%|████▉ | 2923/5920 [44:27<20:57, 2.38it/s, tok/s=14741.4]
Inference: 49%|████▉ | 2923/5920 [44:27<20:57, 2.38it/s, tok/s=14735.9]
Inference: 49%|████▉ | 2924/5920 [44:27<19:13, 2.60it/s, tok/s=14735.9]
Inference: 49%|████▉ | 2924/5920 [44:27<19:13, 2.60it/s, tok/s=14735.0]
Inference: 49%|████▉ | 2925/5920 [44:28<26:16, 1.90it/s, tok/s=14735.0]
Inference: 49%|████▉ | 2925/5920 [44:28<26:16, 1.90it/s, tok/s=14731.1]
Inference: 49%|████▉ | 2926/5920 [44:29<26:28, 1.88it/s, tok/s=14731.1]
Inference: 49%|████▉ | 2926/5920 [44:29<26:28, 1.88it/s, tok/s=14728.7]
Inference: 49%|████▉ | 2927/5920 [44:29<24:22, 2.05it/s, tok/s=14728.7]
Inference: 49%|████▉ | 2927/5920 [44:29<24:22, 2.05it/s, tok/s=14729.0]
Inference: 49%|████▉ | 2928/5920 [44:32<55:59, 1.12s/it, tok/s=14729.0]
Inference: 49%|████▉ | 2928/5920 [44:32<55:59, 1.12s/it, tok/s=14714.5]
Inference: 49%|████▉ | 2929/5920 [44:32<44:46, 1.11it/s, tok/s=14714.5]
Inference: 49%|████▉ | 2929/5920 [44:32<44:46, 1.11it/s, tok/s=14713.6]
Inference: 49%|████▉ | 2930/5920 [44:32<33:45, 1.48it/s, tok/s=14713.6]
Inference: 49%|████▉ | 2930/5920 [44:32<33:45, 1.48it/s, tok/s=14714.5]
Inference: 50%|████▉ | 2931/5920 [44:33<28:09, 1.77it/s, tok/s=14714.5]
Inference: 50%|████▉ | 2931/5920 [44:33<28:09, 1.77it/s, tok/s=14716.7]
Inference: 50%|████▉ | 2932/5920 [44:34<36:40, 1.36it/s, tok/s=14716.7]
Inference: 50%|████▉ | 2932/5920 [44:34<36:40, 1.36it/s, tok/s=14711.2]
Inference: 50%|████▉ | 2933/5920 [44:34<30:00, 1.66it/s, tok/s=14711.2]
Inference: 50%|████▉ | 2933/5920 [44:34<30:00, 1.66it/s, tok/s=14715.4]
Inference: 50%|████▉ | 2934/5920 [44:35<27:21, 1.82it/s, tok/s=14715.4]
Inference: 50%|████▉ | 2934/5920 [44:35<27:21, 1.82it/s, tok/s=14713.6]
Inference: 50%|████▉ | 2935/5920 [44:35<27:21, 1.82it/s, tok/s=14713.8]
Inference: 50%|████▉ | 2936/5920 [44:35<16:54, 2.94it/s, tok/s=14713.8]
Inference: 50%|████▉ | 2936/5920 [44:35<16:54, 2.94it/s, tok/s=14714.3]
Inference: 50%|████▉ | 2937/5920 [44:35<15:07, 3.29it/s, tok/s=14714.3]
Inference: 50%|████▉ | 2937/5920 [44:35<15:07, 3.29it/s, tok/s=14724.0]
Inference: 50%|████▉ | 2938/5920 [44:35<16:26, 3.02it/s, tok/s=14724.0]
Inference: 50%|████▉ | 2938/5920 [44:35<16:26, 3.02it/s, tok/s=14722.3]
Inference: 50%|████▉ | 2939/5920 [44:36<14:12, 3.50it/s, tok/s=14722.3]
Inference: 50%|████▉ | 2939/5920 [44:36<14:12, 3.50it/s, tok/s=14728.8]
Inference: 50%|████▉ | 2940/5920 [44:36<12:32, 3.96it/s, tok/s=14728.8]
Inference: 50%|████▉ | 2940/5920 [44:36<12:32, 3.96it/s, tok/s=14728.6]
Inference: 50%|████▉ | 2941/5920 [44:37<28:40, 1.73it/s, tok/s=14728.6]
Inference: 50%|████▉ | 2941/5920 [44:37<28:40, 1.73it/s, tok/s=14721.7]
Inference: 50%|████▉ | 2942/5920 [44:38<27:24, 1.81it/s, tok/s=14721.7]
Inference: 50%|████▉ | 2942/5920 [44:38<27:24, 1.81it/s, tok/s=14719.9]
Inference: 50%|████▉ | 2943/5920 [44:38<23:04, 2.15it/s, tok/s=14719.9]
Inference: 50%|████▉ | 2943/5920 [44:38<23:04, 2.15it/s, tok/s=14719.0]
Inference: 50%|████▉ | 2944/5920 [44:38<18:08, 2.74it/s, tok/s=14719.0]
Inference: 50%|████▉ | 2944/5920 [44:38<18:08, 2.74it/s, tok/s=14720.2]
Inference: 50%|████▉ | 2945/5920 [44:38<16:18, 3.04it/s, tok/s=14720.2]
Inference: 50%|████▉ | 2945/5920 [44:38<16:18, 3.04it/s, tok/s=14727.4]
Inference: 50%|████▉ | 2946/5920 [44:40<35:17, 1.40it/s, tok/s=14727.4]
Inference: 50%|████▉ | 2946/5920 [44:40<35:17, 1.40it/s, tok/s=14728.7]
Inference: 50%|████▉ | 2947/5920 [44:40<27:33, 1.80it/s, tok/s=14728.7]
Inference: 50%|████▉ | 2947/5920 [44:40<27:33, 1.80it/s, tok/s=14733.8]
Inference: 50%|████▉ | 2948/5920 [44:41<37:24, 1.32it/s, tok/s=14733.8]
Inference: 50%|████▉ | 2948/5920 [44:41<37:24, 1.32it/s, tok/s=14727.5]
Inference: 50%|████▉ | 2949/5920 [44:41<30:27, 1.63it/s, tok/s=14727.5]
Inference: 50%|████▉ | 2949/5920 [44:41<30:27, 1.63it/s, tok/s=14727.2]
Inference: 50%|████▉ | 2950/5920 [44:42<27:55, 1.77it/s, tok/s=14727.2]
Inference: 50%|████▉ | 2950/5920 [44:42<27:55, 1.77it/s, tok/s=14725.9]
Inference: 50%|████▉ | 2951/5920 [44:42<23:27, 2.11it/s, tok/s=14725.9]
Inference: 50%|████▉ | 2951/5920 [44:42<23:27, 2.11it/s, tok/s=14728.9]
Inference: 50%|████▉ | 2952/5920 [44:43<22:39, 2.18it/s, tok/s=14728.9]
Inference: 50%|████▉ | 2952/5920 [44:43<22:39, 2.18it/s, tok/s=14727.2]
Inference: 50%|████▉ | 2953/5920 [44:44<39:26, 1.25it/s, tok/s=14727.2]
Inference: 50%|████▉ | 2953/5920 [44:44<39:26, 1.25it/s, tok/s=14720.3]
Inference: 50%|████▉ | 2954/5920 [44:44<30:05, 1.64it/s, tok/s=14720.3]
Inference: 50%|████▉ | 2954/5920 [44:44<30:05, 1.64it/s, tok/s=14720.3]
Inference: 50%|████▉ | 2955/5920 [44:45<26:11, 1.89it/s, tok/s=14720.3]
Inference: 50%|████▉ | 2955/5920 [44:45<26:11, 1.89it/s, tok/s=14718.9]
Inference: 50%|████▉ | 2956/5920 [44:45<21:27, 2.30it/s, tok/s=14718.9]
Inference: 50%|████▉ | 2956/5920 [44:45<21:27, 2.30it/s, tok/s=14725.6]
Inference: 50%|████▉ | 2957/5920 [44:45<19:03, 2.59it/s, tok/s=14725.6]
Inference: 50%|████▉ | 2957/5920 [44:45<19:03, 2.59it/s, tok/s=14726.7]
Inference: 50%|████▉ | 2958/5920 [44:45<15:55, 3.10it/s, tok/s=14726.7]
Inference: 50%|████▉ | 2958/5920 [44:45<15:55, 3.10it/s, tok/s=14726.4]
Inference: 50%|████▉ | 2959/5920 [44:45<15:54, 3.10it/s, tok/s=14732.8]
Inference: 50%|█████ | 2960/5920 [44:46<11:36, 4.25it/s, tok/s=14732.8]
Inference: 50%|█████ | 2960/5920 [44:46<11:36, 4.25it/s, tok/s=14732.4]
Inference: 50%|█████ | 2961/5920 [44:46<16:09, 3.05it/s, tok/s=14732.4]
Inference: 50%|█████ | 2961/5920 [44:46<16:09, 3.05it/s, tok/s=14736.6]
Inference: 50%|█████ | 2962/5920 [44:46<16:08, 3.05it/s, tok/s=14738.1]
Inference: 50%|█████ | 2963/5920 [44:47<12:13, 4.03it/s, tok/s=14738.1]
Inference: 50%|█████ | 2963/5920 [44:47<12:13, 4.03it/s, tok/s=14743.8]
Inference: 50%|█████ | 2964/5920 [44:47<11:11, 4.40it/s, tok/s=14743.8]
Inference: 50%|█████ | 2964/5920 [44:47<11:11, 4.40it/s, tok/s=14744.3]
Inference: 50%|█████ | 2965/5920 [44:47<17:20, 2.84it/s, tok/s=14744.3]
Inference: 50%|█████ | 2965/5920 [44:47<17:20, 2.84it/s, tok/s=14742.4]
Inference: 50%|█████ | 2966/5920 [44:48<15:21, 3.20it/s, tok/s=14742.4]
Inference: 50%|█████ | 2966/5920 [44:48<15:21, 3.20it/s, tok/s=14745.2]
Inference: 50%|█████ | 2967/5920 [44:48<16:07, 3.05it/s, tok/s=14745.2]
Inference: 50%|█████ | 2967/5920 [44:48<16:07, 3.05it/s, tok/s=14744.6]
Inference: 50%|█████ | 2968/5920 [44:48<14:36, 3.37it/s, tok/s=14744.6]
Inference: 50%|█████ | 2968/5920 [44:48<14:36, 3.37it/s, tok/s=14744.1]
Inference: 50%|█████ | 2969/5920 [44:49<17:22, 2.83it/s, tok/s=14744.1]
Inference: 50%|█████ | 2969/5920 [44:49<17:22, 2.83it/s, tok/s=14742.2]
Inference: 50%|█████ | 2970/5920 [44:50<24:13, 2.03it/s, tok/s=14742.2]
Inference: 50%|█████ | 2970/5920 [44:50<24:13, 2.03it/s, tok/s=14738.0]
Inference: 50%|█████ | 2971/5920 [44:50<21:44, 2.26it/s, tok/s=14738.0]
Inference: 50%|█████ | 2971/5920 [44:50<21:44, 2.26it/s, tok/s=14745.9]
Inference: 50%|█████ | 2972/5920 [44:51<27:31, 1.78it/s, tok/s=14745.9]
Inference: 50%|█████ | 2972/5920 [44:51<27:31, 1.78it/s, tok/s=14741.9]
Inference: 50%|█████ | 2973/5920 [44:52<40:45, 1.21it/s, tok/s=14741.9]
Inference: 50%|█████ | 2973/5920 [44:52<40:45, 1.21it/s, tok/s=14739.3]
Inference: 50%|█████ | 2974/5920 [44:52<40:44, 1.21it/s, tok/s=14740.0]
Inference: 50%|█████ | 2975/5920 [44:52<23:11, 2.12it/s, tok/s=14740.0]
Inference: 50%|█████ | 2975/5920 [44:52<23:11, 2.12it/s, tok/s=14745.4]
Inference: 50%|█████ | 2976/5920 [44:53<24:20, 2.02it/s, tok/s=14745.4]
Inference: 50%|█████ | 2976/5920 [44:53<24:20, 2.02it/s, tok/s=14743.2]
Inference: 50%|█████ | 2977/5920 [44:54<35:13, 1.39it/s, tok/s=14743.2]
Inference: 50%|█████ | 2977/5920 [44:54<35:13, 1.39it/s, tok/s=14737.3]
Inference: 50%|█████ | 2978/5920 [44:54<35:12, 1.39it/s, tok/s=14747.6]
Inference: 50%|█████ | 2979/5920 [44:55<23:34, 2.08it/s, tok/s=14747.6]
Inference: 50%|█████ | 2979/5920 [44:55<23:34, 2.08it/s, tok/s=14746.3]
Inference: 50%|█████ | 2980/5920 [44:55<23:33, 2.08it/s, tok/s=14751.7]
Inference: 50%|█████ | 2981/5920 [44:55<15:59, 3.06it/s, tok/s=14751.7]
Inference: 50%|█████ | 2981/5920 [44:55<15:59, 3.06it/s, tok/s=14751.8]
Inference: 50%|█████ | 2982/5920 [44:56<22:29, 2.18it/s, tok/s=14751.8]
Inference: 50%|█████ | 2982/5920 [44:56<22:29, 2.18it/s, tok/s=14747.0]
Inference: 50%|█████ | 2983/5920 [44:56<22:01, 2.22it/s, tok/s=14747.0]
Inference: 50%|█████ | 2983/5920 [44:56<22:01, 2.22it/s, tok/s=14745.6]
Inference: 50%|█████ | 2984/5920 [44:56<22:01, 2.22it/s, tok/s=14746.0]
Inference: 50%|█████ | 2985/5920 [44:56<16:59, 2.88it/s, tok/s=14746.0]
Inference: 50%|█████ | 2985/5920 [44:56<16:59, 2.88it/s, tok/s=14745.2]
Inference: 50%|█████ | 2986/5920 [44:57<17:23, 2.81it/s, tok/s=14745.2]
Inference: 50%|█████ | 2986/5920 [44:57<17:23, 2.81it/s, tok/s=14749.0]
Inference: 50%|█████ | 2987/5920 [44:57<17:22, 2.81it/s, tok/s=14749.8]
Inference: 50%|█████ | 2988/5920 [44:58<18:18, 2.67it/s, tok/s=14749.8]
Inference: 50%|█████ | 2988/5920 [44:58<18:18, 2.67it/s, tok/s=14746.4]
Inference: 50%|█████ | 2989/5920 [44:58<21:57, 2.22it/s, tok/s=14746.4]
Inference: 50%|█████ | 2989/5920 [44:58<21:57, 2.22it/s, tok/s=14743.1]
Inference: 51%|█████ | 2990/5920 [44:59<23:57, 2.04it/s, tok/s=14743.1]
Inference: 51%|█████ | 2990/5920 [44:59<23:57, 2.04it/s, tok/s=14742.4]
Inference: 51%|█████ | 2991/5920 [44:59<21:26, 2.28it/s, tok/s=14742.4]
Inference: 51%|█████ | 2991/5920 [44:59<21:26, 2.28it/s, tok/s=14747.9]
Inference: 51%|█████ | 2992/5920 [44:59<17:43, 2.75it/s, tok/s=14747.9]
Inference: 51%|█████ | 2992/5920 [44:59<17:43, 2.75it/s, tok/s=14752.4]
Inference: 51%|█████ | 2993/5920 [45:00<16:09, 3.02it/s, tok/s=14752.4]
Inference: 51%|█████ | 2993/5920 [45:00<16:09, 3.02it/s, tok/s=14751.9]
Inference: 51%|█████ | 2994/5920 [45:00<13:49, 3.53it/s, tok/s=14751.9]
Inference: 51%|█████ | 2994/5920 [45:00<13:49, 3.53it/s, tok/s=14751.6]
Inference: 51%|█████ | 2995/5920 [45:00<13:06, 3.72it/s, tok/s=14751.6]
Inference: 51%|█████ | 2995/5920 [45:00<13:06, 3.72it/s, tok/s=14755.6]
Inference: 51%|█████ | 2996/5920 [45:00<11:01, 4.42it/s, tok/s=14755.6]
Inference: 51%|█████ | 2996/5920 [45:00<11:01, 4.42it/s, tok/s=14756.2]
Inference: 51%|█████ | 2997/5920 [45:01<14:14, 3.42it/s, tok/s=14756.2]
Inference: 51%|█████ | 2997/5920 [45:01<14:14, 3.42it/s, tok/s=14759.8]
Inference: 51%|█████ | 2998/5920 [45:01<14:14, 3.42it/s, tok/s=14760.5]
Inference: 51%|█████ | 2999/5920 [45:01<09:59, 4.88it/s, tok/s=14760.5]
Inference: 51%|█████ | 2999/5920 [45:01<09:59, 4.88it/s, tok/s=14760.2]
Inference: 51%|█████ | 3000/5920 [45:01<12:54, 3.77it/s, tok/s=14760.2]
Inference: 51%|█████ | 3000/5920 [45:01<12:54, 3.77it/s, tok/s=14759.5]
Inference: 51%|█████ | 3001/5920 [45:01<12:53, 3.77it/s, tok/s=14760.2]
Inference: 51%|█████ | 3002/5920 [45:01<09:57, 4.88it/s, tok/s=14760.2]
Inference: 51%|█████ | 3002/5920 [45:01<09:57, 4.88it/s, tok/s=14760.1]
Inference: 51%|█████ | 3003/5920 [45:02<13:15, 3.67it/s, tok/s=14760.1]
Inference: 51%|█████ | 3003/5920 [45:02<13:15, 3.67it/s, tok/s=14763.7]
Inference: 51%|█████ | 3004/5920 [45:02<12:31, 3.88it/s, tok/s=14763.7]
Inference: 51%|█████ | 3004/5920 [45:02<12:31, 3.88it/s, tok/s=14764.2]
Inference: 51%|█████ | 3005/5920 [45:03<13:07, 3.70it/s, tok/s=14764.2]
Inference: 51%|█████ | 3005/5920 [45:03<13:07, 3.70it/s, tok/s=14763.1]
Inference: 51%|█████ | 3006/5920 [45:03<13:06, 3.70it/s, tok/s=14764.0]
Inference: 51%|█████ | 3007/5920 [45:04<20:00, 2.43it/s, tok/s=14764.0]
Inference: 51%|█████ | 3007/5920 [45:04<20:00, 2.43it/s, tok/s=14767.7]
Inference: 51%|█████ | 3008/5920 [45:04<17:47, 2.73it/s, tok/s=14767.7]
Inference: 51%|█████ | 3008/5920 [45:04<17:47, 2.73it/s, tok/s=14768.8]
Inference: 51%|█████ | 3009/5920 [45:04<14:43, 3.29it/s, tok/s=14768.8]
Inference: 51%|█████ | 3009/5920 [45:04<14:43, 3.29it/s, tok/s=14769.0]
Inference: 51%|█████ | 3010/5920 [45:04<13:10, 3.68it/s, tok/s=14769.0]
Inference: 51%|█████ | 3010/5920 [45:04<13:10, 3.68it/s, tok/s=14768.7]
Inference: 51%|█████ | 3011/5920 [45:05<15:02, 3.22it/s, tok/s=14768.7]
Inference: 51%|█████ | 3011/5920 [45:05<15:02, 3.22it/s, tok/s=14774.8]
Inference: 51%|█████ | 3012/5920 [45:06<25:54, 1.87it/s, tok/s=14774.8]
Inference: 51%|█████ | 3012/5920 [45:06<25:54, 1.87it/s, tok/s=14779.2]
Inference: 51%|█████ | 3013/5920 [45:06<23:48, 2.04it/s, tok/s=14779.2]
Inference: 51%|█████ | 3013/5920 [45:06<23:48, 2.04it/s, tok/s=14777.9]
Inference: 51%|█████ | 3014/5920 [45:06<18:20, 2.64it/s, tok/s=14777.9]
Inference: 51%|█████ | 3014/5920 [45:06<18:20, 2.64it/s, tok/s=14778.3]
Inference: 51%|█████ | 3015/5920 [45:06<14:24, 3.36it/s, tok/s=14778.3]
Inference: 51%|█████ | 3015/5920 [45:06<14:24, 3.36it/s, tok/s=14788.9]
Inference: 51%|█████ | 3016/5920 [45:08<28:49, 1.68it/s, tok/s=14788.9]
Inference: 51%|█████ | 3016/5920 [45:08<28:49, 1.68it/s, tok/s=14785.9]
Inference: 51%|█████ | 3017/5920 [45:08<24:42, 1.96it/s, tok/s=14785.9]
Inference: 51%|█████ | 3017/5920 [45:08<24:42, 1.96it/s, tok/s=14785.0]
Inference: 51%|█████ | 3018/5920 [45:09<31:52, 1.52it/s, tok/s=14785.0]
Inference: 51%|█████ | 3018/5920 [45:09<31:52, 1.52it/s, tok/s=14786.0]
Inference: 51%|█████ | 3019/5920 [45:09<27:12, 1.78it/s, tok/s=14786.0]
Inference: 51%|█████ | 3019/5920 [45:09<27:12, 1.78it/s, tok/s=14784.8]
Inference: 51%|█████ | 3020/5920 [45:10<23:00, 2.10it/s, tok/s=14784.8]
Inference: 51%|█████ | 3020/5920 [45:10<23:00, 2.10it/s, tok/s=14794.4]
Inference: 51%|█████ | 3021/5920 [45:10<18:14, 2.65it/s, tok/s=14794.4]
Inference: 51%|█████ | 3021/5920 [45:10<18:14, 2.65it/s, tok/s=14795.9]
Inference: 51%|█████ | 3022/5920 [45:10<18:13, 2.65it/s, tok/s=14797.7]
Inference: 51%|█████ | 3023/5920 [45:11<20:02, 2.41it/s, tok/s=14797.7]
Inference: 51%|█████ | 3023/5920 [45:11<20:02, 2.41it/s, tok/s=14803.0]
Inference: 51%|█████ | 3024/5920 [45:11<20:02, 2.41it/s, tok/s=14803.2]
Inference: 51%|█████ | 3025/5920 [45:12<24:34, 1.96it/s, tok/s=14803.2]
Inference: 51%|█████ | 3025/5920 [45:12<24:34, 1.96it/s, tok/s=14797.4]
Inference: 51%|█████ | 3026/5920 [45:12<24:33, 1.96it/s, tok/s=14799.0]
Inference: 51%|█████ | 3027/5920 [45:12<20:23, 2.36it/s, tok/s=14799.0]
Inference: 51%|█████ | 3027/5920 [45:12<20:23, 2.36it/s, tok/s=14798.3]
Inference: 51%|█████ | 3028/5920 [45:13<20:23, 2.36it/s, tok/s=14805.5]
Inference: 51%|█████ | 3029/5920 [45:13<15:03, 3.20it/s, tok/s=14805.5]
Inference: 51%|█████ | 3029/5920 [45:13<15:03, 3.20it/s, tok/s=14806.3]
Inference: 51%|█████ | 3030/5920 [45:13<14:22, 3.35it/s, tok/s=14806.3]
Inference: 51%|█████ | 3030/5920 [45:13<14:22, 3.35it/s, tok/s=14805.5]
Inference: 51%|█████ | 3031/5920 [45:13<14:22, 3.35it/s, tok/s=14806.4]
Inference: 51%|█████ | 3032/5920 [45:13<12:16, 3.92it/s, tok/s=14806.4]
Inference: 51%|█████ | 3032/5920 [45:13<12:16, 3.92it/s, tok/s=14806.3]
Inference: 51%|█████ | 3033/5920 [45:14<18:20, 2.62it/s, tok/s=14806.3]
Inference: 51%|█████ | 3033/5920 [45:14<18:20, 2.62it/s, tok/s=14802.4]
Inference: 51%|█████▏ | 3034/5920 [45:14<16:41, 2.88it/s, tok/s=14802.4]
Inference: 51%|█████▏ | 3034/5920 [45:14<16:41, 2.88it/s, tok/s=14802.2]
Inference: 51%|█████▏ | 3035/5920 [45:16<28:26, 1.69it/s, tok/s=14802.2]
Inference: 51%|█████▏ | 3035/5920 [45:16<28:26, 1.69it/s, tok/s=14795.4]
Inference: 51%|█████▏ | 3036/5920 [45:16<28:26, 1.69it/s, tok/s=14796.2]
Inference: 51%|█████▏ | 3037/5920 [45:16<20:31, 2.34it/s, tok/s=14796.2]
Inference: 51%|█████▏ | 3037/5920 [45:16<20:31, 2.34it/s, tok/s=14805.3]
Inference: 51%|█████▏ | 3038/5920 [45:16<17:58, 2.67it/s, tok/s=14805.3]
Inference: 51%|█████▏ | 3038/5920 [45:16<17:58, 2.67it/s, tok/s=14805.7]
Inference: 51%|█████▏ | 3039/5920 [45:16<17:58, 2.67it/s, tok/s=14805.9]
Inference: 51%|█████▏ | 3040/5920 [45:17<13:29, 3.56it/s, tok/s=14805.9]
Inference: 51%|█████▏ | 3040/5920 [45:17<13:29, 3.56it/s, tok/s=14805.4]
Inference: 51%|█████▏ | 3041/5920 [45:17<11:49, 4.06it/s, tok/s=14805.4]
Inference: 51%|█████▏ | 3041/5920 [45:17<11:49, 4.06it/s, tok/s=14808.9]
Inference: 51%|█████▏ | 3042/5920 [45:17<17:07, 2.80it/s, tok/s=14808.9]
Inference: 51%|█████▏ | 3042/5920 [45:17<17:07, 2.80it/s, tok/s=14805.4]
Inference: 51%|█████▏ | 3043/5920 [45:18<18:55, 2.53it/s, tok/s=14805.4]
Inference: 51%|█████▏ | 3043/5920 [45:18<18:55, 2.53it/s, tok/s=14803.6]
Inference: 51%|█████▏ | 3044/5920 [45:19<23:31, 2.04it/s, tok/s=14803.6]
Inference: 51%|█████▏ | 3044/5920 [45:19<23:31, 2.04it/s, tok/s=14803.6]
Inference: 51%|█████▏ | 3045/5920 [45:19<18:41, 2.56it/s, tok/s=14803.6]
Inference: 51%|█████▏ | 3045/5920 [45:19<18:41, 2.56it/s, tok/s=14809.1]
Inference: 51%|█████▏ | 3046/5920 [45:19<18:47, 2.55it/s, tok/s=14809.1]
Inference: 51%|█████▏ | 3046/5920 [45:19<18:47, 2.55it/s, tok/s=14808.7]
Inference: 51%|█████▏ | 3047/5920 [45:19<15:56, 3.00it/s, tok/s=14808.7]
Inference: 51%|█████▏ | 3047/5920 [45:19<15:56, 3.00it/s, tok/s=14808.0]
Inference: 51%|█████▏ | 3048/5920 [45:19<12:42, 3.76it/s, tok/s=14808.0]
Inference: 51%|█████▏ | 3048/5920 [45:19<12:42, 3.76it/s, tok/s=14807.9]
Inference: 52%|█████▏ | 3049/5920 [45:20<12:08, 3.94it/s, tok/s=14807.9]
Inference: 52%|█████▏ | 3049/5920 [45:20<12:08, 3.94it/s, tok/s=14807.6]
Inference: 52%|█████▏ | 3050/5920 [45:21<30:37, 1.56it/s, tok/s=14807.6]
Inference: 52%|█████▏ | 3050/5920 [45:21<30:37, 1.56it/s, tok/s=14800.0]
Inference: 52%|█████▏ | 3051/5920 [45:21<30:36, 1.56it/s, tok/s=14800.7]
Inference: 52%|█████▏ | 3052/5920 [45:21<17:57, 2.66it/s, tok/s=14800.7]
Inference: 52%|█████▏ | 3052/5920 [45:21<17:57, 2.66it/s, tok/s=14800.9]
Inference: 52%|█████▏ | 3053/5920 [45:22<16:36, 2.88it/s, tok/s=14800.9]
Inference: 52%|█████▏ | 3053/5920 [45:22<16:36, 2.88it/s, tok/s=14806.3]
Inference: 52%|█████▏ | 3054/5920 [45:22<18:37, 2.56it/s, tok/s=14806.3]
Inference: 52%|█████▏ | 3054/5920 [45:22<18:37, 2.56it/s, tok/s=14805.9]
Inference: 52%|█████▏ | 3055/5920 [45:23<18:34, 2.57it/s, tok/s=14805.9]
Inference: 52%|█████▏ | 3055/5920 [45:23<18:34, 2.57it/s, tok/s=14809.2]
Inference: 52%|█████▏ | 3056/5920 [45:23<17:38, 2.71it/s, tok/s=14809.2]
Inference: 52%|█████▏ | 3056/5920 [45:23<17:38, 2.71it/s, tok/s=14808.3]
Inference: 52%|█████▏ | 3057/5920 [45:25<35:36, 1.34it/s, tok/s=14808.3]
Inference: 52%|█████▏ | 3057/5920 [45:25<35:36, 1.34it/s, tok/s=14800.9]
Inference: 52%|█████▏ | 3058/5920 [45:25<32:04, 1.49it/s, tok/s=14800.9]
Inference: 52%|█████▏ | 3058/5920 [45:25<32:04, 1.49it/s, tok/s=14800.4]
Inference: 52%|█████▏ | 3059/5920 [45:26<32:40, 1.46it/s, tok/s=14800.4]
Inference: 52%|█████▏ | 3059/5920 [45:26<32:40, 1.46it/s, tok/s=14797.4]
Inference: 52%|█████▏ | 3060/5920 [45:26<25:04, 1.90it/s, tok/s=14797.4]
Inference: 52%|█████▏ | 3060/5920 [45:26<25:04, 1.90it/s, tok/s=14801.1]
Inference: 52%|█████▏ | 3061/5920 [45:26<23:31, 2.03it/s, tok/s=14801.1]
Inference: 52%|█████▏ | 3061/5920 [45:26<23:31, 2.03it/s, tok/s=14801.9]
Inference: 52%|█████▏ | 3062/5920 [45:28<38:35, 1.23it/s, tok/s=14801.9]
Inference: 52%|█████▏ | 3062/5920 [45:28<38:35, 1.23it/s, tok/s=14795.4]
Inference: 52%|█████▏ | 3063/5920 [45:29<37:37, 1.27it/s, tok/s=14795.4]
Inference: 52%|█████▏ | 3063/5920 [45:29<37:37, 1.27it/s, tok/s=14793.9]
Inference: 52%|█████▏ | 3064/5920 [45:29<37:37, 1.27it/s, tok/s=14794.2]
Inference: 52%|█████▏ | 3065/5920 [45:29<25:35, 1.86it/s, tok/s=14794.2]
Inference: 52%|█████▏ | 3065/5920 [45:29<25:35, 1.86it/s, tok/s=14800.9]
Inference: 52%|█████▏ | 3066/5920 [45:29<20:27, 2.32it/s, tok/s=14800.9]
Inference: 52%|█████▏ | 3066/5920 [45:29<20:27, 2.32it/s, tok/s=14801.5]
Inference: 52%|█████▏ | 3067/5920 [45:30<21:28, 2.21it/s, tok/s=14801.5]
Inference: 52%|█████▏ | 3067/5920 [45:30<21:28, 2.21it/s, tok/s=14799.4]
Inference: 52%|█████▏ | 3068/5920 [45:30<20:30, 2.32it/s, tok/s=14799.4]
Inference: 52%|█████▏ | 3068/5920 [45:30<20:30, 2.32it/s, tok/s=14800.2]
Inference: 52%|█████▏ | 3069/5920 [45:30<20:29, 2.32it/s, tok/s=14800.3]
Inference: 52%|█████▏ | 3070/5920 [45:31<16:50, 2.82it/s, tok/s=14800.3]
Inference: 52%|█████▏ | 3070/5920 [45:31<16:50, 2.82it/s, tok/s=14798.4]
Inference: 52%|█████▏ | 3071/5920 [45:31<21:09, 2.24it/s, tok/s=14798.4]
Inference: 52%|█████▏ | 3071/5920 [45:31<21:09, 2.24it/s, tok/s=14795.6]
Inference: 52%|█████▏ | 3072/5920 [45:31<21:08, 2.24it/s, tok/s=14805.8]
Inference: 52%|█████▏ | 3073/5920 [45:32<14:33, 3.26it/s, tok/s=14805.8]
Inference: 52%|█████▏ | 3073/5920 [45:32<14:33, 3.26it/s, tok/s=14805.3]
Inference: 52%|█████▏ | 3074/5920 [45:32<14:32, 3.26it/s, tok/s=14806.3]
Inference: 52%|█████▏ | 3075/5920 [45:32<12:22, 3.83it/s, tok/s=14806.3]
Inference: 52%|█████▏ | 3075/5920 [45:32<12:22, 3.83it/s, tok/s=14805.1]
Inference: 52%|█████▏ | 3076/5920 [45:32<11:26, 4.14it/s, tok/s=14805.1]
Inference: 52%|█████▏ | 3076/5920 [45:32<11:26, 4.14it/s, tok/s=14814.7]
Inference: 52%|█████▏ | 3077/5920 [45:32<12:06, 3.92it/s, tok/s=14814.7]
Inference: 52%|█████▏ | 3077/5920 [45:32<12:06, 3.92it/s, tok/s=14813.5]
Inference: 52%|█████▏ | 3078/5920 [45:33<13:24, 3.53it/s, tok/s=14813.5]
Inference: 52%|█████▏ | 3078/5920 [45:33<13:24, 3.53it/s, tok/s=14817.2]
Inference: 52%|█████▏ | 3079/5920 [45:33<13:24, 3.53it/s, tok/s=14817.8]
Inference: 52%|█████▏ | 3080/5920 [45:33<09:33, 4.95it/s, tok/s=14817.8]
Inference: 52%|█████▏ | 3080/5920 [45:33<09:33, 4.95it/s, tok/s=14827.5]
Inference: 52%|█████▏ | 3081/5920 [45:34<16:33, 2.86it/s, tok/s=14827.5]
Inference: 52%|█████▏ | 3081/5920 [45:34<16:33, 2.86it/s, tok/s=14829.0]
Inference: 52%|█████▏ | 3082/5920 [45:34<19:04, 2.48it/s, tok/s=14829.0]
Inference: 52%|█████▏ | 3082/5920 [45:34<19:04, 2.48it/s, tok/s=14834.5]
Inference: 52%|█████▏ | 3083/5920 [45:35<21:32, 2.19it/s, tok/s=14834.5]
Inference: 52%|█████▏ | 3083/5920 [45:35<21:32, 2.19it/s, tok/s=14832.2]
Inference: 52%|█████▏ | 3084/5920 [45:35<17:57, 2.63it/s, tok/s=14832.2]
Inference: 52%|█████▏ | 3084/5920 [45:35<17:57, 2.63it/s, tok/s=14832.1]
Inference: 52%|█████▏ | 3085/5920 [45:35<16:20, 2.89it/s, tok/s=14832.1]
Inference: 52%|█████▏ | 3085/5920 [45:35<16:20, 2.89it/s, tok/s=14831.2]
Inference: 52%|█████▏ | 3086/5920 [45:35<16:20, 2.89it/s, tok/s=14831.4]
Inference: 52%|█████▏ | 3087/5920 [45:36<10:44, 4.40it/s, tok/s=14831.4]
Inference: 52%|█████▏ | 3087/5920 [45:36<10:44, 4.40it/s, tok/s=14832.5]
Inference: 52%|█████▏ | 3088/5920 [45:36<10:43, 4.40it/s, tok/s=14833.2]
Inference: 52%|█████▏ | 3089/5920 [45:36<08:10, 5.77it/s, tok/s=14833.2]
Inference: 52%|█████▏ | 3089/5920 [45:36<08:10, 5.77it/s, tok/s=14839.1]
Inference: 52%|█████▏ | 3090/5920 [45:36<08:10, 5.77it/s, tok/s=14845.1]
Inference: 52%|█████▏ | 3091/5920 [45:36<08:31, 5.53it/s, tok/s=14845.1]
Inference: 52%|█████▏ | 3091/5920 [45:36<08:31, 5.53it/s, tok/s=14845.2]
Inference: 52%|█████▏ | 3092/5920 [45:37<11:02, 4.27it/s, tok/s=14845.2]
Inference: 52%|█████▏ | 3092/5920 [45:37<11:02, 4.27it/s, tok/s=14853.9]
Inference: 52%|█████▏ | 3093/5920 [45:38<22:37, 2.08it/s, tok/s=14853.9]
Inference: 52%|█████▏ | 3093/5920 [45:38<22:37, 2.08it/s, tok/s=14847.2]
Inference: 52%|█████▏ | 3094/5920 [45:38<22:36, 2.08it/s, tok/s=14847.7]
Inference: 52%|█████▏ | 3095/5920 [45:39<20:11, 2.33it/s, tok/s=14847.7]
Inference: 52%|█████▏ | 3095/5920 [45:39<20:11, 2.33it/s, tok/s=14844.6]
Inference: 52%|█████▏ | 3096/5920 [45:39<22:17, 2.11it/s, tok/s=14844.6]
Inference: 52%|█████▏ | 3096/5920 [45:39<22:17, 2.11it/s, tok/s=14849.9]
Inference: 52%|█████▏ | 3097/5920 [45:40<26:21, 1.79it/s, tok/s=14849.9]
Inference: 52%|█████▏ | 3097/5920 [45:40<26:21, 1.79it/s, tok/s=14846.2]
Inference: 52%|█████▏ | 3098/5920 [45:40<26:20, 1.79it/s, tok/s=14846.7]
Inference: 52%|█████▏ | 3099/5920 [45:40<17:27, 2.69it/s, tok/s=14846.7]
Inference: 52%|█████▏ | 3099/5920 [45:40<17:27, 2.69it/s, tok/s=14846.7]
Inference: 52%|█████▏ | 3100/5920 [45:41<26:17, 1.79it/s, tok/s=14846.7]
Inference: 52%|█████▏ | 3100/5920 [45:41<26:17, 1.79it/s, tok/s=14840.8]
Inference: 52%|█████▏ | 3101/5920 [45:42<22:04, 2.13it/s, tok/s=14840.8]
Inference: 52%|█████▏ | 3101/5920 [45:42<22:04, 2.13it/s, tok/s=14840.3]
Inference: 52%|█████▏ | 3102/5920 [45:42<18:09, 2.59it/s, tok/s=14840.3]
Inference: 52%|█████▏ | 3102/5920 [45:42<18:09, 2.59it/s, tok/s=14846.3]
Inference: 52%|█████▏ | 3103/5920 [45:42<18:53, 2.49it/s, tok/s=14846.3]
Inference: 52%|█████▏ | 3103/5920 [45:42<18:53, 2.49it/s, tok/s=14851.6]
Inference: 52%|█████▏ | 3104/5920 [45:42<15:11, 3.09it/s, tok/s=14851.6]
Inference: 52%|█████▏ | 3104/5920 [45:42<15:11, 3.09it/s, tok/s=14852.5]
Inference: 52%|█████▏ | 3105/5920 [45:43<22:01, 2.13it/s, tok/s=14852.5]
Inference: 52%|█████▏ | 3105/5920 [45:43<22:01, 2.13it/s, tok/s=14849.8]
Inference: 52%|█████▏ | 3106/5920 [45:44<28:01, 1.67it/s, tok/s=14849.8]
Inference: 52%|█████▏ | 3106/5920 [45:44<28:01, 1.67it/s, tok/s=14846.9]
Inference: 52%|█████▏ | 3107/5920 [45:45<25:27, 1.84it/s, tok/s=14846.9]
Inference: 52%|█████▏ | 3107/5920 [45:45<25:27, 1.84it/s, tok/s=14845.6]
Inference: 52%|█████▎ | 3108/5920 [45:45<23:04, 2.03it/s, tok/s=14845.6]
Inference: 52%|█████▎ | 3108/5920 [45:45<23:04, 2.03it/s, tok/s=14844.7]
Inference: 53%|█████▎ | 3109/5920 [45:45<20:13, 2.32it/s, tok/s=14844.7]
Inference: 53%|█████▎ | 3109/5920 [45:45<20:13, 2.32it/s, tok/s=14852.0]
Inference: 53%|█████▎ | 3110/5920 [45:46<19:57, 2.35it/s, tok/s=14852.0]
Inference: 53%|█████▎ | 3110/5920 [45:46<19:57, 2.35it/s, tok/s=14850.3]
Inference: 53%|█████▎ | 3111/5920 [45:46<16:20, 2.87it/s, tok/s=14850.3]
Inference: 53%|█████▎ | 3111/5920 [45:46<16:20, 2.87it/s, tok/s=14851.2]
Inference: 53%|█████▎ | 3112/5920 [45:46<16:19, 2.87it/s, tok/s=14851.5]
Inference: 53%|█████▎ | 3113/5920 [45:47<23:25, 2.00it/s, tok/s=14851.5]
Inference: 53%|█████▎ | 3113/5920 [45:47<23:25, 2.00it/s, tok/s=14845.4]
Inference: 53%|█████▎ | 3114/5920 [45:47<21:41, 2.16it/s, tok/s=14845.4]
Inference: 53%|█████▎ | 3114/5920 [45:47<21:41, 2.16it/s, tok/s=14844.4]
Inference: 53%|█████▎ | 3115/5920 [45:48<21:41, 2.16it/s, tok/s=14850.8]
Inference: 53%|█████▎ | 3116/5920 [45:48<21:40, 2.16it/s, tok/s=14858.0]
Inference: 53%|█████▎ | 3117/5920 [45:48<14:11, 3.29it/s, tok/s=14858.0]
Inference: 53%|█████▎ | 3117/5920 [45:48<14:11, 3.29it/s, tok/s=14861.9]
Inference: 53%|█████▎ | 3118/5920 [45:48<13:09, 3.55it/s, tok/s=14861.9]
Inference: 53%|█████▎ | 3118/5920 [45:48<13:09, 3.55it/s, tok/s=14863.0]
Inference: 53%|█████▎ | 3119/5920 [45:48<14:00, 3.33it/s, tok/s=14863.0]
Inference: 53%|█████▎ | 3119/5920 [45:48<14:00, 3.33it/s, tok/s=14862.5]
Inference: 53%|█████▎ | 3120/5920 [45:49<12:49, 3.64it/s, tok/s=14862.5]
Inference: 53%|█████▎ | 3120/5920 [45:49<12:49, 3.64it/s, tok/s=14862.4]
Inference: 53%|█████▎ | 3121/5920 [45:49<11:17, 4.13it/s, tok/s=14862.4]
Inference: 53%|█████▎ | 3121/5920 [45:49<11:17, 4.13it/s, tok/s=14866.9]
Inference: 53%|█████▎ | 3122/5920 [45:49<13:03, 3.57it/s, tok/s=14866.9]
Inference: 53%|█████▎ | 3122/5920 [45:49<13:03, 3.57it/s, tok/s=14867.8]
Inference: 53%|█████▎ | 3123/5920 [45:49<10:48, 4.31it/s, tok/s=14867.8]
Inference: 53%|█████▎ | 3123/5920 [45:49<10:48, 4.31it/s, tok/s=14874.8]
Inference: 53%|█████▎ | 3124/5920 [45:49<09:53, 4.71it/s, tok/s=14874.8]
Inference: 53%|█████▎ | 3124/5920 [45:49<09:53, 4.71it/s, tok/s=14874.9]
Inference: 53%|█████▎ | 3125/5920 [45:50<09:53, 4.71it/s, tok/s=14875.3]
Inference: 53%|█████▎ | 3126/5920 [45:50<13:04, 3.56it/s, tok/s=14875.3]
Inference: 53%|█████▎ | 3126/5920 [45:50<13:04, 3.56it/s, tok/s=14877.6]
Inference: 53%|█████▎ | 3127/5920 [45:50<13:04, 3.56it/s, tok/s=14877.6]
Inference: 53%|█████▎ | 3128/5920 [45:52<21:53, 2.13it/s, tok/s=14877.6]
Inference: 53%|█████▎ | 3128/5920 [45:52<21:53, 2.13it/s, tok/s=14870.2]
Inference: 53%|█████▎ | 3129/5920 [45:52<21:35, 2.15it/s, tok/s=14870.2]
Inference: 53%|█████▎ | 3129/5920 [45:52<21:35, 2.15it/s, tok/s=14868.9]
Inference: 53%|█████▎ | 3130/5920 [45:54<33:13, 1.40it/s, tok/s=14868.9]
Inference: 53%|█████▎ | 3130/5920 [45:54<33:13, 1.40it/s, tok/s=14867.2]
Inference: 53%|█████▎ | 3131/5920 [45:54<29:15, 1.59it/s, tok/s=14867.2]
Inference: 53%|█████▎ | 3131/5920 [45:54<29:15, 1.59it/s, tok/s=14865.5]
Inference: 53%|█████▎ | 3132/5920 [45:54<24:24, 1.90it/s, tok/s=14865.5]
Inference: 53%|█████▎ | 3132/5920 [45:54<24:24, 1.90it/s, tok/s=14865.5]
Inference: 53%|█████▎ | 3133/5920 [45:55<28:32, 1.63it/s, tok/s=14865.5]
Inference: 53%|█████▎ | 3133/5920 [45:55<28:32, 1.63it/s, tok/s=14861.6]
Inference: 53%|█████▎ | 3134/5920 [45:55<21:47, 2.13it/s, tok/s=14861.6]
Inference: 53%|█████▎ | 3134/5920 [45:55<21:47, 2.13it/s, tok/s=14868.0]
Inference: 53%|█████▎ | 3135/5920 [45:55<21:47, 2.13it/s, tok/s=14868.1]
Inference: 53%|█████▎ | 3136/5920 [45:55<21:46, 2.13it/s, tok/s=14868.7]
Inference: 53%|█████▎ | 3137/5920 [45:56<12:13, 3.79it/s, tok/s=14868.7]
Inference: 53%|█████▎ | 3137/5920 [45:56<12:13, 3.79it/s, tok/s=14868.7]
Inference: 53%|█████▎ | 3138/5920 [45:56<12:13, 3.79it/s, tok/s=14868.5]
Inference: 53%|█████▎ | 3139/5920 [45:56<11:13, 4.13it/s, tok/s=14868.5]
Inference: 53%|█████▎ | 3139/5920 [45:56<11:13, 4.13it/s, tok/s=14867.3]
Inference: 53%|█████▎ | 3140/5920 [45:56<11:13, 4.13it/s, tok/s=14868.3]
Inference: 53%|█████▎ | 3141/5920 [45:56<09:57, 4.65it/s, tok/s=14868.3]
Inference: 53%|█████▎ | 3141/5920 [45:56<09:57, 4.65it/s, tok/s=14867.7]
Inference: 53%|█████▎ | 3142/5920 [45:57<10:52, 4.26it/s, tok/s=14867.7]
Inference: 53%|█████▎ | 3142/5920 [45:57<10:52, 4.26it/s, tok/s=14866.5]
Inference: 53%|█████▎ | 3143/5920 [45:57<11:03, 4.18it/s, tok/s=14866.5]
Inference: 53%|█████▎ | 3143/5920 [45:57<11:03, 4.18it/s, tok/s=14865.8]
Inference: 53%|█████▎ | 3144/5920 [45:57<11:03, 4.18it/s, tok/s=14875.7]
Inference: 53%|█████▎ | 3145/5920 [45:57<09:24, 4.91it/s, tok/s=14875.7]
Inference: 53%|█████▎ | 3145/5920 [45:57<09:24, 4.91it/s, tok/s=14885.2]
Inference: 53%|█████▎ | 3146/5920 [45:57<09:24, 4.91it/s, tok/s=14886.3]
Inference: 53%|█████▎ | 3147/5920 [45:57<07:36, 6.08it/s, tok/s=14886.3]
Inference: 53%|█████▎ | 3147/5920 [45:57<07:36, 6.08it/s, tok/s=14886.3]
Inference: 53%|█████▎ | 3148/5920 [45:58<10:40, 4.33it/s, tok/s=14886.3]
Inference: 53%|█████▎ | 3148/5920 [45:58<10:40, 4.33it/s, tok/s=14885.0]
Inference: 53%|█████▎ | 3149/5920 [45:59<24:56, 1.85it/s, tok/s=14885.0]
Inference: 53%|█████▎ | 3149/5920 [45:59<24:56, 1.85it/s, tok/s=14877.2]
Inference: 53%|█████▎ | 3150/5920 [46:00<24:20, 1.90it/s, tok/s=14877.2]
Inference: 53%|█████▎ | 3150/5920 [46:00<24:20, 1.90it/s, tok/s=14876.2]
Inference: 53%|█████▎ | 3151/5920 [46:00<19:35, 2.36it/s, tok/s=14876.2]
Inference: 53%|█████▎ | 3151/5920 [46:00<19:35, 2.36it/s, tok/s=14876.5]
Inference: 53%|█████▎ | 3152/5920 [46:01<22:34, 2.04it/s, tok/s=14876.5]
Inference: 53%|█████▎ | 3152/5920 [46:01<22:34, 2.04it/s, tok/s=14873.8]
Inference: 53%|█████▎ | 3153/5920 [46:01<21:38, 2.13it/s, tok/s=14873.8]
Inference: 53%|█████▎ | 3153/5920 [46:01<21:38, 2.13it/s, tok/s=14879.4]
Inference: 53%|█████▎ | 3154/5920 [46:02<20:39, 2.23it/s, tok/s=14879.4]
Inference: 53%|█████▎ | 3154/5920 [46:02<20:39, 2.23it/s, tok/s=14884.6]
Inference: 53%|█████▎ | 3155/5920 [46:02<20:39, 2.23it/s, tok/s=14886.1]
Inference: 53%|█████▎ | 3156/5920 [46:02<14:44, 3.13it/s, tok/s=14886.1]
Inference: 53%|█████▎ | 3156/5920 [46:02<14:44, 3.13it/s, tok/s=14885.6]
Inference: 53%|█████▎ | 3157/5920 [46:02<14:43, 3.13it/s, tok/s=14885.7]
Inference: 53%|█████▎ | 3158/5920 [46:02<11:18, 4.07it/s, tok/s=14885.7]
Inference: 53%|█████▎ | 3158/5920 [46:02<11:18, 4.07it/s, tok/s=14895.4]
Inference: 53%|█████▎ | 3159/5920 [46:02<11:33, 3.98it/s, tok/s=14895.4]
Inference: 53%|█████▎ | 3159/5920 [46:02<11:33, 3.98it/s, tok/s=14895.5]
Inference: 53%|█████▎ | 3160/5920 [46:03<13:34, 3.39it/s, tok/s=14895.5]
Inference: 53%|█████▎ | 3160/5920 [46:03<13:34, 3.39it/s, tok/s=14893.5]
Inference: 53%|█████▎ | 3161/5920 [46:03<13:33, 3.39it/s, tok/s=14894.2]
Inference: 53%|█████▎ | 3162/5920 [46:03<11:25, 4.02it/s, tok/s=14894.2]
Inference: 53%|█████▎ | 3162/5920 [46:03<11:25, 4.02it/s, tok/s=14893.7]
Inference: 53%|█████▎ | 3163/5920 [46:03<11:25, 4.02it/s, tok/s=14894.1]
Inference: 53%|█████▎ | 3164/5920 [46:04<14:19, 3.21it/s, tok/s=14894.1]
Inference: 53%|█████▎ | 3164/5920 [46:04<14:19, 3.21it/s, tok/s=14890.3]
Inference: 53%|█████▎ | 3165/5920 [46:04<13:20, 3.44it/s, tok/s=14890.3]
Inference: 53%|█████▎ | 3165/5920 [46:04<13:20, 3.44it/s, tok/s=14889.8]
Inference: 53%|█████▎ | 3166/5920 [46:05<13:51, 3.31it/s, tok/s=14889.8]
Inference: 53%|█████▎ | 3166/5920 [46:05<13:51, 3.31it/s, tok/s=14896.9]
Inference: 53%|█████▎ | 3167/5920 [46:05<13:51, 3.31it/s, tok/s=14907.8]
Inference: 54%|█████▎ | 3168/5920 [46:06<17:17, 2.65it/s, tok/s=14907.8]
Inference: 54%|█████▎ | 3168/5920 [46:06<17:17, 2.65it/s, tok/s=14905.9]
Inference: 54%|█████▎ | 3169/5920 [46:06<17:17, 2.65it/s, tok/s=14906.1]
Inference: 54%|█████▎ | 3170/5920 [46:06<14:15, 3.21it/s, tok/s=14906.1]
Inference: 54%|█████▎ | 3170/5920 [46:06<14:15, 3.21it/s, tok/s=14905.1]
Inference: 54%|█████▎ | 3171/5920 [46:06<16:15, 2.82it/s, tok/s=14905.1]
Inference: 54%|█████▎ | 3171/5920 [46:06<16:15, 2.82it/s, tok/s=14902.8]
Inference: 54%|█████▎ | 3172/5920 [46:08<25:34, 1.79it/s, tok/s=14902.8]
Inference: 54%|█████▎ | 3172/5920 [46:08<25:34, 1.79it/s, tok/s=14896.8]
Inference: 54%|█████▎ | 3173/5920 [46:09<33:46, 1.36it/s, tok/s=14896.8]
Inference: 54%|█████▎ | 3173/5920 [46:09<33:46, 1.36it/s, tok/s=14890.2]
Inference: 54%|█████▎ | 3174/5920 [46:09<26:36, 1.72it/s, tok/s=14890.2]
Inference: 54%|█████▎ | 3174/5920 [46:09<26:36, 1.72it/s, tok/s=14890.0]
Inference: 54%|█████▎ | 3175/5920 [46:10<27:57, 1.64it/s, tok/s=14890.0]
Inference: 54%|█████▎ | 3175/5920 [46:10<27:57, 1.64it/s, tok/s=14887.1]
Inference: 54%|█████▎ | 3176/5920 [46:10<27:57, 1.64it/s, tok/s=14887.5]
Inference: 54%|█████▎ | 3177/5920 [46:11<23:20, 1.96it/s, tok/s=14887.5]
Inference: 54%|█████▎ | 3177/5920 [46:11<23:20, 1.96it/s, tok/s=14884.4]
Inference: 54%|█████▎ | 3178/5920 [46:11<19:40, 2.32it/s, tok/s=14884.4]
Inference: 54%|█████▎ | 3178/5920 [46:11<19:40, 2.32it/s, tok/s=14885.4]
Inference: 54%|█████▎ | 3179/5920 [46:11<19:32, 2.34it/s, tok/s=14885.4]
Inference: 54%|█████▎ | 3179/5920 [46:11<19:32, 2.34it/s, tok/s=14894.0]
Inference: 54%|█████▎ | 3180/5920 [46:11<16:33, 2.76it/s, tok/s=14894.0]
Inference: 54%|█████▎ | 3180/5920 [46:11<16:33, 2.76it/s, tok/s=14893.6]
Inference: 54%|█████▎ | 3181/5920 [46:12<15:29, 2.95it/s, tok/s=14893.6]
Inference: 54%|█████▎ | 3181/5920 [46:12<15:29, 2.95it/s, tok/s=14893.2]
Inference: 54%|█████▍ | 3182/5920 [46:13<25:00, 1.82it/s, tok/s=14893.2]
Inference: 54%|█████▍ | 3182/5920 [46:13<25:00, 1.82it/s, tok/s=14898.3]
Inference: 54%|█████▍ | 3183/5920 [46:13<20:52, 2.19it/s, tok/s=14898.3]
Inference: 54%|█████▍ | 3183/5920 [46:13<20:52, 2.19it/s, tok/s=14897.7]
Inference: 54%|█████▍ | 3184/5920 [46:13<20:51, 2.19it/s, tok/s=14898.5]
Inference: 54%|█████▍ | 3185/5920 [46:13<16:26, 2.77it/s, tok/s=14898.5]
Inference: 54%|█████▍ | 3185/5920 [46:13<16:26, 2.77it/s, tok/s=14897.0]
Inference: 54%|█████▍ | 3186/5920 [46:14<17:34, 2.59it/s, tok/s=14897.0]
Inference: 54%|█████▍ | 3186/5920 [46:14<17:34, 2.59it/s, tok/s=14895.2]
Inference: 54%|█████▍ | 3187/5920 [46:14<19:48, 2.30it/s, tok/s=14895.2]
Inference: 54%|█████▍ | 3187/5920 [46:14<19:48, 2.30it/s, tok/s=14902.6]
Inference: 54%|█████▍ | 3188/5920 [46:15<19:48, 2.30it/s, tok/s=14903.8]
Inference: 54%|█████▍ | 3189/5920 [46:15<14:45, 3.08it/s, tok/s=14903.8]
Inference: 54%|█████▍ | 3189/5920 [46:15<14:45, 3.08it/s, tok/s=14903.3]
Inference: 54%|█████▍ | 3190/5920 [46:15<13:18, 3.42it/s, tok/s=14903.3]
Inference: 54%|█████▍ | 3190/5920 [46:15<13:18, 3.42it/s, tok/s=14902.7]
Inference: 54%|█████▍ | 3191/5920 [46:15<13:18, 3.42it/s, tok/s=14903.8]
Inference: 54%|█████▍ | 3192/5920 [46:15<13:17, 3.42it/s, tok/s=14905.9]
Inference: 54%|█████▍ | 3193/5920 [46:16<10:36, 4.29it/s, tok/s=14905.9]
Inference: 54%|█████▍ | 3193/5920 [46:16<10:36, 4.29it/s, tok/s=14910.8]
Inference: 54%|█████▍ | 3194/5920 [46:16<14:14, 3.19it/s, tok/s=14910.8]
Inference: 54%|█████▍ | 3194/5920 [46:16<14:14, 3.19it/s, tok/s=14916.3]
Inference: 54%|█████▍ | 3195/5920 [46:16<14:14, 3.19it/s, tok/s=14916.5]
Inference: 54%|█████▍ | 3196/5920 [46:16<10:42, 4.24it/s, tok/s=14916.5]
Inference: 54%|█████▍ | 3196/5920 [46:16<10:42, 4.24it/s, tok/s=14916.9]
Inference: 54%|█████▍ | 3197/5920 [46:16<10:42, 4.24it/s, tok/s=14921.7]
Inference: 54%|█████▍ | 3198/5920 [46:17<13:03, 3.47it/s, tok/s=14921.7]
Inference: 54%|█████▍ | 3198/5920 [46:17<13:03, 3.47it/s, tok/s=14918.2]
Inference: 54%|█████▍ | 3199/5920 [46:18<15:32, 2.92it/s, tok/s=14918.2]
Inference: 54%|█████▍ | 3199/5920 [46:18<15:32, 2.92it/s, tok/s=14922.6]
Inference: 54%|█████▍ | 3200/5920 [46:18<17:39, 2.57it/s, tok/s=14922.6]
Inference: 54%|█████▍ | 3200/5920 [46:18<17:39, 2.57it/s, tok/s=14921.4]
Inference: 54%|█████▍ | 3201/5920 [46:18<14:39, 3.09it/s, tok/s=14921.4]
Inference: 54%|█████▍ | 3201/5920 [46:18<14:39, 3.09it/s, tok/s=14921.7]
Inference: 54%|█████▍ | 3202/5920 [46:19<12:27, 3.64it/s, tok/s=14921.7]
Inference: 54%|█████▍ | 3202/5920 [46:19<12:27, 3.64it/s, tok/s=14922.4]
Inference: 54%|█████▍ | 3203/5920 [46:19<14:14, 3.18it/s, tok/s=14922.4]
Inference: 54%|█████▍ | 3203/5920 [46:19<14:14, 3.18it/s, tok/s=14928.6]
Inference: 54%|█████▍ | 3204/5920 [46:20<24:17, 1.86it/s, tok/s=14928.6]
Inference: 54%|█████▍ | 3204/5920 [46:20<24:17, 1.86it/s, tok/s=14933.4]
Inference: 54%|█████▍ | 3205/5920 [46:21<30:32, 1.48it/s, tok/s=14933.4]
Inference: 54%|█████▍ | 3205/5920 [46:21<30:32, 1.48it/s, tok/s=14928.7]
Inference: 54%|█████▍ | 3206/5920 [46:21<23:39, 1.91it/s, tok/s=14928.7]
Inference: 54%|█████▍ | 3206/5920 [46:21<23:39, 1.91it/s, tok/s=14929.3]
Inference: 54%|█████▍ | 3207/5920 [46:22<24:07, 1.87it/s, tok/s=14929.3]
Inference: 54%|█████▍ | 3207/5920 [46:22<24:07, 1.87it/s, tok/s=14927.3]
Inference: 54%|█████▍ | 3208/5920 [46:23<27:32, 1.64it/s, tok/s=14927.3]
Inference: 54%|█████▍ | 3208/5920 [46:23<27:32, 1.64it/s, tok/s=14929.8]
Inference: 54%|█████▍ | 3209/5920 [46:23<24:43, 1.83it/s, tok/s=14929.8]
Inference: 54%|█████▍ | 3209/5920 [46:23<24:43, 1.83it/s, tok/s=14928.1]
Inference: 54%|█████▍ | 3210/5920 [46:23<24:42, 1.83it/s, tok/s=14929.1]
Inference: 54%|█████▍ | 3211/5920 [46:23<16:18, 2.77it/s, tok/s=14929.1]
Inference: 54%|█████▍ | 3211/5920 [46:23<16:18, 2.77it/s, tok/s=14929.1]
Inference: 54%|█████▍ | 3212/5920 [46:23<13:55, 3.24it/s, tok/s=14929.1]
Inference: 54%|█████▍ | 3212/5920 [46:23<13:55, 3.24it/s, tok/s=14930.2]
Inference: 54%|█████▍ | 3213/5920 [46:24<12:16, 3.67it/s, tok/s=14930.2]
Inference: 54%|█████▍ | 3213/5920 [46:24<12:16, 3.67it/s, tok/s=14930.0]
Inference: 54%|█████▍ | 3214/5920 [46:24<14:34, 3.09it/s, tok/s=14930.0]
Inference: 54%|█████▍ | 3214/5920 [46:24<14:34, 3.09it/s, tok/s=14934.4]
Inference: 54%|█████▍ | 3215/5920 [46:25<20:12, 2.23it/s, tok/s=14934.4]
Inference: 54%|█████▍ | 3215/5920 [46:25<20:12, 2.23it/s, tok/s=14930.7]
Inference: 54%|█████▍ | 3216/5920 [46:25<22:02, 2.05it/s, tok/s=14930.7]
Inference: 54%|█████▍ | 3216/5920 [46:25<22:02, 2.05it/s, tok/s=14928.4]
Inference: 54%|█████▍ | 3217/5920 [46:26<28:12, 1.60it/s, tok/s=14928.4]
Inference: 54%|█████▍ | 3217/5920 [46:26<28:12, 1.60it/s, tok/s=14925.0]
Inference: 54%|█████▍ | 3218/5920 [46:27<25:30, 1.77it/s, tok/s=14925.0]
Inference: 54%|█████▍ | 3218/5920 [46:27<25:30, 1.77it/s, tok/s=14925.0]
Inference: 54%|█████▍ | 3219/5920 [46:27<24:08, 1.87it/s, tok/s=14925.0]
Inference: 54%|█████▍ | 3219/5920 [46:27<24:08, 1.87it/s, tok/s=14931.5]
Inference: 54%|█████▍ | 3220/5920 [46:28<24:15, 1.86it/s, tok/s=14931.5]
Inference: 54%|█████▍ | 3220/5920 [46:28<24:15, 1.86it/s, tok/s=14929.0]
Inference: 54%|█████▍ | 3221/5920 [46:29<28:55, 1.56it/s, tok/s=14929.0]
Inference: 54%|█████▍ | 3221/5920 [46:29<28:55, 1.56it/s, tok/s=14925.0]
Inference: 54%|█████▍ | 3222/5920 [46:29<28:54, 1.56it/s, tok/s=14925.0]
Inference: 54%|█████▍ | 3223/5920 [46:29<22:05, 2.03it/s, tok/s=14925.0]
Inference: 54%|█████▍ | 3223/5920 [46:29<22:05, 2.03it/s, tok/s=14925.3]
Inference: 54%|█████▍ | 3224/5920 [46:30<19:19, 2.33it/s, tok/s=14925.3]
Inference: 54%|█████▍ | 3224/5920 [46:30<19:19, 2.33it/s, tok/s=14931.5]
Inference: 54%|█████▍ | 3225/5920 [46:30<19:19, 2.33it/s, tok/s=14931.8]
Inference: 54%|█████▍ | 3226/5920 [46:30<15:41, 2.86it/s, tok/s=14931.8]
Inference: 54%|█████▍ | 3226/5920 [46:30<15:41, 2.86it/s, tok/s=14930.6]
Inference: 55%|█████▍ | 3227/5920 [46:30<15:41, 2.86it/s, tok/s=14931.8]
Inference: 55%|█████▍ | 3228/5920 [46:30<12:17, 3.65it/s, tok/s=14931.8]
Inference: 55%|█████▍ | 3228/5920 [46:30<12:17, 3.65it/s, tok/s=14939.3]
Inference: 55%|█████▍ | 3229/5920 [46:31<12:56, 3.47it/s, tok/s=14939.3]
Inference: 55%|█████▍ | 3229/5920 [46:31<12:56, 3.47it/s, tok/s=14938.3]
Inference: 55%|█████▍ | 3230/5920 [46:31<16:04, 2.79it/s, tok/s=14938.3]
Inference: 55%|█████▍ | 3230/5920 [46:31<16:04, 2.79it/s, tok/s=14941.3]
Inference: 55%|█████▍ | 3231/5920 [46:32<18:31, 2.42it/s, tok/s=14941.3]
Inference: 55%|█████▍ | 3231/5920 [46:32<18:31, 2.42it/s, tok/s=14944.2]
Inference: 55%|█████▍ | 3232/5920 [46:32<19:54, 2.25it/s, tok/s=14944.2]
Inference: 55%|█████▍ | 3232/5920 [46:32<19:54, 2.25it/s, tok/s=14946.9]
Inference: 55%|█████▍ | 3233/5920 [46:33<17:12, 2.60it/s, tok/s=14946.9]
Inference: 55%|█████▍ | 3233/5920 [46:33<17:12, 2.60it/s, tok/s=14946.4]
Inference: 55%|█████▍ | 3234/5920 [46:34<27:27, 1.63it/s, tok/s=14946.4]
Inference: 55%|█████▍ | 3234/5920 [46:34<27:27, 1.63it/s, tok/s=14948.8]
Inference: 55%|█████▍ | 3235/5920 [46:35<28:41, 1.56it/s, tok/s=14948.8]
Inference: 55%|█████▍ | 3235/5920 [46:35<28:41, 1.56it/s, tok/s=14946.7]
Inference: 55%|█████▍ | 3236/5920 [46:35<23:37, 1.89it/s, tok/s=14946.7]
Inference: 55%|█████▍ | 3236/5920 [46:35<23:37, 1.89it/s, tok/s=14946.3]
Inference: 55%|█████▍ | 3237/5920 [46:35<19:59, 2.24it/s, tok/s=14946.3]
Inference: 55%|█████▍ | 3237/5920 [46:35<19:59, 2.24it/s, tok/s=14945.6]
Inference: 55%|█████▍ | 3238/5920 [46:35<18:33, 2.41it/s, tok/s=14945.6]
Inference: 55%|█████▍ | 3238/5920 [46:35<18:33, 2.41it/s, tok/s=14944.5]
Inference: 55%|█████▍ | 3239/5920 [46:37<33:54, 1.32it/s, tok/s=14944.5]
Inference: 55%|█████▍ | 3239/5920 [46:37<33:54, 1.32it/s, tok/s=14937.7]
Inference: 55%|█████▍ | 3240/5920 [46:37<28:16, 1.58it/s, tok/s=14937.7]
Inference: 55%|█████▍ | 3240/5920 [46:37<28:16, 1.58it/s, tok/s=14936.4]
Inference: 55%|█████▍ | 3241/5920 [46:37<28:16, 1.58it/s, tok/s=14938.4]
Inference: 55%|█████▍ | 3242/5920 [46:37<16:46, 2.66it/s, tok/s=14938.4]
Inference: 55%|█████▍ | 3242/5920 [46:37<16:46, 2.66it/s, tok/s=14938.1]
Inference: 55%|█████▍ | 3243/5920 [46:38<17:36, 2.53it/s, tok/s=14938.1]
Inference: 55%|█████▍ | 3243/5920 [46:38<17:36, 2.53it/s, tok/s=14943.5]
Inference: 55%|█████▍ | 3244/5920 [46:38<15:43, 2.83it/s, tok/s=14943.5]
Inference: 55%|█████▍ | 3244/5920 [46:38<15:43, 2.83it/s, tok/s=14944.5]
Inference: 55%|█████▍ | 3245/5920 [46:38<13:46, 3.24it/s, tok/s=14944.5]
Inference: 55%|█████▍ | 3245/5920 [46:38<13:46, 3.24it/s, tok/s=14944.1]
Inference: 55%|█████▍ | 3246/5920 [46:38<13:46, 3.24it/s, tok/s=14944.9]
Inference: 55%|█████▍ | 3247/5920 [46:38<09:07, 4.88it/s, tok/s=14944.9]
Inference: 55%|█████▍ | 3247/5920 [46:38<09:07, 4.88it/s, tok/s=14945.0]
Inference: 55%|█████▍ | 3248/5920 [46:39<12:27, 3.57it/s, tok/s=14945.0]
Inference: 55%|█████▍ | 3248/5920 [46:39<12:27, 3.57it/s, tok/s=14944.7]
Inference: 55%|█████▍ | 3249/5920 [46:39<10:50, 4.11it/s, tok/s=14944.7]
Inference: 55%|█████▍ | 3249/5920 [46:39<10:50, 4.11it/s, tok/s=14954.6]
Inference: 55%|█████▍ | 3250/5920 [46:40<21:29, 2.07it/s, tok/s=14954.6]
Inference: 55%|█████▍ | 3250/5920 [46:40<21:29, 2.07it/s, tok/s=14949.3]
Inference: 55%|█████▍ | 3251/5920 [46:40<21:28, 2.07it/s, tok/s=14952.5]
Inference: 55%|█████▍ | 3252/5920 [46:40<14:27, 3.08it/s, tok/s=14952.5]
Inference: 55%|█████▍ | 3252/5920 [46:40<14:27, 3.08it/s, tok/s=14952.3]
Inference: 55%|█████▍ | 3253/5920 [46:40<14:27, 3.08it/s, tok/s=14953.0]
Inference: 55%|█████▍ | 3254/5920 [46:41<13:47, 3.22it/s, tok/s=14953.0]
Inference: 55%|█████▍ | 3254/5920 [46:41<13:47, 3.22it/s, tok/s=14952.9]
Inference: 55%|█████▍ | 3255/5920 [46:42<15:38, 2.84it/s, tok/s=14952.9]
Inference: 55%|█████▍ | 3255/5920 [46:42<15:38, 2.84it/s, tok/s=14953.3]
Inference: 55%|█████▌ | 3256/5920 [46:42<15:38, 2.84it/s, tok/s=14954.1]
Inference: 55%|█████▌ | 3257/5920 [46:42<15:37, 2.84it/s, tok/s=14954.6]
Inference: 55%|█████▌ | 3258/5920 [46:42<12:21, 3.59it/s, tok/s=14954.6]
Inference: 55%|█████▌ | 3258/5920 [46:42<12:21, 3.59it/s, tok/s=14952.7]
Inference: 55%|█████▌ | 3259/5920 [46:43<20:49, 2.13it/s, tok/s=14952.7]
Inference: 55%|█████▌ | 3259/5920 [46:43<20:49, 2.13it/s, tok/s=14946.5]
Inference: 55%|█████▌ | 3260/5920 [46:44<17:32, 2.53it/s, tok/s=14946.5]
Inference: 55%|█████▌ | 3260/5920 [46:44<17:32, 2.53it/s, tok/s=14946.5]
Inference: 55%|█████▌ | 3261/5920 [46:45<24:08, 1.84it/s, tok/s=14946.5]
Inference: 55%|█████▌ | 3261/5920 [46:45<24:08, 1.84it/s, tok/s=14941.4]
Inference: 55%|█████▌ | 3262/5920 [46:45<20:08, 2.20it/s, tok/s=14941.4]
Inference: 55%|█████▌ | 3262/5920 [46:45<20:08, 2.20it/s, tok/s=14941.1]
Inference: 55%|█████▌ | 3263/5920 [46:45<16:37, 2.66it/s, tok/s=14941.1]
Inference: 55%|█████▌ | 3263/5920 [46:45<16:37, 2.66it/s, tok/s=14941.2]
Inference: 55%|█████▌ | 3264/5920 [46:45<13:22, 3.31it/s, tok/s=14941.2]
Inference: 55%|█████▌ | 3264/5920 [46:45<13:22, 3.31it/s, tok/s=14947.7]
Inference: 55%|█████▌ | 3265/5920 [46:45<13:21, 3.31it/s, tok/s=14950.7]
Inference: 55%|█████▌ | 3266/5920 [46:45<09:03, 4.89it/s, tok/s=14950.7]
Inference: 55%|█████▌ | 3266/5920 [46:45<09:03, 4.89it/s, tok/s=14951.3]
Inference: 55%|█████▌ | 3267/5920 [46:46<12:45, 3.47it/s, tok/s=14951.3]
Inference: 55%|█████▌ | 3267/5920 [46:46<12:45, 3.47it/s, tok/s=14949.1]
Inference: 55%|█████▌ | 3268/5920 [46:46<11:32, 3.83it/s, tok/s=14949.1]
Inference: 55%|█████▌ | 3268/5920 [46:46<11:32, 3.83it/s, tok/s=14949.1]
Inference: 55%|█████▌ | 3269/5920 [46:46<09:49, 4.50it/s, tok/s=14949.1]
Inference: 55%|█████▌ | 3269/5920 [46:46<09:49, 4.50it/s, tok/s=14949.4]
Inference: 55%|█████▌ | 3270/5920 [46:46<09:48, 4.50it/s, tok/s=14958.4]
Inference: 55%|█████▌ | 3271/5920 [46:46<07:53, 5.60it/s, tok/s=14958.4]
Inference: 55%|█████▌ | 3271/5920 [46:46<07:53, 5.60it/s, tok/s=14959.4]
Inference: 55%|█████▌ | 3272/5920 [46:46<07:53, 5.60it/s, tok/s=14959.5]
Inference: 55%|█████▌ | 3273/5920 [46:46<05:56, 7.42it/s, tok/s=14959.5]
Inference: 55%|█████▌ | 3273/5920 [46:46<05:56, 7.42it/s, tok/s=14960.3]
Inference: 55%|█████▌ | 3274/5920 [46:47<05:56, 7.42it/s, tok/s=14960.5]
Inference: 55%|█████▌ | 3275/5920 [46:47<04:58, 8.85it/s, tok/s=14960.5]
Inference: 55%|█████▌ | 3275/5920 [46:47<04:58, 8.85it/s, tok/s=14960.8]
Inference: 55%|█████▌ | 3276/5920 [46:47<04:58, 8.85it/s, tok/s=14964.9]
Inference: 55%|█████▌ | 3277/5920 [46:47<04:55, 8.94it/s, tok/s=14964.9]
Inference: 55%|█████▌ | 3277/5920 [46:47<04:55, 8.94it/s, tok/s=14966.3]
Inference: 55%|█████▌ | 3278/5920 [46:47<04:55, 8.94it/s, tok/s=14971.4]
Inference: 55%|█████▌ | 3279/5920 [46:48<10:13, 4.30it/s, tok/s=14971.4]
Inference: 55%|█████▌ | 3279/5920 [46:48<10:13, 4.30it/s, tok/s=14970.4]
Inference: 55%|█████▌ | 3280/5920 [46:48<10:55, 4.03it/s, tok/s=14970.4]
Inference: 55%|█████▌ | 3280/5920 [46:48<10:55, 4.03it/s, tok/s=14971.5]
Inference: 55%|█████▌ | 3281/5920 [46:48<09:53, 4.44it/s, tok/s=14971.5]
Inference: 55%|█████▌ | 3281/5920 [46:48<09:53, 4.44it/s, tok/s=14971.2]
Inference: 55%|█████▌ | 3282/5920 [46:48<09:39, 4.55it/s, tok/s=14971.2]
Inference: 55%|█████▌ | 3282/5920 [46:48<09:39, 4.55it/s, tok/s=14970.7]
Inference: 55%|█████▌ | 3283/5920 [46:49<09:18, 4.72it/s, tok/s=14970.7]
Inference: 55%|█████▌ | 3283/5920 [46:49<09:18, 4.72it/s, tok/s=14970.6]
Inference: 55%|█████▌ | 3284/5920 [46:49<12:04, 3.64it/s, tok/s=14970.6]
Inference: 55%|█████▌ | 3284/5920 [46:49<12:04, 3.64it/s, tok/s=14968.9]
Inference: 55%|█████▌ | 3285/5920 [46:49<13:35, 3.23it/s, tok/s=14968.9]
Inference: 55%|█████▌ | 3285/5920 [46:49<13:35, 3.23it/s, tok/s=14967.5]
Inference: 56%|█████▌ | 3286/5920 [46:50<13:25, 3.27it/s, tok/s=14967.5]
Inference: 56%|█████▌ | 3286/5920 [46:50<13:25, 3.27it/s, tok/s=14966.8]
Inference: 56%|█████▌ | 3287/5920 [46:50<13:25, 3.27it/s, tok/s=14969.3]
Inference: 56%|█████▌ | 3288/5920 [46:50<11:21, 3.86it/s, tok/s=14969.3]
Inference: 56%|█████▌ | 3288/5920 [46:50<11:21, 3.86it/s, tok/s=14967.8]
Inference: 56%|█████▌ | 3289/5920 [46:51<13:24, 3.27it/s, tok/s=14967.8]
Inference: 56%|█████▌ | 3289/5920 [46:51<13:24, 3.27it/s, tok/s=14966.1]
Inference: 56%|█████▌ | 3290/5920 [46:51<13:23, 3.27it/s, tok/s=14966.1]
Inference: 56%|█████▌ | 3291/5920 [46:52<16:46, 2.61it/s, tok/s=14966.1]
Inference: 56%|█████▌ | 3291/5920 [46:52<16:46, 2.61it/s, tok/s=14962.3]
Inference: 56%|█████▌ | 3292/5920 [46:52<14:17, 3.06it/s, tok/s=14962.3]
Inference: 56%|█████▌ | 3292/5920 [46:52<14:17, 3.06it/s, tok/s=14966.3]
Inference: 56%|█████▌ | 3293/5920 [46:52<14:17, 3.06it/s, tok/s=14966.7]
Inference: 56%|█████▌ | 3294/5920 [46:52<09:42, 4.51it/s, tok/s=14966.7]
Inference: 56%|█████▌ | 3294/5920 [46:52<09:42, 4.51it/s, tok/s=14973.1]
Inference: 56%|█████▌ | 3295/5920 [46:52<09:42, 4.51it/s, tok/s=14973.5]
Inference: 56%|█████▌ | 3296/5920 [46:52<09:51, 4.44it/s, tok/s=14973.5]
Inference: 56%|█████▌ | 3296/5920 [46:52<09:51, 4.44it/s, tok/s=14972.1]
Inference: 56%|█████▌ | 3297/5920 [46:53<09:50, 4.44it/s, tok/s=14972.1]
Inference: 56%|█████▌ | 3297/5920 [46:53<09:50, 4.44it/s, tok/s=14978.5]
Inference: 56%|█████▌ | 3298/5920 [46:53<12:08, 3.60it/s, tok/s=14978.5]
Inference: 56%|█████▌ | 3298/5920 [46:53<12:08, 3.60it/s, tok/s=14976.7]
Inference: 56%|█████▌ | 3299/5920 [46:53<12:07, 3.60it/s, tok/s=14978.2]
Inference: 56%|█████▌ | 3300/5920 [46:54<13:01, 3.35it/s, tok/s=14978.2]
Inference: 56%|█████▌ | 3300/5920 [46:54<13:01, 3.35it/s, tok/s=14975.4]
Inference: 56%|█████▌ | 3301/5920 [46:54<13:21, 3.27it/s, tok/s=14975.4]
Inference: 56%|█████▌ | 3301/5920 [46:54<13:21, 3.27it/s, tok/s=14974.2]
Inference: 56%|█████▌ | 3302/5920 [46:54<12:57, 3.37it/s, tok/s=14974.2]
Inference: 56%|█████▌ | 3302/5920 [46:54<12:57, 3.37it/s, tok/s=14973.8]
Inference: 56%|█████▌ | 3303/5920 [46:55<12:59, 3.36it/s, tok/s=14973.8]
Inference: 56%|█████▌ | 3303/5920 [46:55<12:59, 3.36it/s, tok/s=14973.4]
Inference: 56%|█████▌ | 3304/5920 [46:55<12:58, 3.36it/s, tok/s=14975.1]
Inference: 56%|█████▌ | 3305/5920 [46:55<09:17, 4.69it/s, tok/s=14975.1]
Inference: 56%|█████▌ | 3305/5920 [46:55<09:17, 4.69it/s, tok/s=14976.7]
Inference: 56%|█████▌ | 3306/5920 [46:56<17:18, 2.52it/s, tok/s=14976.7]
Inference: 56%|█████▌ | 3306/5920 [46:56<17:18, 2.52it/s, tok/s=14972.2]
Inference: 56%|█████▌ | 3307/5920 [46:56<19:33, 2.23it/s, tok/s=14972.2]
Inference: 56%|█████▌ | 3307/5920 [46:56<19:33, 2.23it/s, tok/s=14971.1]
Inference: 56%|█████▌ | 3308/5920 [46:56<19:32, 2.23it/s, tok/s=14971.4]
Inference: 56%|█████▌ | 3309/5920 [46:56<19:32, 2.23it/s, tok/s=14971.9]
Inference: 56%|█████▌ | 3310/5920 [46:56<10:13, 4.26it/s, tok/s=14971.9]
Inference: 56%|█████▌ | 3310/5920 [46:56<10:13, 4.26it/s, tok/s=14976.6]
Inference: 56%|█████▌ | 3311/5920 [46:57<10:12, 4.26it/s, tok/s=14983.0]
Inference: 56%|█████▌ | 3312/5920 [46:57<13:37, 3.19it/s, tok/s=14983.0]
Inference: 56%|█████▌ | 3312/5920 [46:57<13:37, 3.19it/s, tok/s=14979.2]
Inference: 56%|█████▌ | 3313/5920 [46:58<13:11, 3.30it/s, tok/s=14979.2]
Inference: 56%|█████▌ | 3313/5920 [46:58<13:11, 3.30it/s, tok/s=14978.2]
Inference: 56%|█████▌ | 3314/5920 [46:58<11:21, 3.82it/s, tok/s=14978.2]
Inference: 56%|█████▌ | 3314/5920 [46:58<11:21, 3.82it/s, tok/s=14978.3]
Inference: 56%|█████▌ | 3315/5920 [46:58<11:21, 3.82it/s, tok/s=14979.0]
Inference: 56%|█████▌ | 3316/5920 [46:59<15:15, 2.85it/s, tok/s=14979.0]
Inference: 56%|█████▌ | 3316/5920 [46:59<15:15, 2.85it/s, tok/s=14974.5]
Inference: 56%|█████▌ | 3317/5920 [46:59<15:14, 2.85it/s, tok/s=14976.0]
Inference: 56%|█████▌ | 3318/5920 [46:59<11:22, 3.81it/s, tok/s=14976.0]
Inference: 56%|█████▌ | 3318/5920 [46:59<11:22, 3.81it/s, tok/s=14976.0]
Inference: 56%|█████▌ | 3319/5920 [46:59<11:21, 3.81it/s, tok/s=14978.0]
Inference: 56%|█████▌ | 3320/5920 [47:00<12:20, 3.51it/s, tok/s=14978.0]
Inference: 56%|█████▌ | 3320/5920 [47:00<12:20, 3.51it/s, tok/s=14977.0]
Inference: 56%|█████▌ | 3321/5920 [47:00<13:23, 3.23it/s, tok/s=14977.0]
Inference: 56%|█████▌ | 3321/5920 [47:00<13:23, 3.23it/s, tok/s=14975.9]
Inference: 56%|█████▌ | 3322/5920 [47:00<13:23, 3.23it/s, tok/s=14976.0]
Inference: 56%|█████▌ | 3323/5920 [47:01<12:28, 3.47it/s, tok/s=14976.0]
Inference: 56%|█████▌ | 3323/5920 [47:01<12:28, 3.47it/s, tok/s=14977.0]
Inference: 56%|█████▌ | 3324/5920 [47:01<11:18, 3.82it/s, tok/s=14977.0]
Inference: 56%|█████▌ | 3324/5920 [47:01<11:18, 3.82it/s, tok/s=14983.7]
Inference: 56%|█████▌ | 3325/5920 [47:01<12:43, 3.40it/s, tok/s=14983.7]
Inference: 56%|█████▌ | 3325/5920 [47:01<12:43, 3.40it/s, tok/s=14989.0]
Inference: 56%|█████▌ | 3326/5920 [47:01<12:43, 3.40it/s, tok/s=14997.1]
Inference: 56%|█████▌ | 3327/5920 [47:01<09:05, 4.75it/s, tok/s=14997.1]
Inference: 56%|█████▌ | 3327/5920 [47:01<09:05, 4.75it/s, tok/s=14997.3]
Inference: 56%|█████▌ | 3328/5920 [47:01<08:20, 5.18it/s, tok/s=14997.3]
Inference: 56%|█████▌ | 3328/5920 [47:01<08:20, 5.18it/s, tok/s=14997.5]
Inference: 56%|█████▌ | 3329/5920 [47:02<10:04, 4.28it/s, tok/s=14997.5]
Inference: 56%|█████▌ | 3329/5920 [47:02<10:04, 4.28it/s, tok/s=15001.6]
Inference: 56%|█████▋ | 3330/5920 [47:02<10:04, 4.28it/s, tok/s=15002.3]
Inference: 56%|█████▋ | 3331/5920 [47:02<07:27, 5.79it/s, tok/s=15002.3]
Inference: 56%|█████▋ | 3331/5920 [47:02<07:27, 5.79it/s, tok/s=15002.4]
Inference: 56%|█████▋ | 3332/5920 [47:04<25:48, 1.67it/s, tok/s=15002.4]
Inference: 56%|█████▋ | 3332/5920 [47:04<25:48, 1.67it/s, tok/s=14993.2]
Inference: 56%|█████▋ | 3333/5920 [47:04<21:34, 2.00it/s, tok/s=14993.2]
Inference: 56%|█████▋ | 3333/5920 [47:04<21:34, 2.00it/s, tok/s=14993.0]
Inference: 56%|█████▋ | 3334/5920 [47:05<19:31, 2.21it/s, tok/s=14993.0]
Inference: 56%|█████▋ | 3334/5920 [47:05<19:31, 2.21it/s, tok/s=14991.9]
Inference: 56%|█████▋ | 3335/5920 [47:05<16:52, 2.55it/s, tok/s=14991.9]
Inference: 56%|█████▋ | 3335/5920 [47:05<16:52, 2.55it/s, tok/s=14992.9]
Inference: 56%|█████▋ | 3336/5920 [47:05<16:52, 2.55it/s, tok/s=14993.5]
Inference: 56%|█████▋ | 3337/5920 [47:06<22:02, 1.95it/s, tok/s=14993.5]
Inference: 56%|█████▋ | 3337/5920 [47:06<22:02, 1.95it/s, tok/s=14987.5]
Inference: 56%|█████▋ | 3338/5920 [47:06<22:01, 1.95it/s, tok/s=14992.7]
Inference: 56%|█████▋ | 3339/5920 [47:06<15:32, 2.77it/s, tok/s=14992.7]
Inference: 56%|█████▋ | 3339/5920 [47:06<15:32, 2.77it/s, tok/s=14992.2]
Inference: 56%|█████▋ | 3340/5920 [47:08<26:34, 1.62it/s, tok/s=14992.2]
Inference: 56%|█████▋ | 3340/5920 [47:08<26:34, 1.62it/s, tok/s=14985.6]
Inference: 56%|█████▋ | 3341/5920 [47:08<25:04, 1.71it/s, tok/s=14985.6]
Inference: 56%|█████▋ | 3341/5920 [47:08<25:04, 1.71it/s, tok/s=14991.0]
Inference: 56%|█████▋ | 3342/5920 [47:09<22:08, 1.94it/s, tok/s=14991.0]
Inference: 56%|█████▋ | 3342/5920 [47:09<22:08, 1.94it/s, tok/s=14990.3]
Inference: 56%|█████▋ | 3343/5920 [47:10<28:41, 1.50it/s, tok/s=14990.3]
Inference: 56%|█████▋ | 3343/5920 [47:10<28:41, 1.50it/s, tok/s=14986.1]
Inference: 56%|█████▋ | 3344/5920 [47:10<24:11, 1.77it/s, tok/s=14986.1]
Inference: 56%|█████▋ | 3344/5920 [47:10<24:11, 1.77it/s, tok/s=14985.2]
Inference: 57%|█████▋ | 3345/5920 [47:10<18:38, 2.30it/s, tok/s=14985.2]
Inference: 57%|█████▋ | 3345/5920 [47:10<18:38, 2.30it/s, tok/s=14986.0]
Inference: 57%|█████▋ | 3346/5920 [47:11<21:06, 2.03it/s, tok/s=14986.0]
Inference: 57%|█████▋ | 3346/5920 [47:11<21:06, 2.03it/s, tok/s=14991.6]
Inference: 57%|█████▋ | 3347/5920 [47:11<17:14, 2.49it/s, tok/s=14991.6]
Inference: 57%|█████▋ | 3347/5920 [47:11<17:14, 2.49it/s, tok/s=14991.4]
Inference: 57%|█████▋ | 3348/5920 [47:11<17:59, 2.38it/s, tok/s=14991.4]
Inference: 57%|█████▋ | 3348/5920 [47:11<17:59, 2.38it/s, tok/s=14989.5]
Inference: 57%|█████▋ | 3349/5920 [47:12<14:24, 2.97it/s, tok/s=14989.5]
Inference: 57%|█████▋ | 3349/5920 [47:12<14:24, 2.97it/s, tok/s=14990.8]
Inference: 57%|█████▋ | 3350/5920 [47:13<25:39, 1.67it/s, tok/s=14990.8]
Inference: 57%|█████▋ | 3350/5920 [47:13<25:39, 1.67it/s, tok/s=14984.9]
Inference: 57%|█████▋ | 3351/5920 [47:13<21:13, 2.02it/s, tok/s=14984.9]
Inference: 57%|█████▋ | 3351/5920 [47:13<21:13, 2.02it/s, tok/s=14985.5]
Inference: 57%|█████▋ | 3352/5920 [47:13<19:21, 2.21it/s, tok/s=14985.5]
Inference: 57%|█████▋ | 3352/5920 [47:13<19:21, 2.21it/s, tok/s=14994.5]
Inference: 57%|█████▋ | 3353/5920 [47:14<15:31, 2.76it/s, tok/s=14994.5]
Inference: 57%|█████▋ | 3353/5920 [47:14<15:31, 2.76it/s, tok/s=15002.1]
Inference: 57%|█████▋ | 3354/5920 [47:14<13:04, 3.27it/s, tok/s=15002.1]
Inference: 57%|█████▋ | 3354/5920 [47:14<13:04, 3.27it/s, tok/s=15002.0]
Inference: 57%|█████▋ | 3355/5920 [47:14<15:18, 2.79it/s, tok/s=15002.0]
Inference: 57%|█████▋ | 3355/5920 [47:14<15:18, 2.79it/s, tok/s=15000.2]
Inference: 57%|█████▋ | 3356/5920 [47:15<15:00, 2.85it/s, tok/s=15000.2]
Inference: 57%|█████▋ | 3356/5920 [47:15<15:00, 2.85it/s, tok/s=14999.2]
Inference: 57%|█████▋ | 3357/5920 [47:15<14:39, 2.91it/s, tok/s=14999.2]
Inference: 57%|█████▋ | 3357/5920 [47:15<14:39, 2.91it/s, tok/s=14998.0]
Inference: 57%|█████▋ | 3358/5920 [47:15<14:39, 2.91it/s, tok/s=15003.2]
Inference: 57%|█████▋ | 3359/5920 [47:15<11:41, 3.65it/s, tok/s=15003.2]
Inference: 57%|█████▋ | 3359/5920 [47:15<11:41, 3.65it/s, tok/s=15002.2]
Inference: 57%|█████▋ | 3360/5920 [47:15<10:28, 4.07it/s, tok/s=15002.2]
Inference: 57%|█████▋ | 3360/5920 [47:15<10:28, 4.07it/s, tok/s=15003.2]
Inference: 57%|█████▋ | 3361/5920 [47:16<18:57, 2.25it/s, tok/s=15003.2]
Inference: 57%|█████▋ | 3361/5920 [47:16<18:57, 2.25it/s, tok/s=14998.9]
Inference: 57%|█████▋ | 3362/5920 [47:17<17:44, 2.40it/s, tok/s=14998.9]
Inference: 57%|█████▋ | 3362/5920 [47:17<17:44, 2.40it/s, tok/s=14998.1]
Inference: 57%|█████▋ | 3363/5920 [47:17<17:44, 2.40it/s, tok/s=14998.7]
Inference: 57%|█████▋ | 3364/5920 [47:18<19:26, 2.19it/s, tok/s=14998.7]
Inference: 57%|█████▋ | 3364/5920 [47:18<19:26, 2.19it/s, tok/s=14994.5]
Inference: 57%|█████▋ | 3365/5920 [47:18<19:53, 2.14it/s, tok/s=14994.5]
Inference: 57%|█████▋ | 3365/5920 [47:18<19:53, 2.14it/s, tok/s=14993.6]
Inference: 57%|█████▋ | 3366/5920 [47:19<17:55, 2.38it/s, tok/s=14993.6]
Inference: 57%|█████▋ | 3366/5920 [47:19<17:55, 2.38it/s, tok/s=14997.3]
Inference: 57%|█████▋ | 3367/5920 [47:19<17:00, 2.50it/s, tok/s=14997.3]
Inference: 57%|█████▋ | 3367/5920 [47:19<17:00, 2.50it/s, tok/s=14996.1]
Inference: 57%|█████▋ | 3368/5920 [47:20<24:06, 1.76it/s, tok/s=14996.1]
Inference: 57%|█████▋ | 3368/5920 [47:20<24:06, 1.76it/s, tok/s=14996.7]
Inference: 57%|█████▋ | 3369/5920 [47:20<19:14, 2.21it/s, tok/s=14996.7]
Inference: 57%|█████▋ | 3369/5920 [47:20<19:14, 2.21it/s, tok/s=14999.2]
Inference: 57%|█████▋ | 3370/5920 [47:20<19:13, 2.21it/s, tok/s=14999.5]
Inference: 57%|█████▋ | 3371/5920 [47:21<16:43, 2.54it/s, tok/s=14999.5]
Inference: 57%|█████▋ | 3371/5920 [47:21<16:43, 2.54it/s, tok/s=14997.5]
Inference: 57%|█████▋ | 3372/5920 [47:21<14:47, 2.87it/s, tok/s=14997.5]
Inference: 57%|█████▋ | 3372/5920 [47:21<14:47, 2.87it/s, tok/s=14997.9]
Inference: 57%|█████▋ | 3373/5920 [47:21<14:25, 2.94it/s, tok/s=14997.9]
Inference: 57%|█████▋ | 3373/5920 [47:21<14:25, 2.94it/s, tok/s=15003.7]
Inference: 57%|█████▋ | 3374/5920 [47:22<17:50, 2.38it/s, tok/s=15003.7]
Inference: 57%|█████▋ | 3374/5920 [47:22<17:50, 2.38it/s, tok/s=15004.1]
Inference: 57%|█████▋ | 3375/5920 [47:22<17:04, 2.48it/s, tok/s=15004.1]
Inference: 57%|█████▋ | 3375/5920 [47:22<17:04, 2.48it/s, tok/s=15010.5]
Inference: 57%|█████▋ | 3376/5920 [47:22<17:04, 2.48it/s, tok/s=15019.7]
Inference: 57%|█████▋ | 3377/5920 [47:22<10:43, 3.95it/s, tok/s=15019.7]
Inference: 57%|█████▋ | 3377/5920 [47:22<10:43, 3.95it/s, tok/s=15021.9]
Inference: 57%|█████▋ | 3378/5920 [47:22<10:43, 3.95it/s, tok/s=15029.9]
Inference: 57%|█████▋ | 3379/5920 [47:23<08:36, 4.92it/s, tok/s=15029.9]
Inference: 57%|█████▋ | 3379/5920 [47:23<08:36, 4.92it/s, tok/s=15031.9]
Inference: 57%|█████▋ | 3380/5920 [47:23<08:09, 5.19it/s, tok/s=15031.9]
Inference: 57%|█████▋ | 3380/5920 [47:23<08:09, 5.19it/s, tok/s=15031.9]
Inference: 57%|█████▋ | 3381/5920 [47:23<08:09, 5.19it/s, tok/s=15032.5]
Inference: 57%|█████▋ | 3382/5920 [47:24<15:33, 2.72it/s, tok/s=15032.5]
Inference: 57%|█████▋ | 3382/5920 [47:24<15:33, 2.72it/s, tok/s=15026.1]
Inference: 57%|█████▋ | 3383/5920 [47:24<15:32, 2.72it/s, tok/s=15026.5]
Inference: 57%|█████▋ | 3384/5920 [47:25<13:25, 3.15it/s, tok/s=15026.5]
Inference: 57%|█████▋ | 3384/5920 [47:25<13:25, 3.15it/s, tok/s=15033.0]
Inference: 57%|█████▋ | 3385/5920 [47:25<13:25, 3.15it/s, tok/s=15034.7]
Inference: 57%|█████▋ | 3386/5920 [47:25<12:53, 3.27it/s, tok/s=15034.7]
Inference: 57%|█████▋ | 3386/5920 [47:25<12:53, 3.27it/s, tok/s=15032.3]
Inference: 57%|█████▋ | 3387/5920 [47:25<13:39, 3.09it/s, tok/s=15032.3]
Inference: 57%|█████▋ | 3387/5920 [47:25<13:39, 3.09it/s, tok/s=15035.5]
Inference: 57%|█████▋ | 3388/5920 [47:26<11:42, 3.61it/s, tok/s=15035.5]
Inference: 57%|█████▋ | 3388/5920 [47:26<11:42, 3.61it/s, tok/s=15041.4]
Inference: 57%|█████▋ | 3389/5920 [47:26<11:42, 3.61it/s, tok/s=15043.4]
Inference: 57%|█████▋ | 3390/5920 [47:26<08:24, 5.02it/s, tok/s=15043.4]
Inference: 57%|█████▋ | 3390/5920 [47:26<08:24, 5.02it/s, tok/s=15045.6]
Inference: 57%|█████▋ | 3391/5920 [47:26<09:46, 4.31it/s, tok/s=15045.6]
Inference: 57%|█████▋ | 3391/5920 [47:26<09:46, 4.31it/s, tok/s=15046.7]
Inference: 57%|█████▋ | 3392/5920 [47:26<09:51, 4.27it/s, tok/s=15046.7]
Inference: 57%|█████▋ | 3392/5920 [47:26<09:51, 4.27it/s, tok/s=15046.7]
Inference: 57%|█████▋ | 3393/5920 [47:27<09:27, 4.45it/s, tok/s=15046.7]
Inference: 57%|█████▋ | 3393/5920 [47:27<09:27, 4.45it/s, tok/s=15047.1]
Inference: 57%|█████▋ | 3394/5920 [47:27<16:12, 2.60it/s, tok/s=15047.1]
Inference: 57%|█████▋ | 3394/5920 [47:27<16:12, 2.60it/s, tok/s=15045.4]
Inference: 57%|█████▋ | 3395/5920 [47:27<16:11, 2.60it/s, tok/s=15048.1]
Inference: 57%|█████▋ | 3396/5920 [47:28<11:45, 3.58it/s, tok/s=15048.1]
Inference: 57%|█████▋ | 3396/5920 [47:28<11:45, 3.58it/s, tok/s=15047.6]
Inference: 57%|█████▋ | 3397/5920 [47:28<10:00, 4.20it/s, tok/s=15047.6]
Inference: 57%|█████▋ | 3397/5920 [47:28<10:00, 4.20it/s, tok/s=15047.7]
Inference: 57%|█████▋ | 3398/5920 [47:28<09:59, 4.20it/s, tok/s=15048.5]
Inference: 57%|█████▋ | 3399/5920 [47:28<07:09, 5.87it/s, tok/s=15048.5]
Inference: 57%|█████▋ | 3399/5920 [47:28<07:09, 5.87it/s, tok/s=15049.7]
Inference: 57%|█████▋ | 3400/5920 [47:29<12:16, 3.42it/s, tok/s=15049.7]
Inference: 57%|█████▋ | 3400/5920 [47:29<12:16, 3.42it/s, tok/s=15047.0]
Inference: 57%|█████▋ | 3401/5920 [47:29<12:16, 3.42it/s, tok/s=15047.3]
Inference: 57%|█████▋ | 3402/5920 [47:29<08:30, 4.93it/s, tok/s=15047.3]
Inference: 57%|█████▋ | 3402/5920 [47:29<08:30, 4.93it/s, tok/s=15047.8]
Inference: 57%|█████▋ | 3403/5920 [47:29<09:21, 4.49it/s, tok/s=15047.8]
Inference: 57%|█████▋ | 3403/5920 [47:29<09:21, 4.49it/s, tok/s=15049.8]
Inference: 57%|█████▊ | 3404/5920 [47:29<09:25, 4.45it/s, tok/s=15049.8]
Inference: 57%|█████▊ | 3404/5920 [47:29<09:25, 4.45it/s, tok/s=15048.9]
Inference: 58%|█████▊ | 3405/5920 [47:30<12:42, 3.30it/s, tok/s=15048.9]
Inference: 58%|█████▊ | 3405/5920 [47:30<12:42, 3.30it/s, tok/s=15048.0]
Inference: 58%|█████▊ | 3406/5920 [47:30<11:02, 3.80it/s, tok/s=15048.0]
Inference: 58%|█████▊ | 3406/5920 [47:30<11:02, 3.80it/s, tok/s=15052.6]
Inference: 58%|█████▊ | 3407/5920 [47:30<11:01, 3.80it/s, tok/s=15053.4]
Inference: 58%|█████▊ | 3408/5920 [47:31<11:56, 3.50it/s, tok/s=15053.4]
Inference: 58%|█████▊ | 3408/5920 [47:31<11:56, 3.50it/s, tok/s=15050.8]
Inference: 58%|█████▊ | 3409/5920 [47:31<11:56, 3.50it/s, tok/s=15054.1]
Inference: 58%|█████▊ | 3410/5920 [47:31<12:04, 3.46it/s, tok/s=15054.1]
Inference: 58%|█████▊ | 3410/5920 [47:31<12:04, 3.46it/s, tok/s=15053.4]
Inference: 58%|█████▊ | 3411/5920 [47:31<12:04, 3.46it/s, tok/s=15055.7]
Inference: 58%|█████▊ | 3412/5920 [47:31<08:46, 4.76it/s, tok/s=15055.7]
Inference: 58%|█████▊ | 3412/5920 [47:31<08:46, 4.76it/s, tok/s=15058.8]
Inference: 58%|█████▊ | 3413/5920 [47:31<08:46, 4.76it/s, tok/s=15061.1]
Inference: 58%|█████▊ | 3414/5920 [47:31<08:46, 4.76it/s, tok/s=15062.8]
Inference: 58%|█████▊ | 3415/5920 [47:32<08:01, 5.20it/s, tok/s=15062.8]
Inference: 58%|█████▊ | 3415/5920 [47:32<08:01, 5.20it/s, tok/s=15060.9]
Inference: 58%|█████▊ | 3416/5920 [47:32<08:07, 5.13it/s, tok/s=15060.9]
Inference: 58%|█████▊ | 3416/5920 [47:32<08:07, 5.13it/s, tok/s=15062.4]
Inference: 58%|█████▊ | 3417/5920 [47:32<08:07, 5.13it/s, tok/s=15064.6]
Inference: 58%|█████▊ | 3418/5920 [47:33<09:16, 4.50it/s, tok/s=15064.6]
Inference: 58%|█████▊ | 3418/5920 [47:33<09:16, 4.50it/s, tok/s=15064.4]
Inference: 58%|█████▊ | 3419/5920 [47:33<10:54, 3.82it/s, tok/s=15064.4]
Inference: 58%|█████▊ | 3419/5920 [47:33<10:54, 3.82it/s, tok/s=15063.9]
Inference: 58%|█████▊ | 3420/5920 [47:34<13:24, 3.11it/s, tok/s=15063.9]
Inference: 58%|█████▊ | 3420/5920 [47:34<13:24, 3.11it/s, tok/s=15061.7]
Inference: 58%|█████▊ | 3421/5920 [47:34<12:20, 3.37it/s, tok/s=15061.7]
Inference: 58%|█████▊ | 3421/5920 [47:34<12:20, 3.37it/s, tok/s=15061.8]
Inference: 58%|█████▊ | 3422/5920 [47:34<11:53, 3.50it/s, tok/s=15061.8]
Inference: 58%|█████▊ | 3422/5920 [47:34<11:53, 3.50it/s, tok/s=15061.3]
Inference: 58%|█████▊ | 3423/5920 [47:34<10:05, 4.13it/s, tok/s=15061.3]
Inference: 58%|█████▊ | 3423/5920 [47:34<10:05, 4.13it/s, tok/s=15061.5]
Inference: 58%|█████▊ | 3424/5920 [47:34<10:08, 4.10it/s, tok/s=15061.5]
Inference: 58%|█████▊ | 3424/5920 [47:34<10:08, 4.10it/s, tok/s=15060.8]
Inference: 58%|█████▊ | 3425/5920 [47:34<10:08, 4.10it/s, tok/s=15061.1]
Inference: 58%|█████▊ | 3426/5920 [47:35<10:28, 3.97it/s, tok/s=15061.1]
Inference: 58%|█████▊ | 3426/5920 [47:35<10:28, 3.97it/s, tok/s=15060.3]
Inference: 58%|█████▊ | 3427/5920 [47:35<09:36, 4.33it/s, tok/s=15060.3]
Inference: 58%|█████▊ | 3427/5920 [47:35<09:36, 4.33it/s, tok/s=15059.9]
Inference: 58%|█████▊ | 3428/5920 [47:35<09:35, 4.33it/s, tok/s=15062.1]
Inference: 58%|█████▊ | 3429/5920 [47:36<12:48, 3.24it/s, tok/s=15062.1]
Inference: 58%|█████▊ | 3429/5920 [47:36<12:48, 3.24it/s, tok/s=15059.9]
Inference: 58%|█████▊ | 3430/5920 [47:36<12:47, 3.24it/s, tok/s=15059.9]
Inference: 58%|█████▊ | 3431/5920 [47:36<10:46, 3.85it/s, tok/s=15059.9]
Inference: 58%|█████▊ | 3431/5920 [47:36<10:46, 3.85it/s, tok/s=15064.9]
Inference: 58%|█████▊ | 3432/5920 [47:37<11:20, 3.65it/s, tok/s=15064.9]
Inference: 58%|█████▊ | 3432/5920 [47:37<11:20, 3.65it/s, tok/s=15064.4]
Inference: 58%|█████▊ | 3433/5920 [47:37<11:20, 3.65it/s, tok/s=15066.3]
Inference: 58%|█████▊ | 3434/5920 [47:37<11:13, 3.69it/s, tok/s=15066.3]
Inference: 58%|█████▊ | 3434/5920 [47:37<11:13, 3.69it/s, tok/s=15064.1]
Inference: 58%|█████▊ | 3435/5920 [47:38<13:46, 3.01it/s, tok/s=15064.1]
Inference: 58%|█████▊ | 3435/5920 [47:38<13:46, 3.01it/s, tok/s=15064.8]
Inference: 58%|█████▊ | 3436/5920 [47:38<14:47, 2.80it/s, tok/s=15064.8]
Inference: 58%|█████▊ | 3436/5920 [47:38<14:47, 2.80it/s, tok/s=15066.5]
Inference: 58%|█████▊ | 3437/5920 [47:38<12:17, 3.37it/s, tok/s=15066.5]
Inference: 58%|█████▊ | 3437/5920 [47:38<12:17, 3.37it/s, tok/s=15074.9]
Inference: 58%|█████▊ | 3438/5920 [47:39<12:38, 3.27it/s, tok/s=15074.9]
Inference: 58%|█████▊ | 3438/5920 [47:39<12:38, 3.27it/s, tok/s=15078.0]
Inference: 58%|█████▊ | 3439/5920 [47:39<16:26, 2.51it/s, tok/s=15078.0]
Inference: 58%|█████▊ | 3439/5920 [47:39<16:26, 2.51it/s, tok/s=15075.6]
Inference: 58%|█████▊ | 3440/5920 [47:39<16:26, 2.51it/s, tok/s=15077.7]
Inference: 58%|█████▊ | 3441/5920 [47:40<16:40, 2.48it/s, tok/s=15077.7]
Inference: 58%|█████▊ | 3441/5920 [47:40<16:40, 2.48it/s, tok/s=15074.2]
Inference: 58%|█████▊ | 3442/5920 [47:40<16:39, 2.48it/s, tok/s=15074.8]
Inference: 58%|█████▊ | 3443/5920 [47:40<12:46, 3.23it/s, tok/s=15074.8]
Inference: 58%|█████▊ | 3443/5920 [47:40<12:46, 3.23it/s, tok/s=15081.6]
Inference: 58%|█████▊ | 3444/5920 [47:41<13:43, 3.01it/s, tok/s=15081.6]
Inference: 58%|█████▊ | 3444/5920 [47:41<13:43, 3.01it/s, tok/s=15080.1]
Inference: 58%|█████▊ | 3445/5920 [47:41<11:38, 3.54it/s, tok/s=15080.1]
Inference: 58%|█████▊ | 3445/5920 [47:41<11:38, 3.54it/s, tok/s=15080.4]
Inference: 58%|█████▊ | 3446/5920 [47:41<10:00, 4.12it/s, tok/s=15080.4]
Inference: 58%|█████▊ | 3446/5920 [47:41<10:00, 4.12it/s, tok/s=15080.2]
Inference: 58%|█████▊ | 3447/5920 [47:42<15:32, 2.65it/s, tok/s=15080.2]
Inference: 58%|█████▊ | 3447/5920 [47:42<15:32, 2.65it/s, tok/s=15078.1]
Inference: 58%|█████▊ | 3448/5920 [47:42<14:24, 2.86it/s, tok/s=15078.1]
Inference: 58%|█████▊ | 3448/5920 [47:42<14:24, 2.86it/s, tok/s=15077.1]
Inference: 58%|█████▊ | 3449/5920 [47:42<14:24, 2.86it/s, tok/s=15085.4]
Inference: 58%|█████▊ | 3450/5920 [47:42<09:52, 4.17it/s, tok/s=15085.4]
Inference: 58%|█████▊ | 3450/5920 [47:42<09:52, 4.17it/s, tok/s=15086.0]
Inference: 58%|█████▊ | 3451/5920 [47:42<09:51, 4.17it/s, tok/s=15086.7]
Inference: 58%|█████▊ | 3452/5920 [47:43<09:17, 4.43it/s, tok/s=15086.7]
Inference: 58%|█████▊ | 3452/5920 [47:43<09:17, 4.43it/s, tok/s=15088.6]
Inference: 58%|█████▊ | 3453/5920 [47:43<08:59, 4.57it/s, tok/s=15088.6]
Inference: 58%|█████▊ | 3453/5920 [47:43<08:59, 4.57it/s, tok/s=15088.1]
Inference: 58%|█████▊ | 3454/5920 [47:43<08:00, 5.13it/s, tok/s=15088.1]
Inference: 58%|█████▊ | 3454/5920 [47:43<08:00, 5.13it/s, tok/s=15090.0]
Inference: 58%|█████▊ | 3455/5920 [47:43<08:00, 5.13it/s, tok/s=15100.4]
Inference: 58%|█████▊ | 3456/5920 [47:44<10:50, 3.79it/s, tok/s=15100.4]
Inference: 58%|█████▊ | 3456/5920 [47:44<10:50, 3.79it/s, tok/s=15097.5]
Inference: 58%|█████▊ | 3457/5920 [47:45<16:55, 2.43it/s, tok/s=15097.5]
Inference: 58%|█████▊ | 3457/5920 [47:45<16:55, 2.43it/s, tok/s=15093.3]
Inference: 58%|█████▊ | 3458/5920 [47:45<16:55, 2.43it/s, tok/s=15093.8]
Inference: 58%|█████▊ | 3459/5920 [47:45<11:27, 3.58it/s, tok/s=15093.8]
Inference: 58%|█████▊ | 3459/5920 [47:45<11:27, 3.58it/s, tok/s=15095.5]
Inference: 58%|█████▊ | 3460/5920 [47:45<10:33, 3.88it/s, tok/s=15095.5]
Inference: 58%|█████▊ | 3460/5920 [47:45<10:33, 3.88it/s, tok/s=15098.5]
Inference: 58%|█████▊ | 3461/5920 [47:45<10:33, 3.88it/s, tok/s=15099.4]
Inference: 58%|█████▊ | 3462/5920 [47:45<09:06, 4.50it/s, tok/s=15099.4]
Inference: 58%|█████▊ | 3462/5920 [47:45<09:06, 4.50it/s, tok/s=15103.3]
Inference: 58%|█████▊ | 3463/5920 [47:46<11:17, 3.63it/s, tok/s=15103.3]
Inference: 58%|█████▊ | 3463/5920 [47:46<11:17, 3.63it/s, tok/s=15101.7]
Inference: 59%|█████▊ | 3464/5920 [47:46<11:17, 3.63it/s, tok/s=15104.4]
Inference: 59%|█████▊ | 3465/5920 [47:46<11:16, 3.63it/s, tok/s=15105.0]
Inference: 59%|█████▊ | 3466/5920 [47:46<07:04, 5.78it/s, tok/s=15105.0]
Inference: 59%|█████▊ | 3466/5920 [47:46<07:04, 5.78it/s, tok/s=15104.9]
Inference: 59%|█████▊ | 3467/5920 [47:46<07:32, 5.42it/s, tok/s=15104.9]
Inference: 59%|█████▊ | 3467/5920 [47:46<07:32, 5.42it/s, tok/s=15106.9]
Inference: 59%|█████▊ | 3468/5920 [47:46<06:53, 5.93it/s, tok/s=15106.9]
Inference: 59%|█████▊ | 3468/5920 [47:46<06:53, 5.93it/s, tok/s=15107.9]
Inference: 59%|█████▊ | 3469/5920 [47:46<06:53, 5.93it/s, tok/s=15107.9]
Inference: 59%|█████▊ | 3470/5920 [47:46<06:53, 5.93it/s, tok/s=15116.9]
Inference: 59%|█████▊ | 3471/5920 [47:47<08:15, 4.95it/s, tok/s=15116.9]
Inference: 59%|█████▊ | 3471/5920 [47:47<08:15, 4.95it/s, tok/s=15114.0]
Inference: 59%|█████▊ | 3472/5920 [47:47<08:14, 4.95it/s, tok/s=15114.2]
Inference: 59%|█████▊ | 3473/5920 [47:47<06:27, 6.31it/s, tok/s=15114.2]
Inference: 59%|█████▊ | 3473/5920 [47:47<06:27, 6.31it/s, tok/s=15116.1]
Inference: 59%|█████▊ | 3474/5920 [47:47<06:56, 5.87it/s, tok/s=15116.1]
Inference: 59%|█████▊ | 3474/5920 [47:47<06:56, 5.87it/s, tok/s=15116.5]
Inference: 59%|█████▊ | 3475/5920 [47:48<09:29, 4.29it/s, tok/s=15116.5]
Inference: 59%|█████▊ | 3475/5920 [47:48<09:29, 4.29it/s, tok/s=15121.0]
Inference: 59%|█████▊ | 3476/5920 [47:48<12:35, 3.24it/s, tok/s=15121.0]
Inference: 59%|█████▊ | 3476/5920 [47:48<12:35, 3.24it/s, tok/s=15121.0]
Inference: 59%|█████▊ | 3477/5920 [47:48<12:34, 3.24it/s, tok/s=15121.4]
Inference: 59%|█████▉ | 3478/5920 [47:49<09:08, 4.46it/s, tok/s=15121.4]
Inference: 59%|█████▉ | 3478/5920 [47:49<09:08, 4.46it/s, tok/s=15123.0]
Inference: 59%|█████▉ | 3479/5920 [47:49<14:42, 2.77it/s, tok/s=15123.0]
Inference: 59%|█████▉ | 3479/5920 [47:49<14:42, 2.77it/s, tok/s=15119.4]
Inference: 59%|█████▉ | 3480/5920 [47:49<14:41, 2.77it/s, tok/s=15120.4]
Inference: 59%|█████▉ | 3481/5920 [47:50<12:53, 3.15it/s, tok/s=15120.4]
Inference: 59%|█████▉ | 3481/5920 [47:50<12:53, 3.15it/s, tok/s=15118.6]
Inference: 59%|█████▉ | 3482/5920 [47:51<15:59, 2.54it/s, tok/s=15118.6]
Inference: 59%|█████▉ | 3482/5920 [47:51<15:59, 2.54it/s, tok/s=15117.6]
Inference: 59%|█████▉ | 3483/5920 [47:51<14:19, 2.83it/s, tok/s=15117.6]
Inference: 59%|█████▉ | 3483/5920 [47:51<14:19, 2.83it/s, tok/s=15123.1]
Inference: 59%|█████▉ | 3484/5920 [47:51<14:19, 2.83it/s, tok/s=15123.3]
Inference: 59%|█████▉ | 3485/5920 [47:51<11:50, 3.43it/s, tok/s=15123.3]
Inference: 59%|█████▉ | 3485/5920 [47:51<11:50, 3.43it/s, tok/s=15122.1]
Inference: 59%|█████▉ | 3486/5920 [47:51<10:14, 3.96it/s, tok/s=15122.1]
Inference: 59%|█████▉ | 3486/5920 [47:51<10:14, 3.96it/s, tok/s=15122.5]
Inference: 59%|█████▉ | 3487/5920 [47:52<10:00, 4.05it/s, tok/s=15122.5]
Inference: 59%|█████▉ | 3487/5920 [47:52<10:00, 4.05it/s, tok/s=15122.8]
Inference: 59%|█████▉ | 3488/5920 [47:52<13:33, 2.99it/s, tok/s=15122.8]
Inference: 59%|█████▉ | 3488/5920 [47:52<13:33, 2.99it/s, tok/s=15123.8]
Inference: 59%|█████▉ | 3489/5920 [47:52<13:33, 2.99it/s, tok/s=15124.1]
Inference: 59%|█████▉ | 3490/5920 [47:53<11:42, 3.46it/s, tok/s=15124.1]
Inference: 59%|█████▉ | 3490/5920 [47:53<11:42, 3.46it/s, tok/s=15122.8]
Inference: 59%|█████▉ | 3491/5920 [47:53<11:45, 3.44it/s, tok/s=15122.8]
Inference: 59%|█████▉ | 3491/5920 [47:53<11:45, 3.44it/s, tok/s=15121.9]
Inference: 59%|█████▉ | 3492/5920 [47:53<09:51, 4.10it/s, tok/s=15121.9]
Inference: 59%|█████▉ | 3492/5920 [47:53<09:51, 4.10it/s, tok/s=15125.0]
Inference: 59%|█████▉ | 3493/5920 [47:53<09:41, 4.18it/s, tok/s=15125.0]
Inference: 59%|█████▉ | 3493/5920 [47:53<09:41, 4.18it/s, tok/s=15131.8]
Inference: 59%|█████▉ | 3494/5920 [47:54<11:28, 3.52it/s, tok/s=15131.8]
Inference: 59%|█████▉ | 3494/5920 [47:54<11:28, 3.52it/s, tok/s=15130.5]
Inference: 59%|█████▉ | 3495/5920 [47:54<11:28, 3.52it/s, tok/s=15131.2]
Inference: 59%|█████▉ | 3496/5920 [47:54<07:33, 5.35it/s, tok/s=15131.2]
Inference: 59%|█████▉ | 3496/5920 [47:54<07:33, 5.35it/s, tok/s=15136.1]
Inference: 59%|█████▉ | 3497/5920 [47:54<07:32, 5.35it/s, tok/s=15136.4]
Inference: 59%|█████▉ | 3498/5920 [47:54<07:29, 5.39it/s, tok/s=15136.4]
Inference: 59%|█████▉ | 3498/5920 [47:54<07:29, 5.39it/s, tok/s=15137.2]
Inference: 59%|█████▉ | 3499/5920 [47:54<07:14, 5.58it/s, tok/s=15137.2]
Inference: 59%|█████▉ | 3499/5920 [47:54<07:14, 5.58it/s, tok/s=15137.3]
Inference: 59%|█████▉ | 3500/5920 [47:54<07:38, 5.27it/s, tok/s=15137.3]
Inference: 59%|█████▉ | 3500/5920 [47:54<07:38, 5.27it/s, tok/s=15136.9]
Inference: 59%|█████▉ | 3501/5920 [47:55<14:25, 2.80it/s, tok/s=15136.9]
Inference: 59%|█████▉ | 3501/5920 [47:55<14:25, 2.80it/s, tok/s=15132.8]
Inference: 59%|█████▉ | 3502/5920 [47:56<12:43, 3.17it/s, tok/s=15132.8]
Inference: 59%|█████▉ | 3502/5920 [47:56<12:43, 3.17it/s, tok/s=15132.2]
Inference: 59%|█████▉ | 3503/5920 [47:56<12:42, 3.17it/s, tok/s=15132.7]
Inference: 59%|█████▉ | 3504/5920 [47:56<08:19, 4.83it/s, tok/s=15132.7]
Inference: 59%|█████▉ | 3504/5920 [47:56<08:19, 4.83it/s, tok/s=15132.8]
Inference: 59%|█████▉ | 3505/5920 [47:56<09:03, 4.44it/s, tok/s=15132.8]
Inference: 59%|█████▉ | 3505/5920 [47:56<09:03, 4.44it/s, tok/s=15131.9]
Inference: 59%|█████▉ | 3506/5920 [47:56<09:03, 4.44it/s, tok/s=15132.3]
Inference: 59%|█████▉ | 3507/5920 [47:56<07:41, 5.23it/s, tok/s=15132.3]
Inference: 59%|█████▉ | 3507/5920 [47:56<07:41, 5.23it/s, tok/s=15131.4]
Inference: 59%|█████▉ | 3508/5920 [47:56<07:41, 5.23it/s, tok/s=15131.7]
Inference: 59%|█████▉ | 3509/5920 [47:56<06:49, 5.89it/s, tok/s=15131.7]
Inference: 59%|█████▉ | 3509/5920 [47:56<06:49, 5.89it/s, tok/s=15139.4]
Inference: 59%|█████▉ | 3510/5920 [47:57<06:36, 6.08it/s, tok/s=15139.4]
Inference: 59%|█████▉ | 3510/5920 [47:57<06:36, 6.08it/s, tok/s=15139.3]
Inference: 59%|█████▉ | 3511/5920 [47:57<07:51, 5.11it/s, tok/s=15139.3]
Inference: 59%|█████▉ | 3511/5920 [47:57<07:51, 5.11it/s, tok/s=15138.5]
Inference: 59%|█████▉ | 3512/5920 [47:57<07:53, 5.08it/s, tok/s=15138.5]
Inference: 59%|█████▉ | 3512/5920 [47:57<07:53, 5.08it/s, tok/s=15138.1]
Inference: 59%|█████▉ | 3513/5920 [47:57<09:11, 4.37it/s, tok/s=15138.1]
Inference: 59%|█████▉ | 3513/5920 [47:57<09:11, 4.37it/s, tok/s=15142.9]
Inference: 59%|█████▉ | 3514/5920 [47:59<19:35, 2.05it/s, tok/s=15142.9]
Inference: 59%|█████▉ | 3514/5920 [47:59<19:35, 2.05it/s, tok/s=15140.8]
Inference: 59%|█████▉ | 3515/5920 [47:59<16:47, 2.39it/s, tok/s=15140.8]
Inference: 59%|█████▉ | 3515/5920 [47:59<16:47, 2.39it/s, tok/s=15139.9]
Inference: 59%|█████▉ | 3516/5920 [47:59<14:05, 2.84it/s, tok/s=15139.9]
Inference: 59%|█████▉ | 3516/5920 [47:59<14:05, 2.84it/s, tok/s=15139.8]
Inference: 59%|█████▉ | 3517/5920 [47:59<12:39, 3.16it/s, tok/s=15139.8]
Inference: 59%|█████▉ | 3517/5920 [47:59<12:39, 3.16it/s, tok/s=15138.9]
Inference: 59%|█████▉ | 3518/5920 [48:00<13:07, 3.05it/s, tok/s=15138.9]
Inference: 59%|█████▉ | 3518/5920 [48:00<13:07, 3.05it/s, tok/s=15139.3]
Inference: 59%|█████▉ | 3519/5920 [48:00<14:34, 2.74it/s, tok/s=15139.3]
Inference: 59%|█████▉ | 3519/5920 [48:00<14:34, 2.74it/s, tok/s=15138.1]
Inference: 59%|█████▉ | 3520/5920 [48:00<11:30, 3.48it/s, tok/s=15138.1]
Inference: 59%|█████▉ | 3520/5920 [48:00<11:30, 3.48it/s, tok/s=15139.5]
Inference: 59%|█████▉ | 3521/5920 [48:00<11:30, 3.48it/s, tok/s=15141.5]
Inference: 59%|█████▉ | 3522/5920 [48:01<11:19, 3.53it/s, tok/s=15141.5]
Inference: 59%|█████▉ | 3522/5920 [48:01<11:19, 3.53it/s, tok/s=15139.3]
Inference: 60%|█████▉ | 3523/5920 [48:01<09:37, 4.15it/s, tok/s=15139.3]
Inference: 60%|█████▉ | 3523/5920 [48:01<09:37, 4.15it/s, tok/s=15139.3]
Inference: 60%|█████▉ | 3524/5920 [48:01<08:31, 4.69it/s, tok/s=15139.3]
Inference: 60%|█████▉ | 3524/5920 [48:01<08:31, 4.69it/s, tok/s=15140.3]
Inference: 60%|█████▉ | 3525/5920 [48:01<08:30, 4.69it/s, tok/s=15140.6]
Inference: 60%|█████▉ | 3526/5920 [48:01<06:35, 6.05it/s, tok/s=15140.6]
Inference: 60%|█████▉ | 3526/5920 [48:01<06:35, 6.05it/s, tok/s=15141.4]
Inference: 60%|█████▉ | 3527/5920 [48:01<07:25, 5.37it/s, tok/s=15141.4]
Inference: 60%|█████▉ | 3527/5920 [48:01<07:25, 5.37it/s, tok/s=15140.7]
Inference: 60%|█████▉ | 3528/5920 [48:02<08:50, 4.51it/s, tok/s=15140.7]
Inference: 60%|█████▉ | 3528/5920 [48:02<08:50, 4.51it/s, tok/s=15141.9]
Inference: 60%|█████▉ | 3529/5920 [48:02<08:50, 4.51it/s, tok/s=15142.1]
Inference: 60%|█████▉ | 3530/5920 [48:02<09:54, 4.02it/s, tok/s=15142.1]
Inference: 60%|█████▉ | 3530/5920 [48:02<09:54, 4.02it/s, tok/s=15141.5]
Inference: 60%|█████▉ | 3531/5920 [48:02<09:54, 4.02it/s, tok/s=15141.9]
Inference: 60%|█████▉ | 3532/5920 [48:03<07:12, 5.52it/s, tok/s=15141.9]
Inference: 60%|█████▉ | 3532/5920 [48:03<07:12, 5.52it/s, tok/s=15148.2]
Inference: 60%|█████▉ | 3533/5920 [48:03<12:27, 3.19it/s, tok/s=15148.2]
Inference: 60%|█████▉ | 3533/5920 [48:03<12:27, 3.19it/s, tok/s=15144.5]
Inference: 60%|█████▉ | 3534/5920 [48:04<12:47, 3.11it/s, tok/s=15144.5]
Inference: 60%|█████▉ | 3534/5920 [48:04<12:47, 3.11it/s, tok/s=15143.8]
Inference: 60%|█████▉ | 3535/5920 [48:04<13:19, 2.98it/s, tok/s=15143.8]
Inference: 60%|█████▉ | 3535/5920 [48:04<13:19, 2.98it/s, tok/s=15151.7]
Inference: 60%|█████▉ | 3536/5920 [48:04<13:19, 2.98it/s, tok/s=15152.3]
Inference: 60%|█████▉ | 3537/5920 [48:04<09:07, 4.35it/s, tok/s=15152.3]
Inference: 60%|█████▉ | 3537/5920 [48:04<09:07, 4.35it/s, tok/s=15152.7]
Inference: 60%|█████▉ | 3538/5920 [48:05<11:17, 3.52it/s, tok/s=15152.7]
Inference: 60%|█████▉ | 3538/5920 [48:05<11:17, 3.52it/s, tok/s=15152.1]
Inference: 60%|█████▉ | 3539/5920 [48:05<11:40, 3.40it/s, tok/s=15152.1]
Inference: 60%|█████▉ | 3539/5920 [48:05<11:40, 3.40it/s, tok/s=15157.7]
Inference: 60%|█████▉ | 3540/5920 [48:05<09:59, 3.97it/s, tok/s=15157.7]
Inference: 60%|█████▉ | 3540/5920 [48:05<09:59, 3.97it/s, tok/s=15158.2]
Inference: 60%|█████▉ | 3541/5920 [48:05<10:48, 3.67it/s, tok/s=15158.2]
Inference: 60%|█████▉ | 3541/5920 [48:05<10:48, 3.67it/s, tok/s=15157.2]
Inference: 60%|█████▉ | 3542/5920 [48:06<10:48, 3.67it/s, tok/s=15157.4]
Inference: 60%|█████▉ | 3543/5920 [48:06<09:16, 4.27it/s, tok/s=15157.4]
Inference: 60%|█████▉ | 3543/5920 [48:06<09:16, 4.27it/s, tok/s=15156.5]
Inference: 60%|█████▉ | 3544/5920 [48:07<14:21, 2.76it/s, tok/s=15156.5]
Inference: 60%|█████▉ | 3544/5920 [48:07<14:21, 2.76it/s, tok/s=15153.3]
Inference: 60%|█████▉ | 3545/5920 [48:07<14:20, 2.76it/s, tok/s=15153.5]
Inference: 60%|█████▉ | 3546/5920 [48:07<14:20, 2.76it/s, tok/s=15156.6]
Inference: 60%|█████▉ | 3547/5920 [48:07<08:36, 4.59it/s, tok/s=15156.6]
Inference: 60%|█████▉ | 3547/5920 [48:07<08:36, 4.59it/s, tok/s=15156.4]
Inference: 60%|█████▉ | 3548/5920 [48:07<08:36, 4.59it/s, tok/s=15156.3]
Inference: 60%|█████▉ | 3549/5920 [48:07<06:43, 5.87it/s, tok/s=15156.3]
Inference: 60%|█████▉ | 3549/5920 [48:07<06:43, 5.87it/s, tok/s=15156.6]
Inference: 60%|█████▉ | 3550/5920 [48:07<06:43, 5.87it/s, tok/s=15163.4]
Inference: 60%|█████▉ | 3551/5920 [48:08<08:08, 4.85it/s, tok/s=15163.4]
Inference: 60%|█████▉ | 3551/5920 [48:08<08:08, 4.85it/s, tok/s=15168.1]
Inference: 60%|██████ | 3552/5920 [48:08<07:55, 4.98it/s, tok/s=15168.1]
Inference: 60%|██████ | 3552/5920 [48:08<07:55, 4.98it/s, tok/s=15169.3]
Inference: 60%|██████ | 3553/5920 [48:08<08:18, 4.75it/s, tok/s=15169.3]
Inference: 60%|██████ | 3553/5920 [48:08<08:18, 4.75it/s, tok/s=15174.9]
Inference: 60%|██████ | 3554/5920 [48:08<10:21, 3.81it/s, tok/s=15174.9]
Inference: 60%|██████ | 3554/5920 [48:08<10:21, 3.81it/s, tok/s=15173.2]
Inference: 60%|██████ | 3555/5920 [48:09<09:09, 4.31it/s, tok/s=15173.2]
Inference: 60%|██████ | 3555/5920 [48:09<09:09, 4.31it/s, tok/s=15180.9]
Inference: 60%|██████ | 3556/5920 [48:09<14:54, 2.64it/s, tok/s=15180.9]
Inference: 60%|██████ | 3556/5920 [48:09<14:54, 2.64it/s, tok/s=15177.5]
Inference: 60%|██████ | 3557/5920 [48:10<15:26, 2.55it/s, tok/s=15177.5]
Inference: 60%|██████ | 3557/5920 [48:10<15:26, 2.55it/s, tok/s=15182.2]
Inference: 60%|██████ | 3558/5920 [48:10<16:25, 2.40it/s, tok/s=15182.2]
Inference: 60%|██████ | 3558/5920 [48:10<16:25, 2.40it/s, tok/s=15180.1]
Inference: 60%|██████ | 3559/5920 [48:10<12:54, 3.05it/s, tok/s=15180.1]
Inference: 60%|██████ | 3559/5920 [48:10<12:54, 3.05it/s, tok/s=15189.5]
Inference: 60%|██████ | 3560/5920 [48:10<12:53, 3.05it/s, tok/s=15189.9]
Inference: 60%|██████ | 3561/5920 [48:11<09:26, 4.17it/s, tok/s=15189.9]
Inference: 60%|██████ | 3561/5920 [48:11<09:26, 4.17it/s, tok/s=15195.7]
Inference: 60%|██████ | 3562/5920 [48:11<11:36, 3.38it/s, tok/s=15195.7]
Inference: 60%|██████ | 3562/5920 [48:11<11:36, 3.38it/s, tok/s=15193.7]
Inference: 60%|██████ | 3563/5920 [48:11<10:50, 3.62it/s, tok/s=15193.7]
Inference: 60%|██████ | 3563/5920 [48:11<10:50, 3.62it/s, tok/s=15193.0]
Inference: 60%|██████ | 3564/5920 [48:12<15:27, 2.54it/s, tok/s=15193.0]
Inference: 60%|██████ | 3564/5920 [48:12<15:27, 2.54it/s, tok/s=15190.6]
Inference: 60%|██████ | 3565/5920 [48:12<13:17, 2.95it/s, tok/s=15190.6]
Inference: 60%|██████ | 3565/5920 [48:12<13:17, 2.95it/s, tok/s=15190.6]
Inference: 60%|██████ | 3566/5920 [48:12<13:17, 2.95it/s, tok/s=15198.1]
Inference: 60%|██████ | 3567/5920 [48:12<09:22, 4.19it/s, tok/s=15198.1]
Inference: 60%|██████ | 3567/5920 [48:12<09:22, 4.19it/s, tok/s=15198.2]
Inference: 60%|██████ | 3568/5920 [48:13<08:05, 4.84it/s, tok/s=15198.2]
Inference: 60%|██████ | 3568/5920 [48:13<08:05, 4.84it/s, tok/s=15198.3]
Inference: 60%|██████ | 3569/5920 [48:13<10:15, 3.82it/s, tok/s=15198.3]
Inference: 60%|██████ | 3569/5920 [48:13<10:15, 3.82it/s, tok/s=15206.5]
Inference: 60%|██████ | 3570/5920 [48:13<10:14, 3.82it/s, tok/s=15214.1]
Inference: 60%|██████ | 3571/5920 [48:13<07:35, 5.16it/s, tok/s=15214.1]
Inference: 60%|██████ | 3571/5920 [48:13<07:35, 5.16it/s, tok/s=15216.5]
Inference: 60%|██████ | 3572/5920 [48:13<07:56, 4.93it/s, tok/s=15216.5]
Inference: 60%|██████ | 3572/5920 [48:13<07:56, 4.93it/s, tok/s=15217.9]
Inference: 60%|██████ | 3573/5920 [48:14<09:58, 3.92it/s, tok/s=15217.9]
Inference: 60%|██████ | 3573/5920 [48:14<09:58, 3.92it/s, tok/s=15216.4]
Inference: 60%|██████ | 3574/5920 [48:14<09:57, 3.92it/s, tok/s=15217.6]
Inference: 60%|██████ | 3575/5920 [48:14<07:52, 4.96it/s, tok/s=15217.6]
Inference: 60%|██████ | 3575/5920 [48:14<07:52, 4.96it/s, tok/s=15219.1]
Inference: 60%|██████ | 3576/5920 [48:14<07:52, 4.96it/s, tok/s=15226.5]
Inference: 60%|██████ | 3577/5920 [48:14<07:11, 5.43it/s, tok/s=15226.5]
Inference: 60%|██████ | 3577/5920 [48:14<07:11, 5.43it/s, tok/s=15225.5]
Inference: 60%|██████ | 3578/5920 [48:15<09:48, 3.98it/s, tok/s=15225.5]
Inference: 60%|██████ | 3578/5920 [48:15<09:48, 3.98it/s, tok/s=15223.7]
Inference: 60%|██████ | 3579/5920 [48:15<11:07, 3.51it/s, tok/s=15223.7]
Inference: 60%|██████ | 3579/5920 [48:15<11:07, 3.51it/s, tok/s=15222.1]
Inference: 60%|██████ | 3580/5920 [48:16<13:48, 2.82it/s, tok/s=15222.1]
Inference: 60%|██████ | 3580/5920 [48:16<13:48, 2.82it/s, tok/s=15221.0]
Inference: 60%|██████ | 3581/5920 [48:16<11:50, 3.29it/s, tok/s=15221.0]
Inference: 60%|██████ | 3581/5920 [48:16<11:50, 3.29it/s, tok/s=15220.7]
Inference: 61%|██████ | 3582/5920 [48:16<11:50, 3.29it/s, tok/s=15226.6]
Inference: 61%|██████ | 3583/5920 [48:16<09:46, 3.99it/s, tok/s=15226.6]
Inference: 61%|██████ | 3583/5920 [48:16<09:46, 3.99it/s, tok/s=15226.3]
Inference: 61%|██████ | 3584/5920 [48:16<08:29, 4.58it/s, tok/s=15226.3]
Inference: 61%|██████ | 3584/5920 [48:16<08:29, 4.58it/s, tok/s=15226.6]
Inference: 61%|██████ | 3585/5920 [48:17<08:29, 4.58it/s, tok/s=15229.3]
Inference: 61%|██████ | 3586/5920 [48:17<06:16, 6.19it/s, tok/s=15229.3]
Inference: 61%|██████ | 3586/5920 [48:17<06:16, 6.19it/s, tok/s=15229.5]
Inference: 61%|██████ | 3587/5920 [48:17<12:10, 3.19it/s, tok/s=15229.5]
Inference: 61%|██████ | 3587/5920 [48:17<12:10, 3.19it/s, tok/s=15225.9]
Inference: 61%|██████ | 3588/5920 [48:18<12:01, 3.23it/s, tok/s=15225.9]
Inference: 61%|██████ | 3588/5920 [48:18<12:01, 3.23it/s, tok/s=15225.3]
Inference: 61%|██████ | 3589/5920 [48:18<10:12, 3.81it/s, tok/s=15225.3]
Inference: 61%|██████ | 3589/5920 [48:18<10:12, 3.81it/s, tok/s=15225.5]
Inference: 61%|██████ | 3590/5920 [48:18<11:10, 3.48it/s, tok/s=15225.5]
Inference: 61%|██████ | 3590/5920 [48:18<11:10, 3.48it/s, tok/s=15225.3]
Inference: 61%|██████ | 3591/5920 [48:18<11:09, 3.48it/s, tok/s=15228.0]
Inference: 61%|██████ | 3592/5920 [48:19<09:50, 3.94it/s, tok/s=15228.0]
Inference: 61%|██████ | 3592/5920 [48:19<09:50, 3.94it/s, tok/s=15234.5]
Inference: 61%|██████ | 3593/5920 [48:19<10:12, 3.80it/s, tok/s=15234.5]
Inference: 61%|██████ | 3593/5920 [48:19<10:12, 3.80it/s, tok/s=15235.0]
Inference: 61%|██████ | 3594/5920 [48:19<10:46, 3.60it/s, tok/s=15235.0]
Inference: 61%|██████ | 3594/5920 [48:19<10:46, 3.60it/s, tok/s=15234.2]
Inference: 61%|██████ | 3595/5920 [48:19<10:46, 3.60it/s, tok/s=15234.8]
Inference: 61%|██████ | 3596/5920 [48:20<09:16, 4.17it/s, tok/s=15234.8]
Inference: 61%|██████ | 3596/5920 [48:20<09:16, 4.17it/s, tok/s=15233.5]
Inference: 61%|██████ | 3597/5920 [48:20<09:02, 4.29it/s, tok/s=15233.5]
Inference: 61%|██████ | 3597/5920 [48:20<09:02, 4.29it/s, tok/s=15233.9]
Inference: 61%|██████ | 3598/5920 [48:20<10:46, 3.59it/s, tok/s=15233.9]
Inference: 61%|██████ | 3598/5920 [48:20<10:46, 3.59it/s, tok/s=15232.9]
Inference: 61%|██████ | 3599/5920 [48:21<11:51, 3.26it/s, tok/s=15232.9]
Inference: 61%|██████ | 3599/5920 [48:21<11:51, 3.26it/s, tok/s=15231.6]
Inference: 61%|██████ | 3600/5920 [48:21<10:13, 3.78it/s, tok/s=15231.6]
Inference: 61%|██████ | 3600/5920 [48:21<10:13, 3.78it/s, tok/s=15232.8]
Inference: 61%|██████ | 3601/5920 [48:21<08:37, 4.48it/s, tok/s=15232.8]
Inference: 61%|██████ | 3601/5920 [48:21<08:37, 4.48it/s, tok/s=15233.2]
Inference: 61%|██████ | 3602/5920 [48:21<08:37, 4.48it/s, tok/s=15242.7]
Inference: 61%|██████ | 3603/5920 [48:21<09:06, 4.24it/s, tok/s=15242.7]
Inference: 61%|██████ | 3603/5920 [48:21<09:06, 4.24it/s, tok/s=15241.6]
Inference: 61%|██████ | 3604/5920 [48:22<09:11, 4.20it/s, tok/s=15241.6]
Inference: 61%|██████ | 3604/5920 [48:22<09:11, 4.20it/s, tok/s=15241.0]
Inference: 61%|██████ | 3605/5920 [48:22<08:48, 4.38it/s, tok/s=15241.0]
Inference: 61%|██████ | 3605/5920 [48:22<08:48, 4.38it/s, tok/s=15240.9]
Inference: 61%|██████ | 3606/5920 [48:22<10:04, 3.83it/s, tok/s=15240.9]
Inference: 61%|██████ | 3606/5920 [48:22<10:04, 3.83it/s, tok/s=15242.4]
Inference: 61%|██████ | 3607/5920 [48:23<10:20, 3.73it/s, tok/s=15242.4]
Inference: 61%|██████ | 3607/5920 [48:23<10:20, 3.73it/s, tok/s=15244.1]
Inference: 61%|██████ | 3608/5920 [48:23<11:23, 3.38it/s, tok/s=15244.1]
Inference: 61%|██████ | 3608/5920 [48:23<11:23, 3.38it/s, tok/s=15242.8]
Inference: 61%|██████ | 3609/5920 [48:23<13:01, 2.96it/s, tok/s=15242.8]
Inference: 61%|██████ | 3609/5920 [48:23<13:01, 2.96it/s, tok/s=15243.4]
Inference: 61%|██████ | 3610/5920 [48:24<11:26, 3.36it/s, tok/s=15243.4]
Inference: 61%|██████ | 3610/5920 [48:24<11:26, 3.36it/s, tok/s=15243.4]
Inference: 61%|██████ | 3611/5920 [48:24<12:30, 3.08it/s, tok/s=15243.4]
Inference: 61%|██████ | 3611/5920 [48:24<12:30, 3.08it/s, tok/s=15243.0]
Inference: 61%|██████ | 3612/5920 [48:24<11:13, 3.43it/s, tok/s=15243.0]
Inference: 61%|██████ | 3612/5920 [48:24<11:13, 3.43it/s, tok/s=15243.1]
Inference: 61%|██████ | 3613/5920 [48:25<13:54, 2.77it/s, tok/s=15243.1]
Inference: 61%|██████ | 3613/5920 [48:25<13:54, 2.77it/s, tok/s=15241.2]
Inference: 61%|██████ | 3614/5920 [48:25<13:02, 2.95it/s, tok/s=15241.2]
Inference: 61%|██████ | 3614/5920 [48:25<13:02, 2.95it/s, tok/s=15244.1]
Inference: 61%|██████ | 3615/5920 [48:25<11:48, 3.25it/s, tok/s=15244.1]
Inference: 61%|██████ | 3615/5920 [48:25<11:48, 3.25it/s, tok/s=15243.5]
Inference: 61%|██████ | 3616/5920 [48:25<11:47, 3.25it/s, tok/s=15244.8]
Inference: 61%|██████ | 3617/5920 [48:26<10:11, 3.77it/s, tok/s=15244.8]
Inference: 61%|██████ | 3617/5920 [48:26<10:11, 3.77it/s, tok/s=15243.5]
Inference: 61%|██████ | 3618/5920 [48:26<10:07, 3.79it/s, tok/s=15243.5]
Inference: 61%|██████ | 3618/5920 [48:26<10:07, 3.79it/s, tok/s=15242.8]
Inference: 61%|██████ | 3619/5920 [48:27<16:48, 2.28it/s, tok/s=15242.8]
Inference: 61%|██████ | 3619/5920 [48:27<16:48, 2.28it/s, tok/s=15241.9]
Inference: 61%|██████ | 3620/5920 [48:27<17:56, 2.14it/s, tok/s=15241.9]
Inference: 61%|██████ | 3620/5920 [48:27<17:56, 2.14it/s, tok/s=15239.6]
Inference: 61%|██████ | 3621/5920 [48:28<16:16, 2.35it/s, tok/s=15239.6]
Inference: 61%|██████ | 3621/5920 [48:28<16:16, 2.35it/s, tok/s=15239.2]
Inference: 61%|██████ | 3622/5920 [48:28<13:41, 2.80it/s, tok/s=15239.2]
Inference: 61%|██████ | 3622/5920 [48:28<13:41, 2.80it/s, tok/s=15239.3]
Inference: 61%|██████ | 3623/5920 [48:28<13:41, 2.80it/s, tok/s=15242.0]
Inference: 61%|██████ | 3624/5920 [48:31<32:08, 1.19it/s, tok/s=15242.0]
Inference: 61%|██████ | 3624/5920 [48:31<32:08, 1.19it/s, tok/s=15230.8]
Inference: 61%|██████ | 3625/5920 [48:31<27:46, 1.38it/s, tok/s=15230.8]
Inference: 61%|██████ | 3625/5920 [48:31<27:46, 1.38it/s, tok/s=15229.6]
Inference: 61%|██████▏ | 3626/5920 [48:32<30:27, 1.25it/s, tok/s=15229.6]
Inference: 61%|██████▏ | 3626/5920 [48:32<30:27, 1.25it/s, tok/s=15225.1]
Inference: 61%|██████▏ | 3627/5920 [48:32<30:27, 1.25it/s, tok/s=15228.7]
Inference: 61%|██████▏ | 3628/5920 [48:32<18:38, 2.05it/s, tok/s=15228.7]
Inference: 61%|██████▏ | 3628/5920 [48:32<18:38, 2.05it/s, tok/s=15229.0]
Inference: 61%|██████▏ | 3629/5920 [48:32<16:13, 2.35it/s, tok/s=15229.0]
Inference: 61%|██████▏ | 3629/5920 [48:32<16:13, 2.35it/s, tok/s=15228.8]
Inference: 61%|██████▏ | 3630/5920 [48:33<16:12, 2.35it/s, tok/s=15230.7]
Inference: 61%|██████▏ | 3631/5920 [48:33<10:31, 3.62it/s, tok/s=15230.7]
Inference: 61%|██████▏ | 3631/5920 [48:33<10:31, 3.62it/s, tok/s=15231.1]
Inference: 61%|██████▏ | 3632/5920 [48:33<09:33, 3.99it/s, tok/s=15231.1]
Inference: 61%|██████▏ | 3632/5920 [48:33<09:33, 3.99it/s, tok/s=15230.8]
Inference: 61%|██████▏ | 3633/5920 [48:33<09:33, 3.99it/s, tok/s=15233.3]
Inference: 61%|██████▏ | 3634/5920 [48:34<12:10, 3.13it/s, tok/s=15233.3]
Inference: 61%|██████▏ | 3634/5920 [48:34<12:10, 3.13it/s, tok/s=15230.3]
Inference: 61%|██████▏ | 3635/5920 [48:34<10:36, 3.59it/s, tok/s=15230.3]
Inference: 61%|██████▏ | 3635/5920 [48:34<10:36, 3.59it/s, tok/s=15240.1]
Inference: 61%|██████▏ | 3636/5920 [48:34<10:11, 3.74it/s, tok/s=15240.1]
Inference: 61%|██████▏ | 3636/5920 [48:34<10:11, 3.74it/s, tok/s=15245.4]
Inference: 61%|██████▏ | 3637/5920 [48:35<14:37, 2.60it/s, tok/s=15245.4]
Inference: 61%|██████▏ | 3637/5920 [48:35<14:37, 2.60it/s, tok/s=15249.5]
Inference: 61%|██████▏ | 3638/5920 [48:35<13:25, 2.83it/s, tok/s=15249.5]
Inference: 61%|██████▏ | 3638/5920 [48:35<13:25, 2.83it/s, tok/s=15248.5]
Inference: 61%|██████▏ | 3639/5920 [48:35<12:41, 2.99it/s, tok/s=15248.5]
Inference: 61%|██████▏ | 3639/5920 [48:35<12:41, 2.99it/s, tok/s=15248.8]
Inference: 61%|██████▏ | 3640/5920 [48:35<12:12, 3.11it/s, tok/s=15248.8]
Inference: 61%|██████▏ | 3640/5920 [48:35<12:12, 3.11it/s, tok/s=15248.0]
Inference: 62%|██████▏ | 3641/5920 [48:36<12:11, 3.11it/s, tok/s=15255.4]
Inference: 62%|██████▏ | 3642/5920 [48:36<08:13, 4.62it/s, tok/s=15255.4]
Inference: 62%|██████▏ | 3642/5920 [48:36<08:13, 4.62it/s, tok/s=15256.1]
Inference: 62%|██████▏ | 3643/5920 [48:36<08:26, 4.50it/s, tok/s=15256.1]
Inference: 62%|██████▏ | 3643/5920 [48:36<08:26, 4.50it/s, tok/s=15255.7]
Inference: 62%|██████▏ | 3644/5920 [48:36<08:25, 4.50it/s, tok/s=15255.7]
Inference: 62%|██████▏ | 3644/5920 [48:36<08:25, 4.50it/s, tok/s=15255.5]
Inference: 62%|██████▏ | 3645/5920 [48:36<08:25, 4.50it/s, tok/s=15255.9]
Inference: 62%|██████▏ | 3646/5920 [48:36<06:39, 5.70it/s, tok/s=15255.9]
Inference: 62%|██████▏ | 3646/5920 [48:36<06:39, 5.70it/s, tok/s=15256.0]
Inference: 62%|██████▏ | 3647/5920 [48:37<12:50, 2.95it/s, tok/s=15256.0]
Inference: 62%|██████▏ | 3647/5920 [48:37<12:50, 2.95it/s, tok/s=15252.1]
Inference: 62%|██████▏ | 3648/5920 [48:39<27:12, 1.39it/s, tok/s=15252.1]
Inference: 62%|██████▏ | 3648/5920 [48:39<27:12, 1.39it/s, tok/s=15243.4]
Inference: 62%|██████▏ | 3649/5920 [48:39<21:09, 1.79it/s, tok/s=15243.4]
Inference: 62%|██████▏ | 3649/5920 [48:39<21:09, 1.79it/s, tok/s=15243.4]
Inference: 62%|██████▏ | 3650/5920 [48:40<22:14, 1.70it/s, tok/s=15243.4]
Inference: 62%|██████▏ | 3650/5920 [48:40<22:14, 1.70it/s, tok/s=15250.3]
Inference: 62%|██████▏ | 3651/5920 [48:40<22:14, 1.70it/s, tok/s=15250.7]
Inference: 62%|██████▏ | 3652/5920 [48:40<22:13, 1.70it/s, tok/s=15251.7]
Inference: 62%|██████▏ | 3653/5920 [48:40<12:10, 3.10it/s, tok/s=15251.7]
Inference: 62%|██████▏ | 3653/5920 [48:40<12:10, 3.10it/s, tok/s=15256.7]
Inference: 62%|██████▏ | 3654/5920 [48:40<10:39, 3.54it/s, tok/s=15256.7]
Inference: 62%|██████▏ | 3654/5920 [48:40<10:39, 3.54it/s, tok/s=15263.0]
Inference: 62%|██████▏ | 3655/5920 [48:40<10:39, 3.54it/s, tok/s=15263.6]
Inference: 62%|██████▏ | 3656/5920 [48:40<10:38, 3.54it/s, tok/s=15264.2]
Inference: 62%|██████▏ | 3657/5920 [48:40<10:38, 3.54it/s, tok/s=15265.9]
Inference: 62%|██████▏ | 3658/5920 [48:41<08:31, 4.42it/s, tok/s=15265.9]
Inference: 62%|██████▏ | 3658/5920 [48:41<08:31, 4.42it/s, tok/s=15266.4]
Inference: 62%|██████▏ | 3659/5920 [48:41<08:01, 4.69it/s, tok/s=15266.4]
Inference: 62%|██████▏ | 3659/5920 [48:41<08:01, 4.69it/s, tok/s=15266.3]
Inference: 62%|██████▏ | 3660/5920 [48:41<08:01, 4.69it/s, tok/s=15267.3]
Inference: 62%|██████▏ | 3661/5920 [48:41<08:01, 4.69it/s, tok/s=15268.2]
Inference: 62%|██████▏ | 3662/5920 [48:42<07:03, 5.34it/s, tok/s=15268.2]
Inference: 62%|██████▏ | 3662/5920 [48:42<07:03, 5.34it/s, tok/s=15266.9]
Inference: 62%|██████▏ | 3663/5920 [48:42<06:50, 5.50it/s, tok/s=15266.9]
Inference: 62%|██████▏ | 3663/5920 [48:42<06:50, 5.50it/s, tok/s=15267.6]
Inference: 62%|██████▏ | 3664/5920 [48:42<06:33, 5.74it/s, tok/s=15267.6]
Inference: 62%|██████▏ | 3664/5920 [48:42<06:33, 5.74it/s, tok/s=15269.4]
Inference: 62%|██████▏ | 3665/5920 [48:42<06:32, 5.74it/s, tok/s=15269.9]
Inference: 62%|██████▏ | 3666/5920 [48:43<10:40, 3.52it/s, tok/s=15269.9]
Inference: 62%|██████▏ | 3666/5920 [48:43<10:40, 3.52it/s, tok/s=15266.0]
Inference: 62%|██████▏ | 3667/5920 [48:43<10:40, 3.52it/s, tok/s=15266.6]
Inference: 62%|██████▏ | 3668/5920 [48:43<10:03, 3.73it/s, tok/s=15266.6]
Inference: 62%|██████▏ | 3668/5920 [48:43<10:03, 3.73it/s, tok/s=15267.9]
Inference: 62%|██████▏ | 3669/5920 [48:44<09:33, 3.92it/s, tok/s=15267.9]
Inference: 62%|██████▏ | 3669/5920 [48:44<09:33, 3.92it/s, tok/s=15267.6]
Inference: 62%|██████▏ | 3670/5920 [48:44<15:13, 2.46it/s, tok/s=15267.6]
Inference: 62%|██████▏ | 3670/5920 [48:44<15:13, 2.46it/s, tok/s=15263.6]
Inference: 62%|██████▏ | 3671/5920 [48:45<15:13, 2.46it/s, tok/s=15266.4]
Inference: 62%|██████▏ | 3672/5920 [48:45<11:03, 3.39it/s, tok/s=15266.4]
Inference: 62%|██████▏ | 3672/5920 [48:45<11:03, 3.39it/s, tok/s=15266.2]
Inference: 62%|██████▏ | 3673/5920 [48:45<11:03, 3.39it/s, tok/s=15266.2]
Inference: 62%|██████▏ | 3674/5920 [48:45<08:57, 4.17it/s, tok/s=15266.2]
Inference: 62%|██████▏ | 3674/5920 [48:45<08:57, 4.17it/s, tok/s=15265.7]
Inference: 62%|██████▏ | 3675/5920 [48:45<09:14, 4.05it/s, tok/s=15265.7]
Inference: 62%|██████▏ | 3675/5920 [48:45<09:14, 4.05it/s, tok/s=15266.7]
Inference: 62%|██████▏ | 3676/5920 [48:45<08:18, 4.50it/s, tok/s=15266.7]
Inference: 62%|██████▏ | 3676/5920 [48:45<08:18, 4.50it/s, tok/s=15267.2]
Inference: 62%|██████▏ | 3677/5920 [48:45<08:18, 4.50it/s, tok/s=15271.6]
Inference: 62%|██████▏ | 3678/5920 [48:46<06:42, 5.57it/s, tok/s=15271.6]
Inference: 62%|██████▏ | 3678/5920 [48:46<06:42, 5.57it/s, tok/s=15280.8]
Inference: 62%|██████▏ | 3679/5920 [48:46<07:00, 5.33it/s, tok/s=15280.8]
Inference: 62%|██████▏ | 3679/5920 [48:46<07:00, 5.33it/s, tok/s=15282.8]
Inference: 62%|██████▏ | 3680/5920 [48:46<06:48, 5.48it/s, tok/s=15282.8]
Inference: 62%|██████▏ | 3680/5920 [48:46<06:48, 5.48it/s, tok/s=15282.5]
Inference: 62%|██████▏ | 3681/5920 [48:47<11:05, 3.37it/s, tok/s=15282.5]
Inference: 62%|██████▏ | 3681/5920 [48:47<11:05, 3.37it/s, tok/s=15280.2]
Inference: 62%|██████▏ | 3682/5920 [48:47<11:04, 3.37it/s, tok/s=15280.2]
Inference: 62%|██████▏ | 3683/5920 [48:47<12:22, 3.01it/s, tok/s=15280.2]
Inference: 62%|██████▏ | 3683/5920 [48:47<12:22, 3.01it/s, tok/s=15277.9]
Inference: 62%|██████▏ | 3684/5920 [48:48<10:29, 3.55it/s, tok/s=15277.9]
Inference: 62%|██████▏ | 3684/5920 [48:48<10:29, 3.55it/s, tok/s=15278.6]
Inference: 62%|██████▏ | 3685/5920 [48:48<12:00, 3.10it/s, tok/s=15278.6]
Inference: 62%|██████▏ | 3685/5920 [48:48<12:00, 3.10it/s, tok/s=15286.7]
Inference: 62%|██████▏ | 3686/5920 [48:48<09:57, 3.74it/s, tok/s=15286.7]
Inference: 62%|██████▏ | 3686/5920 [48:48<09:57, 3.74it/s, tok/s=15288.6]
Inference: 62%|██████▏ | 3687/5920 [48:48<09:57, 3.74it/s, tok/s=15289.8]
Inference: 62%|██████▏ | 3688/5920 [48:48<08:15, 4.50it/s, tok/s=15289.8]
Inference: 62%|██████▏ | 3688/5920 [48:48<08:15, 4.50it/s, tok/s=15297.9]
Inference: 62%|██████▏ | 3689/5920 [48:48<08:15, 4.50it/s, tok/s=15297.9]
Inference: 62%|██████▏ | 3690/5920 [48:49<06:51, 5.42it/s, tok/s=15297.9]
Inference: 62%|██████▏ | 3690/5920 [48:49<06:51, 5.42it/s, tok/s=15297.6]
Inference: 62%|██████▏ | 3691/5920 [48:49<06:51, 5.42it/s, tok/s=15298.4]
Inference: 62%|██████▏ | 3692/5920 [48:49<07:50, 4.74it/s, tok/s=15298.4]
Inference: 62%|██████▏ | 3692/5920 [48:49<07:50, 4.74it/s, tok/s=15296.2]
Inference: 62%|██████▏ | 3693/5920 [48:49<07:41, 4.83it/s, tok/s=15296.2]
Inference: 62%|██████▏ | 3693/5920 [48:49<07:41, 4.83it/s, tok/s=15299.3]
Inference: 62%|██████▏ | 3694/5920 [48:49<07:40, 4.83it/s, tok/s=15299.5]
Inference: 62%|██████▏ | 3695/5920 [48:50<06:17, 5.89it/s, tok/s=15299.5]
Inference: 62%|██████▏ | 3695/5920 [48:50<06:17, 5.89it/s, tok/s=15300.8]
Inference: 62%|██████▏ | 3696/5920 [48:50<07:23, 5.01it/s, tok/s=15300.8]
Inference: 62%|██████▏ | 3696/5920 [48:50<07:23, 5.01it/s, tok/s=15300.7]
Inference: 62%|██████▏ | 3697/5920 [48:50<09:31, 3.89it/s, tok/s=15300.7]
Inference: 62%|██████▏ | 3697/5920 [48:50<09:31, 3.89it/s, tok/s=15304.4]
Inference: 62%|██████▏ | 3698/5920 [48:50<08:29, 4.36it/s, tok/s=15304.4]
Inference: 62%|██████▏ | 3698/5920 [48:50<08:29, 4.36it/s, tok/s=15304.1]
Inference: 62%|██████▏ | 3699/5920 [48:50<08:29, 4.36it/s, tok/s=15304.6]
Inference: 62%|██████▎ | 3700/5920 [48:51<08:10, 4.52it/s, tok/s=15304.6]
Inference: 62%|██████▎ | 3700/5920 [48:51<08:10, 4.52it/s, tok/s=15303.3]
Inference: 63%|██████▎ | 3701/5920 [48:51<08:09, 4.53it/s, tok/s=15303.3]
Inference: 63%|██████▎ | 3701/5920 [48:51<08:09, 4.53it/s, tok/s=15310.6]
Inference: 63%|██████▎ | 3702/5920 [48:51<08:09, 4.53it/s, tok/s=15316.4]
Inference: 63%|██████▎ | 3703/5920 [48:51<06:14, 5.93it/s, tok/s=15316.4]
Inference: 63%|██████▎ | 3703/5920 [48:51<06:14, 5.93it/s, tok/s=15318.8]
Inference: 63%|██████▎ | 3704/5920 [48:51<06:15, 5.90it/s, tok/s=15318.8]
Inference: 63%|██████▎ | 3704/5920 [48:51<06:15, 5.90it/s, tok/s=15318.5]
Inference: 63%|██████▎ | 3705/5920 [48:51<06:15, 5.90it/s, tok/s=15325.9]
Inference: 63%|██████▎ | 3706/5920 [48:51<06:15, 5.90it/s, tok/s=15326.3]
Inference: 63%|██████▎ | 3707/5920 [48:52<04:19, 8.52it/s, tok/s=15326.3]
Inference: 63%|██████▎ | 3707/5920 [48:52<04:19, 8.52it/s, tok/s=15327.3]
Inference: 63%|██████▎ | 3708/5920 [48:52<04:19, 8.52it/s, tok/s=15328.4]
Inference: 63%|██████▎ | 3709/5920 [48:52<06:11, 5.95it/s, tok/s=15328.4]
Inference: 63%|██████▎ | 3709/5920 [48:52<06:11, 5.95it/s, tok/s=15326.1]
Inference: 63%|██████▎ | 3710/5920 [48:52<06:11, 5.95it/s, tok/s=15335.5]
Inference: 63%|██████▎ | 3711/5920 [48:53<07:56, 4.63it/s, tok/s=15335.5]
Inference: 63%|██████▎ | 3711/5920 [48:53<07:56, 4.63it/s, tok/s=15333.0]
Inference: 63%|██████▎ | 3712/5920 [48:53<07:51, 4.68it/s, tok/s=15333.0]
Inference: 63%|██████▎ | 3712/5920 [48:53<07:51, 4.68it/s, tok/s=15333.1]
Inference: 63%|██████▎ | 3713/5920 [48:53<07:51, 4.68it/s, tok/s=15333.2]
Inference: 63%|██████▎ | 3714/5920 [48:55<18:03, 2.04it/s, tok/s=15333.2]
Inference: 63%|██████▎ | 3714/5920 [48:55<18:03, 2.04it/s, tok/s=15324.6]
Inference: 63%|██████▎ | 3715/5920 [48:55<18:02, 2.04it/s, tok/s=15330.3]
Inference: 63%|██████▎ | 3716/5920 [48:56<15:10, 2.42it/s, tok/s=15330.3]
Inference: 63%|██████▎ | 3716/5920 [48:56<15:10, 2.42it/s, tok/s=15328.6]
Inference: 63%|██████▎ | 3717/5920 [48:56<15:10, 2.42it/s, tok/s=15336.8]
Inference: 63%|██████▎ | 3718/5920 [48:56<11:15, 3.26it/s, tok/s=15336.8]
Inference: 63%|██████▎ | 3718/5920 [48:56<11:15, 3.26it/s, tok/s=15337.7]
Inference: 63%|██████▎ | 3719/5920 [48:56<11:14, 3.26it/s, tok/s=15338.8]
Inference: 63%|██████▎ | 3720/5920 [48:56<09:55, 3.69it/s, tok/s=15338.8]
Inference: 63%|██████▎ | 3720/5920 [48:56<09:55, 3.69it/s, tok/s=15337.8]
Inference: 63%|██████▎ | 3721/5920 [48:56<09:32, 3.84it/s, tok/s=15337.8]
Inference: 63%|██████▎ | 3721/5920 [48:56<09:32, 3.84it/s, tok/s=15337.5]
Inference: 63%|██████▎ | 3722/5920 [48:57<11:31, 3.18it/s, tok/s=15337.5]
Inference: 63%|██████▎ | 3722/5920 [48:57<11:31, 3.18it/s, tok/s=15336.2]
Inference: 63%|██████▎ | 3723/5920 [48:57<13:38, 2.68it/s, tok/s=15336.2]
Inference: 63%|██████▎ | 3723/5920 [48:57<13:38, 2.68it/s, tok/s=15334.2]
Inference: 63%|██████▎ | 3724/5920 [48:58<13:38, 2.68it/s, tok/s=15334.5]
Inference: 63%|██████▎ | 3725/5920 [48:58<09:14, 3.96it/s, tok/s=15334.5]
Inference: 63%|██████▎ | 3725/5920 [48:58<09:14, 3.96it/s, tok/s=15335.1]
Inference: 63%|██████▎ | 3726/5920 [48:58<08:21, 4.37it/s, tok/s=15335.1]
Inference: 63%|██████▎ | 3726/5920 [48:58<08:21, 4.37it/s, tok/s=15335.0]
Inference: 63%|██████▎ | 3727/5920 [48:58<08:35, 4.25it/s, tok/s=15335.0]
Inference: 63%|██████▎ | 3727/5920 [48:58<08:35, 4.25it/s, tok/s=15337.0]
Inference: 63%|██████▎ | 3728/5920 [48:59<11:16, 3.24it/s, tok/s=15337.0]
Inference: 63%|██████▎ | 3728/5920 [48:59<11:16, 3.24it/s, tok/s=15335.8]
Inference: 63%|██████▎ | 3729/5920 [48:59<11:16, 3.24it/s, tok/s=15336.6]
Inference: 63%|██████▎ | 3730/5920 [48:59<09:12, 3.97it/s, tok/s=15336.6]
Inference: 63%|██████▎ | 3730/5920 [48:59<09:12, 3.97it/s, tok/s=15335.5]
Inference: 63%|██████▎ | 3731/5920 [48:59<09:52, 3.69it/s, tok/s=15335.5]
Inference: 63%|██████▎ | 3731/5920 [48:59<09:52, 3.69it/s, tok/s=15334.4]
Inference: 63%|██████▎ | 3732/5920 [49:00<11:42, 3.11it/s, tok/s=15334.4]
Inference: 63%|██████▎ | 3732/5920 [49:00<11:42, 3.11it/s, tok/s=15340.9]
Inference: 63%|██████▎ | 3733/5920 [49:00<11:42, 3.11it/s, tok/s=15341.5]
Inference: 63%|██████▎ | 3734/5920 [49:00<08:29, 4.29it/s, tok/s=15341.5]
Inference: 63%|██████▎ | 3734/5920 [49:00<08:29, 4.29it/s, tok/s=15346.7]
Inference: 63%|██████▎ | 3735/5920 [49:01<12:04, 3.02it/s, tok/s=15346.7]
Inference: 63%|██████▎ | 3735/5920 [49:01<12:04, 3.02it/s, tok/s=15344.0]
Inference: 63%|██████▎ | 3736/5920 [49:01<12:04, 3.02it/s, tok/s=15345.1]
Inference: 63%|██████▎ | 3737/5920 [49:02<18:38, 1.95it/s, tok/s=15345.1]
Inference: 63%|██████▎ | 3737/5920 [49:02<18:38, 1.95it/s, tok/s=15337.9]
Inference: 63%|██████▎ | 3738/5920 [49:02<15:23, 2.36it/s, tok/s=15337.9]
Inference: 63%|██████▎ | 3738/5920 [49:02<15:23, 2.36it/s, tok/s=15338.4]
Inference: 63%|██████▎ | 3739/5920 [49:02<13:38, 2.66it/s, tok/s=15338.4]
Inference: 63%|██████▎ | 3739/5920 [49:02<13:38, 2.66it/s, tok/s=15340.9]
Inference: 63%|██████▎ | 3740/5920 [49:03<13:38, 2.66it/s, tok/s=15341.0]
Inference: 63%|██████▎ | 3741/5920 [49:03<12:40, 2.87it/s, tok/s=15341.0]
Inference: 63%|██████▎ | 3741/5920 [49:03<12:40, 2.87it/s, tok/s=15348.0]
Inference: 63%|██████▎ | 3742/5920 [49:03<12:40, 2.87it/s, tok/s=15352.1]
Inference: 63%|██████▎ | 3743/5920 [49:03<09:57, 3.64it/s, tok/s=15352.1]
Inference: 63%|██████▎ | 3743/5920 [49:03<09:57, 3.64it/s, tok/s=15351.7]
Inference: 63%|██████▎ | 3744/5920 [49:03<09:57, 3.64it/s, tok/s=15352.2]
Inference: 63%|██████▎ | 3745/5920 [49:04<08:12, 4.42it/s, tok/s=15352.2]
Inference: 63%|██████▎ | 3745/5920 [49:04<08:12, 4.42it/s, tok/s=15351.6]
Inference: 63%|██████▎ | 3746/5920 [49:04<08:11, 4.42it/s, tok/s=15352.8]
Inference: 63%|██████▎ | 3747/5920 [49:04<08:24, 4.30it/s, tok/s=15352.8]
Inference: 63%|██████▎ | 3747/5920 [49:04<08:24, 4.30it/s, tok/s=15351.6]
Inference: 63%|██████▎ | 3748/5920 [49:04<08:24, 4.30it/s, tok/s=15352.9]
Inference: 63%|██████▎ | 3749/5920 [49:05<09:19, 3.88it/s, tok/s=15352.9]
Inference: 63%|██████▎ | 3749/5920 [49:05<09:19, 3.88it/s, tok/s=15350.2]
Inference: 63%|██████▎ | 3750/5920 [49:05<09:44, 3.71it/s, tok/s=15350.2]
Inference: 63%|██████▎ | 3750/5920 [49:05<09:44, 3.71it/s, tok/s=15350.1]
Inference: 63%|██████▎ | 3751/5920 [49:05<09:44, 3.71it/s, tok/s=15350.3]
Inference: 63%|██████▎ | 3752/5920 [49:06<13:19, 2.71it/s, tok/s=15350.3]
Inference: 63%|██████▎ | 3752/5920 [49:06<13:19, 2.71it/s, tok/s=15351.3]
Inference: 63%|██████▎ | 3753/5920 [49:06<13:19, 2.71it/s, tok/s=15351.7]
Inference: 63%|██████▎ | 3754/5920 [49:07<11:14, 3.21it/s, tok/s=15351.7]
Inference: 63%|██████▎ | 3754/5920 [49:07<11:14, 3.21it/s, tok/s=15355.0]
Inference: 63%|██████▎ | 3755/5920 [49:07<09:52, 3.66it/s, tok/s=15355.0]
Inference: 63%|██████▎ | 3755/5920 [49:07<09:52, 3.66it/s, tok/s=15354.9]
Inference: 63%|██████▎ | 3756/5920 [49:07<09:28, 3.81it/s, tok/s=15354.9]
Inference: 63%|██████▎ | 3756/5920 [49:07<09:28, 3.81it/s, tok/s=15356.3]
Inference: 63%|██████▎ | 3757/5920 [49:07<08:48, 4.09it/s, tok/s=15356.3]
Inference: 63%|██████▎ | 3757/5920 [49:07<08:48, 4.09it/s, tok/s=15357.3]
Inference: 63%|██████▎ | 3758/5920 [49:07<08:58, 4.01it/s, tok/s=15357.3]
Inference: 63%|██████▎ | 3758/5920 [49:07<08:58, 4.01it/s, tok/s=15359.6]
Inference: 63%|██████▎ | 3759/5920 [49:07<08:58, 4.01it/s, tok/s=15360.9]
Inference: 64%|██████▎ | 3760/5920 [49:08<07:16, 4.95it/s, tok/s=15360.9]
Inference: 64%|██████▎ | 3760/5920 [49:08<07:16, 4.95it/s, tok/s=15360.1]
Inference: 64%|██████▎ | 3761/5920 [49:08<09:02, 3.98it/s, tok/s=15360.1]
Inference: 64%|██████▎ | 3761/5920 [49:08<09:02, 3.98it/s, tok/s=15359.2]
Inference: 64%|██████▎ | 3762/5920 [49:09<11:37, 3.10it/s, tok/s=15359.2]
Inference: 64%|██████▎ | 3762/5920 [49:09<11:37, 3.10it/s, tok/s=15357.5]
Inference: 64%|██████▎ | 3763/5920 [49:09<13:02, 2.76it/s, tok/s=15357.5]
Inference: 64%|██████▎ | 3763/5920 [49:09<13:02, 2.76it/s, tok/s=15358.8]
Inference: 64%|██████▎ | 3764/5920 [49:09<13:02, 2.76it/s, tok/s=15360.9]
Inference: 64%|██████▎ | 3765/5920 [49:09<08:22, 4.29it/s, tok/s=15360.9]
Inference: 64%|██████▎ | 3765/5920 [49:09<08:22, 4.29it/s, tok/s=15362.0]
Inference: 64%|██████▎ | 3766/5920 [49:09<08:04, 4.45it/s, tok/s=15362.0]
Inference: 64%|██████▎ | 3766/5920 [49:09<08:04, 4.45it/s, tok/s=15369.6]
Inference: 64%|██████▎ | 3767/5920 [49:10<08:31, 4.21it/s, tok/s=15369.6]
Inference: 64%|██████▎ | 3767/5920 [49:10<08:31, 4.21it/s, tok/s=15369.5]
Inference: 64%|██████▎ | 3768/5920 [49:10<12:38, 2.84it/s, tok/s=15369.5]
Inference: 64%|██████▎ | 3768/5920 [49:10<12:38, 2.84it/s, tok/s=15366.9]
Inference: 64%|██████▎ | 3769/5920 [49:11<13:40, 2.62it/s, tok/s=15366.9]
Inference: 64%|██████▎ | 3769/5920 [49:11<13:40, 2.62it/s, tok/s=15365.7]
Inference: 64%|██████▎ | 3770/5920 [49:11<13:40, 2.62it/s, tok/s=15370.7]
Inference: 64%|██████▎ | 3771/5920 [49:11<08:44, 4.09it/s, tok/s=15370.7]
Inference: 64%|██████▎ | 3771/5920 [49:11<08:44, 4.09it/s, tok/s=15372.8]
Inference: 64%|██████▎ | 3772/5920 [49:11<08:39, 4.14it/s, tok/s=15372.8]
Inference: 64%|██████▎ | 3772/5920 [49:11<08:39, 4.14it/s, tok/s=15372.2]
Inference: 64%|██████▎ | 3773/5920 [49:12<11:01, 3.25it/s, tok/s=15372.2]
Inference: 64%|██████▎ | 3773/5920 [49:12<11:01, 3.25it/s, tok/s=15370.1]
Inference: 64%|██████▍ | 3774/5920 [49:12<11:01, 3.25it/s, tok/s=15371.0]
Inference: 64%|██████▍ | 3775/5920 [49:12<07:33, 4.73it/s, tok/s=15371.0]
Inference: 64%|██████▍ | 3775/5920 [49:12<07:33, 4.73it/s, tok/s=15371.8]
Inference: 64%|██████▍ | 3776/5920 [49:12<07:11, 4.96it/s, tok/s=15371.8]
Inference: 64%|██████▍ | 3776/5920 [49:12<07:11, 4.96it/s, tok/s=15371.4]
Inference: 64%|██████▍ | 3777/5920 [49:12<08:53, 4.02it/s, tok/s=15371.4]
Inference: 64%|██████▍ | 3777/5920 [49:12<08:53, 4.02it/s, tok/s=15371.8]
Inference: 64%|██████▍ | 3778/5920 [49:12<08:52, 4.02it/s, tok/s=15373.2]
Inference: 64%|██████▍ | 3779/5920 [49:13<06:40, 5.34it/s, tok/s=15373.2]
Inference: 64%|██████▍ | 3779/5920 [49:13<06:40, 5.34it/s, tok/s=15372.9]
Inference: 64%|██████▍ | 3780/5920 [49:13<09:36, 3.71it/s, tok/s=15372.9]
Inference: 64%|██████▍ | 3780/5920 [49:13<09:36, 3.71it/s, tok/s=15371.7]
Inference: 64%|██████▍ | 3781/5920 [49:13<09:36, 3.71it/s, tok/s=15382.0]
Inference: 64%|██████▍ | 3782/5920 [49:14<08:50, 4.03it/s, tok/s=15382.0]
Inference: 64%|██████▍ | 3782/5920 [49:14<08:50, 4.03it/s, tok/s=15383.8]
Inference: 64%|██████▍ | 3783/5920 [49:14<08:49, 4.03it/s, tok/s=15384.6]
Inference: 64%|██████▍ | 3784/5920 [49:14<09:46, 3.64it/s, tok/s=15384.6]
Inference: 64%|██████▍ | 3784/5920 [49:14<09:46, 3.64it/s, tok/s=15382.5]
Inference: 64%|██████▍ | 3785/5920 [49:15<13:03, 2.72it/s, tok/s=15382.5]
Inference: 64%|██████▍ | 3785/5920 [49:15<13:03, 2.72it/s, tok/s=15379.7]
Inference: 64%|██████▍ | 3786/5920 [49:16<15:27, 2.30it/s, tok/s=15379.7]
Inference: 64%|██████▍ | 3786/5920 [49:16<15:27, 2.30it/s, tok/s=15377.9]
Inference: 64%|██████▍ | 3787/5920 [49:16<13:41, 2.60it/s, tok/s=15377.9]
Inference: 64%|██████▍ | 3787/5920 [49:16<13:41, 2.60it/s, tok/s=15377.7]
Inference: 64%|██████▍ | 3788/5920 [49:16<12:27, 2.85it/s, tok/s=15377.7]
Inference: 64%|██████▍ | 3788/5920 [49:16<12:27, 2.85it/s, tok/s=15377.0]
Inference: 64%|██████▍ | 3789/5920 [49:16<10:24, 3.41it/s, tok/s=15377.0]
Inference: 64%|██████▍ | 3789/5920 [49:16<10:24, 3.41it/s, tok/s=15379.6]
Inference: 64%|██████▍ | 3790/5920 [49:16<10:24, 3.41it/s, tok/s=15389.5]
Inference: 64%|██████▍ | 3791/5920 [49:17<09:57, 3.56it/s, tok/s=15389.5]
Inference: 64%|██████▍ | 3791/5920 [49:17<09:57, 3.56it/s, tok/s=15388.2]
Inference: 64%|██████▍ | 3792/5920 [49:17<08:47, 4.04it/s, tok/s=15388.2]
Inference: 64%|██████▍ | 3792/5920 [49:17<08:47, 4.04it/s, tok/s=15388.4]
Inference: 64%|██████▍ | 3793/5920 [49:17<08:47, 4.04it/s, tok/s=15388.7]
Inference: 64%|██████▍ | 3794/5920 [49:17<06:03, 5.84it/s, tok/s=15388.7]
Inference: 64%|██████▍ | 3794/5920 [49:17<06:03, 5.84it/s, tok/s=15394.8]
Inference: 64%|██████▍ | 3795/5920 [49:17<06:49, 5.18it/s, tok/s=15394.8]
Inference: 64%|██████▍ | 3795/5920 [49:17<06:49, 5.18it/s, tok/s=15394.0]
Inference: 64%|██████▍ | 3796/5920 [49:17<06:49, 5.18it/s, tok/s=15402.7]
Inference: 64%|██████▍ | 3797/5920 [49:18<06:38, 5.33it/s, tok/s=15402.7]
Inference: 64%|██████▍ | 3797/5920 [49:18<06:38, 5.33it/s, tok/s=15402.4]
Inference: 64%|██████▍ | 3798/5920 [49:18<10:41, 3.31it/s, tok/s=15402.4]
Inference: 64%|██████▍ | 3798/5920 [49:18<10:41, 3.31it/s, tok/s=15399.7]
Inference: 64%|██████▍ | 3799/5920 [49:19<12:06, 2.92it/s, tok/s=15399.7]
Inference: 64%|██████▍ | 3799/5920 [49:19<12:06, 2.92it/s, tok/s=15399.1]
Inference: 64%|██████▍ | 3800/5920 [49:19<12:06, 2.92it/s, tok/s=15401.7]
Inference: 64%|██████▍ | 3801/5920 [49:19<11:39, 3.03it/s, tok/s=15401.7]
Inference: 64%|██████▍ | 3801/5920 [49:19<11:39, 3.03it/s, tok/s=15399.2]
Inference: 64%|██████▍ | 3802/5920 [49:20<11:01, 3.20it/s, tok/s=15399.2]
Inference: 64%|██████▍ | 3802/5920 [49:20<11:01, 3.20it/s, tok/s=15398.4]
Inference: 64%|██████▍ | 3803/5920 [49:20<11:01, 3.20it/s, tok/s=15398.9]
Inference: 64%|██████▍ | 3804/5920 [49:20<09:26, 3.73it/s, tok/s=15398.9]
Inference: 64%|██████▍ | 3804/5920 [49:20<09:26, 3.73it/s, tok/s=15398.7]
Inference: 64%|██████▍ | 3805/5920 [49:20<09:26, 3.73it/s, tok/s=15399.2]
Inference: 64%|██████▍ | 3806/5920 [49:20<09:26, 3.73it/s, tok/s=15400.6]
Inference: 64%|██████▍ | 3807/5920 [49:20<05:47, 6.09it/s, tok/s=15400.6]
Inference: 64%|██████▍ | 3807/5920 [49:20<05:47, 6.09it/s, tok/s=15400.7]
Inference: 64%|██████▍ | 3808/5920 [49:20<05:46, 6.09it/s, tok/s=15400.8]
Inference: 64%|██████▍ | 3809/5920 [49:21<06:45, 5.21it/s, tok/s=15400.8]
Inference: 64%|██████▍ | 3809/5920 [49:21<06:45, 5.21it/s, tok/s=15399.2]
Inference: 64%|██████▍ | 3810/5920 [49:21<08:01, 4.38it/s, tok/s=15399.2]
Inference: 64%|██████▍ | 3810/5920 [49:21<08:01, 4.38it/s, tok/s=15399.4]
Inference: 64%|██████▍ | 3811/5920 [49:21<07:43, 4.55it/s, tok/s=15399.4]
Inference: 64%|██████▍ | 3811/5920 [49:21<07:43, 4.55it/s, tok/s=15399.2]
Inference: 64%|██████▍ | 3812/5920 [49:22<07:46, 4.52it/s, tok/s=15399.2]
Inference: 64%|██████▍ | 3812/5920 [49:22<07:46, 4.52it/s, tok/s=15398.7]
Inference: 64%|██████▍ | 3813/5920 [49:22<07:46, 4.52it/s, tok/s=15398.8]
Inference: 64%|██████▍ | 3814/5920 [49:22<06:51, 5.11it/s, tok/s=15398.8]
Inference: 64%|██████▍ | 3814/5920 [49:22<06:51, 5.11it/s, tok/s=15398.5]
Inference: 64%|██████▍ | 3815/5920 [49:22<06:45, 5.19it/s, tok/s=15398.5]
Inference: 64%|██████▍ | 3815/5920 [49:22<06:45, 5.19it/s, tok/s=15399.0]
Inference: 64%|██████▍ | 3816/5920 [49:22<07:49, 4.48it/s, tok/s=15399.0]
Inference: 64%|██████▍ | 3816/5920 [49:22<07:49, 4.48it/s, tok/s=15398.0]
Inference: 64%|██████▍ | 3817/5920 [49:23<12:22, 2.83it/s, tok/s=15398.0]
Inference: 64%|██████▍ | 3817/5920 [49:23<12:22, 2.83it/s, tok/s=15397.1]
Inference: 64%|██████▍ | 3818/5920 [49:23<11:42, 2.99it/s, tok/s=15397.1]
Inference: 64%|██████▍ | 3818/5920 [49:23<11:42, 2.99it/s, tok/s=15396.7]
Inference: 65%|██████▍ | 3819/5920 [49:23<11:42, 2.99it/s, tok/s=15397.3]
Inference: 65%|██████▍ | 3820/5920 [49:24<07:49, 4.47it/s, tok/s=15397.3]
Inference: 65%|██████▍ | 3820/5920 [49:24<07:49, 4.47it/s, tok/s=15406.4]
Inference: 65%|██████▍ | 3821/5920 [49:24<11:46, 2.97it/s, tok/s=15406.4]
Inference: 65%|██████▍ | 3821/5920 [49:24<11:46, 2.97it/s, tok/s=15403.3]
Inference: 65%|██████▍ | 3822/5920 [49:24<11:46, 2.97it/s, tok/s=15406.9]
Inference: 65%|██████▍ | 3823/5920 [49:24<11:46, 2.97it/s, tok/s=15409.8]
Inference: 65%|██████▍ | 3824/5920 [49:24<06:32, 5.35it/s, tok/s=15409.8]
Inference: 65%|██████▍ | 3824/5920 [49:24<06:32, 5.35it/s, tok/s=15410.8]
Inference: 65%|██████▍ | 3825/5920 [49:25<06:31, 5.35it/s, tok/s=15406.9]
Inference: 65%|██████▍ | 3826/5920 [49:26<10:59, 3.18it/s, tok/s=15406.9]
Inference: 65%|██████▍ | 3826/5920 [49:26<10:59, 3.18it/s, tok/s=15407.6]
Inference: 65%|██████▍ | 3827/5920 [49:26<11:17, 3.09it/s, tok/s=15407.6]
Inference: 65%|██████▍ | 3827/5920 [49:26<11:17, 3.09it/s, tok/s=15408.1]
Inference: 65%|██████▍ | 3828/5920 [49:26<11:17, 3.09it/s, tok/s=15408.5]
Inference: 65%|██████▍ | 3829/5920 [49:26<10:54, 3.19it/s, tok/s=15408.5]
Inference: 65%|██████▍ | 3829/5920 [49:26<10:54, 3.19it/s, tok/s=15407.2]
Inference: 65%|██████▍ | 3830/5920 [49:27<11:24, 3.05it/s, tok/s=15407.2]
Inference: 65%|██████▍ | 3830/5920 [49:27<11:24, 3.05it/s, tok/s=15406.7]
Inference: 65%|██████▍ | 3831/5920 [49:27<10:56, 3.18it/s, tok/s=15406.7]
Inference: 65%|██████▍ | 3831/5920 [49:27<10:56, 3.18it/s, tok/s=15408.6]
Inference: 65%|██████▍ | 3832/5920 [49:27<10:56, 3.18it/s, tok/s=15409.3]
Inference: 65%|██████▍ | 3833/5920 [49:27<10:56, 3.18it/s, tok/s=15410.1]
Inference: 65%|██████▍ | 3834/5920 [49:27<10:55, 3.18it/s, tok/s=15411.0]
Inference: 65%|██████▍ | 3835/5920 [49:27<06:00, 5.79it/s, tok/s=15411.0]
Inference: 65%|██████▍ | 3835/5920 [49:27<06:00, 5.79it/s, tok/s=15415.6]
Inference: 65%|██████▍ | 3836/5920 [49:28<06:23, 5.44it/s, tok/s=15415.6]
Inference: 65%|██████▍ | 3836/5920 [49:28<06:23, 5.44it/s, tok/s=15415.0]
Inference: 65%|██████▍ | 3837/5920 [49:28<06:22, 5.44it/s, tok/s=15423.4]
Inference: 65%|██████▍ | 3838/5920 [49:28<06:22, 5.44it/s, tok/s=15424.1]
Inference: 65%|██████▍ | 3839/5920 [49:28<04:32, 7.63it/s, tok/s=15424.1]
Inference: 65%|██████▍ | 3839/5920 [49:28<04:32, 7.63it/s, tok/s=15430.3]
Inference: 65%|██████▍ | 3840/5920 [49:28<04:32, 7.63it/s, tok/s=15430.2]
Inference: 65%|██████▍ | 3841/5920 [49:28<04:03, 8.54it/s, tok/s=15430.2]
Inference: 65%|██████▍ | 3841/5920 [49:28<04:03, 8.54it/s, tok/s=15437.3]
Inference: 65%|██████▍ | 3842/5920 [49:28<04:03, 8.54it/s, tok/s=15437.5]
Inference: 65%|██████▍ | 3843/5920 [49:28<04:26, 7.79it/s, tok/s=15437.5]
Inference: 65%|██████▍ | 3843/5920 [49:28<04:26, 7.79it/s, tok/s=15438.9]
Inference: 65%|██████▍ | 3844/5920 [49:29<06:59, 4.95it/s, tok/s=15438.9]
Inference: 65%|██████▍ | 3844/5920 [49:29<06:59, 4.95it/s, tok/s=15439.6]
Inference: 65%|██████▍ | 3845/5920 [49:29<06:59, 4.95it/s, tok/s=15441.2]
Inference: 65%|██████▍ | 3846/5920 [49:29<05:27, 6.34it/s, tok/s=15441.2]
Inference: 65%|██████▍ | 3846/5920 [49:29<05:27, 6.34it/s, tok/s=15441.2]
Inference: 65%|██████▍ | 3847/5920 [49:29<05:27, 6.34it/s, tok/s=15442.5]
Inference: 65%|██████▌ | 3848/5920 [49:29<06:06, 5.66it/s, tok/s=15442.5]
Inference: 65%|██████▌ | 3848/5920 [49:29<06:06, 5.66it/s, tok/s=15449.1]
Inference: 65%|██████▌ | 3849/5920 [49:30<08:40, 3.98it/s, tok/s=15449.1]
Inference: 65%|██████▌ | 3849/5920 [49:30<08:40, 3.98it/s, tok/s=15449.0]
Inference: 65%|██████▌ | 3850/5920 [49:30<08:39, 3.98it/s, tok/s=15449.7]
Inference: 65%|██████▌ | 3851/5920 [49:30<07:00, 4.92it/s, tok/s=15449.7]
Inference: 65%|██████▌ | 3851/5920 [49:30<07:00, 4.92it/s, tok/s=15449.1]
Inference: 65%|██████▌ | 3852/5920 [49:31<08:30, 4.05it/s, tok/s=15449.1]
Inference: 65%|██████▌ | 3852/5920 [49:31<08:30, 4.05it/s, tok/s=15449.2]
Inference: 65%|██████▌ | 3853/5920 [49:31<13:14, 2.60it/s, tok/s=15449.2]
Inference: 65%|██████▌ | 3853/5920 [49:31<13:14, 2.60it/s, tok/s=15445.2]
Inference: 65%|██████▌ | 3854/5920 [49:32<12:06, 2.84it/s, tok/s=15445.2]
Inference: 65%|██████▌ | 3854/5920 [49:32<12:06, 2.84it/s, tok/s=15445.9]
Inference: 65%|██████▌ | 3855/5920 [49:32<11:17, 3.05it/s, tok/s=15445.9]
Inference: 65%|██████▌ | 3855/5920 [49:32<11:17, 3.05it/s, tok/s=15445.2]
Inference: 65%|██████▌ | 3856/5920 [49:32<11:03, 3.11it/s, tok/s=15445.2]
Inference: 65%|██████▌ | 3856/5920 [49:32<11:03, 3.11it/s, tok/s=15444.5]
Inference: 65%|██████▌ | 3857/5920 [49:32<11:03, 3.11it/s, tok/s=15447.9]
Inference: 65%|██████▌ | 3858/5920 [49:32<07:17, 4.71it/s, tok/s=15447.9]
Inference: 65%|██████▌ | 3858/5920 [49:32<07:17, 4.71it/s, tok/s=15449.5]
Inference: 65%|██████▌ | 3859/5920 [49:33<07:42, 4.45it/s, tok/s=15449.5]
Inference: 65%|██████▌ | 3859/5920 [49:33<07:42, 4.45it/s, tok/s=15449.0]
Inference: 65%|██████▌ | 3860/5920 [49:33<07:42, 4.45it/s, tok/s=15449.3]
Inference: 65%|██████▌ | 3861/5920 [49:33<06:12, 5.53it/s, tok/s=15449.3]
Inference: 65%|██████▌ | 3861/5920 [49:33<06:12, 5.53it/s, tok/s=15457.3]
Inference: 65%|██████▌ | 3862/5920 [49:33<06:13, 5.51it/s, tok/s=15457.3]
Inference: 65%|██████▌ | 3862/5920 [49:33<06:13, 5.51it/s, tok/s=15456.7]
Inference: 65%|██████▌ | 3863/5920 [49:34<08:58, 3.82it/s, tok/s=15456.7]
Inference: 65%|██████▌ | 3863/5920 [49:34<08:58, 3.82it/s, tok/s=15454.5]
Inference: 65%|██████▌ | 3864/5920 [49:34<09:04, 3.78it/s, tok/s=15454.5]
Inference: 65%|██████▌ | 3864/5920 [49:34<09:04, 3.78it/s, tok/s=15454.1]
Inference: 65%|██████▌ | 3865/5920 [49:34<09:04, 3.78it/s, tok/s=15456.4]
Inference: 65%|██████▌ | 3866/5920 [49:34<09:04, 3.78it/s, tok/s=15456.7]
Inference: 65%|██████▌ | 3867/5920 [49:34<05:02, 6.80it/s, tok/s=15456.7]
Inference: 65%|██████▌ | 3867/5920 [49:34<05:02, 6.80it/s, tok/s=15457.5]
Inference: 65%|██████▌ | 3868/5920 [49:34<05:01, 6.80it/s, tok/s=15457.8]
Inference: 65%|██████▌ | 3869/5920 [49:35<06:23, 5.35it/s, tok/s=15457.8]
Inference: 65%|██████▌ | 3869/5920 [49:35<06:23, 5.35it/s, tok/s=15455.9]
Inference: 65%|██████▌ | 3870/5920 [49:35<07:39, 4.46it/s, tok/s=15455.9]
Inference: 65%|██████▌ | 3870/5920 [49:35<07:39, 4.46it/s, tok/s=15455.1]
Inference: 65%|██████▌ | 3871/5920 [49:35<08:09, 4.18it/s, tok/s=15455.1]
Inference: 65%|██████▌ | 3871/5920 [49:35<08:09, 4.18it/s, tok/s=15457.9]
Inference: 65%|██████▌ | 3872/5920 [49:36<10:49, 3.15it/s, tok/s=15457.9]
Inference: 65%|██████▌ | 3872/5920 [49:36<10:49, 3.15it/s, tok/s=15457.8]
Inference: 65%|██████▌ | 3873/5920 [49:36<10:49, 3.15it/s, tok/s=15466.1]
Inference: 65%|██████▌ | 3874/5920 [49:36<10:48, 3.15it/s, tok/s=15475.9]
Inference: 65%|██████▌ | 3875/5920 [49:36<06:28, 5.26it/s, tok/s=15475.9]
Inference: 65%|██████▌ | 3875/5920 [49:36<06:28, 5.26it/s, tok/s=15476.2]
Inference: 65%|██████▌ | 3876/5920 [49:36<07:30, 4.53it/s, tok/s=15476.2]
Inference: 65%|██████▌ | 3876/5920 [49:36<07:30, 4.53it/s, tok/s=15474.9]
Inference: 65%|██████▌ | 3877/5920 [49:36<07:30, 4.53it/s, tok/s=15475.4]
Inference: 66%|██████▌ | 3878/5920 [49:37<07:29, 4.54it/s, tok/s=15475.4]
Inference: 66%|██████▌ | 3878/5920 [49:37<07:29, 4.54it/s, tok/s=15474.3]
Inference: 66%|██████▌ | 3879/5920 [49:37<08:27, 4.03it/s, tok/s=15474.3]
Inference: 66%|██████▌ | 3879/5920 [49:37<08:27, 4.03it/s, tok/s=15476.2]
Inference: 66%|██████▌ | 3880/5920 [49:37<08:12, 4.14it/s, tok/s=15476.2]
Inference: 66%|██████▌ | 3880/5920 [49:37<08:12, 4.14it/s, tok/s=15475.5]
Inference: 66%|██████▌ | 3881/5920 [49:38<07:57, 4.27it/s, tok/s=15475.5]
Inference: 66%|██████▌ | 3881/5920 [49:38<07:57, 4.27it/s, tok/s=15477.4]
Inference: 66%|██████▌ | 3882/5920 [49:38<06:53, 4.93it/s, tok/s=15477.4]
Inference: 66%|██████▌ | 3882/5920 [49:38<06:53, 4.93it/s, tok/s=15480.7]
Inference: 66%|██████▌ | 3883/5920 [49:38<06:53, 4.93it/s, tok/s=15481.7]
Inference: 66%|██████▌ | 3884/5920 [49:38<05:26, 6.24it/s, tok/s=15481.7]
Inference: 66%|██████▌ | 3884/5920 [49:38<05:26, 6.24it/s, tok/s=15481.4]
Inference: 66%|██████▌ | 3885/5920 [49:38<08:07, 4.18it/s, tok/s=15481.4]
Inference: 66%|██████▌ | 3885/5920 [49:38<08:07, 4.18it/s, tok/s=15479.7]
Inference: 66%|██████▌ | 3886/5920 [49:39<10:54, 3.11it/s, tok/s=15479.7]
Inference: 66%|██████▌ | 3886/5920 [49:39<10:54, 3.11it/s, tok/s=15478.4]
Inference: 66%|██████▌ | 3887/5920 [49:39<10:54, 3.11it/s, tok/s=15481.0]
Inference: 66%|██████▌ | 3888/5920 [49:39<07:36, 4.45it/s, tok/s=15481.0]
Inference: 66%|██████▌ | 3888/5920 [49:39<07:36, 4.45it/s, tok/s=15480.9]
Inference: 66%|██████▌ | 3889/5920 [49:39<07:38, 4.43it/s, tok/s=15480.9]
Inference: 66%|██████▌ | 3889/5920 [49:39<07:38, 4.43it/s, tok/s=15480.5]
Inference: 66%|██████▌ | 3890/5920 [49:39<07:38, 4.43it/s, tok/s=15484.4]
Inference: 66%|██████▌ | 3891/5920 [49:40<07:00, 4.83it/s, tok/s=15484.4]
Inference: 66%|██████▌ | 3891/5920 [49:40<07:00, 4.83it/s, tok/s=15483.6]
Inference: 66%|██████▌ | 3892/5920 [49:40<11:04, 3.05it/s, tok/s=15483.6]
Inference: 66%|██████▌ | 3892/5920 [49:40<11:04, 3.05it/s, tok/s=15481.8]
Inference: 66%|██████▌ | 3893/5920 [49:41<11:01, 3.06it/s, tok/s=15481.8]
Inference: 66%|██████▌ | 3893/5920 [49:41<11:01, 3.06it/s, tok/s=15482.4]
Inference: 66%|██████▌ | 3894/5920 [49:41<11:06, 3.04it/s, tok/s=15482.4]
Inference: 66%|██████▌ | 3894/5920 [49:41<11:06, 3.04it/s, tok/s=15481.5]
Inference: 66%|██████▌ | 3895/5920 [49:41<09:52, 3.42it/s, tok/s=15481.5]
Inference: 66%|██████▌ | 3895/5920 [49:41<09:52, 3.42it/s, tok/s=15487.4]
Inference: 66%|██████▌ | 3896/5920 [49:41<08:29, 3.97it/s, tok/s=15487.4]
Inference: 66%|██████▌ | 3896/5920 [49:41<08:29, 3.97it/s, tok/s=15489.5]
Inference: 66%|██████▌ | 3897/5920 [49:42<07:23, 4.56it/s, tok/s=15489.5]
Inference: 66%|██████▌ | 3897/5920 [49:42<07:23, 4.56it/s, tok/s=15489.4]
Inference: 66%|██████▌ | 3898/5920 [49:42<11:56, 2.82it/s, tok/s=15489.4]
Inference: 66%|██████▌ | 3898/5920 [49:42<11:56, 2.82it/s, tok/s=15489.8]
Inference: 66%|██████▌ | 3899/5920 [49:42<09:47, 3.44it/s, tok/s=15489.8]
Inference: 66%|██████▌ | 3899/5920 [49:42<09:47, 3.44it/s, tok/s=15489.9]
Inference: 66%|██████▌ | 3900/5920 [49:43<09:19, 3.61it/s, tok/s=15489.9]
Inference: 66%|██████▌ | 3900/5920 [49:43<09:19, 3.61it/s, tok/s=15490.1]
Inference: 66%|██████▌ | 3901/5920 [49:43<10:36, 3.17it/s, tok/s=15490.1]
Inference: 66%|██████▌ | 3901/5920 [49:43<10:36, 3.17it/s, tok/s=15491.7]
Inference: 66%|██████▌ | 3902/5920 [49:43<08:56, 3.76it/s, tok/s=15491.7]
Inference: 66%|██████▌ | 3902/5920 [49:43<08:56, 3.76it/s, tok/s=15492.0]
Inference: 66%|██████▌ | 3903/5920 [49:43<08:55, 3.76it/s, tok/s=15492.9]
Inference: 66%|██████▌ | 3904/5920 [49:43<06:09, 5.45it/s, tok/s=15492.9]
Inference: 66%|██████▌ | 3904/5920 [49:43<06:09, 5.45it/s, tok/s=15492.7]
Inference: 66%|██████▌ | 3905/5920 [49:44<06:07, 5.48it/s, tok/s=15492.7]
Inference: 66%|██████▌ | 3905/5920 [49:44<06:07, 5.48it/s, tok/s=15492.2]
Inference: 66%|██████▌ | 3906/5920 [49:44<06:14, 5.38it/s, tok/s=15492.2]
Inference: 66%|██████▌ | 3906/5920 [49:44<06:14, 5.38it/s, tok/s=15491.7]
Inference: 66%|██████▌ | 3907/5920 [49:44<06:14, 5.38it/s, tok/s=15493.4]
Inference: 66%|██████▌ | 3908/5920 [49:44<06:14, 5.38it/s, tok/s=15501.7]
Inference: 66%|██████▌ | 3909/5920 [49:44<03:44, 8.97it/s, tok/s=15501.7]
Inference: 66%|██████▌ | 3909/5920 [49:44<03:44, 8.97it/s, tok/s=15502.3]
Inference: 66%|██████▌ | 3910/5920 [49:44<03:44, 8.97it/s, tok/s=15503.0]
Inference: 66%|██████▌ | 3911/5920 [49:44<03:43, 8.97it/s, tok/s=15503.1]
Inference: 66%|██████▌ | 3912/5920 [49:44<04:04, 8.21it/s, tok/s=15503.1]
Inference: 66%|██████▌ | 3912/5920 [49:44<04:04, 8.21it/s, tok/s=15502.9]
Inference: 66%|██████▌ | 3913/5920 [49:45<05:16, 6.34it/s, tok/s=15502.9]
Inference: 66%|██████▌ | 3913/5920 [49:45<05:16, 6.34it/s, tok/s=15503.3]
Inference: 66%|██████▌ | 3914/5920 [49:45<06:22, 5.25it/s, tok/s=15503.3]
Inference: 66%|██████▌ | 3914/5920 [49:45<06:22, 5.25it/s, tok/s=15502.2]
Inference: 66%|██████▌ | 3915/5920 [49:45<05:51, 5.70it/s, tok/s=15502.2]
Inference: 66%|██████▌ | 3915/5920 [49:45<05:51, 5.70it/s, tok/s=15502.8]
Inference: 66%|██████▌ | 3916/5920 [49:45<06:26, 5.18it/s, tok/s=15502.8]
Inference: 66%|██████▌ | 3916/5920 [49:45<06:26, 5.18it/s, tok/s=15502.9]
Inference: 66%|██████▌ | 3917/5920 [49:45<06:26, 5.18it/s, tok/s=15503.6]
Inference: 66%|██████▌ | 3918/5920 [49:45<04:33, 7.32it/s, tok/s=15503.6]
Inference: 66%|██████▌ | 3918/5920 [49:45<04:33, 7.32it/s, tok/s=15503.6]
Inference: 66%|██████▌ | 3919/5920 [49:46<04:32, 7.36it/s, tok/s=15503.6]
Inference: 66%|██████▌ | 3919/5920 [49:46<04:32, 7.36it/s, tok/s=15509.2]
Inference: 66%|██████▌ | 3920/5920 [49:46<07:01, 4.75it/s, tok/s=15509.2]
Inference: 66%|██████▌ | 3920/5920 [49:46<07:01, 4.75it/s, tok/s=15515.5]
Inference: 66%|██████▌ | 3921/5920 [49:46<07:01, 4.75it/s, tok/s=15515.9]
Inference: 66%|██████▋ | 3922/5920 [49:46<07:00, 4.75it/s, tok/s=15516.9]
Inference: 66%|██████▋ | 3923/5920 [49:46<04:17, 7.76it/s, tok/s=15516.9]
Inference: 66%|██████▋ | 3923/5920 [49:46<04:17, 7.76it/s, tok/s=15525.1]
Inference: 66%|██████▋ | 3924/5920 [49:47<04:17, 7.76it/s, tok/s=15523.1]
Inference: 66%|██████▋ | 3925/5920 [49:47<05:43, 5.81it/s, tok/s=15523.1]
Inference: 66%|██████▋ | 3925/5920 [49:47<05:43, 5.81it/s, tok/s=15523.9]
Inference: 66%|██████▋ | 3926/5920 [49:47<06:54, 4.81it/s, tok/s=15523.9]
Inference: 66%|██████▋ | 3926/5920 [49:47<06:54, 4.81it/s, tok/s=15526.3]
Inference: 66%|██████▋ | 3927/5920 [49:47<06:54, 4.81it/s, tok/s=15527.5]
Inference: 66%|██████▋ | 3928/5920 [49:47<05:15, 6.32it/s, tok/s=15527.5]
Inference: 66%|██████▋ | 3928/5920 [49:47<05:15, 6.32it/s, tok/s=15527.9]
Inference: 66%|██████▋ | 3929/5920 [49:47<05:15, 6.32it/s, tok/s=15528.7]
Inference: 66%|██████▋ | 3930/5920 [49:48<05:52, 5.65it/s, tok/s=15528.7]
Inference: 66%|██████▋ | 3930/5920 [49:48<05:52, 5.65it/s, tok/s=15527.3]
Inference: 66%|██████▋ | 3931/5920 [49:48<06:28, 5.12it/s, tok/s=15527.3]
Inference: 66%|██████▋ | 3931/5920 [49:48<06:28, 5.12it/s, tok/s=15526.6]
Inference: 66%|██████▋ | 3932/5920 [49:48<06:28, 5.12it/s, tok/s=15527.2]
Inference: 66%|██████▋ | 3933/5920 [49:48<07:42, 4.30it/s, tok/s=15527.2]
Inference: 66%|██████▋ | 3933/5920 [49:48<07:42, 4.30it/s, tok/s=15528.3]
Inference: 66%|██████▋ | 3934/5920 [49:49<07:51, 4.21it/s, tok/s=15528.3]
Inference: 66%|██████▋ | 3934/5920 [49:49<07:51, 4.21it/s, tok/s=15527.5]
Inference: 66%|██████▋ | 3935/5920 [49:49<07:02, 4.69it/s, tok/s=15527.5]
Inference: 66%|██████▋ | 3935/5920 [49:49<07:02, 4.69it/s, tok/s=15527.5]
Inference: 66%|██████▋ | 3936/5920 [49:49<07:02, 4.69it/s, tok/s=15528.9]
Inference: 67%|██████▋ | 3937/5920 [49:49<05:38, 5.86it/s, tok/s=15528.9]
Inference: 67%|██████▋ | 3937/5920 [49:49<05:38, 5.86it/s, tok/s=15528.5]
Inference: 67%|██████▋ | 3938/5920 [49:49<05:38, 5.86it/s, tok/s=15532.3]
Inference: 67%|██████▋ | 3939/5920 [49:49<04:20, 7.60it/s, tok/s=15532.3]
Inference: 67%|██████▋ | 3939/5920 [49:49<04:20, 7.60it/s, tok/s=15532.0]
Inference: 67%|██████▋ | 3940/5920 [49:49<04:15, 7.76it/s, tok/s=15532.0]
Inference: 67%|██████▋ | 3940/5920 [49:49<04:15, 7.76it/s, tok/s=15531.8]
Inference: 67%|██████▋ | 3941/5920 [49:50<11:48, 2.79it/s, tok/s=15531.8]
Inference: 67%|██████▋ | 3941/5920 [49:50<11:48, 2.79it/s, tok/s=15527.3]
Inference: 67%|██████▋ | 3942/5920 [49:50<11:47, 2.79it/s, tok/s=15532.6]
Inference: 67%|██████▋ | 3943/5920 [49:51<07:51, 4.19it/s, tok/s=15532.6]
Inference: 67%|██████▋ | 3943/5920 [49:51<07:51, 4.19it/s, tok/s=15532.9]
Inference: 67%|██████▋ | 3944/5920 [49:51<07:51, 4.19it/s, tok/s=15541.6]
Inference: 67%|██████▋ | 3945/5920 [49:51<07:39, 4.30it/s, tok/s=15541.6]
Inference: 67%|██████▋ | 3945/5920 [49:51<07:39, 4.30it/s, tok/s=15545.4]
Inference: 67%|██████▋ | 3946/5920 [49:51<08:33, 3.84it/s, tok/s=15545.4]
Inference: 67%|██████▋ | 3946/5920 [49:51<08:33, 3.84it/s, tok/s=15544.4]
Inference: 67%|██████▋ | 3947/5920 [49:52<07:38, 4.30it/s, tok/s=15544.4]
Inference: 67%|██████▋ | 3947/5920 [49:52<07:38, 4.30it/s, tok/s=15544.2]
Inference: 67%|██████▋ | 3948/5920 [49:52<07:37, 4.32it/s, tok/s=15544.2]
Inference: 67%|██████▋ | 3948/5920 [49:52<07:37, 4.32it/s, tok/s=15545.2]
Inference: 67%|██████▋ | 3949/5920 [49:52<11:08, 2.95it/s, tok/s=15545.2]
Inference: 67%|██████▋ | 3949/5920 [49:52<11:08, 2.95it/s, tok/s=15542.3]
Inference: 67%|██████▋ | 3950/5920 [49:53<09:38, 3.41it/s, tok/s=15542.3]
Inference: 67%|██████▋ | 3950/5920 [49:53<09:38, 3.41it/s, tok/s=15543.9]
Inference: 67%|██████▋ | 3951/5920 [49:53<07:52, 4.16it/s, tok/s=15543.9]
Inference: 67%|██████▋ | 3951/5920 [49:53<07:52, 4.16it/s, tok/s=15544.7]
Inference: 67%|██████▋ | 3952/5920 [49:53<11:09, 2.94it/s, tok/s=15544.7]
Inference: 67%|██████▋ | 3952/5920 [49:53<11:09, 2.94it/s, tok/s=15542.7]
Inference: 67%|██████▋ | 3953/5920 [49:54<10:37, 3.08it/s, tok/s=15542.7]
Inference: 67%|██████▋ | 3953/5920 [49:54<10:37, 3.08it/s, tok/s=15544.5]
Inference: 67%|██████▋ | 3954/5920 [49:54<12:30, 2.62it/s, tok/s=15544.5]
Inference: 67%|██████▋ | 3954/5920 [49:54<12:30, 2.62it/s, tok/s=15542.4]
Inference: 67%|██████▋ | 3955/5920 [49:54<12:29, 2.62it/s, tok/s=15543.0]
Inference: 67%|██████▋ | 3956/5920 [49:54<07:55, 4.13it/s, tok/s=15543.0]
Inference: 67%|██████▋ | 3956/5920 [49:54<07:55, 4.13it/s, tok/s=15543.1]
Inference: 67%|██████▋ | 3957/5920 [49:54<07:55, 4.13it/s, tok/s=15546.7]
Inference: 67%|██████▋ | 3958/5920 [49:54<05:52, 5.57it/s, tok/s=15546.7]
Inference: 67%|██████▋ | 3958/5920 [49:54<05:52, 5.57it/s, tok/s=15546.3]
Inference: 67%|██████▋ | 3959/5920 [49:55<09:53, 3.31it/s, tok/s=15546.3]
Inference: 67%|██████▋ | 3959/5920 [49:55<09:53, 3.31it/s, tok/s=15543.1]
Inference: 67%|██████▋ | 3960/5920 [49:55<09:52, 3.31it/s, tok/s=15543.2]
Inference: 67%|██████▋ | 3961/5920 [49:55<07:30, 4.34it/s, tok/s=15543.2]
Inference: 67%|██████▋ | 3961/5920 [49:55<07:30, 4.34it/s, tok/s=15542.9]
Inference: 67%|██████▋ | 3962/5920 [49:56<09:28, 3.44it/s, tok/s=15542.9]
Inference: 67%|██████▋ | 3962/5920 [49:56<09:28, 3.44it/s, tok/s=15550.3]
Inference: 67%|██████▋ | 3963/5920 [49:56<09:28, 3.44it/s, tok/s=15553.3]
Inference: 67%|██████▋ | 3964/5920 [49:56<07:53, 4.13it/s, tok/s=15553.3]
Inference: 67%|██████▋ | 3964/5920 [49:56<07:53, 4.13it/s, tok/s=15552.9]
Inference: 67%|██████▋ | 3965/5920 [49:56<07:53, 4.13it/s, tok/s=15554.2]
Inference: 67%|██████▋ | 3966/5920 [49:57<06:58, 4.67it/s, tok/s=15554.2]
Inference: 67%|██████▋ | 3966/5920 [49:57<06:58, 4.67it/s, tok/s=15553.6]
Inference: 67%|██████▋ | 3967/5920 [49:57<09:28, 3.43it/s, tok/s=15553.6]
Inference: 67%|██████▋ | 3967/5920 [49:57<09:28, 3.43it/s, tok/s=15551.3]
Inference: 67%|██████▋ | 3968/5920 [49:57<09:51, 3.30it/s, tok/s=15551.3]
Inference: 67%|██████▋ | 3968/5920 [49:57<09:51, 3.30it/s, tok/s=15550.4]
Inference: 67%|██████▋ | 3969/5920 [49:58<09:00, 3.61it/s, tok/s=15550.4]
Inference: 67%|██████▋ | 3969/5920 [49:58<09:00, 3.61it/s, tok/s=15550.1]
Inference: 67%|██████▋ | 3970/5920 [49:58<07:51, 4.14it/s, tok/s=15550.1]
Inference: 67%|██████▋ | 3970/5920 [49:58<07:51, 4.14it/s, tok/s=15550.9]
Inference: 67%|██████▋ | 3971/5920 [49:58<08:12, 3.96it/s, tok/s=15550.9]
Inference: 67%|██████▋ | 3971/5920 [49:58<08:12, 3.96it/s, tok/s=15551.2]
Inference: 67%|██████▋ | 3972/5920 [49:58<07:22, 4.40it/s, tok/s=15551.2]
Inference: 67%|██████▋ | 3972/5920 [49:58<07:22, 4.40it/s, tok/s=15554.1]
Inference: 67%|██████▋ | 3973/5920 [49:58<07:22, 4.40it/s, tok/s=15554.9]
Inference: 67%|██████▋ | 3974/5920 [49:58<07:22, 4.40it/s, tok/s=15555.4]
Inference: 67%|██████▋ | 3975/5920 [49:58<04:27, 7.28it/s, tok/s=15555.4]
Inference: 67%|██████▋ | 3975/5920 [49:58<04:27, 7.28it/s, tok/s=15555.5]
Inference: 67%|██████▋ | 3976/5920 [49:59<06:29, 5.00it/s, tok/s=15555.5]
Inference: 67%|██████▋ | 3976/5920 [49:59<06:29, 5.00it/s, tok/s=15554.6]
Inference: 67%|██████▋ | 3977/5920 [49:59<07:09, 4.52it/s, tok/s=15554.6]
Inference: 67%|██████▋ | 3977/5920 [49:59<07:09, 4.52it/s, tok/s=15553.8]
Inference: 67%|██████▋ | 3978/5920 [49:59<07:25, 4.36it/s, tok/s=15553.8]
Inference: 67%|██████▋ | 3978/5920 [49:59<07:25, 4.36it/s, tok/s=15555.8]
Inference: 67%|██████▋ | 3979/5920 [50:00<06:42, 4.83it/s, tok/s=15555.8]
Inference: 67%|██████▋ | 3979/5920 [50:00<06:42, 4.83it/s, tok/s=15555.7]
Inference: 67%|██████▋ | 3980/5920 [50:00<08:41, 3.72it/s, tok/s=15555.7]
Inference: 67%|██████▋ | 3980/5920 [50:00<08:41, 3.72it/s, tok/s=15556.1]
Inference: 67%|██████▋ | 3981/5920 [50:00<08:41, 3.72it/s, tok/s=15565.9]
Inference: 67%|██████▋ | 3982/5920 [50:00<08:41, 3.72it/s, tok/s=15567.9]
Inference: 67%|██████▋ | 3983/5920 [50:00<04:41, 6.89it/s, tok/s=15567.9]
Inference: 67%|██████▋ | 3983/5920 [50:00<04:41, 6.89it/s, tok/s=15568.5]
Inference: 67%|██████▋ | 3984/5920 [50:00<04:40, 6.89it/s, tok/s=15577.7]
Inference: 67%|██████▋ | 3985/5920 [50:00<05:14, 6.16it/s, tok/s=15577.7]
Inference: 67%|██████▋ | 3985/5920 [50:00<05:14, 6.16it/s, tok/s=15576.6]
Inference: 67%|██████▋ | 3986/5920 [50:02<14:01, 2.30it/s, tok/s=15576.6]
Inference: 67%|██████▋ | 3986/5920 [50:02<14:01, 2.30it/s, tok/s=15569.1]
Inference: 67%|██████▋ | 3987/5920 [50:02<12:18, 2.62it/s, tok/s=15569.1]
Inference: 67%|██████▋ | 3987/5920 [50:02<12:18, 2.62it/s, tok/s=15569.0]
Inference: 67%|██████▋ | 3988/5920 [50:02<10:31, 3.06it/s, tok/s=15569.0]
Inference: 67%|██████▋ | 3988/5920 [50:02<10:31, 3.06it/s, tok/s=15575.3]
Inference: 67%|██████▋ | 3989/5920 [50:02<10:30, 3.06it/s, tok/s=15576.9]
Inference: 67%|██████▋ | 3990/5920 [50:03<08:15, 3.90it/s, tok/s=15576.9]
Inference: 67%|██████▋ | 3990/5920 [50:03<08:15, 3.90it/s, tok/s=15585.3]
Inference: 67%|██████▋ | 3991/5920 [50:03<08:45, 3.67it/s, tok/s=15585.3]
Inference: 67%|██████▋ | 3991/5920 [50:03<08:45, 3.67it/s, tok/s=15589.5]
Inference: 67%|██████▋ | 3992/5920 [50:03<08:44, 3.67it/s, tok/s=15589.7]
Inference: 67%|██████▋ | 3993/5920 [50:03<08:12, 3.91it/s, tok/s=15589.7]
Inference: 67%|██████▋ | 3993/5920 [50:03<08:12, 3.91it/s, tok/s=15588.6]
Inference: 67%|██████▋ | 3994/5920 [50:04<07:54, 4.06it/s, tok/s=15588.6]
Inference: 67%|██████▋ | 3994/5920 [50:04<07:54, 4.06it/s, tok/s=15588.2]
Inference: 67%|██████▋ | 3995/5920 [50:04<07:54, 4.06it/s, tok/s=15598.1]
Inference: 68%|██████▊ | 3996/5920 [50:04<07:54, 4.06it/s, tok/s=15598.6]
Inference: 68%|██████▊ | 3997/5920 [50:04<05:04, 6.32it/s, tok/s=15598.6]
Inference: 68%|██████▊ | 3997/5920 [50:04<05:04, 6.32it/s, tok/s=15598.6]
Inference: 68%|██████▊ | 3998/5920 [50:04<05:04, 6.32it/s, tok/s=15599.0]
Inference: 68%|██████▊ | 3999/5920 [50:04<04:14, 7.54it/s, tok/s=15599.0]
Inference: 68%|██████▊ | 3999/5920 [50:04<04:14, 7.54it/s, tok/s=15598.9]
Inference: 68%|██████▊ | 4000/5920 [50:04<04:14, 7.54it/s, tok/s=15599.1]
Inference: 68%|██████▊ | 4001/5920 [50:04<03:36, 8.86it/s, tok/s=15599.1]
Inference: 68%|██████▊ | 4001/5920 [50:04<03:36, 8.86it/s, tok/s=15599.6]
Inference: 68%|██████▊ | 4002/5920 [50:04<03:36, 8.86it/s, tok/s=15599.5]
Inference: 68%|██████▊ | 4003/5920 [50:05<04:32, 7.02it/s, tok/s=15599.5]
Inference: 68%|██████▊ | 4003/5920 [50:05<04:32, 7.02it/s, tok/s=15598.5]
Inference: 68%|██████▊ | 4004/5920 [50:05<04:32, 7.02it/s, tok/s=15608.4]
Inference: 68%|██████▊ | 4005/5920 [50:05<03:38, 8.75it/s, tok/s=15608.4]
Inference: 68%|██████▊ | 4005/5920 [50:05<03:38, 8.75it/s, tok/s=15608.9]
Inference: 68%|██████▊ | 4006/5920 [50:05<03:38, 8.75it/s, tok/s=15608.9]
Inference: 68%|██████▊ | 4007/5920 [50:06<06:37, 4.81it/s, tok/s=15608.9]
Inference: 68%|██████▊ | 4007/5920 [50:06<06:37, 4.81it/s, tok/s=15607.3]
Inference: 68%|██████▊ | 4008/5920 [50:06<06:37, 4.81it/s, tok/s=15607.6]
Inference: 68%|██████▊ | 4009/5920 [50:06<06:36, 4.81it/s, tok/s=15607.7]
Inference: 68%|██████▊ | 4010/5920 [50:06<04:35, 6.93it/s, tok/s=15607.7]
Inference: 68%|██████▊ | 4010/5920 [50:06<04:35, 6.93it/s, tok/s=15608.9]
Inference: 68%|██████▊ | 4011/5920 [50:06<04:35, 6.93it/s, tok/s=15608.3]
Inference: 68%|██████▊ | 4012/5920 [50:06<04:54, 6.48it/s, tok/s=15608.3]
Inference: 68%|██████▊ | 4012/5920 [50:06<04:54, 6.48it/s, tok/s=15610.6]
Inference: 68%|██████▊ | 4013/5920 [50:06<04:54, 6.48it/s, tok/s=15610.7]
Inference: 68%|██████▊ | 4014/5920 [50:06<05:19, 5.97it/s, tok/s=15610.7]
Inference: 68%|██████▊ | 4014/5920 [50:06<05:19, 5.97it/s, tok/s=15611.2]
Inference: 68%|██████▊ | 4015/5920 [50:07<06:57, 4.56it/s, tok/s=15611.2]
Inference: 68%|██████▊ | 4015/5920 [50:07<06:57, 4.56it/s, tok/s=15609.5]
Inference: 68%|██████▊ | 4016/5920 [50:07<07:15, 4.37it/s, tok/s=15609.5]
Inference: 68%|██████▊ | 4016/5920 [50:07<07:15, 4.37it/s, tok/s=15608.6]
Inference: 68%|██████▊ | 4017/5920 [50:07<06:54, 4.59it/s, tok/s=15608.6]
Inference: 68%|██████▊ | 4017/5920 [50:07<06:54, 4.59it/s, tok/s=15608.7]
Inference: 68%|██████▊ | 4018/5920 [50:08<07:45, 4.08it/s, tok/s=15608.7]
Inference: 68%|██████▊ | 4018/5920 [50:08<07:45, 4.08it/s, tok/s=15607.4]
Inference: 68%|██████▊ | 4019/5920 [50:08<07:45, 4.08it/s, tok/s=15607.8]
Inference: 68%|██████▊ | 4020/5920 [50:08<05:24, 5.86it/s, tok/s=15607.8]
Inference: 68%|██████▊ | 4020/5920 [50:08<05:24, 5.86it/s, tok/s=15608.1]
Inference: 68%|██████▊ | 4021/5920 [50:08<05:23, 5.86it/s, tok/s=15611.4]
Inference: 68%|██████▊ | 4022/5920 [50:08<05:11, 6.09it/s, tok/s=15611.4]
Inference: 68%|██████▊ | 4022/5920 [50:08<05:11, 6.09it/s, tok/s=15610.5]
Inference: 68%|██████▊ | 4023/5920 [50:09<10:44, 2.94it/s, tok/s=15610.5]
Inference: 68%|██████▊ | 4023/5920 [50:09<10:44, 2.94it/s, tok/s=15606.6]
Inference: 68%|██████▊ | 4024/5920 [50:09<09:23, 3.37it/s, tok/s=15606.6]
Inference: 68%|██████▊ | 4024/5920 [50:09<09:23, 3.37it/s, tok/s=15615.5]
Inference: 68%|██████▊ | 4025/5920 [50:09<09:22, 3.37it/s, tok/s=15616.2]
Inference: 68%|██████▊ | 4026/5920 [50:10<07:33, 4.18it/s, tok/s=15616.2]
Inference: 68%|██████▊ | 4026/5920 [50:10<07:33, 4.18it/s, tok/s=15616.8]
Inference: 68%|██████▊ | 4027/5920 [50:10<07:33, 4.18it/s, tok/s=15618.4]
Inference: 68%|██████▊ | 4028/5920 [50:10<05:31, 5.71it/s, tok/s=15618.4]
Inference: 68%|██████▊ | 4028/5920 [50:10<05:31, 5.71it/s, tok/s=15619.0]
Inference: 68%|██████▊ | 4029/5920 [50:10<05:20, 5.89it/s, tok/s=15619.0]
Inference: 68%|██████▊ | 4029/5920 [50:10<05:20, 5.89it/s, tok/s=15621.0]
Inference: 68%|██████▊ | 4030/5920 [50:10<06:07, 5.15it/s, tok/s=15621.0]
Inference: 68%|██████▊ | 4030/5920 [50:10<06:07, 5.15it/s, tok/s=15621.1]
Inference: 68%|██████▊ | 4031/5920 [50:10<05:52, 5.36it/s, tok/s=15621.1]
Inference: 68%|██████▊ | 4031/5920 [50:10<05:52, 5.36it/s, tok/s=15621.7]
Inference: 68%|██████▊ | 4032/5920 [50:10<05:52, 5.36it/s, tok/s=15622.8]
Inference: 68%|██████▊ | 4033/5920 [50:11<04:41, 6.69it/s, tok/s=15622.8]
Inference: 68%|██████▊ | 4033/5920 [50:11<04:41, 6.69it/s, tok/s=15623.6]
Inference: 68%|██████▊ | 4034/5920 [50:11<07:29, 4.19it/s, tok/s=15623.6]
Inference: 68%|██████▊ | 4034/5920 [50:11<07:29, 4.19it/s, tok/s=15630.8]
Inference: 68%|██████▊ | 4035/5920 [50:11<08:39, 3.63it/s, tok/s=15630.8]
Inference: 68%|██████▊ | 4035/5920 [50:11<08:39, 3.63it/s, tok/s=15629.8]
Inference: 68%|██████▊ | 4036/5920 [50:11<08:38, 3.63it/s, tok/s=15630.5]
Inference: 68%|██████▊ | 4037/5920 [50:11<08:38, 3.63it/s, tok/s=15631.0]
Inference: 68%|██████▊ | 4038/5920 [50:12<05:51, 5.36it/s, tok/s=15631.0]
Inference: 68%|██████▊ | 4038/5920 [50:12<05:51, 5.36it/s, tok/s=15630.7]
Inference: 68%|██████▊ | 4039/5920 [50:12<06:23, 4.91it/s, tok/s=15630.7]
Inference: 68%|██████▊ | 4039/5920 [50:12<06:23, 4.91it/s, tok/s=15631.5]
Inference: 68%|██████▊ | 4040/5920 [50:12<05:42, 5.50it/s, tok/s=15631.5]
Inference: 68%|██████▊ | 4040/5920 [50:12<05:42, 5.50it/s, tok/s=15631.8]
Inference: 68%|██████▊ | 4041/5920 [50:13<07:11, 4.35it/s, tok/s=15631.8]
Inference: 68%|██████▊ | 4041/5920 [50:13<07:11, 4.35it/s, tok/s=15636.9]
Inference: 68%|██████▊ | 4042/5920 [50:13<07:11, 4.35it/s, tok/s=15637.0]
Inference: 68%|██████▊ | 4043/5920 [50:13<06:31, 4.80it/s, tok/s=15637.0]
Inference: 68%|██████▊ | 4043/5920 [50:13<06:31, 4.80it/s, tok/s=15636.0]
Inference: 68%|██████▊ | 4044/5920 [50:13<06:30, 4.80it/s, tok/s=15636.1]
Inference: 68%|██████▊ | 4045/5920 [50:14<07:47, 4.01it/s, tok/s=15636.1]
Inference: 68%|██████▊ | 4045/5920 [50:14<07:47, 4.01it/s, tok/s=15636.6]
Inference: 68%|██████▊ | 4046/5920 [50:15<13:01, 2.40it/s, tok/s=15636.6]
Inference: 68%|██████▊ | 4046/5920 [50:15<13:01, 2.40it/s, tok/s=15631.9]
Inference: 68%|██████▊ | 4047/5920 [50:15<13:00, 2.40it/s, tok/s=15632.0]
Inference: 68%|██████▊ | 4048/5920 [50:15<09:05, 3.43it/s, tok/s=15632.0]
Inference: 68%|██████▊ | 4048/5920 [50:15<09:05, 3.43it/s, tok/s=15633.3]
Inference: 68%|██████▊ | 4049/5920 [50:15<11:15, 2.77it/s, tok/s=15633.3]
Inference: 68%|██████▊ | 4049/5920 [50:15<11:15, 2.77it/s, tok/s=15632.5]
Inference: 68%|██████▊ | 4050/5920 [50:16<09:52, 3.16it/s, tok/s=15632.5]
Inference: 68%|██████▊ | 4050/5920 [50:16<09:52, 3.16it/s, tok/s=15632.7]
Inference: 68%|██████▊ | 4051/5920 [50:16<08:18, 3.75it/s, tok/s=15632.7]
Inference: 68%|██████▊ | 4051/5920 [50:16<08:18, 3.75it/s, tok/s=15633.6]
Inference: 68%|██████▊ | 4052/5920 [50:17<14:31, 2.14it/s, tok/s=15633.6]
Inference: 68%|██████▊ | 4052/5920 [50:17<14:31, 2.14it/s, tok/s=15631.0]
Inference: 68%|██████▊ | 4053/5920 [50:17<14:30, 2.14it/s, tok/s=15631.8]
Inference: 68%|██████▊ | 4054/5920 [50:17<09:49, 3.16it/s, tok/s=15631.8]
Inference: 68%|██████▊ | 4054/5920 [50:17<09:49, 3.16it/s, tok/s=15632.5]
Inference: 68%|██████▊ | 4055/5920 [50:17<09:21, 3.32it/s, tok/s=15632.5]
Inference: 68%|██████▊ | 4055/5920 [50:17<09:21, 3.32it/s, tok/s=15632.3]
Inference: 69%|██████▊ | 4056/5920 [50:17<09:20, 3.32it/s, tok/s=15633.2]
Inference: 69%|██████▊ | 4057/5920 [50:17<09:20, 3.32it/s, tok/s=15634.0]
Inference: 69%|██████▊ | 4058/5920 [50:17<09:20, 3.32it/s, tok/s=15635.5]
Inference: 69%|██████▊ | 4059/5920 [50:17<05:29, 5.66it/s, tok/s=15635.5]
Inference: 69%|██████▊ | 4059/5920 [50:17<05:29, 5.66it/s, tok/s=15635.3]
Inference: 69%|██████▊ | 4060/5920 [50:18<05:28, 5.66it/s, tok/s=15641.2]
Inference: 69%|██████▊ | 4061/5920 [50:18<05:09, 6.01it/s, tok/s=15641.2]
Inference: 69%|██████▊ | 4061/5920 [50:18<05:09, 6.01it/s, tok/s=15642.8]
Inference: 69%|██████▊ | 4062/5920 [50:18<05:08, 6.01it/s, tok/s=15643.6]
Inference: 69%|██████▊ | 4063/5920 [50:18<04:18, 7.19it/s, tok/s=15643.6]
Inference: 69%|██████▊ | 4063/5920 [50:18<04:18, 7.19it/s, tok/s=15646.1]
Inference: 69%|██████▊ | 4064/5920 [50:18<04:44, 6.52it/s, tok/s=15646.1]
Inference: 69%|██████▊ | 4064/5920 [50:18<04:44, 6.52it/s, tok/s=15645.5]
Inference: 69%|██████▊ | 4065/5920 [50:18<04:38, 6.65it/s, tok/s=15645.5]
Inference: 69%|██████▊ | 4065/5920 [50:18<04:38, 6.65it/s, tok/s=15646.5]
Inference: 69%|██████▊ | 4066/5920 [50:18<04:38, 6.65it/s, tok/s=15647.0]
Inference: 69%|██████▊ | 4067/5920 [50:18<04:38, 6.65it/s, tok/s=15648.0]
Inference: 69%|██████▊ | 4068/5920 [50:19<07:56, 3.89it/s, tok/s=15648.0]
Inference: 69%|██████▊ | 4068/5920 [50:19<07:56, 3.89it/s, tok/s=15643.7]
Inference: 69%|██████▊ | 4069/5920 [50:20<08:08, 3.79it/s, tok/s=15643.7]
Inference: 69%|██████▊ | 4069/5920 [50:20<08:08, 3.79it/s, tok/s=15648.8]
Inference: 69%|██████▉ | 4070/5920 [50:21<11:36, 2.66it/s, tok/s=15648.8]
Inference: 69%|██████▉ | 4070/5920 [50:21<11:36, 2.66it/s, tok/s=15645.1]
Inference: 69%|██████▉ | 4071/5920 [50:21<10:09, 3.04it/s, tok/s=15645.1]
Inference: 69%|██████▉ | 4071/5920 [50:21<10:09, 3.04it/s, tok/s=15645.4]
Inference: 69%|██████▉ | 4072/5920 [50:21<09:21, 3.29it/s, tok/s=15645.4]
Inference: 69%|██████▉ | 4072/5920 [50:21<09:21, 3.29it/s, tok/s=15645.5]
Inference: 69%|██████▉ | 4073/5920 [50:21<08:59, 3.42it/s, tok/s=15645.5]
Inference: 69%|██████▉ | 4073/5920 [50:21<08:59, 3.42it/s, tok/s=15644.9]
Inference: 69%|██████▉ | 4074/5920 [50:21<08:58, 3.43it/s, tok/s=15644.9]
Inference: 69%|██████▉ | 4074/5920 [50:21<08:58, 3.43it/s, tok/s=15651.4]
Inference: 69%|██████▉ | 4075/5920 [50:22<11:48, 2.60it/s, tok/s=15651.4]
Inference: 69%|██████▉ | 4075/5920 [50:22<11:48, 2.60it/s, tok/s=15649.0]
Inference: 69%|██████▉ | 4076/5920 [50:22<10:36, 2.90it/s, tok/s=15649.0]
Inference: 69%|██████▉ | 4076/5920 [50:22<10:36, 2.90it/s, tok/s=15648.2]
Inference: 69%|██████▉ | 4077/5920 [50:22<08:49, 3.48it/s, tok/s=15648.2]
Inference: 69%|██████▉ | 4077/5920 [50:22<08:49, 3.48it/s, tok/s=15650.2]
Inference: 69%|██████▉ | 4078/5920 [50:23<10:37, 2.89it/s, tok/s=15650.2]
Inference: 69%|██████▉ | 4078/5920 [50:23<10:37, 2.89it/s, tok/s=15648.1]
Inference: 69%|██████▉ | 4079/5920 [50:23<09:11, 3.34it/s, tok/s=15648.1]
Inference: 69%|██████▉ | 4079/5920 [50:23<09:11, 3.34it/s, tok/s=15647.7]
Inference: 69%|██████▉ | 4080/5920 [50:24<09:57, 3.08it/s, tok/s=15647.7]
Inference: 69%|██████▉ | 4080/5920 [50:24<09:57, 3.08it/s, tok/s=15646.5]
Inference: 69%|██████▉ | 4081/5920 [50:24<11:34, 2.65it/s, tok/s=15646.5]
Inference: 69%|██████▉ | 4081/5920 [50:24<11:34, 2.65it/s, tok/s=15644.7]
Inference: 69%|██████▉ | 4082/5920 [50:24<11:57, 2.56it/s, tok/s=15644.7]
Inference: 69%|██████▉ | 4082/5920 [50:24<11:57, 2.56it/s, tok/s=15644.5]
Inference: 69%|██████▉ | 4083/5920 [50:25<10:36, 2.89it/s, tok/s=15644.5]
Inference: 69%|██████▉ | 4083/5920 [50:25<10:36, 2.89it/s, tok/s=15643.8]
Inference: 69%|██████▉ | 4084/5920 [50:25<10:36, 2.89it/s, tok/s=15646.7]
Inference: 69%|██████▉ | 4085/5920 [50:25<08:03, 3.80it/s, tok/s=15646.7]
Inference: 69%|██████▉ | 4085/5920 [50:25<08:03, 3.80it/s, tok/s=15646.2]
Inference: 69%|██████▉ | 4086/5920 [50:25<08:02, 3.80it/s, tok/s=15655.0]
Inference: 69%|██████▉ | 4087/5920 [50:25<08:02, 3.80it/s, tok/s=15655.5]
Inference: 69%|██████▉ | 4088/5920 [50:25<05:45, 5.31it/s, tok/s=15655.5]
Inference: 69%|██████▉ | 4088/5920 [50:25<05:45, 5.31it/s, tok/s=15654.9]
Inference: 69%|██████▉ | 4089/5920 [50:26<05:50, 5.22it/s, tok/s=15654.9]
Inference: 69%|██████▉ | 4089/5920 [50:26<05:50, 5.22it/s, tok/s=15655.6]
Inference: 69%|██████▉ | 4090/5920 [50:26<05:55, 5.14it/s, tok/s=15655.6]
Inference: 69%|██████▉ | 4090/5920 [50:26<05:55, 5.14it/s, tok/s=15655.0]
Inference: 69%|██████▉ | 4091/5920 [50:26<05:55, 5.14it/s, tok/s=15655.4]
Inference: 69%|██████▉ | 4092/5920 [50:26<06:15, 4.87it/s, tok/s=15655.4]
Inference: 69%|██████▉ | 4092/5920 [50:26<06:15, 4.87it/s, tok/s=15657.0]
Inference: 69%|██████▉ | 4093/5920 [50:26<06:14, 4.87it/s, tok/s=15663.7]
Inference: 69%|██████▉ | 4094/5920 [50:26<04:46, 6.38it/s, tok/s=15663.7]
Inference: 69%|██████▉ | 4094/5920 [50:26<04:46, 6.38it/s, tok/s=15664.0]
Inference: 69%|██████▉ | 4095/5920 [50:26<04:27, 6.83it/s, tok/s=15664.0]
Inference: 69%|██████▉ | 4095/5920 [50:26<04:27, 6.83it/s, tok/s=15664.6]
Inference: 69%|██████▉ | 4096/5920 [50:27<05:10, 5.87it/s, tok/s=15664.6]
Inference: 69%|██████▉ | 4096/5920 [50:27<05:10, 5.87it/s, tok/s=15664.4]
Inference: 69%|██████▉ | 4097/5920 [50:27<05:10, 5.87it/s, tok/s=15666.6]
Inference: 69%|██████▉ | 4098/5920 [50:27<04:34, 6.63it/s, tok/s=15666.6]
Inference: 69%|██████▉ | 4098/5920 [50:27<04:34, 6.63it/s, tok/s=15666.8]
Inference: 69%|██████▉ | 4099/5920 [50:27<04:51, 6.26it/s, tok/s=15666.8]
Inference: 69%|██████▉ | 4099/5920 [50:27<04:51, 6.26it/s, tok/s=15673.5]
Inference: 69%|██████▉ | 4100/5920 [50:27<04:50, 6.26it/s, tok/s=15674.3]
Inference: 69%|██████▉ | 4101/5920 [50:27<03:54, 7.76it/s, tok/s=15674.3]
Inference: 69%|██████▉ | 4101/5920 [50:27<03:54, 7.76it/s, tok/s=15674.4]
Inference: 69%|██████▉ | 4102/5920 [50:27<03:53, 7.78it/s, tok/s=15674.4]
Inference: 69%|██████▉ | 4102/5920 [50:27<03:53, 7.78it/s, tok/s=15682.6]
Inference: 69%|██████▉ | 4103/5920 [50:28<03:53, 7.78it/s, tok/s=15682.5]
Inference: 69%|██████▉ | 4104/5920 [50:28<05:36, 5.40it/s, tok/s=15682.5]
Inference: 69%|██████▉ | 4104/5920 [50:28<05:36, 5.40it/s, tok/s=15687.1]
Inference: 69%|██████▉ | 4105/5920 [50:28<05:47, 5.22it/s, tok/s=15687.1]
Inference: 69%|██████▉ | 4105/5920 [50:28<05:47, 5.22it/s, tok/s=15687.3]
Inference: 69%|██████▉ | 4106/5920 [50:29<07:57, 3.80it/s, tok/s=15687.3]
Inference: 69%|██████▉ | 4106/5920 [50:29<07:57, 3.80it/s, tok/s=15685.3]
Inference: 69%|██████▉ | 4107/5920 [50:29<07:10, 4.21it/s, tok/s=15685.3]
Inference: 69%|██████▉ | 4107/5920 [50:29<07:10, 4.21it/s, tok/s=15685.3]
Inference: 69%|██████▉ | 4108/5920 [50:29<06:25, 4.70it/s, tok/s=15685.3]
Inference: 69%|██████▉ | 4108/5920 [50:29<06:25, 4.70it/s, tok/s=15693.5]
Inference: 69%|██████▉ | 4109/5920 [50:29<06:25, 4.70it/s, tok/s=15693.6]
Inference: 69%|██████▉ | 4110/5920 [50:29<04:42, 6.41it/s, tok/s=15693.6]
Inference: 69%|██████▉ | 4110/5920 [50:29<04:42, 6.41it/s, tok/s=15697.5]
Inference: 69%|██████▉ | 4111/5920 [50:29<05:38, 5.34it/s, tok/s=15697.5]
Inference: 69%|██████▉ | 4111/5920 [50:29<05:38, 5.34it/s, tok/s=15696.8]
Inference: 69%|██████▉ | 4112/5920 [50:30<05:06, 5.90it/s, tok/s=15696.8]
Inference: 69%|██████▉ | 4112/5920 [50:30<05:06, 5.90it/s, tok/s=15699.6]
Inference: 69%|██████▉ | 4113/5920 [50:30<05:06, 5.90it/s, tok/s=15700.5]
Inference: 69%|██████▉ | 4114/5920 [50:30<04:43, 6.37it/s, tok/s=15700.5]
Inference: 69%|██████▉ | 4114/5920 [50:30<04:43, 6.37it/s, tok/s=15702.1]
Inference: 70%|██████▉ | 4115/5920 [50:30<04:49, 6.23it/s, tok/s=15702.1]
Inference: 70%|██████▉ | 4115/5920 [50:30<04:49, 6.23it/s, tok/s=15701.9]
Inference: 70%|██████▉ | 4116/5920 [50:31<09:27, 3.18it/s, tok/s=15701.9]
Inference: 70%|██████▉ | 4116/5920 [50:31<09:27, 3.18it/s, tok/s=15698.2]
Inference: 70%|██████▉ | 4117/5920 [50:31<09:27, 3.18it/s, tok/s=15698.8]
Inference: 70%|██████▉ | 4118/5920 [50:31<07:01, 4.28it/s, tok/s=15698.8]
Inference: 70%|██████▉ | 4118/5920 [50:31<07:01, 4.28it/s, tok/s=15698.7]
Inference: 70%|██████▉ | 4119/5920 [50:31<07:00, 4.28it/s, tok/s=15699.2]
Inference: 70%|██████▉ | 4120/5920 [50:31<07:00, 4.28it/s, tok/s=15699.6]
Inference: 70%|██████▉ | 4121/5920 [50:31<04:44, 6.32it/s, tok/s=15699.6]
Inference: 70%|██████▉ | 4121/5920 [50:31<04:44, 6.32it/s, tok/s=15700.2]
Inference: 70%|██████▉ | 4122/5920 [50:32<06:01, 4.98it/s, tok/s=15700.2]
Inference: 70%|██████▉ | 4122/5920 [50:32<06:01, 4.98it/s, tok/s=15699.1]
Inference: 70%|██████▉ | 4123/5920 [50:32<07:28, 4.00it/s, tok/s=15699.1]
Inference: 70%|██████▉ | 4123/5920 [50:32<07:28, 4.00it/s, tok/s=15704.5]
Inference: 70%|██████▉ | 4124/5920 [50:33<08:30, 3.52it/s, tok/s=15704.5]
Inference: 70%|██████▉ | 4124/5920 [50:33<08:30, 3.52it/s, tok/s=15704.7]
Inference: 70%|██████▉ | 4125/5920 [50:33<07:24, 4.03it/s, tok/s=15704.7]
Inference: 70%|██████▉ | 4125/5920 [50:33<07:24, 4.03it/s, tok/s=15704.9]
Inference: 70%|██████▉ | 4126/5920 [50:33<09:40, 3.09it/s, tok/s=15704.9]
Inference: 70%|██████▉ | 4126/5920 [50:33<09:40, 3.09it/s, tok/s=15703.1]
Inference: 70%|██████▉ | 4127/5920 [50:33<09:40, 3.09it/s, tok/s=15704.2]
Inference: 70%|██████▉ | 4128/5920 [50:33<07:25, 4.02it/s, tok/s=15704.2]
Inference: 70%|██████▉ | 4128/5920 [50:33<07:25, 4.02it/s, tok/s=15704.7]
Inference: 70%|██████▉ | 4129/5920 [50:34<07:25, 4.02it/s, tok/s=15705.1]
Inference: 70%|██████▉ | 4130/5920 [50:34<07:24, 4.02it/s, tok/s=15705.4]
Inference: 70%|██████▉ | 4131/5920 [50:34<07:24, 4.02it/s, tok/s=15710.4]
Inference: 70%|██████▉ | 4132/5920 [50:34<05:12, 5.72it/s, tok/s=15710.4]
Inference: 70%|██████▉ | 4132/5920 [50:34<05:12, 5.72it/s, tok/s=15709.0]
Inference: 70%|██████▉ | 4133/5920 [50:34<05:06, 5.83it/s, tok/s=15709.0]
Inference: 70%|██████▉ | 4133/5920 [50:34<05:06, 5.83it/s, tok/s=15708.9]
Inference: 70%|██████▉ | 4134/5920 [50:34<05:06, 5.83it/s, tok/s=15710.5]
Inference: 70%|██████▉ | 4135/5920 [50:34<04:57, 5.99it/s, tok/s=15710.5]
Inference: 70%|██████▉ | 4135/5920 [50:34<04:57, 5.99it/s, tok/s=15710.7]
Inference: 70%|██████▉ | 4136/5920 [50:35<06:24, 4.64it/s, tok/s=15710.7]
Inference: 70%|██████▉ | 4136/5920 [50:35<06:24, 4.64it/s, tok/s=15710.1]
Inference: 70%|██████▉ | 4137/5920 [50:35<06:24, 4.64it/s, tok/s=15711.3]
Inference: 70%|██████▉ | 4138/5920 [50:35<06:16, 4.74it/s, tok/s=15711.3]
Inference: 70%|██████▉ | 4138/5920 [50:35<06:16, 4.74it/s, tok/s=15711.9]
Inference: 70%|██████▉ | 4139/5920 [50:35<06:15, 4.74it/s, tok/s=15717.1]
Inference: 70%|██████▉ | 4140/5920 [50:35<06:15, 4.74it/s, tok/s=15717.8]
Inference: 70%|██████▉ | 4141/5920 [50:35<06:15, 4.74it/s, tok/s=15718.2]
Inference: 70%|██████▉ | 4142/5920 [50:35<03:43, 7.95it/s, tok/s=15718.2]
Inference: 70%|██████▉ | 4142/5920 [50:35<03:43, 7.95it/s, tok/s=15718.2]
Inference: 70%|██████▉ | 4143/5920 [50:35<03:43, 7.95it/s, tok/s=15719.0]
Inference: 70%|███████ | 4144/5920 [50:36<03:12, 9.24it/s, tok/s=15719.0]
Inference: 70%|███████ | 4144/5920 [50:36<03:12, 9.24it/s, tok/s=15720.3]
Inference: 70%|███████ | 4145/5920 [50:36<03:12, 9.24it/s, tok/s=15720.9]
Inference: 70%|███████ | 4146/5920 [50:36<04:05, 7.23it/s, tok/s=15720.9]
Inference: 70%|███████ | 4146/5920 [50:36<04:05, 7.23it/s, tok/s=15722.5]
Inference: 70%|███████ | 4147/5920 [50:36<04:05, 7.23it/s, tok/s=15722.9]
Inference: 70%|███████ | 4148/5920 [50:36<03:59, 7.41it/s, tok/s=15722.9]
Inference: 70%|███████ | 4148/5920 [50:36<03:59, 7.41it/s, tok/s=15726.4]
Inference: 70%|███████ | 4149/5920 [50:36<04:01, 7.32it/s, tok/s=15726.4]
Inference: 70%|███████ | 4149/5920 [50:36<04:01, 7.32it/s, tok/s=15728.9]
Inference: 70%|███████ | 4150/5920 [50:36<04:01, 7.32it/s, tok/s=15729.8]
Inference: 70%|███████ | 4151/5920 [50:37<03:36, 8.16it/s, tok/s=15729.8]
Inference: 70%|███████ | 4151/5920 [50:37<03:36, 8.16it/s, tok/s=15729.9]
Inference: 70%|███████ | 4152/5920 [50:37<03:37, 8.12it/s, tok/s=15729.9]
Inference: 70%|███████ | 4152/5920 [50:37<03:37, 8.12it/s, tok/s=15729.9]
Inference: 70%|███████ | 4153/5920 [50:37<05:10, 5.69it/s, tok/s=15729.9]
Inference: 70%|███████ | 4153/5920 [50:37<05:10, 5.69it/s, tok/s=15728.6]
Inference: 70%|███████ | 4154/5920 [50:37<05:10, 5.69it/s, tok/s=15729.2]
Inference: 70%|███████ | 4155/5920 [50:37<05:09, 5.69it/s, tok/s=15730.1]
Inference: 70%|███████ | 4156/5920 [50:38<05:47, 5.08it/s, tok/s=15730.1]
Inference: 70%|███████ | 4156/5920 [50:38<05:47, 5.08it/s, tok/s=15735.5]
Inference: 70%|███████ | 4157/5920 [50:38<05:30, 5.34it/s, tok/s=15735.5]
Inference: 70%|███████ | 4157/5920 [50:38<05:30, 5.34it/s, tok/s=15738.8]
Inference: 70%|███████ | 4158/5920 [50:38<06:28, 4.54it/s, tok/s=15738.8]
Inference: 70%|███████ | 4158/5920 [50:38<06:28, 4.54it/s, tok/s=15739.4]
Inference: 70%|███████ | 4159/5920 [50:38<06:27, 4.54it/s, tok/s=15742.2]
Inference: 70%|███████ | 4160/5920 [50:38<05:27, 5.38it/s, tok/s=15742.2]
Inference: 70%|███████ | 4160/5920 [50:38<05:27, 5.38it/s, tok/s=15742.2]
Inference: 70%|███████ | 4161/5920 [50:39<06:56, 4.22it/s, tok/s=15742.2]
Inference: 70%|███████ | 4161/5920 [50:39<06:56, 4.22it/s, tok/s=15743.4]
Inference: 70%|███████ | 4162/5920 [50:39<06:38, 4.41it/s, tok/s=15743.4]
Inference: 70%|███████ | 4162/5920 [50:39<06:38, 4.41it/s, tok/s=15744.3]
Inference: 70%|███████ | 4163/5920 [50:39<06:38, 4.41it/s, tok/s=15744.9]
Inference: 70%|███████ | 4164/5920 [50:39<05:00, 5.84it/s, tok/s=15744.9]
Inference: 70%|███████ | 4164/5920 [50:39<05:00, 5.84it/s, tok/s=15752.8]
Inference: 70%|███████ | 4165/5920 [50:39<05:00, 5.84it/s, tok/s=15753.0]
Inference: 70%|███████ | 4166/5920 [50:39<03:54, 7.48it/s, tok/s=15753.0]
Inference: 70%|███████ | 4166/5920 [50:39<03:54, 7.48it/s, tok/s=15753.7]
Inference: 70%|███████ | 4167/5920 [50:40<04:35, 6.35it/s, tok/s=15753.7]
Inference: 70%|███████ | 4167/5920 [50:40<04:35, 6.35it/s, tok/s=15752.8]
Inference: 70%|███████ | 4168/5920 [50:40<05:02, 5.79it/s, tok/s=15752.8]
Inference: 70%|███████ | 4168/5920 [50:40<05:02, 5.79it/s, tok/s=15752.3]
Inference: 70%|███████ | 4169/5920 [50:40<05:02, 5.79it/s, tok/s=15753.2]
Inference: 70%|███████ | 4170/5920 [50:40<04:07, 7.08it/s, tok/s=15753.2]
Inference: 70%|███████ | 4170/5920 [50:40<04:07, 7.08it/s, tok/s=15753.2]
Inference: 70%|███████ | 4171/5920 [50:40<05:06, 5.71it/s, tok/s=15753.2]
Inference: 70%|███████ | 4171/5920 [50:40<05:06, 5.71it/s, tok/s=15752.3]
Inference: 70%|███████ | 4172/5920 [50:40<05:06, 5.71it/s, tok/s=15752.5]
Inference: 70%|███████ | 4173/5920 [50:41<04:42, 6.17it/s, tok/s=15752.5]
Inference: 70%|███████ | 4173/5920 [50:41<04:42, 6.17it/s, tok/s=15752.5]
Inference: 71%|███████ | 4174/5920 [50:41<04:42, 6.17it/s, tok/s=15752.6]
Inference: 71%|███████ | 4175/5920 [50:41<04:04, 7.13it/s, tok/s=15752.6]
Inference: 71%|███████ | 4175/5920 [50:41<04:04, 7.13it/s, tok/s=15752.4]
Inference: 71%|███████ | 4176/5920 [50:41<06:32, 4.45it/s, tok/s=15752.4]
Inference: 71%|███████ | 4176/5920 [50:41<06:32, 4.45it/s, tok/s=15750.3]
Inference: 71%|███████ | 4177/5920 [50:41<06:32, 4.45it/s, tok/s=15751.4]
Inference: 71%|███████ | 4178/5920 [50:41<06:31, 4.45it/s, tok/s=15751.7]
Inference: 71%|███████ | 4179/5920 [50:42<05:41, 5.10it/s, tok/s=15751.7]
Inference: 71%|███████ | 4179/5920 [50:42<05:41, 5.10it/s, tok/s=15750.0]
Inference: 71%|███████ | 4180/5920 [50:42<05:41, 5.10it/s, tok/s=15750.5]
Inference: 71%|███████ | 4181/5920 [50:42<04:53, 5.93it/s, tok/s=15750.5]
Inference: 71%|███████ | 4181/5920 [50:42<04:53, 5.93it/s, tok/s=15750.1]
Inference: 71%|███████ | 4182/5920 [50:42<04:53, 5.93it/s, tok/s=15750.7]
Inference: 71%|███████ | 4183/5920 [50:42<04:53, 5.93it/s, tok/s=15752.2]
Inference: 71%|███████ | 4184/5920 [50:42<03:55, 7.38it/s, tok/s=15752.2]
Inference: 71%|███████ | 4184/5920 [50:42<03:55, 7.38it/s, tok/s=15752.9]
Inference: 71%|███████ | 4185/5920 [50:42<03:55, 7.38it/s, tok/s=15753.5]
Inference: 71%|███████ | 4186/5920 [50:42<03:55, 7.38it/s, tok/s=15754.0]
Inference: 71%|███████ | 4187/5920 [50:43<03:24, 8.47it/s, tok/s=15754.0]
Inference: 71%|███████ | 4187/5920 [50:43<03:24, 8.47it/s, tok/s=15753.6]
Inference: 71%|███████ | 4188/5920 [50:43<03:24, 8.47it/s, tok/s=15756.5]
Inference: 71%|███████ | 4189/5920 [50:43<02:55, 9.87it/s, tok/s=15756.5]
Inference: 71%|███████ | 4189/5920 [50:43<02:55, 9.87it/s, tok/s=15757.0]
Inference: 71%|███████ | 4190/5920 [50:43<02:55, 9.87it/s, tok/s=15757.7]
Inference: 71%|███████ | 4191/5920 [50:43<02:46, 10.38it/s, tok/s=15757.7]
Inference: 71%|███████ | 4191/5920 [50:43<02:46, 10.38it/s, tok/s=15760.3]
Inference: 71%|███████ | 4192/5920 [50:43<02:46, 10.38it/s, tok/s=15760.4]
Inference: 71%|███████ | 4193/5920 [50:44<04:41, 6.13it/s, tok/s=15760.4]
Inference: 71%|███████ | 4193/5920 [50:44<04:41, 6.13it/s, tok/s=15758.1]
Inference: 71%|███████ | 4194/5920 [50:44<04:30, 6.39it/s, tok/s=15758.1]
Inference: 71%|███████ | 4194/5920 [50:44<04:30, 6.39it/s, tok/s=15758.4]
Inference: 71%|███████ | 4195/5920 [50:44<04:29, 6.39it/s, tok/s=15759.1]
Inference: 71%|███████ | 4196/5920 [50:44<04:12, 6.83it/s, tok/s=15759.1]
Inference: 71%|███████ | 4196/5920 [50:44<04:12, 6.83it/s, tok/s=15758.7]
Inference: 71%|███████ | 4197/5920 [50:44<04:12, 6.83it/s, tok/s=15768.4]
Inference: 71%|███████ | 4198/5920 [50:44<03:34, 8.01it/s, tok/s=15768.4]
Inference: 71%|███████ | 4198/5920 [50:44<03:34, 8.01it/s, tok/s=15769.7]
Inference: 71%|███████ | 4199/5920 [50:44<03:34, 8.01it/s, tok/s=15770.6]
Inference: 71%|███████ | 4200/5920 [50:44<03:34, 8.01it/s, tok/s=15772.8]
Inference: 71%|███████ | 4201/5920 [50:44<03:15, 8.79it/s, tok/s=15772.8]
Inference: 71%|███████ | 4201/5920 [50:44<03:15, 8.79it/s, tok/s=15772.5]
Inference: 71%|███████ | 4202/5920 [50:45<03:15, 8.79it/s, tok/s=15769.1]
Inference: 71%|███████ | 4203/5920 [50:45<05:37, 5.08it/s, tok/s=15769.1]
Inference: 71%|███████ | 4203/5920 [50:45<05:37, 5.08it/s, tok/s=15776.7]
Inference: 71%|███████ | 4204/5920 [50:45<05:37, 5.08it/s, tok/s=15777.0]
Inference: 71%|███████ | 4205/5920 [50:45<04:26, 6.44it/s, tok/s=15777.0]
Inference: 71%|███████ | 4205/5920 [50:45<04:26, 6.44it/s, tok/s=15777.2]
Inference: 71%|███████ | 4206/5920 [50:45<04:26, 6.44it/s, tok/s=15777.3]
Inference: 71%|███████ | 4207/5920 [50:46<04:24, 6.47it/s, tok/s=15777.3]
Inference: 71%|███████ | 4207/5920 [50:46<04:24, 6.47it/s, tok/s=15778.5]
Inference: 71%|███████ | 4208/5920 [50:46<04:11, 6.81it/s, tok/s=15778.5]
Inference: 71%|███████ | 4208/5920 [50:46<04:11, 6.81it/s, tok/s=15779.2]
Inference: 71%|███████ | 4209/5920 [50:46<06:06, 4.67it/s, tok/s=15779.2]
Inference: 71%|███████ | 4209/5920 [50:46<06:06, 4.67it/s, tok/s=15777.0]
Inference: 71%|███████ | 4210/5920 [50:46<05:29, 5.19it/s, tok/s=15777.0]
Inference: 71%|███████ | 4210/5920 [50:46<05:29, 5.19it/s, tok/s=15786.0]
Inference: 71%|███████ | 4211/5920 [50:46<05:29, 5.19it/s, tok/s=15786.3]
Inference: 71%|███████ | 4212/5920 [50:47<05:12, 5.46it/s, tok/s=15786.3]
Inference: 71%|███████ | 4212/5920 [50:47<05:12, 5.46it/s, tok/s=15785.5]
Inference: 71%|███████ | 4213/5920 [50:47<04:50, 5.87it/s, tok/s=15785.5]
Inference: 71%|███████ | 4213/5920 [50:47<04:50, 5.87it/s, tok/s=15785.3]
Inference: 71%|███████ | 4214/5920 [50:47<04:50, 5.87it/s, tok/s=15785.5]
Inference: 71%|███████ | 4215/5920 [50:47<03:34, 7.97it/s, tok/s=15785.5]
Inference: 71%|███████ | 4215/5920 [50:47<03:34, 7.97it/s, tok/s=15790.9]
Inference: 71%|███████ | 4216/5920 [50:47<03:33, 7.97it/s, tok/s=15792.7]
Inference: 71%|███████ | 4217/5920 [50:47<03:06, 9.15it/s, tok/s=15792.7]
Inference: 71%|███████ | 4217/5920 [50:47<03:06, 9.15it/s, tok/s=15796.3]
Inference: 71%|███████▏ | 4218/5920 [50:47<03:05, 9.15it/s, tok/s=15796.8]
Inference: 71%|███████▏ | 4219/5920 [50:47<03:32, 7.99it/s, tok/s=15796.8]
Inference: 71%|███████▏ | 4219/5920 [50:47<03:32, 7.99it/s, tok/s=15800.2]
Inference: 71%|███████▏ | 4220/5920 [50:47<03:44, 7.57it/s, tok/s=15800.2]
Inference: 71%|███████▏ | 4220/5920 [50:47<03:44, 7.57it/s, tok/s=15800.4]
Inference: 71%|███████▏ | 4221/5920 [50:48<03:51, 7.33it/s, tok/s=15800.4]
Inference: 71%|███████▏ | 4221/5920 [50:48<03:51, 7.33it/s, tok/s=15800.0]
Inference: 71%|███████▏ | 4222/5920 [50:48<03:51, 7.33it/s, tok/s=15800.3]
Inference: 71%|███████▏ | 4223/5920 [50:48<05:42, 4.95it/s, tok/s=15800.3]
Inference: 71%|███████▏ | 4223/5920 [50:48<05:42, 4.95it/s, tok/s=15800.1]
Inference: 71%|███████▏ | 4224/5920 [50:48<05:42, 4.95it/s, tok/s=15800.8]
Inference: 71%|███████▏ | 4225/5920 [50:48<04:16, 6.62it/s, tok/s=15800.8]
Inference: 71%|███████▏ | 4225/5920 [50:48<04:16, 6.62it/s, tok/s=15802.2]
Inference: 71%|███████▏ | 4226/5920 [50:48<04:15, 6.62it/s, tok/s=15802.9]
Inference: 71%|███████▏ | 4227/5920 [50:49<04:04, 6.94it/s, tok/s=15802.9]
Inference: 71%|███████▏ | 4227/5920 [50:49<04:04, 6.94it/s, tok/s=15802.5]
Inference: 71%|███████▏ | 4228/5920 [50:49<04:03, 6.94it/s, tok/s=15802.6]
Inference: 71%|███████▏ | 4229/5920 [50:50<07:51, 3.58it/s, tok/s=15802.6]
Inference: 71%|███████▏ | 4229/5920 [50:50<07:51, 3.58it/s, tok/s=15806.4]
Inference: 71%|███████▏ | 4230/5920 [50:50<07:03, 3.99it/s, tok/s=15806.4]
Inference: 71%|███████▏ | 4230/5920 [50:50<07:03, 3.99it/s, tok/s=15807.6]
Inference: 71%|███████▏ | 4231/5920 [50:50<07:03, 3.99it/s, tok/s=15813.2]
Inference: 71%|███████▏ | 4232/5920 [50:50<05:51, 4.80it/s, tok/s=15813.2]
Inference: 71%|███████▏ | 4232/5920 [50:50<05:51, 4.80it/s, tok/s=15813.2]
Inference: 72%|███████▏ | 4233/5920 [50:50<05:51, 4.80it/s, tok/s=15813.5]
Inference: 72%|███████▏ | 4234/5920 [50:50<04:52, 5.77it/s, tok/s=15813.5]
Inference: 72%|███████▏ | 4234/5920 [50:50<04:52, 5.77it/s, tok/s=15812.8]
Inference: 72%|███████▏ | 4235/5920 [50:50<04:38, 6.06it/s, tok/s=15812.8]
Inference: 72%|███████▏ | 4235/5920 [50:50<04:38, 6.06it/s, tok/s=15812.5]
Inference: 72%|███████▏ | 4236/5920 [50:51<04:38, 6.06it/s, tok/s=15822.1]
Inference: 72%|███████▏ | 4237/5920 [50:52<10:44, 2.61it/s, tok/s=15822.1]
Inference: 72%|███████▏ | 4237/5920 [50:52<10:44, 2.61it/s, tok/s=15816.1]
Inference: 72%|███████▏ | 4238/5920 [50:52<09:45, 2.87it/s, tok/s=15816.1]
Inference: 72%|███████▏ | 4238/5920 [50:52<09:45, 2.87it/s, tok/s=15817.8]
Inference: 72%|███████▏ | 4239/5920 [50:52<09:44, 2.87it/s, tok/s=15818.3]
Inference: 72%|███████▏ | 4240/5920 [50:52<07:20, 3.82it/s, tok/s=15818.3]
Inference: 72%|███████▏ | 4240/5920 [50:52<07:20, 3.82it/s, tok/s=15818.1]
Inference: 72%|███████▏ | 4241/5920 [50:53<08:36, 3.25it/s, tok/s=15818.1]
Inference: 72%|███████▏ | 4241/5920 [50:53<08:36, 3.25it/s, tok/s=15816.0]
Inference: 72%|███████▏ | 4242/5920 [50:53<09:23, 2.98it/s, tok/s=15816.0]
Inference: 72%|███████▏ | 4242/5920 [50:53<09:23, 2.98it/s, tok/s=15814.2]
Inference: 72%|███████▏ | 4243/5920 [50:54<09:38, 2.90it/s, tok/s=15814.2]
Inference: 72%|███████▏ | 4243/5920 [50:54<09:38, 2.90it/s, tok/s=15812.7]
Inference: 72%|███████▏ | 4244/5920 [50:54<09:12, 3.04it/s, tok/s=15812.7]
Inference: 72%|███████▏ | 4244/5920 [50:54<09:12, 3.04it/s, tok/s=15812.7]
Inference: 72%|███████▏ | 4245/5920 [50:54<09:11, 3.04it/s, tok/s=15816.4]
Inference: 72%|███████▏ | 4246/5920 [50:54<06:28, 4.31it/s, tok/s=15816.4]
Inference: 72%|███████▏ | 4246/5920 [50:54<06:28, 4.31it/s, tok/s=15816.1]
Inference: 72%|███████▏ | 4247/5920 [50:54<06:28, 4.31it/s, tok/s=15817.4]
Inference: 72%|███████▏ | 4248/5920 [50:55<05:45, 4.84it/s, tok/s=15817.4]
Inference: 72%|███████▏ | 4248/5920 [50:55<05:45, 4.84it/s, tok/s=15816.3]
Inference: 72%|███████▏ | 4249/5920 [50:55<05:45, 4.84it/s, tok/s=15816.7]
Inference: 72%|███████▏ | 4250/5920 [50:55<04:16, 6.50it/s, tok/s=15816.7]
Inference: 72%|███████▏ | 4250/5920 [50:55<04:16, 6.50it/s, tok/s=15818.8]
Inference: 72%|███████▏ | 4251/5920 [50:55<04:16, 6.50it/s, tok/s=15819.0]
Inference: 72%|███████▏ | 4252/5920 [50:55<03:18, 8.39it/s, tok/s=15819.0]
Inference: 72%|███████▏ | 4252/5920 [50:55<03:18, 8.39it/s, tok/s=15820.5]
Inference: 72%|███████▏ | 4253/5920 [50:55<03:18, 8.39it/s, tok/s=15819.6]
Inference: 72%|███████▏ | 4254/5920 [50:55<04:26, 6.24it/s, tok/s=15819.6]
Inference: 72%|███████▏ | 4254/5920 [50:55<04:26, 6.24it/s, tok/s=15820.8]
Inference: 72%|███████▏ | 4255/5920 [50:55<04:26, 6.24it/s, tok/s=15821.3]
Inference: 72%|███████▏ | 4256/5920 [50:56<04:23, 6.31it/s, tok/s=15821.3]
Inference: 72%|███████▏ | 4256/5920 [50:56<04:23, 6.31it/s, tok/s=15822.3]
Inference: 72%|███████▏ | 4257/5920 [50:56<04:27, 6.21it/s, tok/s=15822.3]
Inference: 72%|███████▏ | 4257/5920 [50:56<04:27, 6.21it/s, tok/s=15821.7]
Inference: 72%|███████▏ | 4258/5920 [50:56<04:27, 6.21it/s, tok/s=15823.8]
Inference: 72%|███████▏ | 4259/5920 [50:57<06:23, 4.33it/s, tok/s=15823.8]
Inference: 72%|███████▏ | 4259/5920 [50:57<06:23, 4.33it/s, tok/s=15821.2]
Inference: 72%|███████▏ | 4260/5920 [50:57<06:45, 4.10it/s, tok/s=15821.2]
Inference: 72%|███████▏ | 4260/5920 [50:57<06:45, 4.10it/s, tok/s=15820.6]
Inference: 72%|███████▏ | 4261/5920 [50:57<06:44, 4.10it/s, tok/s=15820.9]
Inference: 72%|███████▏ | 4262/5920 [50:57<06:44, 4.10it/s, tok/s=15821.2]
Inference: 72%|███████▏ | 4263/5920 [50:57<04:21, 6.34it/s, tok/s=15821.2]
Inference: 72%|███████▏ | 4263/5920 [50:57<04:21, 6.34it/s, tok/s=15821.3]
Inference: 72%|███████▏ | 4264/5920 [50:57<05:49, 4.74it/s, tok/s=15821.3]
Inference: 72%|███████▏ | 4264/5920 [50:57<05:49, 4.74it/s, tok/s=15821.0]
Inference: 72%|███████▏ | 4265/5920 [50:57<05:49, 4.74it/s, tok/s=15821.5]
Inference: 72%|███████▏ | 4266/5920 [50:58<05:31, 4.99it/s, tok/s=15821.5]
Inference: 72%|███████▏ | 4266/5920 [50:58<05:31, 4.99it/s, tok/s=15820.4]
Inference: 72%|███████▏ | 4267/5920 [50:58<05:08, 5.35it/s, tok/s=15820.4]
Inference: 72%|███████▏ | 4267/5920 [50:58<05:08, 5.35it/s, tok/s=15824.0]
Inference: 72%|███████▏ | 4268/5920 [50:58<05:08, 5.35it/s, tok/s=15825.2]
Inference: 72%|███████▏ | 4269/5920 [50:58<05:25, 5.08it/s, tok/s=15825.2]
Inference: 72%|███████▏ | 4269/5920 [50:58<05:25, 5.08it/s, tok/s=15824.5]
Inference: 72%|███████▏ | 4270/5920 [50:59<06:20, 4.34it/s, tok/s=15824.5]
Inference: 72%|███████▏ | 4270/5920 [50:59<06:20, 4.34it/s, tok/s=15824.6]
Inference: 72%|███████▏ | 4271/5920 [50:59<06:20, 4.34it/s, tok/s=15825.0]
Inference: 72%|███████▏ | 4272/5920 [50:59<06:19, 4.34it/s, tok/s=15826.3]
Inference: 72%|███████▏ | 4273/5920 [50:59<06:19, 4.34it/s, tok/s=15826.8]
Inference: 72%|███████▏ | 4274/5920 [50:59<03:31, 7.77it/s, tok/s=15826.8]
Inference: 72%|███████▏ | 4274/5920 [50:59<03:31, 7.77it/s, tok/s=15827.5]
Inference: 72%|███████▏ | 4275/5920 [50:59<03:31, 7.77it/s, tok/s=15828.1]
Inference: 72%|███████▏ | 4276/5920 [50:59<03:31, 7.77it/s, tok/s=15828.9]
Inference: 72%|███████▏ | 4277/5920 [50:59<02:42, 10.10it/s, tok/s=15828.9]
Inference: 72%|███████▏ | 4277/5920 [50:59<02:42, 10.10it/s, tok/s=15829.5]
Inference: 72%|███████▏ | 4278/5920 [50:59<02:42, 10.10it/s, tok/s=15829.0]
Inference: 72%|███████▏ | 4279/5920 [51:00<05:09, 5.30it/s, tok/s=15829.0]
Inference: 72%|███████▏ | 4279/5920 [51:00<05:09, 5.30it/s, tok/s=15830.8]
Inference: 72%|███████▏ | 4280/5920 [51:00<05:09, 5.30it/s, tok/s=15832.5]
Inference: 72%|███████▏ | 4281/5920 [51:00<04:30, 6.07it/s, tok/s=15832.5]
Inference: 72%|███████▏ | 4281/5920 [51:00<04:30, 6.07it/s, tok/s=15832.0]
Inference: 72%|███████▏ | 4282/5920 [51:00<04:29, 6.07it/s, tok/s=15832.3]
Inference: 72%|███████▏ | 4283/5920 [51:00<03:52, 7.03it/s, tok/s=15832.3]
Inference: 72%|███████▏ | 4283/5920 [51:00<03:52, 7.03it/s, tok/s=15833.6]
Inference: 72%|███████▏ | 4284/5920 [51:01<03:52, 7.03it/s, tok/s=15832.1]
Inference: 72%|███████▏ | 4285/5920 [51:02<09:26, 2.89it/s, tok/s=15832.1]
Inference: 72%|███████▏ | 4285/5920 [51:02<09:26, 2.89it/s, tok/s=15827.3]
Inference: 72%|███████▏ | 4286/5920 [51:03<10:31, 2.59it/s, tok/s=15827.3]
Inference: 72%|███████▏ | 4286/5920 [51:03<10:31, 2.59it/s, tok/s=15825.2]
Inference: 72%|███████▏ | 4287/5920 [51:03<09:35, 2.84it/s, tok/s=15825.2]
Inference: 72%|███████▏ | 4287/5920 [51:03<09:35, 2.84it/s, tok/s=15824.8]
Inference: 72%|███████▏ | 4288/5920 [51:03<09:35, 2.84it/s, tok/s=15826.1]
Inference: 72%|███████▏ | 4289/5920 [51:03<07:45, 3.50it/s, tok/s=15826.1]
Inference: 72%|███████▏ | 4289/5920 [51:03<07:45, 3.50it/s, tok/s=15825.3]
Inference: 72%|███████▏ | 4290/5920 [51:04<10:24, 2.61it/s, tok/s=15825.3]
Inference: 72%|███████▏ | 4290/5920 [51:04<10:24, 2.61it/s, tok/s=15821.9]
Inference: 72%|███████▏ | 4291/5920 [51:04<09:51, 2.76it/s, tok/s=15821.9]
Inference: 72%|███████▏ | 4291/5920 [51:04<09:51, 2.76it/s, tok/s=15821.7]
Inference: 72%|███████▎ | 4292/5920 [51:04<08:26, 3.21it/s, tok/s=15821.7]
Inference: 72%|███████▎ | 4292/5920 [51:04<08:26, 3.21it/s, tok/s=15821.5]
Inference: 73%|███████▎ | 4293/5920 [51:04<08:26, 3.21it/s, tok/s=15822.3]
Inference: 73%|███████▎ | 4294/5920 [51:04<05:53, 4.60it/s, tok/s=15822.3]
Inference: 73%|███████▎ | 4294/5920 [51:04<05:53, 4.60it/s, tok/s=15824.2]
Inference: 73%|███████▎ | 4295/5920 [51:05<05:53, 4.60it/s, tok/s=15824.9]
Inference: 73%|███████▎ | 4296/5920 [51:05<05:53, 4.60it/s, tok/s=15826.7]
Inference: 73%|███████▎ | 4297/5920 [51:05<03:47, 7.13it/s, tok/s=15826.7]
Inference: 73%|███████▎ | 4297/5920 [51:05<03:47, 7.13it/s, tok/s=15836.1]
Inference: 73%|███████▎ | 4298/5920 [51:05<03:47, 7.13it/s, tok/s=15836.5]
Inference: 73%|███████▎ | 4299/5920 [51:05<04:38, 5.82it/s, tok/s=15836.5]
Inference: 73%|███████▎ | 4299/5920 [51:05<04:38, 5.82it/s, tok/s=15840.3]
Inference: 73%|███████▎ | 4300/5920 [51:05<04:38, 5.82it/s, tok/s=15841.0]
Inference: 73%|███████▎ | 4301/5920 [51:06<04:54, 5.50it/s, tok/s=15841.0]
Inference: 73%|███████▎ | 4301/5920 [51:06<04:54, 5.50it/s, tok/s=15840.9]
Inference: 73%|███████▎ | 4302/5920 [51:06<04:32, 5.94it/s, tok/s=15840.9]
Inference: 73%|███████▎ | 4302/5920 [51:06<04:32, 5.94it/s, tok/s=15841.0]
Inference: 73%|███████▎ | 4303/5920 [51:06<04:32, 5.94it/s, tok/s=15843.6]
Inference: 73%|███████▎ | 4304/5920 [51:06<05:23, 4.99it/s, tok/s=15843.6]
Inference: 73%|███████▎ | 4304/5920 [51:06<05:23, 4.99it/s, tok/s=15841.5]
Inference: 73%|███████▎ | 4305/5920 [51:06<05:50, 4.61it/s, tok/s=15841.5]
Inference: 73%|███████▎ | 4305/5920 [51:06<05:50, 4.61it/s, tok/s=15840.8]
Inference: 73%|███████▎ | 4306/5920 [51:07<05:50, 4.61it/s, tok/s=15840.9]
Inference: 73%|███████▎ | 4307/5920 [51:07<05:50, 4.61it/s, tok/s=15841.6]
Inference: 73%|███████▎ | 4308/5920 [51:07<03:39, 7.34it/s, tok/s=15841.6]
Inference: 73%|███████▎ | 4308/5920 [51:07<03:39, 7.34it/s, tok/s=15842.0]
Inference: 73%|███████▎ | 4309/5920 [51:07<03:39, 7.34it/s, tok/s=15841.4]
Inference: 73%|███████▎ | 4310/5920 [51:07<03:29, 7.68it/s, tok/s=15841.4]
Inference: 73%|███████▎ | 4310/5920 [51:07<03:29, 7.68it/s, tok/s=15844.5]
Inference: 73%|███████▎ | 4311/5920 [51:07<03:29, 7.68it/s, tok/s=15844.4]
Inference: 73%|███████▎ | 4312/5920 [51:07<03:23, 7.91it/s, tok/s=15844.4]
Inference: 73%|███████▎ | 4312/5920 [51:07<03:23, 7.91it/s, tok/s=15844.8]
Inference: 73%|███████▎ | 4313/5920 [51:07<03:23, 7.91it/s, tok/s=15845.0]
Inference: 73%|███████▎ | 4314/5920 [51:07<03:58, 6.74it/s, tok/s=15845.0]
Inference: 73%|███████▎ | 4314/5920 [51:07<03:58, 6.74it/s, tok/s=15844.1]
Inference: 73%|███████▎ | 4315/5920 [51:08<03:58, 6.74it/s, tok/s=15845.6]
Inference: 73%|███████▎ | 4316/5920 [51:08<04:00, 6.66it/s, tok/s=15845.6]
Inference: 73%|███████▎ | 4316/5920 [51:08<04:00, 6.66it/s, tok/s=15845.2]
Inference: 73%|███████▎ | 4317/5920 [51:09<07:04, 3.78it/s, tok/s=15845.2]
Inference: 73%|███████▎ | 4317/5920 [51:09<07:04, 3.78it/s, tok/s=15841.7]
Inference: 73%|███████▎ | 4318/5920 [51:09<07:03, 3.78it/s, tok/s=15842.1]
Inference: 73%|███████▎ | 4319/5920 [51:09<06:37, 4.03it/s, tok/s=15842.1]
Inference: 73%|███████▎ | 4319/5920 [51:09<06:37, 4.03it/s, tok/s=15844.2]
Inference: 73%|███████▎ | 4320/5920 [51:09<06:37, 4.03it/s, tok/s=15846.3]
Inference: 73%|███████▎ | 4321/5920 [51:09<06:36, 4.03it/s, tok/s=15846.7]
Inference: 73%|███████▎ | 4322/5920 [51:09<04:39, 5.72it/s, tok/s=15846.7]
Inference: 73%|███████▎ | 4322/5920 [51:09<04:39, 5.72it/s, tok/s=15846.8]
Inference: 73%|███████▎ | 4323/5920 [51:09<04:39, 5.72it/s, tok/s=15848.6]
Inference: 73%|███████▎ | 4324/5920 [51:09<03:45, 7.07it/s, tok/s=15848.6]
Inference: 73%|███████▎ | 4324/5920 [51:09<03:45, 7.07it/s, tok/s=15850.9]
Inference: 73%|███████▎ | 4325/5920 [51:10<03:45, 7.07it/s, tok/s=15849.2]
Inference: 73%|███████▎ | 4326/5920 [51:10<04:27, 5.96it/s, tok/s=15849.2]
Inference: 73%|███████▎ | 4326/5920 [51:10<04:27, 5.96it/s, tok/s=15849.8]
Inference: 73%|███████▎ | 4327/5920 [51:10<04:27, 5.96it/s, tok/s=15850.6]
Inference: 73%|███████▎ | 4328/5920 [51:10<03:43, 7.13it/s, tok/s=15850.6]
Inference: 73%|███████▎ | 4328/5920 [51:10<03:43, 7.13it/s, tok/s=15852.6]
Inference: 73%|███████▎ | 4329/5920 [51:10<03:43, 7.13it/s, tok/s=15852.7]
Inference: 73%|███████▎ | 4330/5920 [51:11<07:12, 3.68it/s, tok/s=15852.7]
Inference: 73%|███████▎ | 4330/5920 [51:11<07:12, 3.68it/s, tok/s=15849.8]
Inference: 73%|███████▎ | 4331/5920 [51:11<07:12, 3.68it/s, tok/s=15849.9]
Inference: 73%|███████▎ | 4332/5920 [51:11<06:28, 4.08it/s, tok/s=15849.9]
Inference: 73%|███████▎ | 4332/5920 [51:11<06:28, 4.08it/s, tok/s=15849.3]
Inference: 73%|███████▎ | 4333/5920 [51:12<06:26, 4.11it/s, tok/s=15849.3]
Inference: 73%|███████▎ | 4333/5920 [51:12<06:26, 4.11it/s, tok/s=15849.7]
Inference: 73%|███████▎ | 4334/5920 [51:12<07:33, 3.50it/s, tok/s=15849.7]
Inference: 73%|███████▎ | 4334/5920 [51:12<07:33, 3.50it/s, tok/s=15848.2]
Inference: 73%|███████▎ | 4335/5920 [51:12<07:42, 3.43it/s, tok/s=15848.2]
Inference: 73%|███████▎ | 4335/5920 [51:12<07:42, 3.43it/s, tok/s=15847.0]
Inference: 73%|███████▎ | 4336/5920 [51:13<07:42, 3.43it/s, tok/s=15847.4]
Inference: 73%|███████▎ | 4337/5920 [51:13<07:44, 3.40it/s, tok/s=15847.4]
Inference: 73%|███████▎ | 4337/5920 [51:13<07:44, 3.40it/s, tok/s=15845.6]
Inference: 73%|███████▎ | 4338/5920 [51:13<07:44, 3.40it/s, tok/s=15845.8]
Inference: 73%|███████▎ | 4339/5920 [51:13<06:49, 3.86it/s, tok/s=15845.8]
Inference: 73%|███████▎ | 4339/5920 [51:13<06:49, 3.86it/s, tok/s=15845.3]
Inference: 73%|███████▎ | 4340/5920 [51:14<05:59, 4.40it/s, tok/s=15845.3]
Inference: 73%|███████▎ | 4340/5920 [51:14<05:59, 4.40it/s, tok/s=15846.2]
Inference: 73%|███████▎ | 4341/5920 [51:14<05:59, 4.40it/s, tok/s=15846.3]
Inference: 73%|███████▎ | 4342/5920 [51:14<04:45, 5.52it/s, tok/s=15846.3]
Inference: 73%|███████▎ | 4342/5920 [51:14<04:45, 5.52it/s, tok/s=15846.4]
Inference: 73%|███████▎ | 4343/5920 [51:14<06:49, 3.85it/s, tok/s=15846.4]
Inference: 73%|███████▎ | 4343/5920 [51:14<06:49, 3.85it/s, tok/s=15845.5]
Inference: 73%|███████▎ | 4344/5920 [51:15<08:32, 3.07it/s, tok/s=15845.5]
Inference: 73%|███████▎ | 4344/5920 [51:15<08:32, 3.07it/s, tok/s=15843.2]
Inference: 73%|███████▎ | 4345/5920 [51:15<07:07, 3.69it/s, tok/s=15843.2]
Inference: 73%|███████▎ | 4345/5920 [51:15<07:07, 3.69it/s, tok/s=15846.3]
Inference: 73%|███████▎ | 4346/5920 [51:16<09:40, 2.71it/s, tok/s=15846.3]
Inference: 73%|███████▎ | 4346/5920 [51:16<09:40, 2.71it/s, tok/s=15843.7]
Inference: 73%|███████▎ | 4347/5920 [51:16<09:40, 2.71it/s, tok/s=15844.1]
Inference: 73%|███████▎ | 4348/5920 [51:16<07:19, 3.58it/s, tok/s=15844.1]
Inference: 73%|███████▎ | 4348/5920 [51:16<07:19, 3.58it/s, tok/s=15844.1]
Inference: 73%|███████▎ | 4349/5920 [51:16<07:18, 3.58it/s, tok/s=15844.4]
Inference: 73%|███████▎ | 4350/5920 [51:16<05:21, 4.88it/s, tok/s=15844.4]
Inference: 73%|███████▎ | 4350/5920 [51:16<05:21, 4.88it/s, tok/s=15844.7]
Inference: 73%|███████▎ | 4351/5920 [51:16<05:21, 4.88it/s, tok/s=15844.7]
Inference: 73%|███████▎ | 4351/5920 [51:16<05:21, 4.88it/s, tok/s=15848.6]
Inference: 74%|███████▎ | 4352/5920 [51:16<05:21, 4.88it/s, tok/s=15854.0]
Inference: 74%|███████▎ | 4353/5920 [51:17<05:15, 4.96it/s, tok/s=15854.0]
Inference: 74%|███████▎ | 4353/5920 [51:17<05:15, 4.96it/s, tok/s=15853.2]
Inference: 74%|███████▎ | 4354/5920 [51:17<04:50, 5.40it/s, tok/s=15853.2]
Inference: 74%|███████▎ | 4354/5920 [51:17<04:50, 5.40it/s, tok/s=15854.4]
Inference: 74%|███████▎ | 4355/5920 [51:17<04:50, 5.40it/s, tok/s=15854.9]
Inference: 74%|███████▎ | 4356/5920 [51:17<03:56, 6.62it/s, tok/s=15854.9]
Inference: 74%|███████▎ | 4356/5920 [51:17<03:56, 6.62it/s, tok/s=15854.6]
Inference: 74%|███████▎ | 4357/5920 [51:18<06:00, 4.33it/s, tok/s=15854.6]
Inference: 74%|███████▎ | 4357/5920 [51:18<06:00, 4.33it/s, tok/s=15852.5]
Inference: 74%|███████▎ | 4358/5920 [51:18<07:06, 3.67it/s, tok/s=15852.5]
Inference: 74%|███████▎ | 4358/5920 [51:18<07:06, 3.67it/s, tok/s=15850.8]
Inference: 74%|███████▎ | 4359/5920 [51:18<07:05, 3.67it/s, tok/s=15855.7]
Inference: 74%|███████▎ | 4360/5920 [51:18<05:42, 4.56it/s, tok/s=15855.7]
Inference: 74%|███████▎ | 4360/5920 [51:18<05:42, 4.56it/s, tok/s=15860.0]
Inference: 74%|███████▎ | 4361/5920 [51:18<05:42, 4.56it/s, tok/s=15862.2]
Inference: 74%|███████▎ | 4362/5920 [51:18<04:10, 6.22it/s, tok/s=15862.2]
Inference: 74%|███████▎ | 4362/5920 [51:18<04:10, 6.22it/s, tok/s=15862.4]
Inference: 74%|███████▎ | 4363/5920 [51:18<03:56, 6.59it/s, tok/s=15862.4]
Inference: 74%|███████▎ | 4363/5920 [51:18<03:56, 6.59it/s, tok/s=15862.9]
Inference: 74%|███████▎ | 4364/5920 [51:19<04:43, 5.49it/s, tok/s=15862.9]
Inference: 74%|███████▎ | 4364/5920 [51:19<04:43, 5.49it/s, tok/s=15862.0]
Inference: 74%|███████▎ | 4365/5920 [51:19<04:15, 6.08it/s, tok/s=15862.0]
Inference: 74%|███████▎ | 4365/5920 [51:19<04:15, 6.08it/s, tok/s=15865.3]
Inference: 74%|███████▍ | 4366/5920 [51:19<04:15, 6.08it/s, tok/s=15867.8]
Inference: 74%|███████▍ | 4367/5920 [51:19<03:47, 6.83it/s, tok/s=15867.8]
Inference: 74%|███████▍ | 4367/5920 [51:19<03:47, 6.83it/s, tok/s=15867.6]
Inference: 74%|███████▍ | 4368/5920 [51:19<03:58, 6.50it/s, tok/s=15867.6]
Inference: 74%|███████▍ | 4368/5920 [51:19<03:58, 6.50it/s, tok/s=15867.3]
Inference: 74%|███████▍ | 4369/5920 [51:19<03:58, 6.50it/s, tok/s=15868.1]
Inference: 74%|███████▍ | 4370/5920 [51:19<03:58, 6.50it/s, tok/s=15868.7]
Inference: 74%|███████▍ | 4371/5920 [51:20<03:29, 7.38it/s, tok/s=15868.7]
Inference: 74%|███████▍ | 4371/5920 [51:20<03:29, 7.38it/s, tok/s=15868.6]
Inference: 74%|███████▍ | 4372/5920 [51:20<04:09, 6.21it/s, tok/s=15868.6]
Inference: 74%|███████▍ | 4372/5920 [51:20<04:09, 6.21it/s, tok/s=15870.8]
Inference: 74%|███████▍ | 4373/5920 [51:21<11:53, 2.17it/s, tok/s=15870.8]
Inference: 74%|███████▍ | 4373/5920 [51:21<11:53, 2.17it/s, tok/s=15872.3]
Inference: 74%|███████▍ | 4374/5920 [51:22<10:35, 2.43it/s, tok/s=15872.3]
Inference: 74%|███████▍ | 4374/5920 [51:22<10:35, 2.43it/s, tok/s=15871.7]
Inference: 74%|███████▍ | 4375/5920 [51:22<12:57, 1.99it/s, tok/s=15871.7]
Inference: 74%|███████▍ | 4375/5920 [51:22<12:57, 1.99it/s, tok/s=15868.2]
Inference: 74%|███████▍ | 4376/5920 [51:23<11:15, 2.29it/s, tok/s=15868.2]
Inference: 74%|███████▍ | 4376/5920 [51:23<11:15, 2.29it/s, tok/s=15867.9]
Inference: 74%|███████▍ | 4377/5920 [51:23<11:15, 2.29it/s, tok/s=15869.9]
Inference: 74%|███████▍ | 4378/5920 [51:23<06:59, 3.67it/s, tok/s=15869.9]
Inference: 74%|███████▍ | 4378/5920 [51:23<06:59, 3.67it/s, tok/s=15870.6]
Inference: 74%|███████▍ | 4379/5920 [51:23<06:06, 4.20it/s, tok/s=15870.6]
Inference: 74%|███████▍ | 4379/5920 [51:23<06:06, 4.20it/s, tok/s=15871.7]
Inference: 74%|███████▍ | 4380/5920 [51:23<05:24, 4.75it/s, tok/s=15871.7]
Inference: 74%|███████▍ | 4380/5920 [51:23<05:24, 4.75it/s, tok/s=15873.0]
Inference: 74%|███████▍ | 4381/5920 [51:23<06:20, 4.04it/s, tok/s=15873.0]
Inference: 74%|███████▍ | 4381/5920 [51:23<06:20, 4.04it/s, tok/s=15880.1]
Inference: 74%|███████▍ | 4382/5920 [51:24<08:01, 3.19it/s, tok/s=15880.1]
Inference: 74%|███████▍ | 4382/5920 [51:24<08:01, 3.19it/s, tok/s=15878.2]
Inference: 74%|███████▍ | 4383/5920 [51:25<10:45, 2.38it/s, tok/s=15878.2]
Inference: 74%|███████▍ | 4383/5920 [51:25<10:45, 2.38it/s, tok/s=15875.1]
Inference: 74%|███████▍ | 4384/5920 [51:25<10:17, 2.49it/s, tok/s=15875.1]
Inference: 74%|███████▍ | 4384/5920 [51:25<10:17, 2.49it/s, tok/s=15873.7]
Inference: 74%|███████▍ | 4385/5920 [51:25<08:51, 2.89it/s, tok/s=15873.7]
Inference: 74%|███████▍ | 4385/5920 [51:25<08:51, 2.89it/s, tok/s=15879.1]
Inference: 74%|███████▍ | 4386/5920 [51:25<08:50, 2.89it/s, tok/s=15881.2]
Inference: 74%|███████▍ | 4387/5920 [51:26<08:02, 3.18it/s, tok/s=15881.2]
Inference: 74%|███████▍ | 4387/5920 [51:26<08:02, 3.18it/s, tok/s=15879.1]
Inference: 74%|███████▍ | 4388/5920 [51:26<07:22, 3.46it/s, tok/s=15879.1]
Inference: 74%|███████▍ | 4388/5920 [51:26<07:22, 3.46it/s, tok/s=15879.8]
Inference: 74%|███████▍ | 4389/5920 [51:26<06:50, 3.73it/s, tok/s=15879.8]
Inference: 74%|███████▍ | 4389/5920 [51:26<06:50, 3.73it/s, tok/s=15882.6]
Inference: 74%|███████▍ | 4390/5920 [51:27<10:28, 2.43it/s, tok/s=15882.6]
Inference: 74%|███████▍ | 4390/5920 [51:27<10:28, 2.43it/s, tok/s=15880.9]
Inference: 74%|███████▍ | 4391/5920 [51:27<09:28, 2.69it/s, tok/s=15880.9]
Inference: 74%|███████▍ | 4391/5920 [51:27<09:28, 2.69it/s, tok/s=15880.2]
Inference: 74%|███████▍ | 4392/5920 [51:28<09:40, 2.63it/s, tok/s=15880.2]
Inference: 74%|███████▍ | 4392/5920 [51:28<09:40, 2.63it/s, tok/s=15879.2]
Inference: 74%|███████▍ | 4393/5920 [51:28<09:39, 2.63it/s, tok/s=15880.7]
Inference: 74%|███████▍ | 4394/5920 [51:28<06:19, 4.03it/s, tok/s=15880.7]
Inference: 74%|███████▍ | 4394/5920 [51:28<06:19, 4.03it/s, tok/s=15881.1]
Inference: 74%|███████▍ | 4395/5920 [51:28<05:57, 4.26it/s, tok/s=15881.1]
Inference: 74%|███████▍ | 4395/5920 [51:28<05:57, 4.26it/s, tok/s=15880.9]
Inference: 74%|███████▍ | 4396/5920 [51:28<05:57, 4.26it/s, tok/s=15881.2]
Inference: 74%|███████▍ | 4397/5920 [51:28<05:04, 5.00it/s, tok/s=15881.2]
Inference: 74%|███████▍ | 4397/5920 [51:28<05:04, 5.00it/s, tok/s=15881.5]
Inference: 74%|███████▍ | 4398/5920 [51:28<05:04, 5.00it/s, tok/s=15882.4]
Inference: 74%|███████▍ | 4399/5920 [51:29<05:33, 4.56it/s, tok/s=15882.4]
Inference: 74%|███████▍ | 4399/5920 [51:29<05:33, 4.56it/s, tok/s=15883.1]
Inference: 74%|███████▍ | 4400/5920 [51:29<05:33, 4.56it/s, tok/s=15883.4]
Inference: 74%|███████▍ | 4401/5920 [51:29<05:13, 4.85it/s, tok/s=15883.4]
Inference: 74%|███████▍ | 4401/5920 [51:29<05:13, 4.85it/s, tok/s=15882.6]
Inference: 74%|███████▍ | 4402/5920 [51:29<05:13, 4.85it/s, tok/s=15892.1]
Inference: 74%|███████▍ | 4403/5920 [51:29<05:13, 4.85it/s, tok/s=15892.9]
Inference: 74%|███████▍ | 4404/5920 [51:29<03:53, 6.50it/s, tok/s=15892.9]
Inference: 74%|███████▍ | 4404/5920 [51:29<03:53, 6.50it/s, tok/s=15892.7]
Inference: 74%|███████▍ | 4405/5920 [51:30<03:45, 6.72it/s, tok/s=15892.7]
Inference: 74%|███████▍ | 4405/5920 [51:30<03:45, 6.72it/s, tok/s=15892.4]
Inference: 74%|███████▍ | 4406/5920 [51:30<06:37, 3.81it/s, tok/s=15892.4]
Inference: 74%|███████▍ | 4406/5920 [51:30<06:37, 3.81it/s, tok/s=15889.2]
Inference: 74%|███████▍ | 4407/5920 [51:30<06:14, 4.04it/s, tok/s=15889.2]
Inference: 74%|███████▍ | 4407/5920 [51:30<06:14, 4.04it/s, tok/s=15889.2]
Inference: 74%|███████▍ | 4408/5920 [51:31<05:35, 4.51it/s, tok/s=15889.2]
Inference: 74%|███████▍ | 4408/5920 [51:31<05:35, 4.51it/s, tok/s=15889.4]
Inference: 74%|███████▍ | 4409/5920 [51:31<05:35, 4.51it/s, tok/s=15889.7]
Inference: 74%|███████▍ | 4410/5920 [51:31<04:05, 6.14it/s, tok/s=15889.7]
Inference: 74%|███████▍ | 4410/5920 [51:31<04:05, 6.14it/s, tok/s=15890.0]
Inference: 75%|███████▍ | 4411/5920 [51:31<05:00, 5.03it/s, tok/s=15890.0]
Inference: 75%|███████▍ | 4411/5920 [51:31<05:00, 5.03it/s, tok/s=15889.0]
Inference: 75%|███████▍ | 4412/5920 [51:32<09:15, 2.71it/s, tok/s=15889.0]
Inference: 75%|███████▍ | 4412/5920 [51:32<09:15, 2.71it/s, tok/s=15885.4]
Inference: 75%|███████▍ | 4413/5920 [51:32<09:15, 2.71it/s, tok/s=15886.6]
Inference: 75%|███████▍ | 4414/5920 [51:32<08:01, 3.13it/s, tok/s=15886.6]
Inference: 75%|███████▍ | 4414/5920 [51:32<08:01, 3.13it/s, tok/s=15885.5]
Inference: 75%|███████▍ | 4415/5920 [51:33<08:01, 3.13it/s, tok/s=15885.7]
Inference: 75%|███████▍ | 4416/5920 [51:33<05:41, 4.41it/s, tok/s=15885.7]
Inference: 75%|███████▍ | 4416/5920 [51:33<05:41, 4.41it/s, tok/s=15894.1]
Inference: 75%|███████▍ | 4417/5920 [51:33<07:37, 3.28it/s, tok/s=15894.1]
Inference: 75%|███████▍ | 4417/5920 [51:33<07:37, 3.28it/s, tok/s=15892.4]
Inference: 75%|███████▍ | 4418/5920 [51:34<09:47, 2.56it/s, tok/s=15892.4]
Inference: 75%|███████▍ | 4418/5920 [51:34<09:47, 2.56it/s, tok/s=15890.2]
Inference: 75%|███████▍ | 4419/5920 [51:34<08:54, 2.81it/s, tok/s=15890.2]
Inference: 75%|███████▍ | 4419/5920 [51:34<08:54, 2.81it/s, tok/s=15889.6]
Inference: 75%|███████▍ | 4420/5920 [51:34<08:53, 2.81it/s, tok/s=15890.9]
Inference: 75%|███████▍ | 4421/5920 [51:34<05:55, 4.22it/s, tok/s=15890.9]
Inference: 75%|███████▍ | 4421/5920 [51:34<05:55, 4.22it/s, tok/s=15891.6]
Inference: 75%|███████▍ | 4422/5920 [51:34<05:55, 4.22it/s, tok/s=15897.2]
Inference: 75%|███████▍ | 4423/5920 [51:34<04:34, 5.45it/s, tok/s=15897.2]
Inference: 75%|███████▍ | 4423/5920 [51:34<04:34, 5.45it/s, tok/s=15897.4]
Inference: 75%|███████▍ | 4424/5920 [51:35<05:03, 4.93it/s, tok/s=15897.4]
Inference: 75%|███████▍ | 4424/5920 [51:35<05:03, 4.93it/s, tok/s=15898.6]
Inference: 75%|███████▍ | 4425/5920 [51:35<05:03, 4.93it/s, tok/s=15898.8]
Inference: 75%|███████▍ | 4426/5920 [51:35<04:24, 5.64it/s, tok/s=15898.8]
Inference: 75%|███████▍ | 4426/5920 [51:35<04:24, 5.64it/s, tok/s=15907.6]
Inference: 75%|███████▍ | 4427/5920 [51:35<05:20, 4.65it/s, tok/s=15907.6]
Inference: 75%|███████▍ | 4427/5920 [51:35<05:20, 4.65it/s, tok/s=15908.1]
Inference: 75%|███████▍ | 4428/5920 [51:36<06:18, 3.94it/s, tok/s=15908.1]
Inference: 75%|███████▍ | 4428/5920 [51:36<06:18, 3.94it/s, tok/s=15907.3]
Inference: 75%|███████▍ | 4429/5920 [51:36<07:22, 3.37it/s, tok/s=15907.3]
Inference: 75%|███████▍ | 4429/5920 [51:36<07:22, 3.37it/s, tok/s=15910.7]
Inference: 75%|███████▍ | 4430/5920 [51:36<06:25, 3.86it/s, tok/s=15910.7]
Inference: 75%|███████▍ | 4430/5920 [51:36<06:25, 3.86it/s, tok/s=15911.0]
Inference: 75%|███████▍ | 4431/5920 [51:36<05:26, 4.56it/s, tok/s=15911.0]
Inference: 75%|███████▍ | 4431/5920 [51:36<05:26, 4.56it/s, tok/s=15911.1]
Inference: 75%|███████▍ | 4432/5920 [51:37<04:55, 5.04it/s, tok/s=15911.1]
Inference: 75%|███████▍ | 4432/5920 [51:37<04:55, 5.04it/s, tok/s=15911.4]
Inference: 75%|███████▍ | 4433/5920 [51:37<04:59, 4.96it/s, tok/s=15911.4]
Inference: 75%|███████▍ | 4433/5920 [51:37<04:59, 4.96it/s, tok/s=15910.8]
Inference: 75%|███████▍ | 4434/5920 [51:37<05:55, 4.18it/s, tok/s=15910.8]
Inference: 75%|███████▍ | 4434/5920 [51:37<05:55, 4.18it/s, tok/s=15909.9]
Inference: 75%|███████▍ | 4435/5920 [51:37<05:33, 4.46it/s, tok/s=15909.9]
Inference: 75%|███████▍ | 4435/5920 [51:37<05:33, 4.46it/s, tok/s=15910.0]
Inference: 75%|███████▍ | 4436/5920 [51:38<07:21, 3.36it/s, tok/s=15910.0]
Inference: 75%|███████▍ | 4436/5920 [51:38<07:21, 3.36it/s, tok/s=15908.2]
Inference: 75%|███████▍ | 4437/5920 [51:38<07:21, 3.36it/s, tok/s=15916.1]
Inference: 75%|███████▍ | 4438/5920 [51:38<04:48, 5.14it/s, tok/s=15916.1]
Inference: 75%|███████▍ | 4438/5920 [51:38<04:48, 5.14it/s, tok/s=15920.3]
Inference: 75%|███████▍ | 4439/5920 [51:38<05:05, 4.85it/s, tok/s=15920.3]
Inference: 75%|███████▍ | 4439/5920 [51:38<05:05, 4.85it/s, tok/s=15919.6]
Inference: 75%|███████▌ | 4440/5920 [51:38<04:48, 5.13it/s, tok/s=15919.6]
Inference: 75%|███████▌ | 4440/5920 [51:38<04:48, 5.13it/s, tok/s=15926.2]
Inference: 75%|███████▌ | 4441/5920 [51:39<07:03, 3.49it/s, tok/s=15926.2]
Inference: 75%|███████▌ | 4441/5920 [51:39<07:03, 3.49it/s, tok/s=15930.6]
Inference: 75%|███████▌ | 4442/5920 [51:39<07:36, 3.24it/s, tok/s=15930.6]
Inference: 75%|███████▌ | 4442/5920 [51:39<07:36, 3.24it/s, tok/s=15929.4]
Inference: 75%|███████▌ | 4443/5920 [51:39<07:36, 3.24it/s, tok/s=15930.0]
Inference: 75%|███████▌ | 4444/5920 [51:40<06:59, 3.52it/s, tok/s=15930.0]
Inference: 75%|███████▌ | 4444/5920 [51:40<06:59, 3.52it/s, tok/s=15931.4]
Inference: 75%|███████▌ | 4445/5920 [51:40<06:59, 3.52it/s, tok/s=15932.2]
Inference: 75%|███████▌ | 4446/5920 [51:40<06:59, 3.52it/s, tok/s=15932.9]
Inference: 75%|███████▌ | 4447/5920 [51:40<04:28, 5.49it/s, tok/s=15932.9]
Inference: 75%|███████▌ | 4447/5920 [51:40<04:28, 5.49it/s, tok/s=15933.5]
Inference: 75%|███████▌ | 4448/5920 [51:40<04:23, 5.58it/s, tok/s=15933.5]
Inference: 75%|███████▌ | 4448/5920 [51:40<04:23, 5.58it/s, tok/s=15933.1]
Inference: 75%|███████▌ | 4449/5920 [51:40<04:23, 5.58it/s, tok/s=15934.3]
Inference: 75%|███████▌ | 4450/5920 [51:40<04:23, 5.58it/s, tok/s=15934.6]
Inference: 75%|███████▌ | 4451/5920 [51:40<04:23, 5.58it/s, tok/s=15935.2]
Inference: 75%|███████▌ | 4452/5920 [51:40<02:42, 9.01it/s, tok/s=15935.2]
Inference: 75%|███████▌ | 4452/5920 [51:40<02:42, 9.01it/s, tok/s=15935.4]
Inference: 75%|███████▌ | 4453/5920 [51:41<02:42, 9.01it/s, tok/s=15937.3]
Inference: 75%|███████▌ | 4454/5920 [51:41<03:13, 7.57it/s, tok/s=15937.3]
Inference: 75%|███████▌ | 4454/5920 [51:41<03:13, 7.57it/s, tok/s=15938.4]
Inference: 75%|███████▌ | 4455/5920 [51:41<03:13, 7.57it/s, tok/s=15941.0]
Inference: 75%|███████▌ | 4456/5920 [51:41<04:15, 5.73it/s, tok/s=15941.0]
Inference: 75%|███████▌ | 4456/5920 [51:41<04:15, 5.73it/s, tok/s=15938.8]
Inference: 75%|███████▌ | 4457/5920 [51:41<04:15, 5.73it/s, tok/s=15941.8]
Inference: 75%|███████▌ | 4458/5920 [51:41<04:15, 5.73it/s, tok/s=15950.5]
Inference: 75%|███████▌ | 4459/5920 [51:42<03:36, 6.76it/s, tok/s=15950.5]
Inference: 75%|███████▌ | 4459/5920 [51:42<03:36, 6.76it/s, tok/s=15949.9]
Inference: 75%|███████▌ | 4460/5920 [51:42<04:25, 5.49it/s, tok/s=15949.9]
Inference: 75%|███████▌ | 4460/5920 [51:42<04:25, 5.49it/s, tok/s=15949.8]
Inference: 75%|███████▌ | 4461/5920 [51:42<05:32, 4.39it/s, tok/s=15949.8]
Inference: 75%|███████▌ | 4461/5920 [51:42<05:32, 4.39it/s, tok/s=15948.9]
Inference: 75%|███████▌ | 4462/5920 [51:43<04:58, 4.89it/s, tok/s=15948.9]
Inference: 75%|███████▌ | 4462/5920 [51:43<04:58, 4.89it/s, tok/s=15948.6]
Inference: 75%|███████▌ | 4463/5920 [51:43<04:43, 5.14it/s, tok/s=15948.6]
Inference: 75%|███████▌ | 4463/5920 [51:43<04:43, 5.14it/s, tok/s=15948.7]
Inference: 75%|███████▌ | 4464/5920 [51:43<07:06, 3.41it/s, tok/s=15948.7]
Inference: 75%|███████▌ | 4464/5920 [51:43<07:06, 3.41it/s, tok/s=15951.2]
Inference: 75%|███████▌ | 4465/5920 [51:43<07:06, 3.41it/s, tok/s=15952.4]
Inference: 75%|███████▌ | 4466/5920 [51:44<07:14, 3.35it/s, tok/s=15952.4]
Inference: 75%|███████▌ | 4466/5920 [51:44<07:14, 3.35it/s, tok/s=15950.1]
Inference: 75%|███████▌ | 4467/5920 [51:44<07:13, 3.35it/s, tok/s=15953.8]
Inference: 75%|███████▌ | 4468/5920 [51:44<05:20, 4.53it/s, tok/s=15953.8]
Inference: 75%|███████▌ | 4468/5920 [51:44<05:20, 4.53it/s, tok/s=15954.7]
Inference: 75%|███████▌ | 4469/5920 [51:44<05:20, 4.53it/s, tok/s=15955.4]
Inference: 76%|███████▌ | 4470/5920 [51:44<04:01, 6.01it/s, tok/s=15955.4]
Inference: 76%|███████▌ | 4470/5920 [51:44<04:01, 6.01it/s, tok/s=15956.9]
Inference: 76%|███████▌ | 4471/5920 [51:45<04:58, 4.85it/s, tok/s=15956.9]
Inference: 76%|███████▌ | 4471/5920 [51:45<04:58, 4.85it/s, tok/s=15956.3]
Inference: 76%|███████▌ | 4472/5920 [51:45<07:09, 3.37it/s, tok/s=15956.3]
Inference: 76%|███████▌ | 4472/5920 [51:45<07:09, 3.37it/s, tok/s=15953.7]
Inference: 76%|███████▌ | 4473/5920 [51:45<07:09, 3.37it/s, tok/s=15954.7]
Inference: 76%|███████▌ | 4474/5920 [51:45<05:25, 4.45it/s, tok/s=15954.7]
Inference: 76%|███████▌ | 4474/5920 [51:45<05:25, 4.45it/s, tok/s=15957.0]
Inference: 76%|███████▌ | 4475/5920 [51:45<05:25, 4.45it/s, tok/s=15958.3]
Inference: 76%|███████▌ | 4476/5920 [51:45<05:24, 4.45it/s, tok/s=15959.5]
Inference: 76%|███████▌ | 4477/5920 [51:46<03:21, 7.17it/s, tok/s=15959.5]
Inference: 76%|███████▌ | 4477/5920 [51:46<03:21, 7.17it/s, tok/s=15963.5]
Inference: 76%|███████▌ | 4478/5920 [51:46<03:21, 7.17it/s, tok/s=15964.2]
Inference: 76%|███████▌ | 4479/5920 [51:46<02:57, 8.10it/s, tok/s=15964.2]
Inference: 76%|███████▌ | 4479/5920 [51:46<02:57, 8.10it/s, tok/s=15965.0]
Inference: 76%|███████▌ | 4480/5920 [51:46<02:57, 8.10it/s, tok/s=15965.3]
Inference: 76%|███████▌ | 4481/5920 [51:46<03:18, 7.25it/s, tok/s=15965.3]
Inference: 76%|███████▌ | 4481/5920 [51:46<03:18, 7.25it/s, tok/s=15965.0]
Inference: 76%|███████▌ | 4482/5920 [51:46<03:18, 7.25it/s, tok/s=15968.4]
Inference: 76%|███████▌ | 4483/5920 [51:46<02:41, 8.89it/s, tok/s=15968.4]
Inference: 76%|███████▌ | 4483/5920 [51:46<02:41, 8.89it/s, tok/s=15975.6]
Inference: 76%|███████▌ | 4484/5920 [51:46<02:41, 8.89it/s, tok/s=15978.2]
Inference: 76%|███████▌ | 4485/5920 [51:47<03:16, 7.29it/s, tok/s=15978.2]
Inference: 76%|███████▌ | 4485/5920 [51:47<03:16, 7.29it/s, tok/s=15978.6]
Inference: 76%|███████▌ | 4486/5920 [51:47<03:16, 7.29it/s, tok/s=15978.1]
Inference: 76%|███████▌ | 4487/5920 [51:47<03:57, 6.02it/s, tok/s=15978.1]
Inference: 76%|███████▌ | 4487/5920 [51:47<03:57, 6.02it/s, tok/s=15978.2]
Inference: 76%|███████▌ | 4488/5920 [51:48<05:57, 4.00it/s, tok/s=15978.2]
Inference: 76%|███████▌ | 4488/5920 [51:48<05:57, 4.00it/s, tok/s=15976.4]
Inference: 76%|███████▌ | 4489/5920 [51:48<07:25, 3.21it/s, tok/s=15976.4]
Inference: 76%|███████▌ | 4489/5920 [51:48<07:25, 3.21it/s, tok/s=15974.9]
Inference: 76%|███████▌ | 4490/5920 [51:48<06:37, 3.60it/s, tok/s=15974.9]
Inference: 76%|███████▌ | 4490/5920 [51:48<06:37, 3.60it/s, tok/s=15978.9]
Inference: 76%|███████▌ | 4491/5920 [51:48<06:37, 3.60it/s, tok/s=15979.9]
Inference: 76%|███████▌ | 4492/5920 [51:49<05:49, 4.08it/s, tok/s=15979.9]
Inference: 76%|███████▌ | 4492/5920 [51:49<05:49, 4.08it/s, tok/s=15978.6]
Inference: 76%|███████▌ | 4493/5920 [51:49<05:49, 4.08it/s, tok/s=15979.2]
Inference: 76%|███████▌ | 4494/5920 [51:49<05:02, 4.72it/s, tok/s=15979.2]
Inference: 76%|███████▌ | 4494/5920 [51:49<05:02, 4.72it/s, tok/s=15978.3]
Inference: 76%|███████▌ | 4495/5920 [51:49<05:01, 4.72it/s, tok/s=15980.1]
Inference: 76%|███████▌ | 4496/5920 [51:50<06:29, 3.65it/s, tok/s=15980.1]
Inference: 76%|███████▌ | 4496/5920 [51:50<06:29, 3.65it/s, tok/s=15978.7]
Inference: 76%|███████▌ | 4497/5920 [51:50<06:08, 3.86it/s, tok/s=15978.7]
Inference: 76%|███████▌ | 4497/5920 [51:50<06:08, 3.86it/s, tok/s=15978.6]
Inference: 76%|███████▌ | 4498/5920 [51:50<05:50, 4.06it/s, tok/s=15978.6]
Inference: 76%|███████▌ | 4498/5920 [51:50<05:50, 4.06it/s, tok/s=15979.6]
Inference: 76%|███████▌ | 4499/5920 [51:50<05:50, 4.06it/s, tok/s=15980.0]
Inference: 76%|███████▌ | 4500/5920 [51:50<04:20, 5.44it/s, tok/s=15980.0]
Inference: 76%|███████▌ | 4500/5920 [51:50<04:20, 5.44it/s, tok/s=15983.4]
Inference: 76%|███████▌ | 4501/5920 [51:51<06:54, 3.42it/s, tok/s=15983.4]
Inference: 76%|███████▌ | 4501/5920 [51:51<06:54, 3.42it/s, tok/s=15981.5]
Inference: 76%|███████▌ | 4502/5920 [51:52<07:40, 3.08it/s, tok/s=15981.5]
Inference: 76%|███████▌ | 4502/5920 [51:52<07:40, 3.08it/s, tok/s=15980.1]
Inference: 76%|███████▌ | 4503/5920 [51:52<07:45, 3.04it/s, tok/s=15980.1]
Inference: 76%|███████▌ | 4503/5920 [51:52<07:45, 3.04it/s, tok/s=15979.1]
Inference: 76%|███████▌ | 4504/5920 [51:52<07:51, 3.00it/s, tok/s=15979.1]
Inference: 76%|███████▌ | 4504/5920 [51:52<07:51, 3.00it/s, tok/s=15978.2]
Inference: 76%|███████▌ | 4505/5920 [51:52<06:21, 3.71it/s, tok/s=15978.2]
Inference: 76%|███████▌ | 4505/5920 [51:52<06:21, 3.71it/s, tok/s=15985.7]
Inference: 76%|███████▌ | 4506/5920 [51:53<07:23, 3.19it/s, tok/s=15985.7]
Inference: 76%|███████▌ | 4506/5920 [51:53<07:23, 3.19it/s, tok/s=15984.4]
Inference: 76%|███████▌ | 4507/5920 [51:53<08:23, 2.81it/s, tok/s=15984.4]
Inference: 76%|███████▌ | 4507/5920 [51:53<08:23, 2.81it/s, tok/s=15983.4]
Inference: 76%|███████▌ | 4508/5920 [51:54<10:25, 2.26it/s, tok/s=15983.4]
Inference: 76%|███████▌ | 4508/5920 [51:54<10:25, 2.26it/s, tok/s=15982.1]
Inference: 76%|███████▌ | 4509/5920 [51:54<10:25, 2.26it/s, tok/s=15983.7]
Inference: 76%|███████▌ | 4510/5920 [51:54<06:54, 3.40it/s, tok/s=15983.7]
Inference: 76%|███████▌ | 4510/5920 [51:54<06:54, 3.40it/s, tok/s=15983.9]
Inference: 76%|███████▌ | 4511/5920 [51:54<07:10, 3.28it/s, tok/s=15983.9]
Inference: 76%|███████▌ | 4511/5920 [51:54<07:10, 3.28it/s, tok/s=15983.2]
Inference: 76%|███████▌ | 4512/5920 [51:55<10:46, 2.18it/s, tok/s=15983.2]
Inference: 76%|███████▌ | 4512/5920 [51:55<10:46, 2.18it/s, tok/s=15980.5]
Inference: 76%|███████▌ | 4513/5920 [51:56<10:35, 2.21it/s, tok/s=15980.5]
Inference: 76%|███████▌ | 4513/5920 [51:56<10:35, 2.21it/s, tok/s=15979.4]
Inference: 76%|███████▋ | 4514/5920 [51:56<10:05, 2.32it/s, tok/s=15979.4]
Inference: 76%|███████▋ | 4514/5920 [51:56<10:05, 2.32it/s, tok/s=15978.3]
Inference: 76%|███████▋ | 4515/5920 [51:56<09:41, 2.42it/s, tok/s=15978.3]
Inference: 76%|███████▋ | 4515/5920 [51:56<09:41, 2.42it/s, tok/s=15983.6]
Inference: 76%|███████▋ | 4516/5920 [51:57<09:41, 2.42it/s, tok/s=15984.9]
Inference: 76%|███████▋ | 4517/5920 [51:57<06:03, 3.86it/s, tok/s=15984.9]
Inference: 76%|███████▋ | 4517/5920 [51:57<06:03, 3.86it/s, tok/s=15986.1]
Inference: 76%|███████▋ | 4518/5920 [51:57<05:50, 4.00it/s, tok/s=15986.1]
Inference: 76%|███████▋ | 4518/5920 [51:57<05:50, 4.00it/s, tok/s=15985.9]
Inference: 76%|███████▋ | 4519/5920 [51:57<05:50, 4.00it/s, tok/s=15988.1]
Inference: 76%|███████▋ | 4520/5920 [51:57<05:02, 4.63it/s, tok/s=15988.1]
Inference: 76%|███████▋ | 4520/5920 [51:57<05:02, 4.63it/s, tok/s=15987.0]
Inference: 76%|███████▋ | 4521/5920 [51:58<06:46, 3.44it/s, tok/s=15987.0]
Inference: 76%|███████▋ | 4521/5920 [51:58<06:46, 3.44it/s, tok/s=15986.1]
Inference: 76%|███████▋ | 4522/5920 [51:58<07:58, 2.92it/s, tok/s=15986.1]
Inference: 76%|███████▋ | 4522/5920 [51:58<07:58, 2.92it/s, tok/s=15985.2]
Inference: 76%|███████▋ | 4523/5920 [51:59<08:18, 2.80it/s, tok/s=15985.2]
Inference: 76%|███████▋ | 4523/5920 [51:59<08:18, 2.80it/s, tok/s=15984.9]
Inference: 76%|███████▋ | 4524/5920 [51:59<08:18, 2.80it/s, tok/s=15985.2]
Inference: 76%|███████▋ | 4525/5920 [51:59<07:30, 3.09it/s, tok/s=15985.2]
Inference: 76%|███████▋ | 4525/5920 [51:59<07:30, 3.09it/s, tok/s=15984.3]
Inference: 76%|███████▋ | 4526/5920 [51:59<07:30, 3.09it/s, tok/s=15988.5]
Inference: 76%|███████▋ | 4527/5920 [51:59<07:30, 3.09it/s, tok/s=15991.1]
Inference: 76%|███████▋ | 4528/5920 [52:00<05:44, 4.04it/s, tok/s=15991.1]
Inference: 76%|███████▋ | 4528/5920 [52:00<05:44, 4.04it/s, tok/s=15991.8]
Inference: 77%|███████▋ | 4529/5920 [52:00<06:13, 3.72it/s, tok/s=15991.8]
Inference: 77%|███████▋ | 4529/5920 [52:00<06:13, 3.72it/s, tok/s=15991.8]
Inference: 77%|███████▋ | 4530/5920 [52:00<05:28, 4.24it/s, tok/s=15991.8]
Inference: 77%|███████▋ | 4530/5920 [52:00<05:28, 4.24it/s, tok/s=15992.2]
Inference: 77%|███████▋ | 4531/5920 [52:00<04:55, 4.70it/s, tok/s=15992.2]
Inference: 77%|███████▋ | 4531/5920 [52:00<04:55, 4.70it/s, tok/s=15992.3]
Inference: 77%|███████▋ | 4532/5920 [52:01<09:00, 2.57it/s, tok/s=15992.3]
Inference: 77%|███████▋ | 4532/5920 [52:01<09:00, 2.57it/s, tok/s=15989.3]
Inference: 77%|███████▋ | 4533/5920 [52:02<13:01, 1.77it/s, tok/s=15989.3]
Inference: 77%|███████▋ | 4533/5920 [52:02<13:01, 1.77it/s, tok/s=15985.0]
Inference: 77%|███████▋ | 4534/5920 [52:03<13:26, 1.72it/s, tok/s=15985.0]
Inference: 77%|███████▋ | 4534/5920 [52:03<13:26, 1.72it/s, tok/s=15983.7]
Inference: 77%|███████▋ | 4535/5920 [52:03<11:30, 2.01it/s, tok/s=15983.7]
Inference: 77%|███████▋ | 4535/5920 [52:03<11:30, 2.01it/s, tok/s=15983.2]
Inference: 77%|███████▋ | 4536/5920 [52:04<10:56, 2.11it/s, tok/s=15983.2]
Inference: 77%|███████▋ | 4536/5920 [52:04<10:56, 2.11it/s, tok/s=15982.3]
Inference: 77%|███████▋ | 4537/5920 [52:04<10:56, 2.11it/s, tok/s=15982.6]
Inference: 77%|███████▋ | 4538/5920 [52:04<08:26, 2.73it/s, tok/s=15982.6]
Inference: 77%|███████▋ | 4538/5920 [52:04<08:26, 2.73it/s, tok/s=15986.8]
Inference: 77%|███████▋ | 4539/5920 [52:04<08:26, 2.73it/s, tok/s=15987.3]
Inference: 77%|███████▋ | 4540/5920 [52:04<08:26, 2.73it/s, tok/s=15988.4]
Inference: 77%|███████▋ | 4541/5920 [52:05<06:54, 3.33it/s, tok/s=15988.4]
Inference: 77%|███████▋ | 4541/5920 [52:05<06:54, 3.33it/s, tok/s=15987.5]
Inference: 77%|███████▋ | 4542/5920 [52:05<07:36, 3.02it/s, tok/s=15987.5]
Inference: 77%|███████▋ | 4542/5920 [52:05<07:36, 3.02it/s, tok/s=15989.2]
Inference: 77%|███████▋ | 4543/5920 [52:06<08:04, 2.84it/s, tok/s=15989.2]
Inference: 77%|███████▋ | 4543/5920 [52:06<08:04, 2.84it/s, tok/s=15988.3]
Inference: 77%|███████▋ | 4544/5920 [52:06<08:04, 2.84it/s, tok/s=15989.4]
Inference: 77%|███████▋ | 4545/5920 [52:06<08:04, 2.84it/s, tok/s=15989.9]
Inference: 77%|███████▋ | 4546/5920 [52:06<04:59, 4.59it/s, tok/s=15989.9]
Inference: 77%|███████▋ | 4546/5920 [52:06<04:59, 4.59it/s, tok/s=15992.6]
Inference: 77%|███████▋ | 4547/5920 [52:06<04:51, 4.71it/s, tok/s=15992.6]
Inference: 77%|███████▋ | 4547/5920 [52:06<04:51, 4.71it/s, tok/s=15992.2]
Inference: 77%|███████▋ | 4548/5920 [52:06<04:51, 4.71it/s, tok/s=15996.5]
Inference: 77%|███████▋ | 4549/5920 [52:06<04:24, 5.18it/s, tok/s=15996.5]
Inference: 77%|███████▋ | 4549/5920 [52:06<04:24, 5.18it/s, tok/s=15996.2]
Inference: 77%|███████▋ | 4550/5920 [52:06<04:24, 5.18it/s, tok/s=15997.9]
Inference: 77%|███████▋ | 4551/5920 [52:07<03:59, 5.70it/s, tok/s=15997.9]
Inference: 77%|███████▋ | 4551/5920 [52:07<03:59, 5.70it/s, tok/s=15997.4]
Inference: 77%|███████▋ | 4552/5920 [52:07<03:54, 5.84it/s, tok/s=15997.4]
Inference: 77%|███████▋ | 4552/5920 [52:07<03:54, 5.84it/s, tok/s=15997.2]
Inference: 77%|███████▋ | 4553/5920 [52:07<04:45, 4.79it/s, tok/s=15997.2]
Inference: 77%|███████▋ | 4553/5920 [52:07<04:45, 4.79it/s, tok/s=15996.5]
Inference: 77%|███████▋ | 4554/5920 [52:07<04:12, 5.42it/s, tok/s=15996.5]
Inference: 77%|███████▋ | 4554/5920 [52:07<04:12, 5.42it/s, tok/s=15998.5]
Inference: 77%|███████▋ | 4555/5920 [52:07<04:16, 5.32it/s, tok/s=15998.5]
Inference: 77%|███████▋ | 4555/5920 [52:07<04:16, 5.32it/s, tok/s=16006.3]
Inference: 77%|███████▋ | 4556/5920 [52:08<06:08, 3.70it/s, tok/s=16006.3]
Inference: 77%|███████▋ | 4556/5920 [52:08<06:08, 3.70it/s, tok/s=16005.4]
Inference: 77%|███████▋ | 4557/5920 [52:09<12:43, 1.78it/s, tok/s=16005.4]
Inference: 77%|███████▋ | 4557/5920 [52:09<12:43, 1.78it/s, tok/s=15999.8]
Inference: 77%|███████▋ | 4558/5920 [52:09<10:21, 2.19it/s, tok/s=15999.8]
Inference: 77%|███████▋ | 4558/5920 [52:09<10:21, 2.19it/s, tok/s=15999.5]
Inference: 77%|███████▋ | 4559/5920 [52:10<10:38, 2.13it/s, tok/s=15999.5]
Inference: 77%|███████▋ | 4559/5920 [52:10<10:38, 2.13it/s, tok/s=15998.0]
Inference: 77%|███████▋ | 4560/5920 [52:10<09:10, 2.47it/s, tok/s=15998.0]
Inference: 77%|███████▋ | 4560/5920 [52:10<09:10, 2.47it/s, tok/s=15998.2]
Inference: 77%|███████▋ | 4561/5920 [52:11<09:42, 2.33it/s, tok/s=15998.2]
Inference: 77%|███████▋ | 4561/5920 [52:11<09:42, 2.33it/s, tok/s=16003.7]
Inference: 77%|███████▋ | 4562/5920 [52:11<07:36, 2.97it/s, tok/s=16003.7]
Inference: 77%|███████▋ | 4562/5920 [52:11<07:36, 2.97it/s, tok/s=16012.7]
Inference: 77%|███████▋ | 4563/5920 [52:11<08:25, 2.68it/s, tok/s=16012.7]
Inference: 77%|███████▋ | 4563/5920 [52:11<08:25, 2.68it/s, tok/s=16014.6]
Inference: 77%|███████▋ | 4564/5920 [52:11<06:56, 3.26it/s, tok/s=16014.6]
Inference: 77%|███████▋ | 4564/5920 [52:11<06:56, 3.26it/s, tok/s=16014.5]
Inference: 77%|███████▋ | 4565/5920 [52:12<08:55, 2.53it/s, tok/s=16014.5]
Inference: 77%|███████▋ | 4565/5920 [52:12<08:55, 2.53it/s, tok/s=16013.8]
Inference: 77%|███████▋ | 4566/5920 [52:12<08:32, 2.64it/s, tok/s=16013.8]
Inference: 77%|███████▋ | 4566/5920 [52:12<08:32, 2.64it/s, tok/s=16021.5]
Inference: 77%|███████▋ | 4567/5920 [52:13<08:08, 2.77it/s, tok/s=16021.5]
Inference: 77%|███████▋ | 4567/5920 [52:13<08:08, 2.77it/s, tok/s=16020.6]
Inference: 77%|███████▋ | 4568/5920 [52:13<08:07, 2.77it/s, tok/s=16020.7]
Inference: 77%|███████▋ | 4569/5920 [52:14<09:29, 2.37it/s, tok/s=16020.7]
Inference: 77%|███████▋ | 4569/5920 [52:14<09:29, 2.37it/s, tok/s=16025.6]
Inference: 77%|███████▋ | 4570/5920 [52:14<09:29, 2.37it/s, tok/s=16026.5]
Inference: 77%|███████▋ | 4571/5920 [52:14<06:23, 3.52it/s, tok/s=16026.5]
Inference: 77%|███████▋ | 4571/5920 [52:14<06:23, 3.52it/s, tok/s=16027.0]
Inference: 77%|███████▋ | 4572/5920 [52:14<06:25, 3.50it/s, tok/s=16027.0]
Inference: 77%|███████▋ | 4572/5920 [52:14<06:25, 3.50it/s, tok/s=16026.4]
Inference: 77%|███████▋ | 4573/5920 [52:14<05:33, 4.03it/s, tok/s=16026.4]
Inference: 77%|███████▋ | 4573/5920 [52:14<05:33, 4.03it/s, tok/s=16026.9]
Inference: 77%|███████▋ | 4574/5920 [52:15<05:56, 3.78it/s, tok/s=16026.9]
Inference: 77%|███████▋ | 4574/5920 [52:15<05:56, 3.78it/s, tok/s=16026.7]
Inference: 77%|███████▋ | 4575/5920 [52:15<08:57, 2.50it/s, tok/s=16026.7]
Inference: 77%|███████▋ | 4575/5920 [52:15<08:57, 2.50it/s, tok/s=16025.8]
Inference: 77%|███████▋ | 4576/5920 [52:16<08:07, 2.76it/s, tok/s=16025.8]
Inference: 77%|███████▋ | 4576/5920 [52:16<08:07, 2.76it/s, tok/s=16027.5]
Inference: 77%|███████▋ | 4577/5920 [52:16<09:54, 2.26it/s, tok/s=16027.5]
Inference: 77%|███████▋ | 4577/5920 [52:16<09:54, 2.26it/s, tok/s=16024.9]
Inference: 77%|███████▋ | 4578/5920 [52:17<08:54, 2.51it/s, tok/s=16024.9]
Inference: 77%|███████▋ | 4578/5920 [52:17<08:54, 2.51it/s, tok/s=16028.0]
Inference: 77%|███████▋ | 4579/5920 [52:17<08:55, 2.50it/s, tok/s=16028.0]
Inference: 77%|███████▋ | 4579/5920 [52:17<08:55, 2.50it/s, tok/s=16029.2]
Inference: 77%|███████▋ | 4580/5920 [52:18<15:02, 1.49it/s, tok/s=16029.2]
Inference: 77%|███████▋ | 4580/5920 [52:18<15:02, 1.49it/s, tok/s=16023.7]
Inference: 77%|███████▋ | 4581/5920 [52:18<15:01, 1.49it/s, tok/s=16024.3]
Inference: 77%|███████▋ | 4582/5920 [52:19<10:04, 2.21it/s, tok/s=16024.3]
Inference: 77%|███████▋ | 4582/5920 [52:19<10:04, 2.21it/s, tok/s=16024.4]
Inference: 77%|███████▋ | 4583/5920 [52:19<08:13, 2.71it/s, tok/s=16024.4]
Inference: 77%|███████▋ | 4583/5920 [52:19<08:13, 2.71it/s, tok/s=16025.3]
Inference: 77%|███████▋ | 4584/5920 [52:19<06:56, 3.21it/s, tok/s=16025.3]
Inference: 77%|███████▋ | 4584/5920 [52:19<06:56, 3.21it/s, tok/s=16025.7]
Inference: 77%|███████▋ | 4585/5920 [52:19<06:55, 3.21it/s, tok/s=16026.3]
Inference: 77%|███████▋ | 4586/5920 [52:19<04:52, 4.57it/s, tok/s=16026.3]
Inference: 77%|███████▋ | 4586/5920 [52:19<04:52, 4.57it/s, tok/s=16035.0]
Inference: 77%|███████▋ | 4587/5920 [52:19<04:27, 4.98it/s, tok/s=16035.0]
Inference: 77%|███████▋ | 4587/5920 [52:19<04:27, 4.98it/s, tok/s=16036.0]
Inference: 78%|███████▊ | 4588/5920 [52:19<04:27, 4.98it/s, tok/s=16039.2]
Inference: 78%|███████▊ | 4589/5920 [52:19<04:27, 4.98it/s, tok/s=16041.0]
Inference: 78%|███████▊ | 4590/5920 [52:20<04:00, 5.53it/s, tok/s=16041.0]
Inference: 78%|███████▊ | 4590/5920 [52:20<04:00, 5.53it/s, tok/s=16041.9]
Inference: 78%|███████▊ | 4591/5920 [52:20<04:00, 5.53it/s, tok/s=16042.8]
Inference: 78%|███████▊ | 4592/5920 [52:20<03:27, 6.40it/s, tok/s=16042.8]
Inference: 78%|███████▊ | 4592/5920 [52:20<03:27, 6.40it/s, tok/s=16043.4]
Inference: 78%|███████▊ | 4593/5920 [52:20<03:27, 6.40it/s, tok/s=16044.1]
Inference: 78%|███████▊ | 4594/5920 [52:20<03:06, 7.13it/s, tok/s=16044.1]
Inference: 78%|███████▊ | 4594/5920 [52:20<03:06, 7.13it/s, tok/s=16046.1]
Inference: 78%|███████▊ | 4595/5920 [52:21<06:21, 3.47it/s, tok/s=16046.1]
Inference: 78%|███████▊ | 4595/5920 [52:21<06:21, 3.47it/s, tok/s=16041.8]
Inference: 78%|███████▊ | 4596/5920 [52:21<06:51, 3.22it/s, tok/s=16041.8]
Inference: 78%|███████▊ | 4596/5920 [52:21<06:51, 3.22it/s, tok/s=16044.3]
Inference: 78%|███████▊ | 4597/5920 [52:22<06:35, 3.35it/s, tok/s=16044.3]
Inference: 78%|███████▊ | 4597/5920 [52:22<06:35, 3.35it/s, tok/s=16044.1]
Inference: 78%|███████▊ | 4598/5920 [52:22<07:30, 2.93it/s, tok/s=16044.1]
Inference: 78%|███████▊ | 4598/5920 [52:22<07:30, 2.93it/s, tok/s=16042.9]
Inference: 78%|███████▊ | 4599/5920 [52:22<06:11, 3.55it/s, tok/s=16042.9]
Inference: 78%|███████▊ | 4599/5920 [52:22<06:11, 3.55it/s, tok/s=16043.3]
Inference: 78%|███████▊ | 4600/5920 [52:23<06:01, 3.65it/s, tok/s=16043.3]
Inference: 78%|███████▊ | 4600/5920 [52:23<06:01, 3.65it/s, tok/s=16042.4]
Inference: 78%|███████▊ | 4601/5920 [52:23<05:35, 3.93it/s, tok/s=16042.4]
Inference: 78%|███████▊ | 4601/5920 [52:23<05:35, 3.93it/s, tok/s=16042.3]
Inference: 78%|███████▊ | 4602/5920 [52:23<07:21, 2.98it/s, tok/s=16042.3]
Inference: 78%|███████▊ | 4602/5920 [52:23<07:21, 2.98it/s, tok/s=16041.7]
Inference: 78%|███████▊ | 4603/5920 [52:23<07:21, 2.98it/s, tok/s=16041.9]
Inference: 78%|███████▊ | 4604/5920 [52:23<04:56, 4.45it/s, tok/s=16041.9]
Inference: 78%|███████▊ | 4604/5920 [52:23<04:56, 4.45it/s, tok/s=16042.1]
Inference: 78%|███████▊ | 4605/5920 [52:24<06:16, 3.49it/s, tok/s=16042.1]
Inference: 78%|███████▊ | 4605/5920 [52:24<06:16, 3.49it/s, tok/s=16040.5]
Inference: 78%|███████▊ | 4606/5920 [52:24<05:34, 3.93it/s, tok/s=16040.5]
Inference: 78%|███████▊ | 4606/5920 [52:24<05:34, 3.93it/s, tok/s=16040.1]
Inference: 78%|███████▊ | 4607/5920 [52:24<05:10, 4.23it/s, tok/s=16040.1]
Inference: 78%|███████▊ | 4607/5920 [52:24<05:10, 4.23it/s, tok/s=16041.4]
Inference: 78%|███████▊ | 4608/5920 [52:25<04:49, 4.53it/s, tok/s=16041.4]
Inference: 78%|███████▊ | 4608/5920 [52:25<04:49, 4.53it/s, tok/s=16041.6]
Inference: 78%|███████▊ | 4609/5920 [52:25<04:35, 4.75it/s, tok/s=16041.6]
Inference: 78%|███████▊ | 4609/5920 [52:25<04:35, 4.75it/s, tok/s=16042.2]
Inference: 78%|███████▊ | 4610/5920 [52:25<04:35, 4.75it/s, tok/s=16044.0]
Inference: 78%|███████▊ | 4611/5920 [52:26<06:57, 3.14it/s, tok/s=16044.0]
Inference: 78%|███████▊ | 4611/5920 [52:26<06:57, 3.14it/s, tok/s=16040.0]
Inference: 78%|███████▊ | 4612/5920 [52:27<11:45, 1.85it/s, tok/s=16040.0]
Inference: 78%|███████▊ | 4612/5920 [52:27<11:45, 1.85it/s, tok/s=16035.4]
Inference: 78%|███████▊ | 4613/5920 [52:27<11:45, 1.85it/s, tok/s=16039.0]
Inference: 78%|███████▊ | 4614/5920 [52:27<07:26, 2.92it/s, tok/s=16039.0]
Inference: 78%|███████▊ | 4614/5920 [52:27<07:26, 2.92it/s, tok/s=16039.4]
Inference: 78%|███████▊ | 4615/5920 [52:28<10:38, 2.04it/s, tok/s=16039.4]
Inference: 78%|███████▊ | 4615/5920 [52:28<10:38, 2.04it/s, tok/s=16035.1]
Inference: 78%|███████▊ | 4616/5920 [52:30<17:50, 1.22it/s, tok/s=16035.1]
Inference: 78%|███████▊ | 4616/5920 [52:30<17:50, 1.22it/s, tok/s=16027.0]
Inference: 78%|███████▊ | 4617/5920 [52:30<17:04, 1.27it/s, tok/s=16027.0]
Inference: 78%|███████▊ | 4617/5920 [52:30<17:04, 1.27it/s, tok/s=16024.4]
Inference: 78%|███████▊ | 4618/5920 [52:31<14:00, 1.55it/s, tok/s=16024.4]
Inference: 78%|███████▊ | 4618/5920 [52:31<14:00, 1.55it/s, tok/s=16024.0]
Inference: 78%|███████▊ | 4619/5920 [52:31<13:59, 1.55it/s, tok/s=16024.4]
Inference: 78%|███████▊ | 4620/5920 [52:31<10:12, 2.12it/s, tok/s=16024.4]
Inference: 78%|███████▊ | 4620/5920 [52:31<10:12, 2.12it/s, tok/s=16027.0]
Inference: 78%|███████▊ | 4621/5920 [52:31<08:29, 2.55it/s, tok/s=16027.0]
Inference: 78%|███████▊ | 4621/5920 [52:31<08:29, 2.55it/s, tok/s=16026.7]
Inference: 78%|███████▊ | 4622/5920 [52:32<07:39, 2.82it/s, tok/s=16026.7]
Inference: 78%|███████▊ | 4622/5920 [52:32<07:39, 2.82it/s, tok/s=16027.4]
Inference: 78%|███████▊ | 4623/5920 [52:32<07:39, 2.82it/s, tok/s=16027.8]
Inference: 78%|███████▊ | 4624/5920 [52:32<07:34, 2.85it/s, tok/s=16027.8]
Inference: 78%|███████▊ | 4624/5920 [52:32<07:34, 2.85it/s, tok/s=16026.0]
Inference: 78%|███████▊ | 4625/5920 [52:33<06:58, 3.09it/s, tok/s=16026.0]
Inference: 78%|███████▊ | 4625/5920 [52:33<06:58, 3.09it/s, tok/s=16025.9]
Inference: 78%|███████▊ | 4626/5920 [52:33<06:58, 3.09it/s, tok/s=16027.8]
Inference: 78%|███████▊ | 4627/5920 [52:33<06:58, 3.09it/s, tok/s=16029.1]
Inference: 78%|███████▊ | 4628/5920 [52:33<05:27, 3.95it/s, tok/s=16029.1]
Inference: 78%|███████▊ | 4628/5920 [52:33<05:27, 3.95it/s, tok/s=16028.3]
Inference: 78%|███████▊ | 4629/5920 [52:33<05:27, 3.95it/s, tok/s=16028.6]
Inference: 78%|███████▊ | 4630/5920 [52:33<05:26, 3.95it/s, tok/s=16035.2]
Inference: 78%|███████▊ | 4631/5920 [52:34<05:18, 4.05it/s, tok/s=16035.2]
Inference: 78%|███████▊ | 4631/5920 [52:34<05:18, 4.05it/s, tok/s=16038.0]
Inference: 78%|███████▊ | 4632/5920 [52:34<05:17, 4.05it/s, tok/s=16038.6]
Inference: 78%|███████▊ | 4633/5920 [52:35<07:42, 2.78it/s, tok/s=16038.6]
Inference: 78%|███████▊ | 4633/5920 [52:35<07:42, 2.78it/s, tok/s=16033.5]
Inference: 78%|███████▊ | 4634/5920 [52:35<07:05, 3.02it/s, tok/s=16033.5]
Inference: 78%|███████▊ | 4634/5920 [52:35<07:05, 3.02it/s, tok/s=16033.8]
Inference: 78%|███████▊ | 4635/5920 [52:36<08:09, 2.62it/s, tok/s=16033.8]
Inference: 78%|███████▊ | 4635/5920 [52:36<08:09, 2.62it/s, tok/s=16032.3]
Inference: 78%|███████▊ | 4636/5920 [52:37<10:16, 2.08it/s, tok/s=16032.3]
Inference: 78%|███████▊ | 4636/5920 [52:37<10:16, 2.08it/s, tok/s=16030.7]
Inference: 78%|███████▊ | 4637/5920 [52:37<10:16, 2.08it/s, tok/s=16034.3]
Inference: 78%|███████▊ | 4638/5920 [52:37<07:13, 2.96it/s, tok/s=16034.3]
Inference: 78%|███████▊ | 4638/5920 [52:37<07:13, 2.96it/s, tok/s=16036.1]
Inference: 78%|███████▊ | 4639/5920 [52:37<06:25, 3.32it/s, tok/s=16036.1]
Inference: 78%|███████▊ | 4639/5920 [52:37<06:25, 3.32it/s, tok/s=16042.4]
Inference: 78%|███████▊ | 4640/5920 [52:38<07:28, 2.85it/s, tok/s=16042.4]
Inference: 78%|███████▊ | 4640/5920 [52:38<07:28, 2.85it/s, tok/s=16042.2]
Inference: 78%|███████▊ | 4641/5920 [52:38<07:29, 2.85it/s, tok/s=16042.2]
Inference: 78%|███████▊ | 4641/5920 [52:38<07:29, 2.85it/s, tok/s=16044.7]
Inference: 78%|███████▊ | 4642/5920 [52:38<07:24, 2.88it/s, tok/s=16044.7]
Inference: 78%|███████▊ | 4642/5920 [52:38<07:24, 2.88it/s, tok/s=16043.7]
Inference: 78%|███████▊ | 4643/5920 [52:38<07:23, 2.88it/s, tok/s=16046.3]
Inference: 78%|███████▊ | 4644/5920 [52:39<05:22, 3.96it/s, tok/s=16046.3]
Inference: 78%|███████▊ | 4644/5920 [52:39<05:22, 3.96it/s, tok/s=16047.2]
Inference: 78%|███████▊ | 4645/5920 [52:39<04:55, 4.31it/s, tok/s=16047.2]
Inference: 78%|███████▊ | 4645/5920 [52:39<04:55, 4.31it/s, tok/s=16046.9]
Inference: 78%|███████▊ | 4646/5920 [52:39<04:44, 4.48it/s, tok/s=16046.9]
Inference: 78%|███████▊ | 4646/5920 [52:39<04:44, 4.48it/s, tok/s=16046.3]
Inference: 78%|███████▊ | 4647/5920 [52:39<04:52, 4.35it/s, tok/s=16046.3]
Inference: 78%|███████▊ | 4647/5920 [52:39<04:52, 4.35it/s, tok/s=16045.6]
Inference: 79%|███████▊ | 4648/5920 [52:39<04:52, 4.35it/s, tok/s=16052.2]
Inference: 79%|███████▊ | 4649/5920 [52:39<04:52, 4.35it/s, tok/s=16052.3]
Inference: 79%|███████▊ | 4650/5920 [52:39<04:52, 4.35it/s, tok/s=16053.1]
Inference: 79%|███████▊ | 4651/5920 [52:39<02:26, 8.69it/s, tok/s=16053.1]
Inference: 79%|███████▊ | 4651/5920 [52:39<02:26, 8.69it/s, tok/s=16053.3]
Inference: 79%|███████▊ | 4652/5920 [52:39<02:25, 8.69it/s, tok/s=16053.8]
Inference: 79%|███████▊ | 4653/5920 [52:40<04:39, 4.53it/s, tok/s=16053.8]
Inference: 79%|███████▊ | 4653/5920 [52:40<04:39, 4.53it/s, tok/s=16052.6]
Inference: 79%|███████▊ | 4654/5920 [52:40<04:18, 4.90it/s, tok/s=16052.6]
Inference: 79%|███████▊ | 4654/5920 [52:40<04:18, 4.90it/s, tok/s=16053.0]
Inference: 79%|███████▊ | 4655/5920 [52:41<04:06, 5.14it/s, tok/s=16053.0]
Inference: 79%|███████▊ | 4655/5920 [52:41<04:06, 5.14it/s, tok/s=16052.6]
Inference: 79%|███████▊ | 4656/5920 [52:41<06:19, 3.33it/s, tok/s=16052.6]
Inference: 79%|███████▊ | 4656/5920 [52:41<06:19, 3.33it/s, tok/s=16051.1]
Inference: 79%|███████▊ | 4657/5920 [52:41<05:20, 3.94it/s, tok/s=16051.1]
Inference: 79%|███████▊ | 4657/5920 [52:41<05:20, 3.94it/s, tok/s=16052.2]
Inference: 79%|███████▊ | 4658/5920 [52:41<05:19, 3.94it/s, tok/s=16052.3]
Inference: 79%|███████▊ | 4659/5920 [52:42<04:18, 4.88it/s, tok/s=16052.3]
Inference: 79%|███████▊ | 4659/5920 [52:42<04:18, 4.88it/s, tok/s=16053.1]
Inference: 79%|███████▊ | 4660/5920 [52:42<05:26, 3.86it/s, tok/s=16053.1]
Inference: 79%|███████▊ | 4660/5920 [52:42<05:26, 3.86it/s, tok/s=16052.9]
Inference: 79%|███████▊ | 4661/5920 [52:42<05:26, 3.86it/s, tok/s=16056.6]
Inference: 79%|███████▉ | 4662/5920 [52:42<04:48, 4.36it/s, tok/s=16056.6]
Inference: 79%|███████▉ | 4662/5920 [52:42<04:48, 4.36it/s, tok/s=16055.4]
Inference: 79%|███████▉ | 4663/5920 [52:42<04:48, 4.36it/s, tok/s=16055.7]
Inference: 79%|███████▉ | 4664/5920 [52:43<03:38, 5.74it/s, tok/s=16055.7]
Inference: 79%|███████▉ | 4664/5920 [52:43<03:38, 5.74it/s, tok/s=16056.7]
Inference: 79%|███████▉ | 4665/5920 [52:43<05:52, 3.56it/s, tok/s=16056.7]
Inference: 79%|███████▉ | 4665/5920 [52:43<05:52, 3.56it/s, tok/s=16054.4]
Inference: 79%|███████▉ | 4666/5920 [52:43<05:52, 3.56it/s, tok/s=16054.4]
Inference: 79%|███████▉ | 4667/5920 [52:44<06:10, 3.38it/s, tok/s=16054.4]
Inference: 79%|███████▉ | 4667/5920 [52:44<06:10, 3.38it/s, tok/s=16053.6]
Inference: 79%|███████▉ | 4668/5920 [52:44<05:22, 3.88it/s, tok/s=16053.6]
Inference: 79%|███████▉ | 4668/5920 [52:44<05:22, 3.88it/s, tok/s=16055.4]
Inference: 79%|███████▉ | 4669/5920 [52:44<05:27, 3.82it/s, tok/s=16055.4]
Inference: 79%|███████▉ | 4669/5920 [52:44<05:27, 3.82it/s, tok/s=16055.2]
Inference: 79%|███████▉ | 4670/5920 [52:44<05:27, 3.82it/s, tok/s=16055.3]
Inference: 79%|███████▉ | 4671/5920 [52:45<04:33, 4.57it/s, tok/s=16055.3]
Inference: 79%|███████▉ | 4671/5920 [52:45<04:33, 4.57it/s, tok/s=16054.7]
Inference: 79%|███████▉ | 4672/5920 [52:45<04:48, 4.32it/s, tok/s=16054.7]
Inference: 79%|███████▉ | 4672/5920 [52:45<04:48, 4.32it/s, tok/s=16055.8]
Inference: 79%|███████▉ | 4673/5920 [52:45<04:34, 4.55it/s, tok/s=16055.8]
Inference: 79%|███████▉ | 4673/5920 [52:45<04:34, 4.55it/s, tok/s=16055.5]
Inference: 79%|███████▉ | 4674/5920 [52:45<04:22, 4.75it/s, tok/s=16055.5]
Inference: 79%|███████▉ | 4674/5920 [52:45<04:22, 4.75it/s, tok/s=16055.6]
Inference: 79%|███████▉ | 4675/5920 [52:46<10:17, 2.02it/s, tok/s=16055.6]
Inference: 79%|███████▉ | 4675/5920 [52:46<10:17, 2.02it/s, tok/s=16049.6]
Inference: 79%|███████▉ | 4676/5920 [52:46<10:17, 2.02it/s, tok/s=16050.8]
Inference: 79%|███████▉ | 4677/5920 [52:47<06:22, 3.25it/s, tok/s=16050.8]
Inference: 79%|███████▉ | 4677/5920 [52:47<06:22, 3.25it/s, tok/s=16051.7]
Inference: 79%|███████▉ | 4678/5920 [52:47<05:25, 3.82it/s, tok/s=16051.7]
Inference: 79%|███████▉ | 4678/5920 [52:47<05:25, 3.82it/s, tok/s=16058.9]
Inference: 79%|███████▉ | 4679/5920 [52:47<07:33, 2.74it/s, tok/s=16058.9]
Inference: 79%|███████▉ | 4679/5920 [52:47<07:33, 2.74it/s, tok/s=16060.2]
Inference: 79%|███████▉ | 4680/5920 [52:48<06:36, 3.13it/s, tok/s=16060.2]
Inference: 79%|███████▉ | 4680/5920 [52:48<06:36, 3.13it/s, tok/s=16062.1]
Inference: 79%|███████▉ | 4681/5920 [52:48<06:36, 3.13it/s, tok/s=16064.4]
Inference: 79%|███████▉ | 4682/5920 [52:48<06:35, 3.13it/s, tok/s=16065.3]
Inference: 79%|███████▉ | 4683/5920 [52:48<03:58, 5.18it/s, tok/s=16065.3]
Inference: 79%|███████▉ | 4683/5920 [52:48<03:58, 5.18it/s, tok/s=16074.0]
Inference: 79%|███████▉ | 4684/5920 [52:48<04:32, 4.54it/s, tok/s=16074.0]
Inference: 79%|███████▉ | 4684/5920 [52:48<04:32, 4.54it/s, tok/s=16075.6]
Inference: 79%|███████▉ | 4685/5920 [52:48<04:33, 4.52it/s, tok/s=16075.6]
Inference: 79%|███████▉ | 4685/5920 [52:48<04:33, 4.52it/s, tok/s=16075.7]
Inference: 79%|███████▉ | 4686/5920 [52:49<04:32, 4.53it/s, tok/s=16075.7]
Inference: 79%|███████▉ | 4686/5920 [52:49<04:32, 4.53it/s, tok/s=16075.1]
Inference: 79%|███████▉ | 4687/5920 [52:49<05:59, 3.43it/s, tok/s=16075.1]
Inference: 79%|███████▉ | 4687/5920 [52:49<05:59, 3.43it/s, tok/s=16073.1]
Inference: 79%|███████▉ | 4688/5920 [52:49<04:57, 4.14it/s, tok/s=16073.1]
Inference: 79%|███████▉ | 4688/5920 [52:49<04:57, 4.14it/s, tok/s=16075.8]
Inference: 79%|███████▉ | 4689/5920 [52:50<07:46, 2.64it/s, tok/s=16075.8]
Inference: 79%|███████▉ | 4689/5920 [52:50<07:46, 2.64it/s, tok/s=16074.3]
Inference: 79%|███████▉ | 4690/5920 [52:50<07:14, 2.83it/s, tok/s=16074.3]
Inference: 79%|███████▉ | 4690/5920 [52:50<07:14, 2.83it/s, tok/s=16074.1]
Inference: 79%|███████▉ | 4691/5920 [52:51<07:36, 2.69it/s, tok/s=16074.1]
Inference: 79%|███████▉ | 4691/5920 [52:51<07:36, 2.69it/s, tok/s=16073.9]
Inference: 79%|███████▉ | 4692/5920 [52:51<08:41, 2.36it/s, tok/s=16073.9]
Inference: 79%|███████▉ | 4692/5920 [52:51<08:41, 2.36it/s, tok/s=16072.0]
Inference: 79%|███████▉ | 4693/5920 [52:51<07:23, 2.77it/s, tok/s=16072.0]
Inference: 79%|███████▉ | 4693/5920 [52:51<07:23, 2.77it/s, tok/s=16073.4]
Inference: 79%|███████▉ | 4694/5920 [52:52<07:27, 2.74it/s, tok/s=16073.4]
Inference: 79%|███████▉ | 4694/5920 [52:52<07:27, 2.74it/s, tok/s=16080.1]
Inference: 79%|███████▉ | 4695/5920 [52:52<07:26, 2.74it/s, tok/s=16080.3]
Inference: 79%|███████▉ | 4696/5920 [52:52<06:44, 3.02it/s, tok/s=16080.3]
Inference: 79%|███████▉ | 4696/5920 [52:52<06:44, 3.02it/s, tok/s=16078.0]
Inference: 79%|███████▉ | 4697/5920 [52:54<11:40, 1.75it/s, tok/s=16078.0]
Inference: 79%|███████▉ | 4697/5920 [52:54<11:40, 1.75it/s, tok/s=16072.0]
Inference: 79%|███████▉ | 4698/5920 [52:54<09:36, 2.12it/s, tok/s=16072.0]
Inference: 79%|███████▉ | 4698/5920 [52:54<09:36, 2.12it/s, tok/s=16071.5]
Inference: 79%|███████▉ | 4699/5920 [52:54<07:56, 2.56it/s, tok/s=16071.5]
Inference: 79%|███████▉ | 4699/5920 [52:54<07:56, 2.56it/s, tok/s=16071.5]
Inference: 79%|███████▉ | 4700/5920 [52:54<06:39, 3.05it/s, tok/s=16071.5]
Inference: 79%|███████▉ | 4700/5920 [52:54<06:39, 3.05it/s, tok/s=16071.2]
Inference: 79%|███████▉ | 4701/5920 [52:54<05:58, 3.40it/s, tok/s=16071.2]
Inference: 79%|███████▉ | 4701/5920 [52:54<05:58, 3.40it/s, tok/s=16076.7]
Inference: 79%|███████▉ | 4702/5920 [52:55<05:10, 3.92it/s, tok/s=16076.7]
Inference: 79%|███████▉ | 4702/5920 [52:55<05:10, 3.92it/s, tok/s=16078.7]
Inference: 79%|███████▉ | 4703/5920 [52:55<05:03, 4.01it/s, tok/s=16078.7]
Inference: 79%|███████▉ | 4703/5920 [52:55<05:03, 4.01it/s, tok/s=16079.5]
Inference: 79%|███████▉ | 4704/5920 [52:55<05:03, 4.01it/s, tok/s=16079.5]
Inference: 79%|███████▉ | 4705/5920 [52:55<03:39, 5.54it/s, tok/s=16079.5]
Inference: 79%|███████▉ | 4705/5920 [52:55<03:39, 5.54it/s, tok/s=16079.9]
Inference: 79%|███████▉ | 4706/5920 [52:55<03:32, 5.72it/s, tok/s=16079.9]
Inference: 79%|███████▉ | 4706/5920 [52:55<03:32, 5.72it/s, tok/s=16082.0]
Inference: 80%|███████▉ | 4707/5920 [52:55<04:25, 4.57it/s, tok/s=16082.0]
Inference: 80%|███████▉ | 4707/5920 [52:55<04:25, 4.57it/s, tok/s=16082.0]
Inference: 80%|███████▉ | 4708/5920 [52:56<04:38, 4.35it/s, tok/s=16082.0]
Inference: 80%|███████▉ | 4708/5920 [52:56<04:38, 4.35it/s, tok/s=16081.6]
Inference: 80%|███████▉ | 4709/5920 [52:56<04:04, 4.95it/s, tok/s=16081.6]
Inference: 80%|███████▉ | 4709/5920 [52:56<04:04, 4.95it/s, tok/s=16081.5]
Inference: 80%|███████▉ | 4710/5920 [52:56<03:31, 5.72it/s, tok/s=16081.5]
Inference: 80%|███████▉ | 4710/5920 [52:56<03:31, 5.72it/s, tok/s=16090.5]
Inference: 80%|███████▉ | 4711/5920 [52:56<03:47, 5.32it/s, tok/s=16090.5]
Inference: 80%|███████▉ | 4711/5920 [52:56<03:47, 5.32it/s, tok/s=16091.4]
Inference: 80%|███████▉ | 4712/5920 [52:57<07:50, 2.57it/s, tok/s=16091.4]
Inference: 80%|███████▉ | 4712/5920 [52:57<07:50, 2.57it/s, tok/s=16089.8]
Inference: 80%|███████▉ | 4713/5920 [52:58<09:45, 2.06it/s, tok/s=16089.8]
Inference: 80%|███████▉ | 4713/5920 [52:58<09:45, 2.06it/s, tok/s=16087.2]
Inference: 80%|███████▉ | 4714/5920 [52:58<09:45, 2.06it/s, tok/s=16087.7]
Inference: 80%|███████▉ | 4715/5920 [52:58<09:44, 2.06it/s, tok/s=16090.2]
Inference: 80%|███████▉ | 4716/5920 [52:58<05:50, 3.43it/s, tok/s=16090.2]
Inference: 80%|███████▉ | 4716/5920 [52:58<05:50, 3.43it/s, tok/s=16091.2]
Inference: 80%|███████▉ | 4717/5920 [52:59<06:27, 3.11it/s, tok/s=16091.2]
Inference: 80%|███████▉ | 4717/5920 [52:59<06:27, 3.11it/s, tok/s=16089.9]
Inference: 80%|███████▉ | 4718/5920 [52:59<06:27, 3.11it/s, tok/s=16091.1]
Inference: 80%|███████▉ | 4719/5920 [52:59<04:35, 4.36it/s, tok/s=16091.1]
Inference: 80%|███████▉ | 4719/5920 [52:59<04:35, 4.36it/s, tok/s=16091.3]
Inference: 80%|███████▉ | 4720/5920 [52:59<04:43, 4.24it/s, tok/s=16091.3]
Inference: 80%|███████▉ | 4720/5920 [52:59<04:43, 4.24it/s, tok/s=16094.5]
Inference: 80%|███████▉ | 4721/5920 [52:59<04:20, 4.61it/s, tok/s=16094.5]
Inference: 80%|███████▉ | 4721/5920 [52:59<04:20, 4.61it/s, tok/s=16097.8]
Inference: 80%|███████▉ | 4722/5920 [53:01<09:49, 2.03it/s, tok/s=16097.8]
Inference: 80%|███████▉ | 4722/5920 [53:01<09:49, 2.03it/s, tok/s=16093.0]
Inference: 80%|███████▉ | 4723/5920 [53:01<09:49, 2.03it/s, tok/s=16094.3]
Inference: 80%|███████▉ | 4724/5920 [53:01<07:40, 2.60it/s, tok/s=16094.3]
Inference: 80%|███████▉ | 4724/5920 [53:01<07:40, 2.60it/s, tok/s=16093.7]
Inference: 80%|███████▉ | 4725/5920 [53:01<07:40, 2.60it/s, tok/s=16099.8]
Inference: 80%|███████▉ | 4726/5920 [53:01<05:18, 3.75it/s, tok/s=16099.8]
Inference: 80%|███████▉ | 4726/5920 [53:01<05:18, 3.75it/s, tok/s=16100.1]
Inference: 80%|███████▉ | 4727/5920 [53:01<05:11, 3.83it/s, tok/s=16100.1]
Inference: 80%|███████▉ | 4727/5920 [53:01<05:11, 3.83it/s, tok/s=16099.4]
Inference: 80%|███████▉ | 4728/5920 [53:01<04:32, 4.38it/s, tok/s=16099.4]
Inference: 80%|███████▉ | 4728/5920 [53:01<04:32, 4.38it/s, tok/s=16100.1]
Inference: 80%|███████▉ | 4729/5920 [53:02<04:32, 4.38it/s, tok/s=16100.2]
Inference: 80%|███████▉ | 4730/5920 [53:02<04:09, 4.77it/s, tok/s=16100.2]
Inference: 80%|███████▉ | 4730/5920 [53:02<04:09, 4.77it/s, tok/s=16102.9]
Inference: 80%|███████▉ | 4731/5920 [53:03<08:08, 2.43it/s, tok/s=16102.9]
Inference: 80%|███████▉ | 4731/5920 [53:03<08:08, 2.43it/s, tok/s=16097.7]
Inference: 80%|███████▉ | 4732/5920 [53:03<08:08, 2.43it/s, tok/s=16099.0]
Inference: 80%|███████▉ | 4733/5920 [53:04<07:04, 2.79it/s, tok/s=16099.0]
Inference: 80%|███████▉ | 4733/5920 [53:04<07:04, 2.79it/s, tok/s=16100.2]
Inference: 80%|███████▉ | 4734/5920 [53:04<07:04, 2.79it/s, tok/s=16101.7]
Inference: 80%|███████▉ | 4735/5920 [53:04<05:24, 3.65it/s, tok/s=16101.7]
Inference: 80%|███████▉ | 4735/5920 [53:04<05:24, 3.65it/s, tok/s=16101.2]
Inference: 80%|████████ | 4736/5920 [53:05<07:51, 2.51it/s, tok/s=16101.2]
Inference: 80%|████████ | 4736/5920 [53:05<07:51, 2.51it/s, tok/s=16098.1]
Inference: 80%|████████ | 4737/5920 [53:05<06:39, 2.96it/s, tok/s=16098.1]
Inference: 80%|████████ | 4737/5920 [53:05<06:39, 2.96it/s, tok/s=16098.5]
Inference: 80%|████████ | 4738/5920 [53:06<09:50, 2.00it/s, tok/s=16098.5]
Inference: 80%|████████ | 4738/5920 [53:06<09:50, 2.00it/s, tok/s=16095.2]
Inference: 80%|████████ | 4739/5920 [53:06<10:14, 1.92it/s, tok/s=16095.2]
Inference: 80%|████████ | 4739/5920 [53:06<10:14, 1.92it/s, tok/s=16092.9]
Inference: 80%|████████ | 4740/5920 [53:07<09:25, 2.09it/s, tok/s=16092.9]
Inference: 80%|████████ | 4740/5920 [53:07<09:25, 2.09it/s, tok/s=16100.5]
Inference: 80%|████████ | 4741/5920 [53:07<09:24, 2.09it/s, tok/s=16110.0]
Inference: 80%|████████ | 4742/5920 [53:07<09:24, 2.09it/s, tok/s=16113.2]
Inference: 80%|████████ | 4743/5920 [53:07<05:06, 3.83it/s, tok/s=16113.2]
Inference: 80%|████████ | 4743/5920 [53:07<05:06, 3.83it/s, tok/s=16112.8]
Inference: 80%|████████ | 4744/5920 [53:07<04:38, 4.23it/s, tok/s=16112.8]
Inference: 80%|████████ | 4744/5920 [53:07<04:38, 4.23it/s, tok/s=16113.3]
Inference: 80%|████████ | 4745/5920 [53:07<04:37, 4.23it/s, tok/s=16113.8]
Inference: 80%|████████ | 4746/5920 [53:07<03:24, 5.74it/s, tok/s=16113.8]
Inference: 80%|████████ | 4746/5920 [53:07<03:24, 5.74it/s, tok/s=16114.1]
Inference: 80%|████████ | 4747/5920 [53:07<03:24, 5.74it/s, tok/s=16114.5]
Inference: 80%|████████ | 4748/5920 [53:08<03:08, 6.21it/s, tok/s=16114.5]
Inference: 80%|████████ | 4748/5920 [53:08<03:08, 6.21it/s, tok/s=16114.2]
Inference: 80%|████████ | 4749/5920 [53:08<03:58, 4.92it/s, tok/s=16114.2]
Inference: 80%|████████ | 4749/5920 [53:08<03:58, 4.92it/s, tok/s=16121.7]
Inference: 80%|████████ | 4750/5920 [53:09<07:13, 2.70it/s, tok/s=16121.7]
Inference: 80%|████████ | 4750/5920 [53:09<07:13, 2.70it/s, tok/s=16118.4]
Inference: 80%|████████ | 4751/5920 [53:09<08:32, 2.28it/s, tok/s=16118.4]
Inference: 80%|████████ | 4751/5920 [53:09<08:32, 2.28it/s, tok/s=16118.0]
Inference: 80%|████████ | 4752/5920 [53:10<08:31, 2.28it/s, tok/s=16124.0]
Inference: 80%|████████ | 4753/5920 [53:10<05:46, 3.36it/s, tok/s=16124.0]
Inference: 80%|████████ | 4753/5920 [53:10<05:46, 3.36it/s, tok/s=16125.0]
Inference: 80%|████████ | 4754/5920 [53:10<05:46, 3.36it/s, tok/s=16125.1]
Inference: 80%|████████ | 4755/5920 [53:10<05:46, 3.36it/s, tok/s=16128.2]
Inference: 80%|████████ | 4756/5920 [53:10<03:29, 5.55it/s, tok/s=16128.2]
Inference: 80%|████████ | 4756/5920 [53:10<03:29, 5.55it/s, tok/s=16137.4]
Inference: 80%|████████ | 4757/5920 [53:10<03:29, 5.55it/s, tok/s=16138.2]
Inference: 80%|████████ | 4758/5920 [53:10<03:29, 5.55it/s, tok/s=16138.5]
Inference: 80%|████████ | 4759/5920 [53:10<03:45, 5.16it/s, tok/s=16138.5]
Inference: 80%|████████ | 4759/5920 [53:10<03:45, 5.16it/s, tok/s=16136.4]
Inference: 80%|████████ | 4760/5920 [53:11<03:38, 5.31it/s, tok/s=16136.4]
Inference: 80%|████████ | 4760/5920 [53:11<03:38, 5.31it/s, tok/s=16137.2]
Inference: 80%|████████ | 4761/5920 [53:11<03:38, 5.31it/s, tok/s=16138.2]
Inference: 80%|████████ | 4762/5920 [53:11<03:38, 5.31it/s, tok/s=16140.1]
Inference: 80%|████████ | 4763/5920 [53:11<03:37, 5.31it/s, tok/s=16140.9]
Inference: 80%|████████ | 4764/5920 [53:11<03:37, 5.31it/s, tok/s=16141.6]
Inference: 80%|████████ | 4765/5920 [53:11<01:56, 9.93it/s, tok/s=16141.6]
Inference: 80%|████████ | 4765/5920 [53:11<01:56, 9.93it/s, tok/s=16142.0]
Inference: 81%|████████ | 4766/5920 [53:11<01:56, 9.93it/s, tok/s=16143.1]
Inference: 81%|████████ | 4767/5920 [53:11<02:18, 8.33it/s, tok/s=16143.1]
Inference: 81%|████████ | 4767/5920 [53:11<02:18, 8.33it/s, tok/s=16144.0]
Inference: 81%|████████ | 4768/5920 [53:11<02:18, 8.33it/s, tok/s=16144.6]
Inference: 81%|████████ | 4769/5920 [53:11<02:11, 8.73it/s, tok/s=16144.6]
Inference: 81%|████████ | 4769/5920 [53:11<02:11, 8.73it/s, tok/s=16145.9]
Inference: 81%|████████ | 4770/5920 [53:12<02:11, 8.73it/s, tok/s=16150.1]
Inference: 81%|████████ | 4771/5920 [53:12<03:25, 5.58it/s, tok/s=16150.1]
Inference: 81%|████████ | 4771/5920 [53:12<03:25, 5.58it/s, tok/s=16149.6]
Inference: 81%|████████ | 4772/5920 [53:13<03:25, 5.58it/s, tok/s=16154.2]
Inference: 81%|████████ | 4773/5920 [53:13<04:34, 4.17it/s, tok/s=16154.2]
Inference: 81%|████████ | 4773/5920 [53:13<04:34, 4.17it/s, tok/s=16154.6]
Inference: 81%|████████ | 4774/5920 [53:13<04:40, 4.08it/s, tok/s=16154.6]
Inference: 81%|████████ | 4774/5920 [53:13<04:40, 4.08it/s, tok/s=16156.3]
Inference: 81%|████████ | 4775/5920 [53:13<04:40, 4.08it/s, tok/s=16157.9]
Inference: 81%|████████ | 4776/5920 [53:13<03:40, 5.20it/s, tok/s=16157.9]
Inference: 81%|████████ | 4776/5920 [53:13<03:40, 5.20it/s, tok/s=16161.5]
Inference: 81%|████████ | 4777/5920 [53:13<03:22, 5.64it/s, tok/s=16161.5]
Inference: 81%|████████ | 4777/5920 [53:13<03:22, 5.64it/s, tok/s=16162.1]
Inference: 81%|████████ | 4778/5920 [53:14<03:13, 5.89it/s, tok/s=16162.1]
Inference: 81%|████████ | 4778/5920 [53:14<03:13, 5.89it/s, tok/s=16162.7]
Inference: 81%|████████ | 4779/5920 [53:14<05:32, 3.44it/s, tok/s=16162.7]
Inference: 81%|████████ | 4779/5920 [53:14<05:32, 3.44it/s, tok/s=16161.0]
Inference: 81%|████████ | 4780/5920 [53:14<05:09, 3.68it/s, tok/s=16161.0]
Inference: 81%|████████ | 4780/5920 [53:14<05:09, 3.68it/s, tok/s=16161.2]
Inference: 81%|████████ | 4781/5920 [53:15<04:37, 4.11it/s, tok/s=16161.2]
Inference: 81%|████████ | 4781/5920 [53:15<04:37, 4.11it/s, tok/s=16161.8]
Inference: 81%|████████ | 4782/5920 [53:15<06:31, 2.90it/s, tok/s=16161.8]
Inference: 81%|████████ | 4782/5920 [53:15<06:31, 2.90it/s, tok/s=16159.2]
Inference: 81%|████████ | 4783/5920 [53:15<06:08, 3.08it/s, tok/s=16159.2]
Inference: 81%|████████ | 4783/5920 [53:15<06:08, 3.08it/s, tok/s=16159.5]
Inference: 81%|████████ | 4784/5920 [53:16<05:04, 3.73it/s, tok/s=16159.5]
Inference: 81%|████████ | 4784/5920 [53:16<05:04, 3.73it/s, tok/s=16161.6]
Inference: 81%|████████ | 4785/5920 [53:16<05:04, 3.73it/s, tok/s=16162.4]
Inference: 81%|████████ | 4786/5920 [53:16<04:14, 4.46it/s, tok/s=16162.4]
Inference: 81%|████████ | 4786/5920 [53:16<04:14, 4.46it/s, tok/s=16161.4]
Inference: 81%|████████ | 4787/5920 [53:16<04:13, 4.46it/s, tok/s=16161.7]
Inference: 81%|████████ | 4788/5920 [53:16<04:13, 4.46it/s, tok/s=16163.1]
Inference: 81%|████████ | 4789/5920 [53:16<02:54, 6.49it/s, tok/s=16163.1]
Inference: 81%|████████ | 4789/5920 [53:16<02:54, 6.49it/s, tok/s=16167.8]
Inference: 81%|████████ | 4790/5920 [53:16<02:54, 6.49it/s, tok/s=16168.2]
Inference: 81%|████████ | 4791/5920 [53:17<03:07, 6.03it/s, tok/s=16168.2]
Inference: 81%|████████ | 4791/5920 [53:17<03:07, 6.03it/s, tok/s=16168.3]
Inference: 81%|████████ | 4792/5920 [53:17<03:16, 5.73it/s, tok/s=16168.3]
Inference: 81%|████████ | 4792/5920 [53:17<03:16, 5.73it/s, tok/s=16167.8]
Inference: 81%|████████ | 4793/5920 [53:17<03:15, 5.78it/s, tok/s=16167.8]
Inference: 81%|████████ | 4793/5920 [53:17<03:15, 5.78it/s, tok/s=16167.5]
Inference: 81%|████████ | 4794/5920 [53:17<03:14, 5.78it/s, tok/s=16168.8]
Inference: 81%|████████ | 4795/5920 [53:17<02:50, 6.60it/s, tok/s=16168.8]
Inference: 81%|████████ | 4795/5920 [53:17<02:50, 6.60it/s, tok/s=16168.1]
Inference: 81%|████████ | 4796/5920 [53:17<02:50, 6.59it/s, tok/s=16168.1]
Inference: 81%|████████ | 4796/5920 [53:17<02:50, 6.59it/s, tok/s=16170.8]
Inference: 81%|████████ | 4797/5920 [53:18<04:37, 4.04it/s, tok/s=16170.8]
Inference: 81%|████████ | 4797/5920 [53:18<04:37, 4.04it/s, tok/s=16168.9]
Inference: 81%|████████ | 4798/5920 [53:18<04:37, 4.04it/s, tok/s=16169.1]
Inference: 81%|████████ | 4799/5920 [53:18<03:53, 4.80it/s, tok/s=16169.1]
Inference: 81%|████████ | 4799/5920 [53:18<03:53, 4.80it/s, tok/s=16168.5]
Inference: 81%|████████ | 4800/5920 [53:18<03:32, 5.28it/s, tok/s=16168.5]
Inference: 81%|████████ | 4800/5920 [53:18<03:32, 5.28it/s, tok/s=16169.8]
Inference: 81%|████████ | 4801/5920 [53:18<03:32, 5.28it/s, tok/s=16171.0]
Inference: 81%|████████ | 4802/5920 [53:18<02:46, 6.72it/s, tok/s=16171.0]
Inference: 81%|████████ | 4802/5920 [53:18<02:46, 6.72it/s, tok/s=16179.6]
Inference: 81%|████████ | 4803/5920 [53:19<02:46, 6.72it/s, tok/s=16179.5]
Inference: 81%|████████ | 4804/5920 [53:19<02:46, 6.72it/s, tok/s=16181.0]
Inference: 81%|████████ | 4805/5920 [53:19<03:22, 5.50it/s, tok/s=16181.0]
Inference: 81%|████████ | 4805/5920 [53:19<03:22, 5.50it/s, tok/s=16179.6]
Inference: 81%|████████ | 4806/5920 [53:20<05:22, 3.46it/s, tok/s=16179.6]
Inference: 81%|████████ | 4806/5920 [53:20<05:22, 3.46it/s, tok/s=16177.8]
Inference: 81%|████████ | 4807/5920 [53:21<08:00, 2.31it/s, tok/s=16177.8]
Inference: 81%|████████ | 4807/5920 [53:21<08:00, 2.31it/s, tok/s=16175.4]
Inference: 81%|████████ | 4808/5920 [53:21<08:00, 2.31it/s, tok/s=16176.2]
Inference: 81%|████████ | 4809/5920 [53:22<07:17, 2.54it/s, tok/s=16176.2]
Inference: 81%|████████ | 4809/5920 [53:22<07:17, 2.54it/s, tok/s=16173.9]
Inference: 81%|████████▏ | 4810/5920 [53:22<07:17, 2.54it/s, tok/s=16174.3]
Inference: 81%|████████▏ | 4811/5920 [53:22<06:22, 2.90it/s, tok/s=16174.3]
Inference: 81%|████████▏ | 4811/5920 [53:22<06:22, 2.90it/s, tok/s=16172.8]
Inference: 81%|████████▏ | 4812/5920 [53:23<07:43, 2.39it/s, tok/s=16172.8]
Inference: 81%|████████▏ | 4812/5920 [53:23<07:43, 2.39it/s, tok/s=16171.0]
Inference: 81%|████████▏ | 4813/5920 [53:23<06:39, 2.77it/s, tok/s=16171.0]
Inference: 81%|████████▏ | 4813/5920 [53:23<06:39, 2.77it/s, tok/s=16173.2]
Inference: 81%|████████▏ | 4814/5920 [53:23<05:40, 3.25it/s, tok/s=16173.2]
Inference: 81%|████████▏ | 4814/5920 [53:23<05:40, 3.25it/s, tok/s=16174.0]
Inference: 81%|████████▏ | 4815/5920 [53:23<05:49, 3.17it/s, tok/s=16174.0]
Inference: 81%|████████▏ | 4815/5920 [53:23<05:49, 3.17it/s, tok/s=16176.8]
Inference: 81%|████████▏ | 4816/5920 [53:24<05:48, 3.17it/s, tok/s=16178.2]
Inference: 81%|████████▏ | 4817/5920 [53:24<04:17, 4.29it/s, tok/s=16178.2]
Inference: 81%|████████▏ | 4817/5920 [53:24<04:17, 4.29it/s, tok/s=16178.4]
Inference: 81%|████████▏ | 4818/5920 [53:24<04:52, 3.77it/s, tok/s=16178.4]
Inference: 81%|████████▏ | 4818/5920 [53:24<04:52, 3.77it/s, tok/s=16178.7]
Inference: 81%|████████▏ | 4819/5920 [53:24<04:51, 3.77it/s, tok/s=16180.4]
Inference: 81%|████████▏ | 4820/5920 [53:24<03:46, 4.85it/s, tok/s=16180.4]
Inference: 81%|████████▏ | 4820/5920 [53:24<03:46, 4.85it/s, tok/s=16181.8]
Inference: 81%|████████▏ | 4821/5920 [53:24<03:42, 4.94it/s, tok/s=16181.8]
Inference: 81%|████████▏ | 4821/5920 [53:24<03:42, 4.94it/s, tok/s=16181.4]
Inference: 81%|████████▏ | 4822/5920 [53:25<04:01, 4.55it/s, tok/s=16181.4]
Inference: 81%|████████▏ | 4822/5920 [53:25<04:01, 4.55it/s, tok/s=16183.2]
Inference: 81%|████████▏ | 4823/5920 [53:25<04:53, 3.74it/s, tok/s=16183.2]
Inference: 81%|████████▏ | 4823/5920 [53:25<04:53, 3.74it/s, tok/s=16181.8]
Inference: 81%|████████▏ | 4824/5920 [53:25<04:53, 3.74it/s, tok/s=16183.4]
Inference: 82%|████████▏ | 4825/5920 [53:26<07:21, 2.48it/s, tok/s=16183.4]
Inference: 82%|████████▏ | 4825/5920 [53:26<07:21, 2.48it/s, tok/s=16179.2]
Inference: 82%|████████▏ | 4826/5920 [53:26<06:15, 2.91it/s, tok/s=16179.2]
Inference: 82%|████████▏ | 4826/5920 [53:26<06:15, 2.91it/s, tok/s=16182.8]
Inference: 82%|████████▏ | 4827/5920 [53:27<06:24, 2.84it/s, tok/s=16182.8]
Inference: 82%|████████▏ | 4827/5920 [53:27<06:24, 2.84it/s, tok/s=16181.7]
Inference: 82%|████████▏ | 4828/5920 [53:27<06:24, 2.84it/s, tok/s=16183.9]
Inference: 82%|████████▏ | 4829/5920 [53:27<05:54, 3.08it/s, tok/s=16183.9]
Inference: 82%|████████▏ | 4829/5920 [53:27<05:54, 3.08it/s, tok/s=16184.2]
Inference: 82%|████████▏ | 4830/5920 [53:27<05:54, 3.08it/s, tok/s=16185.1]
Inference: 82%|████████▏ | 4831/5920 [53:28<04:47, 3.79it/s, tok/s=16185.1]
Inference: 82%|████████▏ | 4831/5920 [53:28<04:47, 3.79it/s, tok/s=16185.1]
Inference: 82%|████████▏ | 4832/5920 [53:28<05:25, 3.35it/s, tok/s=16185.1]
Inference: 82%|████████▏ | 4832/5920 [53:28<05:25, 3.35it/s, tok/s=16183.5]
Inference: 82%|████████▏ | 4833/5920 [53:28<04:53, 3.70it/s, tok/s=16183.5]
Inference: 82%|████████▏ | 4833/5920 [53:28<04:53, 3.70it/s, tok/s=16184.8]
Inference: 82%|████████▏ | 4834/5920 [53:28<04:53, 3.70it/s, tok/s=16185.1]
Inference: 82%|████████▏ | 4835/5920 [53:28<04:53, 3.70it/s, tok/s=16185.5]
Inference: 82%|████████▏ | 4836/5920 [53:29<03:04, 5.86it/s, tok/s=16185.5]
Inference: 82%|████████▏ | 4836/5920 [53:29<03:04, 5.86it/s, tok/s=16185.6]
Inference: 82%|████████▏ | 4837/5920 [53:29<03:18, 5.44it/s, tok/s=16185.6]
Inference: 82%|████████▏ | 4837/5920 [53:29<03:18, 5.44it/s, tok/s=16187.5]
Inference: 82%|████████▏ | 4838/5920 [53:29<03:09, 5.71it/s, tok/s=16187.5]
Inference: 82%|████████▏ | 4838/5920 [53:29<03:09, 5.71it/s, tok/s=16189.1]
Inference: 82%|████████▏ | 4839/5920 [53:29<04:45, 3.79it/s, tok/s=16189.1]
Inference: 82%|████████▏ | 4839/5920 [53:29<04:45, 3.79it/s, tok/s=16187.1]
Inference: 82%|████████▏ | 4840/5920 [53:30<04:12, 4.28it/s, tok/s=16187.1]
Inference: 82%|████████▏ | 4840/5920 [53:30<04:12, 4.28it/s, tok/s=16189.1]
Inference: 82%|████████▏ | 4841/5920 [53:30<04:12, 4.28it/s, tok/s=16190.1]
Inference: 82%|████████▏ | 4842/5920 [53:30<03:14, 5.54it/s, tok/s=16190.1]
Inference: 82%|████████▏ | 4842/5920 [53:30<03:14, 5.54it/s, tok/s=16189.8]
Inference: 82%|████████▏ | 4843/5920 [53:30<03:51, 4.65it/s, tok/s=16189.8]
Inference: 82%|████████▏ | 4843/5920 [53:30<03:51, 4.65it/s, tok/s=16189.3]
Inference: 82%|████████▏ | 4844/5920 [53:30<03:51, 4.65it/s, tok/s=16190.6]
Inference: 82%|████████▏ | 4845/5920 [53:30<03:51, 4.65it/s, tok/s=16191.4]
Inference: 82%|████████▏ | 4846/5920 [53:30<02:49, 6.32it/s, tok/s=16191.4]
Inference: 82%|████████▏ | 4846/5920 [53:30<02:49, 6.32it/s, tok/s=16191.1]
Inference: 82%|████████▏ | 4847/5920 [53:31<02:54, 6.16it/s, tok/s=16191.1]
Inference: 82%|████████▏ | 4847/5920 [53:31<02:54, 6.16it/s, tok/s=16190.6]
Inference: 82%|████████▏ | 4848/5920 [53:31<03:08, 5.68it/s, tok/s=16190.6]
Inference: 82%|████████▏ | 4848/5920 [53:31<03:08, 5.68it/s, tok/s=16193.3]
Inference: 82%|████████▏ | 4849/5920 [53:31<03:07, 5.72it/s, tok/s=16193.3]
Inference: 82%|████████▏ | 4849/5920 [53:31<03:07, 5.72it/s, tok/s=16196.4]
Inference: 82%|████████▏ | 4850/5920 [53:31<03:56, 4.53it/s, tok/s=16196.4]
Inference: 82%|████████▏ | 4850/5920 [53:31<03:56, 4.53it/s, tok/s=16195.2]
Inference: 82%|████████▏ | 4851/5920 [53:32<05:41, 3.13it/s, tok/s=16195.2]
Inference: 82%|████████▏ | 4851/5920 [53:32<05:41, 3.13it/s, tok/s=16193.3]
Inference: 82%|████████▏ | 4852/5920 [53:32<05:57, 2.99it/s, tok/s=16193.3]
Inference: 82%|████████▏ | 4852/5920 [53:32<05:57, 2.99it/s, tok/s=16192.0]
Inference: 82%|████████▏ | 4853/5920 [53:33<06:24, 2.78it/s, tok/s=16192.0]
Inference: 82%|████████▏ | 4853/5920 [53:33<06:24, 2.78it/s, tok/s=16190.3]
Inference: 82%|████████▏ | 4854/5920 [53:33<06:23, 2.78it/s, tok/s=16191.3]
Inference: 82%|████████▏ | 4855/5920 [53:33<05:08, 3.45it/s, tok/s=16191.3]
Inference: 82%|████████▏ | 4855/5920 [53:33<05:08, 3.45it/s, tok/s=16190.9]
Inference: 82%|████████▏ | 4856/5920 [53:33<05:08, 3.45it/s, tok/s=16191.3]
Inference: 82%|████████▏ | 4857/5920 [53:33<03:29, 5.06it/s, tok/s=16191.3]
Inference: 82%|████████▏ | 4857/5920 [53:33<03:29, 5.06it/s, tok/s=16194.4]
Inference: 82%|████████▏ | 4858/5920 [53:34<03:28, 5.10it/s, tok/s=16194.4]
Inference: 82%|████████▏ | 4858/5920 [53:34<03:28, 5.10it/s, tok/s=16198.1]
Inference: 82%|████████▏ | 4859/5920 [53:34<03:28, 5.10it/s, tok/s=16201.6]
Inference: 82%|████████▏ | 4860/5920 [53:34<02:45, 6.39it/s, tok/s=16201.6]
Inference: 82%|████████▏ | 4860/5920 [53:34<02:45, 6.39it/s, tok/s=16201.7]
Inference: 82%|████████▏ | 4861/5920 [53:34<02:41, 6.54it/s, tok/s=16201.7]
Inference: 82%|████████▏ | 4861/5920 [53:34<02:41, 6.54it/s, tok/s=16202.3]
Inference: 82%|████████▏ | 4862/5920 [53:34<02:41, 6.54it/s, tok/s=16203.7]
Inference: 82%|████████▏ | 4863/5920 [53:34<02:22, 7.42it/s, tok/s=16203.7]
Inference: 82%|████████▏ | 4863/5920 [53:34<02:22, 7.42it/s, tok/s=16203.3]
Inference: 82%|████████▏ | 4864/5920 [53:35<04:11, 4.19it/s, tok/s=16203.3]
Inference: 82%|████████▏ | 4864/5920 [53:35<04:11, 4.19it/s, tok/s=16209.6]
Inference: 82%|████████▏ | 4865/5920 [53:35<04:24, 3.99it/s, tok/s=16209.6]
Inference: 82%|████████▏ | 4865/5920 [53:35<04:24, 3.99it/s, tok/s=16209.5]
Inference: 82%|████████▏ | 4866/5920 [53:35<04:51, 3.62it/s, tok/s=16209.5]
Inference: 82%|████████▏ | 4866/5920 [53:35<04:51, 3.62it/s, tok/s=16208.3]
Inference: 82%|████████▏ | 4867/5920 [53:35<04:50, 3.62it/s, tok/s=16208.8]
Inference: 82%|████████▏ | 4868/5920 [53:36<03:33, 4.93it/s, tok/s=16208.8]
Inference: 82%|████████▏ | 4868/5920 [53:36<03:33, 4.93it/s, tok/s=16208.7]
Inference: 82%|████████▏ | 4869/5920 [53:36<03:33, 4.93it/s, tok/s=16209.2]
Inference: 82%|████████▏ | 4870/5920 [53:36<03:33, 4.93it/s, tok/s=16209.8]
Inference: 82%|████████▏ | 4871/5920 [53:36<02:11, 8.01it/s, tok/s=16209.8]
Inference: 82%|████████▏ | 4871/5920 [53:36<02:11, 8.01it/s, tok/s=16211.4]
Inference: 82%|████████▏ | 4872/5920 [53:36<02:10, 8.01it/s, tok/s=16212.2]
Inference: 82%|████████▏ | 4873/5920 [53:36<03:14, 5.38it/s, tok/s=16212.2]
Inference: 82%|████████▏ | 4873/5920 [53:36<03:14, 5.38it/s, tok/s=16210.7]
Inference: 82%|████████▏ | 4874/5920 [53:36<03:14, 5.38it/s, tok/s=16213.8]
Inference: 82%|████████▏ | 4875/5920 [53:37<02:58, 5.87it/s, tok/s=16213.8]
Inference: 82%|████████▏ | 4875/5920 [53:37<02:58, 5.87it/s, tok/s=16213.9]
Inference: 82%|████████▏ | 4876/5920 [53:37<03:05, 5.62it/s, tok/s=16213.9]
Inference: 82%|████████▏ | 4876/5920 [53:37<03:05, 5.62it/s, tok/s=16214.7]
Inference: 82%|████████▏ | 4877/5920 [53:37<04:19, 4.02it/s, tok/s=16214.7]
Inference: 82%|████████▏ | 4877/5920 [53:37<04:19, 4.02it/s, tok/s=16213.2]
Inference: 82%|████████▏ | 4878/5920 [53:38<05:59, 2.90it/s, tok/s=16213.2]
Inference: 82%|████████▏ | 4878/5920 [53:38<05:59, 2.90it/s, tok/s=16210.8]
Inference: 82%|████████▏ | 4879/5920 [53:38<05:59, 2.90it/s, tok/s=16211.2]
Inference: 82%|████████▏ | 4880/5920 [53:38<04:03, 4.27it/s, tok/s=16211.2]
Inference: 82%|████████▏ | 4880/5920 [53:38<04:03, 4.27it/s, tok/s=16211.4]
Inference: 82%|████████▏ | 4881/5920 [53:39<06:09, 2.82it/s, tok/s=16211.4]
Inference: 82%|████████▏ | 4881/5920 [53:39<06:09, 2.82it/s, tok/s=16208.1]
Inference: 82%|████████▏ | 4882/5920 [53:40<08:24, 2.06it/s, tok/s=16208.1]
Inference: 82%|████████▏ | 4882/5920 [53:40<08:24, 2.06it/s, tok/s=16209.8]
Inference: 82%|████████▏ | 4883/5920 [53:41<09:49, 1.76it/s, tok/s=16209.8]
Inference: 82%|████████▏ | 4883/5920 [53:41<09:49, 1.76it/s, tok/s=16211.1]
Inference: 82%|████████▎ | 4884/5920 [53:41<09:48, 1.76it/s, tok/s=16211.6]
Inference: 83%|████████▎ | 4885/5920 [53:41<06:15, 2.76it/s, tok/s=16211.6]
Inference: 83%|████████▎ | 4885/5920 [53:41<06:15, 2.76it/s, tok/s=16211.5]
Inference: 83%|████████▎ | 4886/5920 [53:41<05:26, 3.17it/s, tok/s=16211.5]
Inference: 83%|████████▎ | 4886/5920 [53:41<05:26, 3.17it/s, tok/s=16211.5]
Inference: 83%|████████▎ | 4887/5920 [53:41<05:25, 3.17it/s, tok/s=16214.4]
Inference: 83%|████████▎ | 4888/5920 [53:41<05:25, 3.17it/s, tok/s=16216.2]
Inference: 83%|████████▎ | 4889/5920 [53:41<03:17, 5.22it/s, tok/s=16216.2]
Inference: 83%|████████▎ | 4889/5920 [53:41<03:17, 5.22it/s, tok/s=16217.8]
Inference: 83%|████████▎ | 4890/5920 [53:41<03:17, 5.22it/s, tok/s=16218.2]
Inference: 83%|████████▎ | 4891/5920 [53:41<03:16, 5.22it/s, tok/s=16219.5]
Inference: 83%|████████▎ | 4892/5920 [53:42<03:28, 4.94it/s, tok/s=16219.5]
Inference: 83%|████████▎ | 4892/5920 [53:42<03:28, 4.94it/s, tok/s=16217.1]
Inference: 83%|████████▎ | 4893/5920 [53:42<03:37, 4.73it/s, tok/s=16217.1]
Inference: 83%|████████▎ | 4893/5920 [53:42<03:37, 4.73it/s, tok/s=16217.0]
Inference: 83%|████████▎ | 4894/5920 [53:43<05:13, 3.27it/s, tok/s=16217.0]
Inference: 83%|████████▎ | 4894/5920 [53:43<05:13, 3.27it/s, tok/s=16214.3]
Inference: 83%|████████▎ | 4895/5920 [53:43<05:00, 3.41it/s, tok/s=16214.3]
Inference: 83%|████████▎ | 4895/5920 [53:43<05:00, 3.41it/s, tok/s=16213.6]
Inference: 83%|████████▎ | 4896/5920 [53:43<05:00, 3.41it/s, tok/s=16219.4]
Inference: 83%|████████▎ | 4897/5920 [53:43<04:59, 3.41it/s, tok/s=16219.8]
Inference: 83%|████████▎ | 4898/5920 [53:43<03:19, 5.12it/s, tok/s=16219.8]
Inference: 83%|████████▎ | 4898/5920 [53:43<03:19, 5.12it/s, tok/s=16221.8]
Inference: 83%|████████▎ | 4899/5920 [53:45<06:46, 2.51it/s, tok/s=16221.8]
Inference: 83%|████████▎ | 4899/5920 [53:45<06:46, 2.51it/s, tok/s=16216.0]
Inference: 83%|████████▎ | 4900/5920 [53:45<06:29, 2.62it/s, tok/s=16216.0]
Inference: 83%|████████▎ | 4900/5920 [53:45<06:29, 2.62it/s, tok/s=16222.6]
Inference: 83%|████████▎ | 4901/5920 [53:45<05:32, 3.07it/s, tok/s=16222.6]
Inference: 83%|████████▎ | 4901/5920 [53:45<05:32, 3.07it/s, tok/s=16222.7]
Inference: 83%|████████▎ | 4902/5920 [53:45<05:31, 3.07it/s, tok/s=16223.9]
Inference: 83%|████████▎ | 4903/5920 [53:45<04:59, 3.39it/s, tok/s=16223.9]
Inference: 83%|████████▎ | 4903/5920 [53:45<04:59, 3.39it/s, tok/s=16224.4]
Inference: 83%|████████▎ | 4904/5920 [53:46<07:44, 2.19it/s, tok/s=16224.4]
Inference: 83%|████████▎ | 4904/5920 [53:47<07:44, 2.19it/s, tok/s=16219.7]
Inference: 83%|████████▎ | 4905/5920 [53:47<07:44, 2.19it/s, tok/s=16224.7]
Inference: 83%|████████▎ | 4906/5920 [53:47<07:43, 2.19it/s, tok/s=16225.4]
Inference: 83%|████████▎ | 4907/5920 [53:47<05:22, 3.14it/s, tok/s=16225.4]
Inference: 83%|████████▎ | 4907/5920 [53:47<05:22, 3.14it/s, tok/s=16223.6]
Inference: 83%|████████▎ | 4908/5920 [53:47<05:22, 3.14it/s, tok/s=16225.0]
Inference: 83%|████████▎ | 4909/5920 [53:47<04:08, 4.07it/s, tok/s=16225.0]
Inference: 83%|████████▎ | 4909/5920 [53:47<04:08, 4.07it/s, tok/s=16226.3]
Inference: 83%|████████▎ | 4910/5920 [53:47<03:58, 4.24it/s, tok/s=16226.3]
Inference: 83%|████████▎ | 4910/5920 [53:47<03:58, 4.24it/s, tok/s=16227.1]
Inference: 83%|████████▎ | 4911/5920 [53:47<03:58, 4.24it/s, tok/s=16227.5]
Inference: 83%|████████▎ | 4912/5920 [53:48<02:59, 5.62it/s, tok/s=16227.5]
Inference: 83%|████████▎ | 4912/5920 [53:48<02:59, 5.62it/s, tok/s=16228.2]
Inference: 83%|████████▎ | 4913/5920 [53:48<02:59, 5.62it/s, tok/s=16228.5]
Inference: 83%|████████▎ | 4914/5920 [53:48<02:20, 7.18it/s, tok/s=16228.5]
Inference: 83%|████████▎ | 4914/5920 [53:48<02:20, 7.18it/s, tok/s=16231.4]
Inference: 83%|████████▎ | 4915/5920 [53:48<02:20, 7.18it/s, tok/s=16231.8]
Inference: 83%|████████▎ | 4916/5920 [53:48<02:21, 7.11it/s, tok/s=16231.8]
Inference: 83%|████████▎ | 4916/5920 [53:48<02:21, 7.11it/s, tok/s=16231.3]
Inference: 83%|████████▎ | 4917/5920 [53:48<02:15, 7.39it/s, tok/s=16231.3]
Inference: 83%|████████▎ | 4917/5920 [53:48<02:15, 7.39it/s, tok/s=16231.3]
Inference: 83%|████████▎ | 4918/5920 [53:49<04:20, 3.85it/s, tok/s=16231.3]
Inference: 83%|████████▎ | 4918/5920 [53:49<04:20, 3.85it/s, tok/s=16228.9]
Inference: 83%|████████▎ | 4919/5920 [53:49<03:54, 4.27it/s, tok/s=16228.9]
Inference: 83%|████████▎ | 4919/5920 [53:49<03:54, 4.27it/s, tok/s=16229.3]
Inference: 83%|████████▎ | 4920/5920 [53:49<03:34, 4.66it/s, tok/s=16229.3]
Inference: 83%|████████▎ | 4920/5920 [53:49<03:34, 4.66it/s, tok/s=16229.4]
Inference: 83%|████████▎ | 4921/5920 [53:49<03:34, 4.66it/s, tok/s=16231.7]
Inference: 83%|████████▎ | 4922/5920 [53:49<03:34, 4.66it/s, tok/s=16232.8]
Inference: 83%|████████▎ | 4923/5920 [53:49<02:02, 8.15it/s, tok/s=16232.8]
Inference: 83%|████████▎ | 4923/5920 [53:49<02:02, 8.15it/s, tok/s=16233.7]
Inference: 83%|████████▎ | 4924/5920 [53:49<02:02, 8.15it/s, tok/s=16234.3]
Inference: 83%|████████▎ | 4925/5920 [53:50<03:27, 4.79it/s, tok/s=16234.3]
Inference: 83%|████████▎ | 4925/5920 [53:50<03:27, 4.79it/s, tok/s=16235.8]
Inference: 83%|████████▎ | 4926/5920 [53:50<04:28, 3.71it/s, tok/s=16235.8]
Inference: 83%|████████▎ | 4926/5920 [53:50<04:28, 3.71it/s, tok/s=16234.4]
Inference: 83%|████████▎ | 4927/5920 [53:51<04:29, 3.69it/s, tok/s=16234.4]
Inference: 83%|████████▎ | 4927/5920 [53:51<04:29, 3.69it/s, tok/s=16235.2]
Inference: 83%|████████▎ | 4928/5920 [53:51<05:52, 2.82it/s, tok/s=16235.2]
Inference: 83%|████████▎ | 4928/5920 [53:51<05:52, 2.82it/s, tok/s=16232.8]
Inference: 83%|████████▎ | 4929/5920 [53:52<05:29, 3.01it/s, tok/s=16232.8]
Inference: 83%|████████▎ | 4929/5920 [53:52<05:29, 3.01it/s, tok/s=16232.0]
Inference: 83%|████████▎ | 4930/5920 [53:52<04:46, 3.46it/s, tok/s=16232.0]
Inference: 83%|████████▎ | 4930/5920 [53:52<04:46, 3.46it/s, tok/s=16232.3]
Inference: 83%|████████▎ | 4931/5920 [53:52<04:27, 3.69it/s, tok/s=16232.3]
Inference: 83%|████████▎ | 4931/5920 [53:52<04:27, 3.69it/s, tok/s=16238.3]
Inference: 83%|████████▎ | 4932/5920 [53:52<04:08, 3.97it/s, tok/s=16238.3]
Inference: 83%|████████▎ | 4932/5920 [53:52<04:08, 3.97it/s, tok/s=16237.8]
Inference: 83%|████████▎ | 4933/5920 [53:53<04:31, 3.64it/s, tok/s=16237.8]
Inference: 83%|████████▎ | 4933/5920 [53:53<04:31, 3.64it/s, tok/s=16237.7]
Inference: 83%|████████▎ | 4934/5920 [53:53<04:25, 3.71it/s, tok/s=16237.7]
Inference: 83%|████████▎ | 4934/5920 [53:53<04:25, 3.71it/s, tok/s=16237.9]
Inference: 83%|████████▎ | 4935/5920 [53:53<04:44, 3.47it/s, tok/s=16237.9]
Inference: 83%|████████▎ | 4935/5920 [53:53<04:44, 3.47it/s, tok/s=16237.3]
Inference: 83%|████████▎ | 4936/5920 [53:53<04:43, 3.47it/s, tok/s=16237.9]
Inference: 83%|████████▎ | 4937/5920 [53:53<03:08, 5.22it/s, tok/s=16237.9]
Inference: 83%|████████▎ | 4937/5920 [53:53<03:08, 5.22it/s, tok/s=16238.1]
Inference: 83%|████████▎ | 4938/5920 [53:53<03:00, 5.45it/s, tok/s=16238.1]
Inference: 83%|████████▎ | 4938/5920 [53:53<03:00, 5.45it/s, tok/s=16238.5]
Inference: 83%|████████▎ | 4939/5920 [53:54<03:40, 4.46it/s, tok/s=16238.5]
Inference: 83%|████████▎ | 4939/5920 [53:54<03:40, 4.46it/s, tok/s=16240.3]
Inference: 83%|████████▎ | 4940/5920 [53:54<03:15, 5.02it/s, tok/s=16240.3]
Inference: 83%|████████▎ | 4940/5920 [53:54<03:15, 5.02it/s, tok/s=16245.0]
Inference: 83%|████████▎ | 4941/5920 [53:54<03:14, 5.02it/s, tok/s=16246.0]
Inference: 83%|████████▎ | 4942/5920 [53:54<03:29, 4.66it/s, tok/s=16246.0]
Inference: 83%|████████▎ | 4942/5920 [53:54<03:29, 4.66it/s, tok/s=16244.2]
Inference: 83%|████████▎ | 4943/5920 [53:55<03:12, 5.06it/s, tok/s=16244.2]
Inference: 83%|████████▎ | 4943/5920 [53:55<03:12, 5.06it/s, tok/s=16246.7]
Inference: 84%|████████▎ | 4944/5920 [53:55<03:12, 5.06it/s, tok/s=16248.4]
Inference: 84%|████████▎ | 4945/5920 [53:55<03:39, 4.44it/s, tok/s=16248.4]
Inference: 84%|████████▎ | 4945/5920 [53:55<03:39, 4.44it/s, tok/s=16246.7]
Inference: 84%|████████▎ | 4946/5920 [53:56<04:21, 3.73it/s, tok/s=16246.7]
Inference: 84%|████████▎ | 4946/5920 [53:56<04:21, 3.73it/s, tok/s=16246.0]
Inference: 84%|████████▎ | 4947/5920 [53:56<04:21, 3.73it/s, tok/s=16246.2]
Inference: 84%|████████▎ | 4948/5920 [53:56<03:16, 4.94it/s, tok/s=16246.2]
Inference: 84%|████████▎ | 4948/5920 [53:56<03:16, 4.94it/s, tok/s=16249.4]
Inference: 84%|████████▎ | 4949/5920 [53:56<03:29, 4.63it/s, tok/s=16249.4]
Inference: 84%|████████▎ | 4949/5920 [53:56<03:29, 4.63it/s, tok/s=16252.2]
Inference: 84%|████████▎ | 4950/5920 [53:56<03:04, 5.25it/s, tok/s=16252.2]
Inference: 84%|████████▎ | 4950/5920 [53:56<03:04, 5.25it/s, tok/s=16255.2]
Inference: 84%|████████▎ | 4951/5920 [53:56<03:04, 5.25it/s, tok/s=16255.3]
Inference: 84%|████████▎ | 4952/5920 [53:57<03:51, 4.17it/s, tok/s=16255.3]
Inference: 84%|████████▎ | 4952/5920 [53:57<03:51, 4.17it/s, tok/s=16252.9]
Inference: 84%|████████▎ | 4953/5920 [53:57<04:19, 3.73it/s, tok/s=16252.9]
Inference: 84%|████████▎ | 4953/5920 [53:57<04:19, 3.73it/s, tok/s=16253.7]
Inference: 84%|████████▎ | 4954/5920 [53:57<04:19, 3.73it/s, tok/s=16254.4]
Inference: 84%|████████▎ | 4955/5920 [53:57<03:54, 4.11it/s, tok/s=16254.4]
Inference: 84%|████████▎ | 4955/5920 [53:57<03:54, 4.11it/s, tok/s=16253.5]
Inference: 84%|████████▎ | 4956/5920 [53:58<03:54, 4.11it/s, tok/s=16254.7]
Inference: 84%|████████▎ | 4957/5920 [53:58<03:08, 5.11it/s, tok/s=16254.7]
Inference: 84%|████████▎ | 4957/5920 [53:58<03:08, 5.11it/s, tok/s=16254.9]
Inference: 84%|████████▍ | 4958/5920 [53:58<03:03, 5.25it/s, tok/s=16254.9]
Inference: 84%|████████▍ | 4958/5920 [53:58<03:03, 5.25it/s, tok/s=16255.2]
Inference: 84%|████████▍ | 4959/5920 [53:58<03:48, 4.20it/s, tok/s=16255.2]
Inference: 84%|████████▍ | 4959/5920 [53:58<03:48, 4.20it/s, tok/s=16258.8]
Inference: 84%|████████▍ | 4960/5920 [53:58<03:31, 4.54it/s, tok/s=16258.8]
Inference: 84%|████████▍ | 4960/5920 [53:58<03:31, 4.54it/s, tok/s=16259.6]
Inference: 84%|████████▍ | 4961/5920 [53:59<03:33, 4.50it/s, tok/s=16259.6]
Inference: 84%|████████▍ | 4961/5920 [53:59<03:33, 4.50it/s, tok/s=16259.2]
Inference: 84%|████████▍ | 4962/5920 [53:59<03:25, 4.65it/s, tok/s=16259.2]
Inference: 84%|████████▍ | 4962/5920 [53:59<03:25, 4.65it/s, tok/s=16262.4]
Inference: 84%|████████▍ | 4963/5920 [53:59<04:28, 3.56it/s, tok/s=16262.4]
Inference: 84%|████████▍ | 4963/5920 [53:59<04:28, 3.56it/s, tok/s=16260.5]
Inference: 84%|████████▍ | 4964/5920 [54:00<05:22, 2.96it/s, tok/s=16260.5]
Inference: 84%|████████▍ | 4964/5920 [54:00<05:22, 2.96it/s, tok/s=16258.6]
Inference: 84%|████████▍ | 4965/5920 [54:00<04:21, 3.65it/s, tok/s=16258.6]
Inference: 84%|████████▍ | 4965/5920 [54:00<04:21, 3.65it/s, tok/s=16259.4]
Inference: 84%|████████▍ | 4966/5920 [54:00<04:05, 3.89it/s, tok/s=16259.4]
Inference: 84%|████████▍ | 4966/5920 [54:00<04:05, 3.89it/s, tok/s=16258.8]
Inference: 84%|████████▍ | 4967/5920 [54:00<04:19, 3.68it/s, tok/s=16258.8]
Inference: 84%|████████▍ | 4967/5920 [54:00<04:19, 3.68it/s, tok/s=16257.6]
Inference: 84%|████████▍ | 4968/5920 [54:00<04:18, 3.68it/s, tok/s=16258.6]
Inference: 84%|████████▍ | 4969/5920 [54:01<03:55, 4.03it/s, tok/s=16258.6]
Inference: 84%|████████▍ | 4969/5920 [54:01<03:55, 4.03it/s, tok/s=16257.5]
Inference: 84%|████████▍ | 4970/5920 [54:01<03:31, 4.49it/s, tok/s=16257.5]
Inference: 84%|████████▍ | 4970/5920 [54:01<03:31, 4.49it/s, tok/s=16257.2]
Inference: 84%|████████▍ | 4971/5920 [54:01<03:31, 4.49it/s, tok/s=16257.5]
Inference: 84%|████████▍ | 4972/5920 [54:01<03:31, 4.49it/s, tok/s=16258.4]
Inference: 84%|████████▍ | 4973/5920 [54:01<02:29, 6.32it/s, tok/s=16258.4]
Inference: 84%|████████▍ | 4973/5920 [54:01<02:29, 6.32it/s, tok/s=16259.9]
Inference: 84%|████████▍ | 4974/5920 [54:01<02:31, 6.26it/s, tok/s=16259.9]
Inference: 84%|████████▍ | 4974/5920 [54:01<02:31, 6.26it/s, tok/s=16260.5]
Inference: 84%|████████▍ | 4975/5920 [54:01<02:31, 6.26it/s, tok/s=16261.2]
Inference: 84%|████████▍ | 4976/5920 [54:02<03:54, 4.03it/s, tok/s=16261.2]
Inference: 84%|████████▍ | 4976/5920 [54:02<03:54, 4.03it/s, tok/s=16259.8]
Inference: 84%|████████▍ | 4977/5920 [54:03<06:37, 2.38it/s, tok/s=16259.8]
Inference: 84%|████████▍ | 4977/5920 [54:03<06:37, 2.38it/s, tok/s=16256.5]
Inference: 84%|████████▍ | 4978/5920 [54:04<05:58, 2.63it/s, tok/s=16256.5]
Inference: 84%|████████▍ | 4978/5920 [54:04<05:58, 2.63it/s, tok/s=16257.9]
Inference: 84%|████████▍ | 4979/5920 [54:04<06:35, 2.38it/s, tok/s=16257.9]
Inference: 84%|████████▍ | 4979/5920 [54:04<06:35, 2.38it/s, tok/s=16257.3]
Inference: 84%|████████▍ | 4980/5920 [54:04<05:19, 2.94it/s, tok/s=16257.3]
Inference: 84%|████████▍ | 4980/5920 [54:04<05:19, 2.94it/s, tok/s=16258.5]
Inference: 84%|████████▍ | 4981/5920 [54:04<05:19, 2.94it/s, tok/s=16258.8]
Inference: 84%|████████▍ | 4982/5920 [54:06<09:21, 1.67it/s, tok/s=16258.8]
Inference: 84%|████████▍ | 4982/5920 [54:06<09:21, 1.67it/s, tok/s=16251.9]
Inference: 84%|████████▍ | 4983/5920 [54:06<07:57, 1.96it/s, tok/s=16251.9]
Inference: 84%|████████▍ | 4983/5920 [54:06<07:57, 1.96it/s, tok/s=16254.0]
Inference: 84%|████████▍ | 4984/5920 [54:07<08:13, 1.90it/s, tok/s=16254.0]
Inference: 84%|████████▍ | 4984/5920 [54:07<08:13, 1.90it/s, tok/s=16252.0]
Inference: 84%|████████▍ | 4985/5920 [54:07<08:12, 1.90it/s, tok/s=16252.3]
Inference: 84%|████████▍ | 4986/5920 [54:07<05:13, 2.98it/s, tok/s=16252.3]
Inference: 84%|████████▍ | 4986/5920 [54:07<05:13, 2.98it/s, tok/s=16259.0]
Inference: 84%|████████▍ | 4987/5920 [54:07<05:13, 2.98it/s, tok/s=16259.8]
Inference: 84%|████████▍ | 4988/5920 [54:07<03:45, 4.13it/s, tok/s=16259.8]
Inference: 84%|████████▍ | 4988/5920 [54:07<03:45, 4.13it/s, tok/s=16260.2]
Inference: 84%|████████▍ | 4989/5920 [54:07<03:45, 4.13it/s, tok/s=16261.3]
Inference: 84%|████████▍ | 4990/5920 [54:08<03:55, 3.96it/s, tok/s=16261.3]
Inference: 84%|████████▍ | 4990/5920 [54:08<03:55, 3.96it/s, tok/s=16259.2]
Inference: 84%|████████▍ | 4991/5920 [54:08<04:41, 3.30it/s, tok/s=16259.2]
Inference: 84%|████████▍ | 4991/5920 [54:08<04:41, 3.30it/s, tok/s=16257.9]
Inference: 84%|████████▍ | 4992/5920 [54:09<05:07, 3.02it/s, tok/s=16257.9]
Inference: 84%|████████▍ | 4992/5920 [54:09<05:07, 3.02it/s, tok/s=16256.1]
Inference: 84%|████████▍ | 4993/5920 [54:09<04:17, 3.60it/s, tok/s=16256.1]
Inference: 84%|████████▍ | 4993/5920 [54:09<04:17, 3.60it/s, tok/s=16257.9]
Inference: 84%|████████▍ | 4994/5920 [54:09<05:01, 3.07it/s, tok/s=16257.9]
Inference: 84%|████████▍ | 4994/5920 [54:09<05:01, 3.07it/s, tok/s=16256.2]
Inference: 84%|████████▍ | 4995/5920 [54:09<04:05, 3.77it/s, tok/s=16256.2]
Inference: 84%|████████▍ | 4995/5920 [54:09<04:05, 3.77it/s, tok/s=16256.5]
Inference: 84%|████████▍ | 4996/5920 [54:09<04:05, 3.77it/s, tok/s=16258.9]
Inference: 84%|████████▍ | 4997/5920 [54:09<04:04, 3.77it/s, tok/s=16259.5]
Inference: 84%|████████▍ | 4998/5920 [54:10<02:22, 6.47it/s, tok/s=16259.5]
Inference: 84%|████████▍ | 4998/5920 [54:10<02:22, 6.47it/s, tok/s=16266.4]
Inference: 84%|████████▍ | 4999/5920 [54:11<04:51, 3.16it/s, tok/s=16266.4]
Inference: 84%|████████▍ | 4999/5920 [54:11<04:51, 3.16it/s, tok/s=16262.3]
Inference: 84%|████████▍ | 5000/5920 [54:12<07:31, 2.04it/s, tok/s=16262.3]
Inference: 84%|████████▍ | 5000/5920 [54:12<07:31, 2.04it/s, tok/s=16258.2]
Inference: 84%|████████▍ | 5001/5920 [54:12<06:30, 2.35it/s, tok/s=16258.2]
Inference: 84%|████████▍ | 5001/5920 [54:12<06:30, 2.35it/s, tok/s=16258.1]
Inference: 84%|████████▍ | 5002/5920 [54:12<06:17, 2.43it/s, tok/s=16258.1]
Inference: 84%|████████▍ | 5002/5920 [54:12<06:17, 2.43it/s, tok/s=16256.7]
Inference: 85%|████████▍ | 5003/5920 [54:13<06:19, 2.42it/s, tok/s=16256.7]
Inference: 85%|████████▍ | 5003/5920 [54:13<06:19, 2.42it/s, tok/s=16255.4]
Inference: 85%|████████▍ | 5004/5920 [54:13<06:19, 2.42it/s, tok/s=16256.0]
Inference: 85%|████████▍ | 5005/5920 [54:13<04:18, 3.54it/s, tok/s=16256.0]
Inference: 85%|████████▍ | 5005/5920 [54:13<04:18, 3.54it/s, tok/s=16255.8]
Inference: 85%|████████▍ | 5006/5920 [54:13<04:48, 3.17it/s, tok/s=16255.8]
Inference: 85%|████████▍ | 5006/5920 [54:13<04:48, 3.17it/s, tok/s=16255.4]
Inference: 85%|████████▍ | 5007/5920 [54:14<04:37, 3.29it/s, tok/s=16255.4]
Inference: 85%|████████▍ | 5007/5920 [54:14<04:37, 3.29it/s, tok/s=16255.0]
Inference: 85%|████████▍ | 5008/5920 [54:14<05:49, 2.61it/s, tok/s=16255.0]
Inference: 85%|████████▍ | 5008/5920 [54:14<05:49, 2.61it/s, tok/s=16255.0]
Inference: 85%|████████▍ | 5009/5920 [54:15<07:24, 2.05it/s, tok/s=16255.0]
Inference: 85%|████████▍ | 5009/5920 [54:15<07:24, 2.05it/s, tok/s=16253.8]
Inference: 85%|████████▍ | 5010/5920 [54:15<06:04, 2.49it/s, tok/s=16253.8]
Inference: 85%|████████▍ | 5010/5920 [54:15<06:04, 2.49it/s, tok/s=16254.4]
Inference: 85%|████████▍ | 5011/5920 [54:16<06:29, 2.33it/s, tok/s=16254.4]
Inference: 85%|████████▍ | 5011/5920 [54:16<06:29, 2.33it/s, tok/s=16254.9]
Inference: 85%|████████▍ | 5012/5920 [54:16<06:15, 2.42it/s, tok/s=16254.9]
Inference: 85%|████████▍ | 5012/5920 [54:16<06:15, 2.42it/s, tok/s=16253.9]
Inference: 85%|████████▍ | 5013/5920 [54:16<06:26, 2.34it/s, tok/s=16253.9]
Inference: 85%|████████▍ | 5013/5920 [54:16<06:26, 2.34it/s, tok/s=16252.3]
Inference: 85%|████████▍ | 5014/5920 [54:17<05:10, 2.92it/s, tok/s=16252.3]
Inference: 85%|████████▍ | 5014/5920 [54:17<05:10, 2.92it/s, tok/s=16252.4]
Inference: 85%|████████▍ | 5015/5920 [54:17<06:22, 2.37it/s, tok/s=16252.4]
Inference: 85%|████████▍ | 5015/5920 [54:17<06:22, 2.37it/s, tok/s=16249.8]
Inference: 85%|████████▍ | 5016/5920 [54:17<06:21, 2.37it/s, tok/s=16251.0]
Inference: 85%|████████▍ | 5017/5920 [54:17<04:09, 3.62it/s, tok/s=16251.0]
Inference: 85%|████████▍ | 5017/5920 [54:17<04:09, 3.62it/s, tok/s=16251.0]
Inference: 85%|████████▍ | 5018/5920 [54:17<04:09, 3.62it/s, tok/s=16259.8]
Inference: 85%|████████▍ | 5019/5920 [54:18<03:28, 4.32it/s, tok/s=16259.8]
Inference: 85%|████████▍ | 5019/5920 [54:18<03:28, 4.32it/s, tok/s=16260.2]
Inference: 85%|████████▍ | 5020/5920 [54:18<04:41, 3.20it/s, tok/s=16260.2]
Inference: 85%|████████▍ | 5020/5920 [54:18<04:41, 3.20it/s, tok/s=16258.3]
Inference: 85%|████████▍ | 5021/5920 [54:18<04:06, 3.65it/s, tok/s=16258.3]
Inference: 85%|████████▍ | 5021/5920 [54:18<04:06, 3.65it/s, tok/s=16264.9]
Inference: 85%|████████▍ | 5022/5920 [54:19<04:36, 3.25it/s, tok/s=16264.9]
Inference: 85%|████████▍ | 5022/5920 [54:19<04:36, 3.25it/s, tok/s=16270.3]
Inference: 85%|████████▍ | 5023/5920 [54:19<04:35, 3.25it/s, tok/s=16272.0]
Inference: 85%|████████▍ | 5024/5920 [54:19<04:12, 3.56it/s, tok/s=16272.0]
Inference: 85%|████████▍ | 5024/5920 [54:19<04:12, 3.56it/s, tok/s=16271.0]
Inference: 85%|████████▍ | 5025/5920 [54:20<03:48, 3.92it/s, tok/s=16271.0]
Inference: 85%|████████▍ | 5025/5920 [54:20<03:48, 3.92it/s, tok/s=16272.8]
Inference: 85%|████████▍ | 5026/5920 [54:20<03:47, 3.92it/s, tok/s=16273.2]
Inference: 85%|████████▍ | 5027/5920 [54:20<02:52, 5.17it/s, tok/s=16273.2]
Inference: 85%|████████▍ | 5027/5920 [54:20<02:52, 5.17it/s, tok/s=16273.4]
Inference: 85%|████████▍ | 5028/5920 [54:20<03:18, 4.50it/s, tok/s=16273.4]
Inference: 85%|████████▍ | 5028/5920 [54:20<03:18, 4.50it/s, tok/s=16272.6]
Inference: 85%|████████▍ | 5029/5920 [54:20<03:02, 4.88it/s, tok/s=16272.6]
Inference: 85%|████████▍ | 5029/5920 [54:20<03:02, 4.88it/s, tok/s=16274.1]
Inference: 85%|████████▍ | 5030/5920 [54:21<04:02, 3.67it/s, tok/s=16274.1]
Inference: 85%|████████▍ | 5030/5920 [54:21<04:02, 3.67it/s, tok/s=16273.7]
Inference: 85%|████████▍ | 5031/5920 [54:21<04:32, 3.26it/s, tok/s=16273.7]
Inference: 85%|████████▍ | 5031/5920 [54:21<04:32, 3.26it/s, tok/s=16275.7]
Inference: 85%|████████▌ | 5032/5920 [54:22<08:05, 1.83it/s, tok/s=16275.7]
Inference: 85%|████████▌ | 5032/5920 [54:22<08:05, 1.83it/s, tok/s=16271.1]
Inference: 85%|████████▌ | 5033/5920 [54:23<06:42, 2.20it/s, tok/s=16271.1]
Inference: 85%|████████▌ | 5033/5920 [54:23<06:42, 2.20it/s, tok/s=16270.4]
Inference: 85%|████████▌ | 5034/5920 [54:23<06:42, 2.20it/s, tok/s=16271.7]
Inference: 85%|████████▌ | 5035/5920 [54:23<06:41, 2.20it/s, tok/s=16273.3]
Inference: 85%|████████▌ | 5036/5920 [54:23<04:56, 2.99it/s, tok/s=16273.3]
Inference: 85%|████████▌ | 5036/5920 [54:23<04:56, 2.99it/s, tok/s=16271.3]
Inference: 85%|████████▌ | 5037/5920 [54:24<06:23, 2.30it/s, tok/s=16271.3]
Inference: 85%|████████▌ | 5037/5920 [54:24<06:23, 2.30it/s, tok/s=16268.0]
Inference: 85%|████████▌ | 5038/5920 [54:24<06:17, 2.34it/s, tok/s=16268.0]
Inference: 85%|████████▌ | 5038/5920 [54:24<06:17, 2.34it/s, tok/s=16270.0]
Inference: 85%|████████▌ | 5039/5920 [54:25<05:42, 2.57it/s, tok/s=16270.0]
Inference: 85%|████████▌ | 5039/5920 [54:25<05:42, 2.57it/s, tok/s=16269.4]
Inference: 85%|████████▌ | 5040/5920 [54:25<06:32, 2.24it/s, tok/s=16269.4]
Inference: 85%|████████▌ | 5040/5920 [54:25<06:32, 2.24it/s, tok/s=16267.4]
Inference: 85%|████████▌ | 5041/5920 [54:25<06:32, 2.24it/s, tok/s=16270.3]
Inference: 85%|████████▌ | 5042/5920 [54:26<04:39, 3.14it/s, tok/s=16270.3]
Inference: 85%|████████▌ | 5042/5920 [54:26<04:39, 3.14it/s, tok/s=16270.0]
Inference: 85%|████████▌ | 5043/5920 [54:26<04:53, 2.99it/s, tok/s=16270.0]
Inference: 85%|████████▌ | 5043/5920 [54:26<04:53, 2.99it/s, tok/s=16269.4]
Inference: 85%|████████▌ | 5044/5920 [54:26<04:52, 2.99it/s, tok/s=16270.2]
Inference: 85%|████████▌ | 5045/5920 [54:26<04:52, 2.99it/s, tok/s=16270.6]
Inference: 85%|████████▌ | 5046/5920 [54:26<03:14, 4.48it/s, tok/s=16270.6]
Inference: 85%|████████▌ | 5046/5920 [54:26<03:14, 4.48it/s, tok/s=16272.8]
Inference: 85%|████████▌ | 5047/5920 [54:26<03:14, 4.48it/s, tok/s=16273.9]
Inference: 85%|████████▌ | 5048/5920 [54:26<02:33, 5.67it/s, tok/s=16273.9]
Inference: 85%|████████▌ | 5048/5920 [54:26<02:33, 5.67it/s, tok/s=16278.2]
Inference: 85%|████████▌ | 5049/5920 [54:27<03:37, 4.01it/s, tok/s=16278.2]
Inference: 85%|████████▌ | 5049/5920 [54:27<03:37, 4.01it/s, tok/s=16278.4]
Inference: 85%|████████▌ | 5050/5920 [54:27<03:36, 4.01it/s, tok/s=16279.1]
Inference: 85%|████████▌ | 5051/5920 [54:27<03:04, 4.72it/s, tok/s=16279.1]
Inference: 85%|████████▌ | 5051/5920 [54:27<03:04, 4.72it/s, tok/s=16278.6]
Inference: 85%|████████▌ | 5052/5920 [54:27<03:03, 4.72it/s, tok/s=16280.4]
Inference: 85%|████████▌ | 5053/5920 [54:28<03:26, 4.20it/s, tok/s=16280.4]
Inference: 85%|████████▌ | 5053/5920 [54:28<03:26, 4.20it/s, tok/s=16278.6]
Inference: 85%|████████▌ | 5054/5920 [54:28<03:26, 4.20it/s, tok/s=16279.5]
Inference: 85%|████████▌ | 5055/5920 [54:28<02:48, 5.12it/s, tok/s=16279.5]
Inference: 85%|████████▌ | 5055/5920 [54:28<02:48, 5.12it/s, tok/s=16279.8]
Inference: 85%|████████▌ | 5056/5920 [54:29<03:20, 4.32it/s, tok/s=16279.8]
Inference: 85%|████████▌ | 5056/5920 [54:29<03:20, 4.32it/s, tok/s=16280.5]
Inference: 85%|████████▌ | 5057/5920 [54:29<02:58, 4.83it/s, tok/s=16280.5]
Inference: 85%|████████▌ | 5057/5920 [54:29<02:58, 4.83it/s, tok/s=16281.7]
Inference: 85%|████████▌ | 5058/5920 [54:29<02:58, 4.83it/s, tok/s=16282.3]
Inference: 85%|████████▌ | 5059/5920 [54:29<02:16, 6.33it/s, tok/s=16282.3]
Inference: 85%|████████▌ | 5059/5920 [54:29<02:16, 6.33it/s, tok/s=16283.1]
Inference: 85%|████████▌ | 5060/5920 [54:29<02:15, 6.33it/s, tok/s=16284.0]
Inference: 85%|████████▌ | 5061/5920 [54:29<01:53, 7.55it/s, tok/s=16284.0]
Inference: 85%|████████▌ | 5061/5920 [54:29<01:53, 7.55it/s, tok/s=16285.3]
Inference: 86%|████████▌ | 5062/5920 [54:29<01:59, 7.16it/s, tok/s=16285.3]
Inference: 86%|████████▌ | 5062/5920 [54:29<01:59, 7.16it/s, tok/s=16286.0]
Inference: 86%|████████▌ | 5063/5920 [54:29<02:23, 5.98it/s, tok/s=16286.0]
Inference: 86%|████████▌ | 5063/5920 [54:29<02:23, 5.98it/s, tok/s=16287.4]
Inference: 86%|████████▌ | 5064/5920 [54:29<02:23, 5.98it/s, tok/s=16290.3]
Inference: 86%|████████▌ | 5065/5920 [54:29<02:22, 5.98it/s, tok/s=16292.8]
Inference: 86%|████████▌ | 5066/5920 [54:30<02:01, 7.01it/s, tok/s=16292.8]
Inference: 86%|████████▌ | 5066/5920 [54:30<02:01, 7.01it/s, tok/s=16296.0]
Inference: 86%|████████▌ | 5067/5920 [54:30<02:01, 7.01it/s, tok/s=16296.3]
Inference: 86%|████████▌ | 5068/5920 [54:30<01:49, 7.76it/s, tok/s=16296.3]
Inference: 86%|████████▌ | 5068/5920 [54:30<01:49, 7.76it/s, tok/s=16302.0]
Inference: 86%|████████▌ | 5069/5920 [54:30<01:49, 7.76it/s, tok/s=16302.5]
Inference: 86%|████████▌ | 5070/5920 [54:30<02:06, 6.74it/s, tok/s=16302.5]
Inference: 86%|████████▌ | 5070/5920 [54:30<02:06, 6.74it/s, tok/s=16301.2]
Inference: 86%|████████▌ | 5071/5920 [54:30<02:05, 6.74it/s, tok/s=16304.1]
Inference: 86%|████████▌ | 5072/5920 [54:30<01:43, 8.23it/s, tok/s=16304.1]
Inference: 86%|████████▌ | 5072/5920 [54:30<01:43, 8.23it/s, tok/s=16309.4]
Inference: 86%|████████▌ | 5073/5920 [54:31<01:42, 8.23it/s, tok/s=16311.8]
Inference: 86%|████████▌ | 5074/5920 [54:31<02:08, 6.60it/s, tok/s=16311.8]
Inference: 86%|████████▌ | 5074/5920 [54:31<02:08, 6.60it/s, tok/s=16312.4]
Inference: 86%|████████▌ | 5075/5920 [54:31<02:22, 5.92it/s, tok/s=16312.4]
Inference: 86%|████████▌ | 5075/5920 [54:31<02:22, 5.92it/s, tok/s=16312.1]
Inference: 86%|████████▌ | 5076/5920 [54:31<02:19, 6.04it/s, tok/s=16312.1]
Inference: 86%|████████▌ | 5076/5920 [54:31<02:19, 6.04it/s, tok/s=16312.7]
Inference: 86%|████████▌ | 5077/5920 [54:31<02:19, 6.04it/s, tok/s=16314.0]
Inference: 86%|████████▌ | 5078/5920 [54:31<02:19, 6.04it/s, tok/s=16315.5]
Inference: 86%|████████▌ | 5079/5920 [54:31<01:38, 8.56it/s, tok/s=16315.5]
Inference: 86%|████████▌ | 5079/5920 [54:31<01:38, 8.56it/s, tok/s=16315.8]
Inference: 86%|████████▌ | 5080/5920 [54:32<01:38, 8.56it/s, tok/s=16316.4]
Inference: 86%|████████▌ | 5081/5920 [54:32<01:56, 7.21it/s, tok/s=16316.4]
Inference: 86%|████████▌ | 5081/5920 [54:32<01:56, 7.21it/s, tok/s=16316.6]
Inference: 86%|████████▌ | 5082/5920 [54:32<01:56, 7.22it/s, tok/s=16316.6]
Inference: 86%|████████▌ | 5082/5920 [54:32<01:56, 7.22it/s, tok/s=16318.9]
Inference: 86%|████████▌ | 5083/5920 [54:32<02:11, 6.36it/s, tok/s=16318.9]
Inference: 86%|████████▌ | 5083/5920 [54:32<02:11, 6.36it/s, tok/s=16318.7]
Inference: 86%|████████▌ | 5084/5920 [54:32<02:11, 6.36it/s, tok/s=16319.5]
Inference: 86%|████████▌ | 5085/5920 [54:33<03:04, 4.53it/s, tok/s=16319.5]
Inference: 86%|████████▌ | 5085/5920 [54:33<03:04, 4.53it/s, tok/s=16321.4]
Inference: 86%|████████▌ | 5086/5920 [54:33<03:55, 3.54it/s, tok/s=16321.4]
Inference: 86%|████████▌ | 5086/5920 [54:33<03:55, 3.54it/s, tok/s=16319.5]
Inference: 86%|████████▌ | 5087/5920 [54:34<04:03, 3.41it/s, tok/s=16319.5]
Inference: 86%|████████▌ | 5087/5920 [54:34<04:03, 3.41it/s, tok/s=16318.8]
Inference: 86%|████████▌ | 5088/5920 [54:34<04:03, 3.41it/s, tok/s=16327.1]
Inference: 86%|████████▌ | 5089/5920 [54:34<04:32, 3.05it/s, tok/s=16327.1]
Inference: 86%|████████▌ | 5089/5920 [54:34<04:32, 3.05it/s, tok/s=16329.4]
Inference: 86%|████████▌ | 5090/5920 [54:34<04:32, 3.05it/s, tok/s=16334.4]
Inference: 86%|████████▌ | 5091/5920 [54:35<03:32, 3.90it/s, tok/s=16334.4]
Inference: 86%|████████▌ | 5091/5920 [54:35<03:32, 3.90it/s, tok/s=16334.9]
Inference: 86%|████████▌ | 5092/5920 [54:35<03:34, 3.86it/s, tok/s=16334.9]
Inference: 86%|████████▌ | 5092/5920 [54:35<03:34, 3.86it/s, tok/s=16336.4]
Inference: 86%|████████▌ | 5093/5920 [54:35<03:34, 3.86it/s, tok/s=16336.7]
Inference: 86%|████████▌ | 5094/5920 [54:35<03:24, 4.04it/s, tok/s=16336.7]
Inference: 86%|████████▌ | 5094/5920 [54:35<03:24, 4.04it/s, tok/s=16337.1]
Inference: 86%|████████▌ | 5095/5920 [54:36<04:17, 3.20it/s, tok/s=16337.1]
Inference: 86%|████████▌ | 5095/5920 [54:36<04:17, 3.20it/s, tok/s=16341.7]
Inference: 86%|████████▌ | 5096/5920 [54:36<04:17, 3.20it/s, tok/s=16343.3]
Inference: 86%|████████▌ | 5097/5920 [54:36<03:25, 4.00it/s, tok/s=16343.3]
Inference: 86%|████████▌ | 5097/5920 [54:36<03:25, 4.00it/s, tok/s=16346.1]
Inference: 86%|████████▌ | 5098/5920 [54:36<03:06, 4.41it/s, tok/s=16346.1]
Inference: 86%|████████▌ | 5098/5920 [54:36<03:06, 4.41it/s, tok/s=16346.0]
Inference: 86%|████████▌ | 5099/5920 [54:37<04:32, 3.01it/s, tok/s=16346.0]
Inference: 86%|████████▌ | 5099/5920 [54:37<04:32, 3.01it/s, tok/s=16343.5]
Inference: 86%|████████▌ | 5100/5920 [54:37<04:34, 2.99it/s, tok/s=16343.5]
Inference: 86%|████████▌ | 5100/5920 [54:37<04:34, 2.99it/s, tok/s=16346.2]
Inference: 86%|████████▌ | 5101/5920 [54:38<03:59, 3.42it/s, tok/s=16346.2]
Inference: 86%|████████▌ | 5101/5920 [54:38<03:59, 3.42it/s, tok/s=16346.3]
Inference: 86%|████████▌ | 5102/5920 [54:38<03:25, 3.99it/s, tok/s=16346.3]
Inference: 86%|████████▌ | 5102/5920 [54:38<03:25, 3.99it/s, tok/s=16346.5]
Inference: 86%|████████▌ | 5103/5920 [54:38<04:38, 2.94it/s, tok/s=16346.5]
Inference: 86%|████████▌ | 5103/5920 [54:38<04:38, 2.94it/s, tok/s=16346.3]
Inference: 86%|████████▌ | 5104/5920 [54:38<04:37, 2.94it/s, tok/s=16346.8]
Inference: 86%|████████▌ | 5105/5920 [54:39<03:36, 3.76it/s, tok/s=16346.8]
Inference: 86%|████████▌ | 5105/5920 [54:39<03:36, 3.76it/s, tok/s=16346.2]
Inference: 86%|████████▋ | 5106/5920 [54:39<03:10, 4.28it/s, tok/s=16346.2]
Inference: 86%|████████▋ | 5106/5920 [54:39<03:10, 4.28it/s, tok/s=16347.3]
Inference: 86%|████████▋ | 5107/5920 [54:39<02:49, 4.78it/s, tok/s=16347.3]
Inference: 86%|████████▋ | 5107/5920 [54:39<02:49, 4.78it/s, tok/s=16347.2]
Inference: 86%|████████▋ | 5108/5920 [54:39<03:34, 3.79it/s, tok/s=16347.2]
Inference: 86%|████████▋ | 5108/5920 [54:39<03:34, 3.79it/s, tok/s=16346.4]
Inference: 86%|████████▋ | 5109/5920 [54:40<03:47, 3.57it/s, tok/s=16346.4]
Inference: 86%|████████▋ | 5109/5920 [54:40<03:47, 3.57it/s, tok/s=16345.2]
Inference: 86%|████████▋ | 5110/5920 [54:40<04:18, 3.13it/s, tok/s=16345.2]
Inference: 86%|████████▋ | 5110/5920 [54:40<04:18, 3.13it/s, tok/s=16344.4]
Inference: 86%|████████▋ | 5111/5920 [54:41<04:53, 2.76it/s, tok/s=16344.4]
Inference: 86%|████████▋ | 5111/5920 [54:41<04:53, 2.76it/s, tok/s=16343.0]
Inference: 86%|████████▋ | 5112/5920 [54:41<05:27, 2.47it/s, tok/s=16343.0]
Inference: 86%|████████▋ | 5112/5920 [54:41<05:27, 2.47it/s, tok/s=16341.0]
Inference: 86%|████████▋ | 5113/5920 [54:41<05:18, 2.54it/s, tok/s=16341.0]
Inference: 86%|████████▋ | 5113/5920 [54:41<05:18, 2.54it/s, tok/s=16342.1]
Inference: 86%|████████▋ | 5114/5920 [54:42<04:29, 2.99it/s, tok/s=16342.1]
Inference: 86%|████████▋ | 5114/5920 [54:42<04:29, 2.99it/s, tok/s=16342.3]
Inference: 86%|████████▋ | 5115/5920 [54:42<04:29, 2.99it/s, tok/s=16345.5]
Inference: 86%|████████▋ | 5116/5920 [54:42<04:17, 3.12it/s, tok/s=16345.5]
Inference: 86%|████████▋ | 5116/5920 [54:42<04:17, 3.12it/s, tok/s=16347.4]
Inference: 86%|████████▋ | 5117/5920 [54:43<04:58, 2.69it/s, tok/s=16347.4]
Inference: 86%|████████▋ | 5117/5920 [54:43<04:58, 2.69it/s, tok/s=16346.5]
Inference: 86%|████████▋ | 5118/5920 [54:43<04:57, 2.69it/s, tok/s=16348.4]
Inference: 86%|████████▋ | 5119/5920 [54:43<03:13, 4.14it/s, tok/s=16348.4]
Inference: 86%|████████▋ | 5119/5920 [54:43<03:13, 4.14it/s, tok/s=16352.3]
Inference: 86%|████████▋ | 5120/5920 [54:44<07:18, 1.82it/s, tok/s=16352.3]
Inference: 86%|████████▋ | 5120/5920 [54:44<07:18, 1.82it/s, tok/s=16345.0]
Inference: 87%|████████▋ | 5121/5920 [54:45<06:00, 2.21it/s, tok/s=16345.0]
Inference: 87%|████████▋ | 5121/5920 [54:45<06:00, 2.21it/s, tok/s=16347.5]
Inference: 87%|████████▋ | 5122/5920 [54:45<05:58, 2.23it/s, tok/s=16347.5]
Inference: 87%|████████▋ | 5122/5920 [54:45<05:58, 2.23it/s, tok/s=16346.6]
Inference: 87%|████████▋ | 5123/5920 [54:45<05:57, 2.23it/s, tok/s=16347.9]
Inference: 87%|████████▋ | 5124/5920 [54:45<03:58, 3.34it/s, tok/s=16347.9]
Inference: 87%|████████▋ | 5124/5920 [54:45<03:58, 3.34it/s, tok/s=16349.2]
Inference: 87%|████████▋ | 5125/5920 [54:45<03:36, 3.67it/s, tok/s=16349.2]
Inference: 87%|████████▋ | 5125/5920 [54:45<03:36, 3.67it/s, tok/s=16348.8]
Inference: 87%|████████▋ | 5126/5920 [54:46<04:58, 2.66it/s, tok/s=16348.8]
Inference: 87%|████████▋ | 5126/5920 [54:46<04:58, 2.66it/s, tok/s=16346.2]
Inference: 87%|████████▋ | 5127/5920 [54:46<04:58, 2.66it/s, tok/s=16346.8]
Inference: 87%|████████▋ | 5128/5920 [54:46<04:58, 2.66it/s, tok/s=16349.4]
Inference: 87%|████████▋ | 5129/5920 [54:47<03:44, 3.53it/s, tok/s=16349.4]
Inference: 87%|████████▋ | 5129/5920 [54:47<03:44, 3.53it/s, tok/s=16348.2]
Inference: 87%|████████▋ | 5130/5920 [54:47<03:22, 3.90it/s, tok/s=16348.2]
Inference: 87%|████████▋ | 5130/5920 [54:47<03:22, 3.90it/s, tok/s=16348.4]
Inference: 87%|████████▋ | 5131/5920 [54:48<04:47, 2.74it/s, tok/s=16348.4]
Inference: 87%|████████▋ | 5131/5920 [54:48<04:47, 2.74it/s, tok/s=16348.1]
Inference: 87%|████████▋ | 5132/5920 [54:48<04:47, 2.74it/s, tok/s=16348.7]
Inference: 87%|████████▋ | 5133/5920 [54:48<03:23, 3.87it/s, tok/s=16348.7]
Inference: 87%|████████▋ | 5133/5920 [54:48<03:23, 3.87it/s, tok/s=16350.7]
Inference: 87%|████████▋ | 5134/5920 [54:48<03:23, 3.87it/s, tok/s=16351.3]
Inference: 87%|████████▋ | 5135/5920 [54:49<04:02, 3.23it/s, tok/s=16351.3]
Inference: 87%|████████▋ | 5135/5920 [54:49<04:02, 3.23it/s, tok/s=16354.2]
Inference: 87%|████████▋ | 5136/5920 [54:49<03:47, 3.44it/s, tok/s=16354.2]
Inference: 87%|████████▋ | 5136/5920 [54:49<03:47, 3.44it/s, tok/s=16354.2]
Inference: 87%|████████▋ | 5137/5920 [54:49<04:50, 2.69it/s, tok/s=16354.2]
Inference: 87%|████████▋ | 5137/5920 [54:49<04:50, 2.69it/s, tok/s=16351.5]
Inference: 87%|████████▋ | 5138/5920 [54:50<05:51, 2.23it/s, tok/s=16351.5]
Inference: 87%|████████▋ | 5138/5920 [54:50<05:51, 2.23it/s, tok/s=16351.5]
Inference: 87%|████████▋ | 5139/5920 [54:50<05:09, 2.53it/s, tok/s=16351.5]
Inference: 87%|████████▋ | 5139/5920 [54:50<05:09, 2.53it/s, tok/s=16351.5]
Inference: 87%|████████▋ | 5140/5920 [54:51<05:33, 2.34it/s, tok/s=16351.5]
Inference: 87%|████████▋ | 5140/5920 [54:51<05:33, 2.34it/s, tok/s=16351.6]
Inference: 87%|████████▋ | 5141/5920 [54:52<07:25, 1.75it/s, tok/s=16351.6]
Inference: 87%|████████▋ | 5141/5920 [54:52<07:25, 1.75it/s, tok/s=16348.5]
Inference: 87%|████████▋ | 5142/5920 [54:52<07:22, 1.76it/s, tok/s=16348.5]
Inference: 87%|████████▋ | 5142/5920 [54:52<07:22, 1.76it/s, tok/s=16346.4]
Inference: 87%|████████▋ | 5143/5920 [54:53<07:22, 1.76it/s, tok/s=16347.2]
Inference: 87%|████████▋ | 5144/5920 [54:53<05:56, 2.18it/s, tok/s=16347.2]
Inference: 87%|████████▋ | 5144/5920 [54:53<05:56, 2.18it/s, tok/s=16345.6]
Inference: 87%|████████▋ | 5145/5920 [54:54<07:55, 1.63it/s, tok/s=16345.6]
Inference: 87%|████████▋ | 5145/5920 [54:54<07:55, 1.63it/s, tok/s=16341.4]
Inference: 87%|████████▋ | 5146/5920 [54:54<06:14, 2.07it/s, tok/s=16341.4]
Inference: 87%|████████▋ | 5146/5920 [54:54<06:14, 2.07it/s, tok/s=16342.7]
Inference: 87%|████████▋ | 5147/5920 [54:55<08:31, 1.51it/s, tok/s=16342.7]
Inference: 87%|████████▋ | 5147/5920 [54:55<08:31, 1.51it/s, tok/s=16340.2]
Inference: 87%|████████▋ | 5148/5920 [54:56<09:42, 1.32it/s, tok/s=16340.2]
Inference: 87%|████████▋ | 5148/5920 [54:56<09:42, 1.32it/s, tok/s=16336.3]
Inference: 87%|████████▋ | 5149/5920 [54:56<09:42, 1.32it/s, tok/s=16337.8]
Inference: 87%|████████▋ | 5150/5920 [54:57<07:02, 1.82it/s, tok/s=16337.8]
Inference: 87%|████████▋ | 5150/5920 [54:57<07:02, 1.82it/s, tok/s=16338.1]
Inference: 87%|████████▋ | 5151/5920 [54:58<08:13, 1.56it/s, tok/s=16338.1]
Inference: 87%|████████▋ | 5151/5920 [54:58<08:13, 1.56it/s, tok/s=16336.0]
Inference: 87%|████████▋ | 5152/5920 [54:58<08:12, 1.56it/s, tok/s=16339.3]
Inference: 87%|████████▋ | 5153/5920 [54:58<05:41, 2.24it/s, tok/s=16339.3]
Inference: 87%|████████▋ | 5153/5920 [54:58<05:41, 2.24it/s, tok/s=16339.2]
Inference: 87%|████████▋ | 5154/5920 [54:58<05:41, 2.24it/s, tok/s=16342.2]
Inference: 87%|████████▋ | 5155/5920 [54:58<03:57, 3.23it/s, tok/s=16342.2]
Inference: 87%|████████▋ | 5155/5920 [54:58<03:57, 3.23it/s, tok/s=16343.3]
Inference: 87%|████████▋ | 5156/5920 [54:59<03:52, 3.29it/s, tok/s=16343.3]
Inference: 87%|████████▋ | 5156/5920 [54:59<03:52, 3.29it/s, tok/s=16343.1]
Inference: 87%|████████▋ | 5157/5920 [55:00<05:20, 2.38it/s, tok/s=16343.1]
Inference: 87%|████████▋ | 5157/5920 [55:00<05:20, 2.38it/s, tok/s=16343.6]
Inference: 87%|████████▋ | 5158/5920 [55:00<05:07, 2.48it/s, tok/s=16343.6]
Inference: 87%|████████▋ | 5158/5920 [55:00<05:07, 2.48it/s, tok/s=16343.7]
Inference: 87%|████████▋ | 5159/5920 [55:00<05:06, 2.48it/s, tok/s=16346.1]
Inference: 87%|████████▋ | 5160/5920 [55:01<05:25, 2.33it/s, tok/s=16346.1]
Inference: 87%|████████▋ | 5160/5920 [55:01<05:25, 2.33it/s, tok/s=16346.1]
Inference: 87%|████████▋ | 5161/5920 [55:01<05:26, 2.33it/s, tok/s=16346.1]
Inference: 87%|████████▋ | 5161/5920 [55:01<05:26, 2.33it/s, tok/s=16345.8]
Inference: 87%|████████▋ | 5162/5920 [55:01<05:25, 2.33it/s, tok/s=16350.6]
Inference: 87%|████████▋ | 5163/5920 [55:02<04:31, 2.78it/s, tok/s=16350.6]
Inference: 87%|████████▋ | 5163/5920 [55:02<04:31, 2.78it/s, tok/s=16349.0]
Inference: 87%|████████▋ | 5164/5920 [55:02<04:52, 2.58it/s, tok/s=16349.0]
Inference: 87%|████████▋ | 5164/5920 [55:02<04:52, 2.58it/s, tok/s=16351.7]
Inference: 87%|████████▋ | 5165/5920 [55:03<05:00, 2.51it/s, tok/s=16351.7]
Inference: 87%|████████▋ | 5165/5920 [55:03<05:00, 2.51it/s, tok/s=16351.2]
Inference: 87%|████████▋ | 5166/5920 [55:03<04:14, 2.96it/s, tok/s=16351.2]
Inference: 87%|████████▋ | 5166/5920 [55:03<04:14, 2.96it/s, tok/s=16352.4]
Inference: 87%|████████▋ | 5167/5920 [55:03<04:14, 2.96it/s, tok/s=16355.1]
Inference: 87%|████████▋ | 5168/5920 [55:03<03:20, 3.75it/s, tok/s=16355.1]
Inference: 87%|████████▋ | 5168/5920 [55:03<03:20, 3.75it/s, tok/s=16356.3]
Inference: 87%|████████▋ | 5169/5920 [55:04<04:19, 2.90it/s, tok/s=16356.3]
Inference: 87%|████████▋ | 5169/5920 [55:04<04:19, 2.90it/s, tok/s=16356.9]
Inference: 87%|████████▋ | 5170/5920 [55:04<04:53, 2.55it/s, tok/s=16356.9]
Inference: 87%|████████▋ | 5170/5920 [55:04<04:53, 2.55it/s, tok/s=16358.5]
Inference: 87%|████████▋ | 5171/5920 [55:06<07:50, 1.59it/s, tok/s=16358.5]
Inference: 87%|████████▋ | 5171/5920 [55:06<07:50, 1.59it/s, tok/s=16353.1]
Inference: 87%|████████▋ | 5172/5920 [55:06<07:52, 1.58it/s, tok/s=16353.1]
Inference: 87%|████████▋ | 5172/5920 [55:06<07:52, 1.58it/s, tok/s=16354.2]
Inference: 87%|████████▋ | 5173/5920 [55:07<08:45, 1.42it/s, tok/s=16354.2]
Inference: 87%|████████▋ | 5173/5920 [55:07<08:45, 1.42it/s, tok/s=16350.8]
Inference: 87%|████████▋ | 5174/5920 [55:08<09:02, 1.38it/s, tok/s=16350.8]
Inference: 87%|████████▋ | 5174/5920 [55:08<09:02, 1.38it/s, tok/s=16356.0]
Inference: 87%|████████▋ | 5175/5920 [55:08<07:32, 1.65it/s, tok/s=16356.0]
Inference: 87%|████████▋ | 5175/5920 [55:08<07:32, 1.65it/s, tok/s=16357.8]
Inference: 87%|████████▋ | 5176/5920 [55:08<05:48, 2.13it/s, tok/s=16357.8]
Inference: 87%|████████▋ | 5176/5920 [55:08<05:48, 2.13it/s, tok/s=16365.6]
Inference: 87%|████████▋ | 5177/5920 [55:09<05:23, 2.30it/s, tok/s=16365.6]
Inference: 87%|████████▋ | 5177/5920 [55:09<05:23, 2.30it/s, tok/s=16366.0]
Inference: 87%|████████▋ | 5178/5920 [55:09<05:03, 2.44it/s, tok/s=16366.0]
Inference: 87%|████████▋ | 5178/5920 [55:09<05:03, 2.44it/s, tok/s=16365.2]
Inference: 87%|████████▋ | 5179/5920 [55:09<04:56, 2.50it/s, tok/s=16365.2]
Inference: 87%|████████▋ | 5179/5920 [55:09<04:56, 2.50it/s, tok/s=16365.2]
Inference: 88%|████████▊ | 5180/5920 [55:10<05:35, 2.21it/s, tok/s=16365.2]
Inference: 88%|████████▊ | 5180/5920 [55:10<05:35, 2.21it/s, tok/s=16371.0]
Inference: 88%|████████▊ | 5181/5920 [55:10<05:34, 2.21it/s, tok/s=16375.2]
Inference: 88%|████████▊ | 5182/5920 [55:10<03:40, 3.35it/s, tok/s=16375.2]
Inference: 88%|████████▊ | 5182/5920 [55:10<03:40, 3.35it/s, tok/s=16381.9]
Inference: 88%|████████▊ | 5183/5920 [55:10<03:12, 3.82it/s, tok/s=16381.9]
Inference: 88%|████████▊ | 5183/5920 [55:10<03:12, 3.82it/s, tok/s=16383.0]
Inference: 88%|████████▊ | 5184/5920 [55:11<04:10, 2.94it/s, tok/s=16383.0]
Inference: 88%|████████▊ | 5184/5920 [55:11<04:10, 2.94it/s, tok/s=16382.6]
Inference: 88%|████████▊ | 5185/5920 [55:11<04:25, 2.77it/s, tok/s=16382.6]
Inference: 88%|████████▊ | 5185/5920 [55:11<04:25, 2.77it/s, tok/s=16384.6]
Inference: 88%|████████▊ | 5186/5920 [55:12<03:55, 3.12it/s, tok/s=16384.6]
Inference: 88%|████████▊ | 5186/5920 [55:12<03:55, 3.12it/s, tok/s=16384.6]
Inference: 88%|████████▊ | 5187/5920 [55:12<03:24, 3.58it/s, tok/s=16384.6]
Inference: 88%|████████▊ | 5187/5920 [55:12<03:24, 3.58it/s, tok/s=16385.9]
Inference: 88%|████████▊ | 5188/5920 [55:12<03:24, 3.58it/s, tok/s=16386.6]
Inference: 88%|████████▊ | 5189/5920 [55:12<03:24, 3.58it/s, tok/s=16387.1]
Inference: 88%|████████▊ | 5190/5920 [55:13<04:31, 2.68it/s, tok/s=16387.1]
Inference: 88%|████████▊ | 5190/5920 [55:13<04:31, 2.68it/s, tok/s=16385.6]
Inference: 88%|████████▊ | 5191/5920 [55:14<05:24, 2.25it/s, tok/s=16385.6]
Inference: 88%|████████▊ | 5191/5920 [55:14<05:24, 2.25it/s, tok/s=16383.1]
Inference: 88%|████████▊ | 5192/5920 [55:14<04:47, 2.53it/s, tok/s=16383.1]
Inference: 88%|████████▊ | 5192/5920 [55:14<04:47, 2.53it/s, tok/s=16383.5]
Inference: 88%|████████▊ | 5193/5920 [55:14<04:47, 2.53it/s, tok/s=16384.9]
Inference: 88%|████████▊ | 5194/5920 [55:15<06:02, 2.00it/s, tok/s=16384.9]
Inference: 88%|████████▊ | 5194/5920 [55:15<06:02, 2.00it/s, tok/s=16382.7]
Inference: 88%|████████▊ | 5195/5920 [55:16<06:06, 1.98it/s, tok/s=16382.7]
Inference: 88%|████████▊ | 5195/5920 [55:16<06:06, 1.98it/s, tok/s=16381.9]
Inference: 88%|████████▊ | 5196/5920 [55:16<06:05, 1.98it/s, tok/s=16382.8]
Inference: 88%|████████▊ | 5197/5920 [55:16<06:05, 1.98it/s, tok/s=16391.0]
Inference: 88%|████████▊ | 5198/5920 [55:17<04:59, 2.41it/s, tok/s=16391.0]
Inference: 88%|████████▊ | 5198/5920 [55:17<04:59, 2.41it/s, tok/s=16388.3]
Inference: 88%|████████▊ | 5199/5920 [55:17<04:17, 2.80it/s, tok/s=16388.3]
Inference: 88%|████████▊ | 5199/5920 [55:17<04:17, 2.80it/s, tok/s=16389.1]
Inference: 88%|████████▊ | 5200/5920 [55:17<04:27, 2.69it/s, tok/s=16389.1]
Inference: 88%|████████▊ | 5200/5920 [55:17<04:27, 2.69it/s, tok/s=16388.1]
Inference: 88%|████████▊ | 5201/5920 [55:18<04:42, 2.54it/s, tok/s=16388.1]
Inference: 88%|████████▊ | 5201/5920 [55:18<04:42, 2.54it/s, tok/s=16390.9]
Inference: 88%|████████▊ | 5202/5920 [55:18<05:00, 2.39it/s, tok/s=16390.9]
Inference: 88%|████████▊ | 5202/5920 [55:18<05:00, 2.39it/s, tok/s=16389.5]
Inference: 88%|████████▊ | 5203/5920 [55:19<05:28, 2.19it/s, tok/s=16389.5]
Inference: 88%|████████▊ | 5203/5920 [55:19<05:28, 2.19it/s, tok/s=16387.7]
Inference: 88%|████████▊ | 5204/5920 [55:20<06:05, 1.96it/s, tok/s=16387.7]
Inference: 88%|████████▊ | 5204/5920 [55:20<06:05, 1.96it/s, tok/s=16385.8]
Inference: 88%|████████▊ | 5205/5920 [55:20<04:48, 2.48it/s, tok/s=16385.8]
Inference: 88%|████████▊ | 5205/5920 [55:20<04:48, 2.48it/s, tok/s=16388.5]
Inference: 88%|████████▊ | 5206/5920 [55:20<04:48, 2.48it/s, tok/s=16389.6]
Inference: 88%|████████▊ | 5207/5920 [55:20<03:28, 3.41it/s, tok/s=16389.6]
Inference: 88%|████████▊ | 5207/5920 [55:20<03:28, 3.41it/s, tok/s=16389.6]
Inference: 88%|████████▊ | 5208/5920 [55:21<04:18, 2.76it/s, tok/s=16389.6]
Inference: 88%|████████▊ | 5208/5920 [55:21<04:18, 2.76it/s, tok/s=16388.3]
Inference: 88%|████████▊ | 5209/5920 [55:21<04:45, 2.49it/s, tok/s=16388.3]
Inference: 88%|████████▊ | 5209/5920 [55:21<04:45, 2.49it/s, tok/s=16387.8]
Inference: 88%|████████▊ | 5210/5920 [55:22<04:41, 2.52it/s, tok/s=16387.8]
Inference: 88%|████████▊ | 5210/5920 [55:22<04:41, 2.52it/s, tok/s=16387.1]
Inference: 88%|████████▊ | 5211/5920 [55:22<06:04, 1.95it/s, tok/s=16387.1]
Inference: 88%|████████▊ | 5211/5920 [55:22<06:04, 1.95it/s, tok/s=16384.0]
Inference: 88%|████████▊ | 5212/5920 [55:22<06:03, 1.95it/s, tok/s=16386.5]
Inference: 88%|████████▊ | 5213/5920 [55:23<04:56, 2.38it/s, tok/s=16386.5]
Inference: 88%|████████▊ | 5213/5920 [55:23<04:56, 2.38it/s, tok/s=16386.3]
Inference: 88%|████████▊ | 5214/5920 [55:23<04:32, 2.59it/s, tok/s=16386.3]
Inference: 88%|████████▊ | 5214/5920 [55:23<04:32, 2.59it/s, tok/s=16389.2]
Inference: 88%|████████▊ | 5215/5920 [55:24<04:47, 2.46it/s, tok/s=16389.2]
Inference: 88%|████████▊ | 5215/5920 [55:24<04:47, 2.46it/s, tok/s=16388.7]
Inference: 88%|████████▊ | 5216/5920 [55:24<04:45, 2.46it/s, tok/s=16388.7]
Inference: 88%|████████▊ | 5216/5920 [55:24<04:45, 2.46it/s, tok/s=16388.6]
Inference: 88%|████████▊ | 5217/5920 [55:24<04:06, 2.85it/s, tok/s=16388.6]
Inference: 88%|████████▊ | 5217/5920 [55:24<04:06, 2.85it/s, tok/s=16390.8]
Inference: 88%|████████▊ | 5218/5920 [55:26<07:54, 1.48it/s, tok/s=16390.8]
Inference: 88%|████████▊ | 5218/5920 [55:26<07:54, 1.48it/s, tok/s=16388.8]
Inference: 88%|████████▊ | 5219/5920 [55:26<07:07, 1.64it/s, tok/s=16388.8]
Inference: 88%|████████▊ | 5219/5920 [55:26<07:07, 1.64it/s, tok/s=16388.5]
Inference: 88%|████████▊ | 5220/5920 [55:27<09:22, 1.24it/s, tok/s=16388.5]
Inference: 88%|████████▊ | 5220/5920 [55:27<09:22, 1.24it/s, tok/s=16384.3]
Inference: 88%|████████▊ | 5221/5920 [55:28<08:27, 1.38it/s, tok/s=16384.3]
Inference: 88%|████████▊ | 5221/5920 [55:28<08:27, 1.38it/s, tok/s=16384.7]
Inference: 88%|████████▊ | 5222/5920 [55:29<10:50, 1.07it/s, tok/s=16384.7]
Inference: 88%|████████▊ | 5222/5920 [55:29<10:50, 1.07it/s, tok/s=16378.8]
Inference: 88%|████████▊ | 5223/5920 [55:30<10:49, 1.07it/s, tok/s=16381.9]
Inference: 88%|████████▊ | 5224/5920 [55:30<06:38, 1.75it/s, tok/s=16381.9]
Inference: 88%|████████▊ | 5224/5920 [55:30<06:38, 1.75it/s, tok/s=16388.4]
Inference: 88%|████████▊ | 5225/5920 [55:30<05:28, 2.11it/s, tok/s=16388.4]
Inference: 88%|████████▊ | 5225/5920 [55:30<05:28, 2.11it/s, tok/s=16388.9]
Inference: 88%|████████▊ | 5226/5920 [55:30<05:28, 2.11it/s, tok/s=16392.3]
Inference: 88%|████████▊ | 5227/5920 [55:30<04:21, 2.65it/s, tok/s=16392.3]
Inference: 88%|████████▊ | 5227/5920 [55:30<04:21, 2.65it/s, tok/s=16393.1]
Inference: 88%|████████▊ | 5228/5920 [55:31<04:08, 2.78it/s, tok/s=16393.1]
Inference: 88%|████████▊ | 5228/5920 [55:31<04:08, 2.78it/s, tok/s=16393.9]
Inference: 88%|████████▊ | 5229/5920 [55:31<03:47, 3.03it/s, tok/s=16393.9]
Inference: 88%|████████▊ | 5229/5920 [55:31<03:47, 3.03it/s, tok/s=16396.8]
Inference: 88%|████████▊ | 5230/5920 [55:31<03:47, 3.03it/s, tok/s=16400.5]
Inference: 88%|████████▊ | 5231/5920 [55:31<03:03, 3.76it/s, tok/s=16400.5]
Inference: 88%|████████▊ | 5231/5920 [55:31<03:03, 3.76it/s, tok/s=16401.4]
Inference: 88%|████████▊ | 5232/5920 [55:32<03:55, 2.92it/s, tok/s=16401.4]
Inference: 88%|████████▊ | 5232/5920 [55:32<03:55, 2.92it/s, tok/s=16403.5]
Inference: 88%|████████▊ | 5233/5920 [55:32<03:59, 2.87it/s, tok/s=16403.5]
Inference: 88%|████████▊ | 5233/5920 [55:32<03:59, 2.87it/s, tok/s=16407.5]
Inference: 88%|████████▊ | 5234/5920 [55:33<03:59, 2.86it/s, tok/s=16407.5]
Inference: 88%|████████▊ | 5234/5920 [55:33<03:59, 2.86it/s, tok/s=16406.9]
Inference: 88%|████████▊ | 5235/5920 [55:33<04:07, 2.77it/s, tok/s=16406.9]
Inference: 88%|████████▊ | 5235/5920 [55:33<04:07, 2.77it/s, tok/s=16406.3]
Inference: 88%|████████▊ | 5236/5920 [55:33<03:17, 3.47it/s, tok/s=16406.3]
Inference: 88%|████████▊ | 5236/5920 [55:33<03:17, 3.47it/s, tok/s=16408.2]
Inference: 88%|████████▊ | 5237/5920 [55:35<07:27, 1.53it/s, tok/s=16408.2]
Inference: 88%|████████▊ | 5237/5920 [55:35<07:27, 1.53it/s, tok/s=16402.6]
Inference: 88%|████████▊ | 5238/5920 [55:36<08:37, 1.32it/s, tok/s=16402.6]
Inference: 88%|████████▊ | 5238/5920 [55:36<08:37, 1.32it/s, tok/s=16399.3]
Inference: 88%|████████▊ | 5239/5920 [55:36<08:36, 1.32it/s, tok/s=16401.4]
Inference: 89%|████████▊ | 5240/5920 [55:37<09:14, 1.23it/s, tok/s=16401.4]
Inference: 89%|████████▊ | 5240/5920 [55:37<09:14, 1.23it/s, tok/s=16394.5]
Inference: 89%|████████▊ | 5241/5920 [55:38<07:47, 1.45it/s, tok/s=16394.5]
Inference: 89%|████████▊ | 5241/5920 [55:38<07:47, 1.45it/s, tok/s=16394.8]
Inference: 89%|████████▊ | 5242/5920 [55:38<07:34, 1.49it/s, tok/s=16394.8]
Inference: 89%|████████▊ | 5242/5920 [55:38<07:34, 1.49it/s, tok/s=16393.3]
Inference: 89%|████████▊ | 5243/5920 [55:39<06:11, 1.82it/s, tok/s=16393.3]
Inference: 89%|████████▊ | 5243/5920 [55:39<06:11, 1.82it/s, tok/s=16395.2]
Inference: 89%|████████▊ | 5244/5920 [55:39<06:33, 1.72it/s, tok/s=16395.2]
Inference: 89%|████████▊ | 5244/5920 [55:39<06:33, 1.72it/s, tok/s=16393.0]
Inference: 89%|████████▊ | 5245/5920 [55:40<06:49, 1.65it/s, tok/s=16393.0]
Inference: 89%|████████▊ | 5245/5920 [55:40<06:49, 1.65it/s, tok/s=16392.1]
Inference: 89%|████████▊ | 5246/5920 [55:41<06:48, 1.65it/s, tok/s=16392.1]
Inference: 89%|████████▊ | 5246/5920 [55:41<06:48, 1.65it/s, tok/s=16391.1]
Inference: 89%|████████▊ | 5247/5920 [55:41<06:17, 1.78it/s, tok/s=16391.1]
Inference: 89%|████████▊ | 5247/5920 [55:41<06:17, 1.78it/s, tok/s=16392.9]
Inference: 89%|████████▊ | 5248/5920 [55:42<06:58, 1.61it/s, tok/s=16392.9]
Inference: 89%|████████▊ | 5248/5920 [55:42<06:58, 1.61it/s, tok/s=16390.1]
Inference: 89%|████████▊ | 5249/5920 [55:42<05:40, 1.97it/s, tok/s=16390.1]
Inference: 89%|████████▊ | 5249/5920 [55:42<05:40, 1.97it/s, tok/s=16397.2]
Inference: 89%|████████▊ | 5250/5920 [55:42<05:28, 2.04it/s, tok/s=16397.2]
Inference: 89%|████████▊ | 5250/5920 [55:42<05:28, 2.04it/s, tok/s=16401.1]
Inference: 89%|████████▊ | 5251/5920 [55:43<04:30, 2.47it/s, tok/s=16401.1]
Inference: 89%|████████▊ | 5251/5920 [55:43<04:30, 2.47it/s, tok/s=16400.6]
Inference: 89%|████████▊ | 5252/5920 [55:43<03:37, 3.07it/s, tok/s=16400.6]
Inference: 89%|████████▊ | 5252/5920 [55:43<03:37, 3.07it/s, tok/s=16403.6]
Inference: 89%|████████▊ | 5253/5920 [55:43<02:55, 3.79it/s, tok/s=16403.6]
Inference: 89%|████████▊ | 5253/5920 [55:43<02:55, 3.79it/s, tok/s=16404.0]
Inference: 89%|████████▉ | 5254/5920 [55:43<02:55, 3.79it/s, tok/s=16405.7]
Inference: 89%|████████▉ | 5255/5920 [55:43<02:26, 4.55it/s, tok/s=16405.7]
Inference: 89%|████████▉ | 5255/5920 [55:43<02:26, 4.55it/s, tok/s=16404.9]
Inference: 89%|████████▉ | 5256/5920 [55:43<02:07, 5.21it/s, tok/s=16404.9]
Inference: 89%|████████▉ | 5256/5920 [55:43<02:07, 5.21it/s, tok/s=16407.9]
Inference: 89%|████████▉ | 5257/5920 [55:44<02:39, 4.16it/s, tok/s=16407.9]
Inference: 89%|████████▉ | 5257/5920 [55:44<02:39, 4.16it/s, tok/s=16406.9]
Inference: 89%|████████▉ | 5258/5920 [55:44<03:30, 3.14it/s, tok/s=16406.9]
Inference: 89%|████████▉ | 5258/5920 [55:44<03:30, 3.14it/s, tok/s=16406.4]
Inference: 89%|████████▉ | 5259/5920 [55:45<05:05, 2.16it/s, tok/s=16406.4]
Inference: 89%|████████▉ | 5259/5920 [55:45<05:05, 2.16it/s, tok/s=16407.3]
Inference: 89%|████████▉ | 5260/5920 [55:45<04:28, 2.46it/s, tok/s=16407.3]
Inference: 89%|████████▉ | 5260/5920 [55:45<04:28, 2.46it/s, tok/s=16411.5]
Inference: 89%|████████▉ | 5261/5920 [55:46<03:43, 2.95it/s, tok/s=16411.5]
Inference: 89%|████████▉ | 5261/5920 [55:46<03:43, 2.95it/s, tok/s=16412.1]
Inference: 89%|████████▉ | 5262/5920 [55:46<03:56, 2.78it/s, tok/s=16412.1]
Inference: 89%|████████▉ | 5262/5920 [55:46<03:56, 2.78it/s, tok/s=16412.2]
Inference: 89%|████████▉ | 5263/5920 [55:47<05:07, 2.14it/s, tok/s=16412.2]
Inference: 89%|████████▉ | 5263/5920 [55:47<05:07, 2.14it/s, tok/s=16410.4]
Inference: 89%|████████▉ | 5264/5920 [55:47<04:17, 2.54it/s, tok/s=16410.4]
Inference: 89%|████████▉ | 5264/5920 [55:47<04:17, 2.54it/s, tok/s=16413.4]
Inference: 89%|████████▉ | 5265/5920 [55:47<03:22, 3.24it/s, tok/s=16413.4]
Inference: 89%|████████▉ | 5265/5920 [55:47<03:22, 3.24it/s, tok/s=16414.2]
Inference: 89%|████████▉ | 5266/5920 [55:47<03:10, 3.43it/s, tok/s=16414.2]
Inference: 89%|████████▉ | 5266/5920 [55:47<03:10, 3.43it/s, tok/s=16415.3]
Inference: 89%|████████▉ | 5267/5920 [55:50<11:34, 1.06s/it, tok/s=16415.3]
Inference: 89%|████████▉ | 5267/5920 [55:50<11:34, 1.06s/it, tok/s=16403.4]
Inference: 89%|████████▉ | 5268/5920 [55:50<09:01, 1.20it/s, tok/s=16403.4]
Inference: 89%|████████▉ | 5268/5920 [55:50<09:01, 1.20it/s, tok/s=16403.6]
Inference: 89%|████████▉ | 5269/5920 [55:51<07:27, 1.46it/s, tok/s=16403.6]
Inference: 89%|████████▉ | 5269/5920 [55:51<07:27, 1.46it/s, tok/s=16407.2]
Inference: 89%|████████▉ | 5270/5920 [55:51<06:58, 1.55it/s, tok/s=16407.2]
Inference: 89%|████████▉ | 5270/5920 [55:51<06:58, 1.55it/s, tok/s=16405.5]
Inference: 89%|████████▉ | 5271/5920 [55:52<08:28, 1.28it/s, tok/s=16405.5]
Inference: 89%|████████▉ | 5271/5920 [55:52<08:28, 1.28it/s, tok/s=16401.3]
Inference: 89%|████████▉ | 5272/5920 [55:53<07:37, 1.42it/s, tok/s=16401.3]
Inference: 89%|████████▉ | 5272/5920 [55:53<07:37, 1.42it/s, tok/s=16401.5]
Inference: 89%|████████▉ | 5273/5920 [55:53<07:36, 1.42it/s, tok/s=16406.1]
Inference: 89%|████████▉ | 5274/5920 [55:53<07:36, 1.42it/s, tok/s=16412.2]
Inference: 89%|████████▉ | 5275/5920 [55:53<07:35, 1.42it/s, tok/s=16416.6]
Inference: 89%|████████▉ | 5276/5920 [55:53<03:11, 3.37it/s, tok/s=16416.6]
Inference: 89%|████████▉ | 5276/5920 [55:53<03:11, 3.37it/s, tok/s=16418.1]
Inference: 89%|████████▉ | 5277/5920 [55:54<03:20, 3.20it/s, tok/s=16418.1]
Inference: 89%|████████▉ | 5277/5920 [55:54<03:20, 3.20it/s, tok/s=16420.5]
Inference: 89%|████████▉ | 5278/5920 [55:54<03:28, 3.08it/s, tok/s=16420.5]
Inference: 89%|████████▉ | 5278/5920 [55:54<03:28, 3.08it/s, tok/s=16421.5]
Inference: 89%|████████▉ | 5279/5920 [55:54<03:52, 2.76it/s, tok/s=16421.5]
Inference: 89%|████████▉ | 5279/5920 [55:54<03:52, 2.76it/s, tok/s=16421.4]
Inference: 89%|████████▉ | 5280/5920 [55:55<03:11, 3.34it/s, tok/s=16421.4]
Inference: 89%|████████▉ | 5280/5920 [55:55<03:11, 3.34it/s, tok/s=16424.3]
Inference: 89%|████████▉ | 5281/5920 [55:55<04:10, 2.55it/s, tok/s=16424.3]
Inference: 89%|████████▉ | 5281/5920 [55:55<04:10, 2.55it/s, tok/s=16428.8]
Inference: 89%|████████▉ | 5282/5920 [55:56<04:04, 2.61it/s, tok/s=16428.8]
Inference: 89%|████████▉ | 5282/5920 [55:56<04:04, 2.61it/s, tok/s=16428.6]
Inference: 89%|████████▉ | 5283/5920 [55:56<04:14, 2.50it/s, tok/s=16428.6]
Inference: 89%|████████▉ | 5283/5920 [55:56<04:14, 2.50it/s, tok/s=16428.8]
Inference: 89%|████████▉ | 5284/5920 [55:57<06:21, 1.67it/s, tok/s=16428.8]
Inference: 89%|████████▉ | 5284/5920 [55:57<06:21, 1.67it/s, tok/s=16425.1]
Inference: 89%|████████▉ | 5285/5920 [55:58<06:57, 1.52it/s, tok/s=16425.1]
Inference: 89%|████████▉ | 5285/5920 [55:58<06:57, 1.52it/s, tok/s=16422.8]
Inference: 89%|████████▉ | 5286/5920 [56:00<11:06, 1.05s/it, tok/s=16422.8]
Inference: 89%|████████▉ | 5286/5920 [56:00<11:06, 1.05s/it, tok/s=16414.4]
Inference: 89%|████████▉ | 5287/5920 [56:00<08:41, 1.21it/s, tok/s=16414.4]
Inference: 89%|████████▉ | 5287/5920 [56:00<08:41, 1.21it/s, tok/s=16415.2]
Inference: 89%|████████▉ | 5288/5920 [56:01<09:16, 1.14it/s, tok/s=16415.2]
Inference: 89%|████████▉ | 5288/5920 [56:01<09:16, 1.14it/s, tok/s=16411.8]
Inference: 89%|████████▉ | 5289/5920 [56:01<07:32, 1.40it/s, tok/s=16411.8]
Inference: 89%|████████▉ | 5289/5920 [56:01<07:32, 1.40it/s, tok/s=16412.4]
Inference: 89%|████████▉ | 5290/5920 [56:02<07:26, 1.41it/s, tok/s=16412.4]
Inference: 89%|████████▉ | 5290/5920 [56:02<07:26, 1.41it/s, tok/s=16411.2]
Inference: 89%|████████▉ | 5291/5920 [56:03<06:20, 1.65it/s, tok/s=16411.2]
Inference: 89%|████████▉ | 5291/5920 [56:03<06:20, 1.65it/s, tok/s=16412.2]
Inference: 89%|████████▉ | 5292/5920 [56:04<08:38, 1.21it/s, tok/s=16412.2]
Inference: 89%|████████▉ | 5292/5920 [56:04<08:38, 1.21it/s, tok/s=16408.6]
Inference: 89%|████████▉ | 5293/5920 [56:04<06:29, 1.61it/s, tok/s=16408.6]
Inference: 89%|████████▉ | 5293/5920 [56:04<06:29, 1.61it/s, tok/s=16409.3]
Inference: 89%|████████▉ | 5294/5920 [56:04<06:28, 1.61it/s, tok/s=16411.1]
Inference: 89%|████████▉ | 5295/5920 [56:04<04:17, 2.42it/s, tok/s=16411.1]
Inference: 89%|████████▉ | 5295/5920 [56:04<04:17, 2.42it/s, tok/s=16418.7]
Inference: 89%|████████▉ | 5296/5920 [56:05<03:35, 2.89it/s, tok/s=16418.7]
Inference: 89%|████████▉ | 5296/5920 [56:05<03:35, 2.89it/s, tok/s=16420.3]
Inference: 89%|████████▉ | 5297/5920 [56:05<03:24, 3.05it/s, tok/s=16420.3]
Inference: 89%|████████▉ | 5297/5920 [56:05<03:24, 3.05it/s, tok/s=16419.8]
Inference: 89%|████████▉ | 5298/5920 [56:06<05:27, 1.90it/s, tok/s=16419.8]
Inference: 89%|████████▉ | 5298/5920 [56:06<05:27, 1.90it/s, tok/s=16419.0]
Inference: 90%|████████▉ | 5299/5920 [56:06<04:24, 2.34it/s, tok/s=16419.0]
Inference: 90%|████████▉ | 5299/5920 [56:06<04:24, 2.34it/s, tok/s=16419.9]
Inference: 90%|████████▉ | 5300/5920 [56:06<04:24, 2.34it/s, tok/s=16421.7]
Inference: 90%|████████▉ | 5301/5920 [56:06<03:30, 2.94it/s, tok/s=16421.7]
Inference: 90%|████████▉ | 5301/5920 [56:06<03:30, 2.94it/s, tok/s=16420.8]
Inference: 90%|████████▉ | 5302/5920 [56:06<03:30, 2.94it/s, tok/s=16422.5]
Inference: 90%|████████▉ | 5303/5920 [56:07<03:30, 2.94it/s, tok/s=16423.7]
Inference: 90%|████████▉ | 5304/5920 [56:07<03:29, 2.94it/s, tok/s=16425.4]
Inference: 90%|████████▉ | 5305/5920 [56:07<01:55, 5.31it/s, tok/s=16425.4]
Inference: 90%|████████▉ | 5305/5920 [56:07<01:55, 5.31it/s, tok/s=16425.7]
Inference: 90%|████████▉ | 5306/5920 [56:07<02:11, 4.68it/s, tok/s=16425.7]
Inference: 90%|████████▉ | 5306/5920 [56:07<02:11, 4.68it/s, tok/s=16424.6]
Inference: 90%|████████▉ | 5307/5920 [56:07<02:05, 4.87it/s, tok/s=16424.6]
Inference: 90%|████████▉ | 5307/5920 [56:07<02:05, 4.87it/s, tok/s=16427.8]
Inference: 90%|████████▉ | 5308/5920 [56:07<02:05, 4.87it/s, tok/s=16431.0]
Inference: 90%|████████▉ | 5309/5920 [56:10<05:36, 1.81it/s, tok/s=16431.0]
Inference: 90%|████████▉ | 5309/5920 [56:10<05:36, 1.81it/s, tok/s=16423.6]
Inference: 90%|████████▉ | 5310/5920 [56:10<05:18, 1.92it/s, tok/s=16423.6]
Inference: 90%|████████▉ | 5310/5920 [56:10<05:18, 1.92it/s, tok/s=16422.4]
Inference: 90%|████████▉ | 5311/5920 [56:10<04:21, 2.33it/s, tok/s=16422.4]
Inference: 90%|████████▉ | 5311/5920 [56:10<04:21, 2.33it/s, tok/s=16422.4]
Inference: 90%|████████▉ | 5312/5920 [56:10<03:50, 2.64it/s, tok/s=16422.4]
Inference: 90%|████████▉ | 5312/5920 [56:10<03:50, 2.64it/s, tok/s=16425.8]
Inference: 90%|████████▉ | 5313/5920 [56:11<04:22, 2.31it/s, tok/s=16425.8]
Inference: 90%|████████▉ | 5313/5920 [56:11<04:22, 2.31it/s, tok/s=16429.1]
Inference: 90%|████████▉ | 5314/5920 [56:11<03:41, 2.74it/s, tok/s=16429.1]
Inference: 90%|████████▉ | 5314/5920 [56:11<03:41, 2.74it/s, tok/s=16431.9]
Inference: 90%|████████▉ | 5315/5920 [56:11<03:40, 2.74it/s, tok/s=16433.5]
Inference: 90%|████████▉ | 5316/5920 [56:11<03:40, 2.74it/s, tok/s=16435.1]
Inference: 90%|████████▉ | 5317/5920 [56:11<02:08, 4.70it/s, tok/s=16435.1]
Inference: 90%|████████▉ | 5317/5920 [56:11<02:08, 4.70it/s, tok/s=16435.3]
Inference: 90%|████████▉ | 5318/5920 [56:12<02:43, 3.68it/s, tok/s=16435.3]
Inference: 90%|████████▉ | 5318/5920 [56:12<02:43, 3.68it/s, tok/s=16434.2]
Inference: 90%|████████▉ | 5319/5920 [56:12<02:32, 3.93it/s, tok/s=16434.2]
Inference: 90%|████████▉ | 5319/5920 [56:12<02:32, 3.93it/s, tok/s=16437.6]
Inference: 90%|████████▉ | 5320/5920 [56:12<02:28, 4.05it/s, tok/s=16437.6]
Inference: 90%|████████▉ | 5320/5920 [56:12<02:28, 4.05it/s, tok/s=16439.6]
Inference: 90%|████████▉ | 5321/5920 [56:13<03:10, 3.15it/s, tok/s=16439.6]
Inference: 90%|████████▉ | 5321/5920 [56:13<03:10, 3.15it/s, tok/s=16438.0]
Inference: 90%|████████▉ | 5322/5920 [56:13<02:55, 3.40it/s, tok/s=16438.0]
Inference: 90%|████████▉ | 5322/5920 [56:13<02:55, 3.40it/s, tok/s=16439.5]
Inference: 90%|████████▉ | 5323/5920 [56:14<03:36, 2.75it/s, tok/s=16439.5]
Inference: 90%|████████▉ | 5323/5920 [56:14<03:36, 2.75it/s, tok/s=16440.8]
Inference: 90%|████████▉ | 5324/5920 [56:14<03:04, 3.23it/s, tok/s=16440.8]
Inference: 90%|████████▉ | 5324/5920 [56:14<03:04, 3.23it/s, tok/s=16440.5]
Inference: 90%|████████▉ | 5325/5920 [56:15<05:21, 1.85it/s, tok/s=16440.5]
Inference: 90%|████████▉ | 5325/5920 [56:15<05:21, 1.85it/s, tok/s=16444.0]
Inference: 90%|████████▉ | 5326/5920 [56:16<06:06, 1.62it/s, tok/s=16444.0]
Inference: 90%|████████▉ | 5326/5920 [56:16<06:06, 1.62it/s, tok/s=16442.2]
Inference: 90%|████████▉ | 5327/5920 [56:17<07:32, 1.31it/s, tok/s=16442.2]
Inference: 90%|████████▉ | 5327/5920 [56:17<07:32, 1.31it/s, tok/s=16439.4]
Inference: 90%|█████████ | 5328/5920 [56:18<09:57, 1.01s/it, tok/s=16439.4]
Inference: 90%|█████████ | 5328/5920 [56:18<09:57, 1.01s/it, tok/s=16432.3]
Inference: 90%|█████████ | 5329/5920 [56:18<09:56, 1.01s/it, tok/s=16436.2]
Inference: 90%|█████████ | 5330/5920 [56:19<05:54, 1.67it/s, tok/s=16436.2]
Inference: 90%|█████████ | 5330/5920 [56:19<05:54, 1.67it/s, tok/s=16436.3]
Inference: 90%|█████████ | 5331/5920 [56:19<05:21, 1.83it/s, tok/s=16436.3]
Inference: 90%|█████████ | 5331/5920 [56:19<05:21, 1.83it/s, tok/s=16443.4]
Inference: 90%|█████████ | 5332/5920 [56:19<04:30, 2.17it/s, tok/s=16443.4]
Inference: 90%|█████████ | 5332/5920 [56:19<04:30, 2.17it/s, tok/s=16443.5]
Inference: 90%|█████████ | 5333/5920 [56:19<03:54, 2.50it/s, tok/s=16443.5]
Inference: 90%|█████████ | 5333/5920 [56:19<03:54, 2.50it/s, tok/s=16446.2]
Inference: 90%|█████████ | 5334/5920 [56:20<03:43, 2.62it/s, tok/s=16446.2]
Inference: 90%|█████████ | 5334/5920 [56:20<03:43, 2.62it/s, tok/s=16445.9]
Inference: 90%|█████████ | 5335/5920 [56:20<03:31, 2.77it/s, tok/s=16445.9]
Inference: 90%|█████████ | 5335/5920 [56:20<03:31, 2.77it/s, tok/s=16447.1]
Inference: 90%|█████████ | 5336/5920 [56:20<03:31, 2.77it/s, tok/s=16448.8]
Inference: 90%|█████████ | 5337/5920 [56:21<04:22, 2.22it/s, tok/s=16448.8]
Inference: 90%|█████████ | 5337/5920 [56:21<04:22, 2.22it/s, tok/s=16449.0]
Inference: 90%|█████████ | 5338/5920 [56:22<04:38, 2.09it/s, tok/s=16449.0]
Inference: 90%|█████████ | 5338/5920 [56:22<04:38, 2.09it/s, tok/s=16453.1]
Inference: 90%|█████████ | 5339/5920 [56:22<04:37, 2.09it/s, tok/s=16457.4]
Inference: 90%|█████████ | 5340/5920 [56:23<04:58, 1.94it/s, tok/s=16457.4]
Inference: 90%|█████████ | 5340/5920 [56:23<04:58, 1.94it/s, tok/s=16453.6]
Inference: 90%|█████████ | 5341/5920 [56:23<04:18, 2.24it/s, tok/s=16453.6]
Inference: 90%|█████████ | 5341/5920 [56:23<04:18, 2.24it/s, tok/s=16454.3]
Inference: 90%|█████████ | 5342/5920 [56:23<04:17, 2.24it/s, tok/s=16456.6]
Inference: 90%|█████████ | 5343/5920 [56:23<04:17, 2.24it/s, tok/s=16459.0]
Inference: 90%|█████████ | 5344/5920 [56:24<03:39, 2.63it/s, tok/s=16459.0]
Inference: 90%|█████████ | 5344/5920 [56:24<03:39, 2.63it/s, tok/s=16458.8]
Inference: 90%|█████████ | 5345/5920 [56:25<05:00, 1.91it/s, tok/s=16458.8]
Inference: 90%|█████████ | 5345/5920 [56:25<05:00, 1.91it/s, tok/s=16457.7]
Inference: 90%|█████████ | 5346/5920 [56:25<04:21, 2.19it/s, tok/s=16457.7]
Inference: 90%|█████████ | 5346/5920 [56:25<04:21, 2.19it/s, tok/s=16458.0]
Inference: 90%|█████████ | 5347/5920 [56:27<06:39, 1.43it/s, tok/s=16458.0]
Inference: 90%|█████████ | 5347/5920 [56:27<06:39, 1.43it/s, tok/s=16452.5]
Inference: 90%|█████████ | 5348/5920 [56:27<06:10, 1.54it/s, tok/s=16452.5]
Inference: 90%|█████████ | 5348/5920 [56:27<06:10, 1.54it/s, tok/s=16450.8]
Inference: 90%|█████████ | 5349/5920 [56:28<04:58, 1.91it/s, tok/s=16450.8]
Inference: 90%|█████████ | 5349/5920 [56:28<04:58, 1.91it/s, tok/s=16450.5]
Inference: 90%|█████████ | 5350/5920 [56:28<05:02, 1.88it/s, tok/s=16450.5]
Inference: 90%|█████████ | 5350/5920 [56:28<05:02, 1.88it/s, tok/s=16448.2]
Inference: 90%|█████████ | 5351/5920 [56:28<05:01, 1.88it/s, tok/s=16450.6]
Inference: 90%|█████████ | 5352/5920 [56:29<03:45, 2.52it/s, tok/s=16450.6]
Inference: 90%|█████████ | 5352/5920 [56:29<03:45, 2.52it/s, tok/s=16449.4]
Inference: 90%|█████████ | 5353/5920 [56:29<03:44, 2.52it/s, tok/s=16453.7]
Inference: 90%|█████████ | 5354/5920 [56:30<05:04, 1.86it/s, tok/s=16453.7]
Inference: 90%|█████████ | 5354/5920 [56:30<05:04, 1.86it/s, tok/s=16447.5]
Inference: 90%|█████████ | 5355/5920 [56:30<04:11, 2.25it/s, tok/s=16447.5]
Inference: 90%|█████████ | 5355/5920 [56:30<04:11, 2.25it/s, tok/s=16449.5]
Inference: 90%|█████████ | 5356/5920 [56:30<03:40, 2.56it/s, tok/s=16449.5]
Inference: 90%|█████████ | 5356/5920 [56:30<03:40, 2.56it/s, tok/s=16450.0]
Inference: 90%|█████████ | 5357/5920 [56:31<03:09, 2.97it/s, tok/s=16450.0]
Inference: 90%|█████████ | 5357/5920 [56:31<03:09, 2.97it/s, tok/s=16452.4]
Inference: 91%|█████████ | 5358/5920 [56:31<03:08, 2.97it/s, tok/s=16454.4]
Inference: 91%|█████████ | 5359/5920 [56:31<03:08, 2.97it/s, tok/s=16456.8]
Inference: 91%|█████████ | 5360/5920 [56:31<02:19, 4.00it/s, tok/s=16456.8]
Inference: 91%|█████████ | 5360/5920 [56:31<02:19, 4.00it/s, tok/s=16456.3]
Inference: 91%|█████████ | 5361/5920 [56:32<03:36, 2.58it/s, tok/s=16456.3]
Inference: 91%|█████████ | 5361/5920 [56:32<03:36, 2.58it/s, tok/s=16455.5]
Inference: 91%|█████████ | 5362/5920 [56:32<03:22, 2.76it/s, tok/s=16455.5]
Inference: 91%|█████████ | 5362/5920 [56:32<03:22, 2.76it/s, tok/s=16456.2]
Inference: 91%|█████████ | 5363/5920 [56:33<03:14, 2.86it/s, tok/s=16456.2]
Inference: 91%|█████████ | 5363/5920 [56:33<03:14, 2.86it/s, tok/s=16457.5]
Inference: 91%|█████████ | 5364/5920 [56:33<03:13, 2.87it/s, tok/s=16457.5]
Inference: 91%|█████████ | 5364/5920 [56:33<03:13, 2.87it/s, tok/s=16458.0]
Inference: 91%|█████████ | 5365/5920 [56:33<03:13, 2.87it/s, tok/s=16462.4]
Inference: 91%|█████████ | 5366/5920 [56:33<02:29, 3.70it/s, tok/s=16462.4]
Inference: 91%|█████████ | 5366/5920 [56:33<02:29, 3.70it/s, tok/s=16465.0]
Inference: 91%|█████████ | 5367/5920 [56:33<02:19, 3.97it/s, tok/s=16465.0]
Inference: 91%|█████████ | 5367/5920 [56:33<02:19, 3.97it/s, tok/s=16464.6]
Inference: 91%|█████████ | 5368/5920 [56:34<02:18, 3.97it/s, tok/s=16469.3]
Inference: 91%|█████████ | 5369/5920 [56:34<01:35, 5.74it/s, tok/s=16469.3]
Inference: 91%|█████████ | 5369/5920 [56:34<01:35, 5.74it/s, tok/s=16472.0]
Inference: 91%|█████████ | 5370/5920 [56:34<02:28, 3.69it/s, tok/s=16472.0]
Inference: 91%|█████████ | 5370/5920 [56:34<02:28, 3.69it/s, tok/s=16470.1]
Inference: 91%|█████████ | 5371/5920 [56:35<03:12, 2.85it/s, tok/s=16470.1]
Inference: 91%|█████████ | 5371/5920 [56:35<03:12, 2.85it/s, tok/s=16467.6]
Inference: 91%|█████████ | 5372/5920 [56:35<03:12, 2.85it/s, tok/s=16475.4]
Inference: 91%|█████████ | 5373/5920 [56:35<02:12, 4.13it/s, tok/s=16475.4]
Inference: 91%|█████████ | 5373/5920 [56:35<02:12, 4.13it/s, tok/s=16476.8]
Inference: 91%|█████████ | 5374/5920 [56:36<03:05, 2.94it/s, tok/s=16476.8]
Inference: 91%|█████████ | 5374/5920 [56:36<03:05, 2.94it/s, tok/s=16476.3]
Inference: 91%|█████████ | 5375/5920 [56:36<03:05, 2.94it/s, tok/s=16476.5]
Inference: 91%|█████████ | 5376/5920 [56:36<02:23, 3.79it/s, tok/s=16476.5]
Inference: 91%|█████████ | 5376/5920 [56:36<02:23, 3.79it/s, tok/s=16479.1]
Inference: 91%|█████████ | 5377/5920 [56:36<02:23, 3.79it/s, tok/s=16482.7]
Inference: 91%|█████████ | 5378/5920 [56:36<02:22, 3.79it/s, tok/s=16485.7]
Inference: 91%|█████████ | 5379/5920 [56:37<02:03, 4.38it/s, tok/s=16485.7]
Inference: 91%|█████████ | 5379/5920 [56:37<02:03, 4.38it/s, tok/s=16486.4]
Inference: 91%|█████████ | 5380/5920 [56:37<02:17, 3.93it/s, tok/s=16486.4]
Inference: 91%|█████████ | 5380/5920 [56:37<02:17, 3.93it/s, tok/s=16486.2]
Inference: 91%|█████████ | 5381/5920 [56:37<02:17, 3.93it/s, tok/s=16491.9]
Inference: 91%|█████████ | 5382/5920 [56:37<01:46, 5.08it/s, tok/s=16491.9]
Inference: 91%|█████████ | 5382/5920 [56:37<01:46, 5.08it/s, tok/s=16494.0]
Inference: 91%|█████████ | 5383/5920 [56:38<03:38, 2.46it/s, tok/s=16494.0]
Inference: 91%|█████████ | 5383/5920 [56:38<03:38, 2.46it/s, tok/s=16489.7]
Inference: 91%|█████████ | 5384/5920 [56:39<04:10, 2.14it/s, tok/s=16489.7]
Inference: 91%|█████████ | 5384/5920 [56:39<04:10, 2.14it/s, tok/s=16488.0]
Inference: 91%|█████████ | 5385/5920 [56:41<06:24, 1.39it/s, tok/s=16488.0]
Inference: 91%|█████████ | 5385/5920 [56:41<06:24, 1.39it/s, tok/s=16483.0]
Inference: 91%|█████████ | 5386/5920 [56:42<07:10, 1.24it/s, tok/s=16483.0]
Inference: 91%|█████████ | 5386/5920 [56:42<07:10, 1.24it/s, tok/s=16482.8]
Inference: 91%|█████████ | 5387/5920 [56:42<06:22, 1.39it/s, tok/s=16482.8]
Inference: 91%|█████████ | 5387/5920 [56:42<06:22, 1.39it/s, tok/s=16483.7]
Inference: 91%|█████████ | 5388/5920 [56:42<05:10, 1.72it/s, tok/s=16483.7]
Inference: 91%|█████████ | 5388/5920 [56:42<05:10, 1.72it/s, tok/s=16483.2]
Inference: 91%|█████████ | 5389/5920 [56:43<05:05, 1.74it/s, tok/s=16483.2]
Inference: 91%|█████████ | 5389/5920 [56:43<05:05, 1.74it/s, tok/s=16481.7]
Inference: 91%|█████████ | 5390/5920 [56:43<04:53, 1.80it/s, tok/s=16481.7]
Inference: 91%|█████████ | 5390/5920 [56:43<04:53, 1.80it/s, tok/s=16480.9]
Inference: 91%|█████████ | 5391/5920 [56:44<05:11, 1.70it/s, tok/s=16480.9]
Inference: 91%|█████████ | 5391/5920 [56:44<05:11, 1.70it/s, tok/s=16481.4]
Inference: 91%|█████████ | 5392/5920 [56:45<06:06, 1.44it/s, tok/s=16481.4]
Inference: 91%|█████████ | 5392/5920 [56:45<06:06, 1.44it/s, tok/s=16479.8]
Inference: 91%|█████████ | 5393/5920 [56:45<05:21, 1.64it/s, tok/s=16479.8]
Inference: 91%|█████████ | 5393/5920 [56:45<05:21, 1.64it/s, tok/s=16478.6]
Inference: 91%|█████████ | 5394/5920 [56:47<06:36, 1.33it/s, tok/s=16478.6]
Inference: 91%|█████████ | 5394/5920 [56:47<06:36, 1.33it/s, tok/s=16478.6]
Inference: 91%|█████████ | 5395/5920 [56:47<05:10, 1.69it/s, tok/s=16478.6]
Inference: 91%|█████████ | 5395/5920 [56:47<05:10, 1.69it/s, tok/s=16481.4]
Inference: 91%|█████████ | 5396/5920 [56:47<05:09, 1.69it/s, tok/s=16482.1]
Inference: 91%|█████████ | 5397/5920 [56:47<03:33, 2.45it/s, tok/s=16482.1]
Inference: 91%|█████████ | 5397/5920 [56:47<03:33, 2.45it/s, tok/s=16488.9]
Inference: 91%|█████████ | 5398/5920 [56:48<03:57, 2.19it/s, tok/s=16488.9]
Inference: 91%|█████████ | 5398/5920 [56:48<03:57, 2.19it/s, tok/s=16490.0]
Inference: 91%|█████████ | 5399/5920 [56:48<04:36, 1.88it/s, tok/s=16490.0]
Inference: 91%|█████████ | 5399/5920 [56:48<04:36, 1.88it/s, tok/s=16487.5]
Inference: 91%|█████████ | 5400/5920 [56:49<04:07, 2.10it/s, tok/s=16487.5]
Inference: 91%|█████████ | 5400/5920 [56:49<04:07, 2.10it/s, tok/s=16486.7]
Inference: 91%|█████████ | 5401/5920 [56:49<03:59, 2.17it/s, tok/s=16486.7]
Inference: 91%|█████████ | 5401/5920 [56:49<03:59, 2.17it/s, tok/s=16485.0]
Inference: 91%|█████████▏| 5402/5920 [56:50<04:06, 2.10it/s, tok/s=16485.0]
Inference: 91%|█████████▏| 5402/5920 [56:50<04:06, 2.10it/s, tok/s=16483.4]
Inference: 91%|█████████▏| 5403/5920 [56:50<03:28, 2.48it/s, tok/s=16483.4]
Inference: 91%|█████████▏| 5403/5920 [56:50<03:28, 2.48it/s, tok/s=16485.1]
Inference: 91%|█████████▏| 5404/5920 [56:50<03:32, 2.42it/s, tok/s=16485.1]
Inference: 91%|█████████▏| 5404/5920 [56:50<03:32, 2.42it/s, tok/s=16485.5]
Inference: 91%|█████████▏| 5405/5920 [56:51<03:12, 2.68it/s, tok/s=16485.5]
Inference: 91%|█████████▏| 5405/5920 [56:51<03:12, 2.68it/s, tok/s=16485.4]
Inference: 91%|█████████▏| 5406/5920 [56:51<03:11, 2.68it/s, tok/s=16485.5]
Inference: 91%|█████████▏| 5407/5920 [56:51<02:33, 3.34it/s, tok/s=16485.5]
Inference: 91%|█████████▏| 5407/5920 [56:51<02:33, 3.34it/s, tok/s=16485.8]
Inference: 91%|█████████▏| 5408/5920 [56:51<02:36, 3.27it/s, tok/s=16485.8]
Inference: 91%|█████████▏| 5408/5920 [56:51<02:36, 3.27it/s, tok/s=16486.3]
Inference: 91%|█████████▏| 5409/5920 [56:51<02:36, 3.27it/s, tok/s=16487.9]
Inference: 91%|█████████▏| 5410/5920 [56:52<02:15, 3.77it/s, tok/s=16487.9]
Inference: 91%|█████████▏| 5410/5920 [56:52<02:15, 3.77it/s, tok/s=16488.3]
Inference: 91%|█████████▏| 5411/5920 [56:52<02:34, 3.29it/s, tok/s=16488.3]
Inference: 91%|█████████▏| 5411/5920 [56:52<02:34, 3.29it/s, tok/s=16486.8]
Inference: 91%|█████████▏| 5412/5920 [56:53<02:31, 3.36it/s, tok/s=16486.8]
Inference: 91%|█████████▏| 5412/5920 [56:53<02:31, 3.36it/s, tok/s=16486.0]
Inference: 91%|█████████▏| 5413/5920 [56:53<02:30, 3.36it/s, tok/s=16490.2]
Inference: 91%|█████████▏| 5414/5920 [56:53<02:30, 3.36it/s, tok/s=16491.9]
Inference: 91%|█████████▏| 5415/5920 [56:53<01:25, 5.91it/s, tok/s=16491.9]
Inference: 91%|█████████▏| 5415/5920 [56:53<01:25, 5.91it/s, tok/s=16492.9]
Inference: 91%|█████████▏| 5416/5920 [56:53<01:40, 5.01it/s, tok/s=16492.9]
Inference: 91%|█████████▏| 5416/5920 [56:53<01:40, 5.01it/s, tok/s=16491.7]
Inference: 92%|█████████▏| 5417/5920 [56:55<04:43, 1.77it/s, tok/s=16491.7]
Inference: 92%|█████████▏| 5417/5920 [56:55<04:43, 1.77it/s, tok/s=16483.2]
Inference: 92%|█████████▏| 5418/5920 [56:55<04:43, 1.77it/s, tok/s=16483.6]
Inference: 92%|█████████▏| 5419/5920 [56:55<03:10, 2.63it/s, tok/s=16483.6]
Inference: 92%|█████████▏| 5419/5920 [56:55<03:10, 2.63it/s, tok/s=16484.0]
Inference: 92%|█████████▏| 5420/5920 [56:56<03:42, 2.25it/s, tok/s=16484.0]
Inference: 92%|█████████▏| 5420/5920 [56:56<03:42, 2.25it/s, tok/s=16481.7]
Inference: 92%|█████████▏| 5421/5920 [56:56<03:41, 2.25it/s, tok/s=16483.6]
Inference: 92%|█████████▏| 5422/5920 [56:56<03:10, 2.61it/s, tok/s=16483.6]
Inference: 92%|█████████▏| 5422/5920 [56:56<03:10, 2.61it/s, tok/s=16482.6]
Inference: 92%|█████████▏| 5423/5920 [56:56<03:10, 2.61it/s, tok/s=16486.2]
Inference: 92%|█████████▏| 5424/5920 [56:57<02:24, 3.43it/s, tok/s=16486.2]
Inference: 92%|█████████▏| 5424/5920 [56:57<02:24, 3.43it/s, tok/s=16488.8]
Inference: 92%|█████████▏| 5425/5920 [56:57<02:24, 3.43it/s, tok/s=16491.6]
Inference: 92%|█████████▏| 5426/5920 [56:57<02:54, 2.83it/s, tok/s=16491.6]
Inference: 92%|█████████▏| 5426/5920 [56:57<02:54, 2.83it/s, tok/s=16489.5]
Inference: 92%|█████████▏| 5427/5920 [56:57<02:54, 2.83it/s, tok/s=16493.2]
Inference: 92%|█████████▏| 5428/5920 [56:58<02:12, 3.70it/s, tok/s=16493.2]
Inference: 92%|█████████▏| 5428/5920 [56:58<02:12, 3.70it/s, tok/s=16493.0]
Inference: 92%|█████████▏| 5429/5920 [56:58<02:12, 3.70it/s, tok/s=16497.7]
Inference: 92%|█████████▏| 5430/5920 [56:58<02:23, 3.41it/s, tok/s=16497.7]
Inference: 92%|█████████▏| 5430/5920 [56:58<02:23, 3.41it/s, tok/s=16495.3]
Inference: 92%|█████████▏| 5431/5920 [56:58<02:23, 3.41it/s, tok/s=16498.2]
Inference: 92%|█████████▏| 5432/5920 [56:59<01:54, 4.26it/s, tok/s=16498.2]
Inference: 92%|█████████▏| 5432/5920 [56:59<01:54, 4.26it/s, tok/s=16499.4]
Inference: 92%|█████████▏| 5433/5920 [56:59<02:09, 3.76it/s, tok/s=16499.4]
Inference: 92%|█████████▏| 5433/5920 [56:59<02:09, 3.76it/s, tok/s=16500.3]
Inference: 92%|█████████▏| 5434/5920 [57:00<02:43, 2.98it/s, tok/s=16500.3]
Inference: 92%|█████████▏| 5434/5920 [57:00<02:43, 2.98it/s, tok/s=16502.1]
Inference: 92%|█████████▏| 5435/5920 [57:00<02:21, 3.44it/s, tok/s=16502.1]
Inference: 92%|█████████▏| 5435/5920 [57:00<02:21, 3.44it/s, tok/s=16502.1]
Inference: 92%|█████████▏| 5436/5920 [57:00<02:20, 3.44it/s, tok/s=16502.3]
Inference: 92%|█████████▏| 5437/5920 [57:00<01:42, 4.70it/s, tok/s=16502.3]
Inference: 92%|█████████▏| 5437/5920 [57:00<01:42, 4.70it/s, tok/s=16503.7]
Inference: 92%|█████████▏| 5438/5920 [57:00<02:06, 3.80it/s, tok/s=16503.7]
Inference: 92%|█████████▏| 5438/5920 [57:00<02:06, 3.80it/s, tok/s=16502.3]
Inference: 92%|█████████▏| 5439/5920 [57:00<02:06, 3.80it/s, tok/s=16502.7]
Inference: 92%|█████████▏| 5440/5920 [57:02<03:17, 2.43it/s, tok/s=16502.7]
Inference: 92%|█████████▏| 5440/5920 [57:02<03:17, 2.43it/s, tok/s=16497.4]
Inference: 92%|█████████▏| 5441/5920 [57:03<04:22, 1.82it/s, tok/s=16497.4]
Inference: 92%|█████████▏| 5441/5920 [57:03<04:22, 1.82it/s, tok/s=16494.6]
Inference: 92%|█████████▏| 5442/5920 [57:03<04:22, 1.82it/s, tok/s=16496.3]
Inference: 92%|█████████▏| 5443/5920 [57:03<04:21, 1.82it/s, tok/s=16497.8]
Inference: 92%|█████████▏| 5444/5920 [57:04<03:19, 2.39it/s, tok/s=16497.8]
Inference: 92%|█████████▏| 5444/5920 [57:04<03:19, 2.39it/s, tok/s=16496.3]
Inference: 92%|█████████▏| 5445/5920 [57:04<03:19, 2.39it/s, tok/s=16496.6]
Inference: 92%|█████████▏| 5446/5920 [57:04<02:52, 2.75it/s, tok/s=16496.6]
Inference: 92%|█████████▏| 5446/5920 [57:04<02:52, 2.75it/s, tok/s=16495.0]
Inference: 92%|█████████▏| 5447/5920 [57:05<03:08, 2.50it/s, tok/s=16495.0]
Inference: 92%|█████████▏| 5447/5920 [57:05<03:08, 2.50it/s, tok/s=16493.1]
Inference: 92%|█████████▏| 5448/5920 [57:05<03:08, 2.50it/s, tok/s=16493.9]
Inference: 92%|█████████▏| 5449/5920 [57:05<02:28, 3.17it/s, tok/s=16493.9]
Inference: 92%|█████████▏| 5449/5920 [57:05<02:28, 3.17it/s, tok/s=16498.6]
Inference: 92%|█████████▏| 5450/5920 [57:05<02:28, 3.17it/s, tok/s=16499.0]
Inference: 92%|█████████▏| 5451/5920 [57:05<01:52, 4.15it/s, tok/s=16499.0]
Inference: 92%|█████████▏| 5451/5920 [57:05<01:52, 4.15it/s, tok/s=16500.6]
Inference: 92%|█████████▏| 5452/5920 [57:05<01:46, 4.40it/s, tok/s=16500.6]
Inference: 92%|█████████▏| 5452/5920 [57:05<01:46, 4.40it/s, tok/s=16500.2]
Inference: 92%|█████████▏| 5453/5920 [57:05<01:46, 4.40it/s, tok/s=16500.6]
Inference: 92%|█████████▏| 5454/5920 [57:06<02:01, 3.85it/s, tok/s=16500.6]
Inference: 92%|█████████▏| 5454/5920 [57:06<02:01, 3.85it/s, tok/s=16498.8]
Inference: 92%|█████████▏| 5455/5920 [57:06<02:18, 3.35it/s, tok/s=16498.8]
Inference: 92%|█████████▏| 5455/5920 [57:06<02:18, 3.35it/s, tok/s=16497.2]
Inference: 92%|█████████▏| 5456/5920 [57:07<03:02, 2.54it/s, tok/s=16497.2]
Inference: 92%|█████████▏| 5456/5920 [57:07<03:02, 2.54it/s, tok/s=16498.7]
Inference: 92%|█████████▏| 5457/5920 [57:08<03:18, 2.33it/s, tok/s=16498.7]
Inference: 92%|█████████▏| 5457/5920 [57:08<03:18, 2.33it/s, tok/s=16499.1]
Inference: 92%|█████████▏| 5458/5920 [57:08<02:58, 2.59it/s, tok/s=16499.1]
Inference: 92%|█████████▏| 5458/5920 [57:08<02:58, 2.59it/s, tok/s=16500.9]
Inference: 92%|█████████▏| 5459/5920 [57:08<02:57, 2.59it/s, tok/s=16502.5]
Inference: 92%|█████████▏| 5460/5920 [57:09<03:03, 2.51it/s, tok/s=16502.5]
Inference: 92%|█████████▏| 5460/5920 [57:09<03:03, 2.51it/s, tok/s=16499.2]
Inference: 92%|█████████▏| 5461/5920 [57:09<03:16, 2.33it/s, tok/s=16499.2]
Inference: 92%|█████████▏| 5461/5920 [57:09<03:16, 2.33it/s, tok/s=16497.4]
Inference: 92%|█████████▏| 5462/5920 [57:09<02:40, 2.85it/s, tok/s=16497.4]
Inference: 92%|█████████▏| 5462/5920 [57:09<02:40, 2.85it/s, tok/s=16501.2]
Inference: 92%|█████████▏| 5463/5920 [57:09<02:40, 2.85it/s, tok/s=16503.1]
Inference: 92%|█████████▏| 5464/5920 [57:10<02:01, 3.75it/s, tok/s=16503.1]
Inference: 92%|█████████▏| 5464/5920 [57:10<02:01, 3.75it/s, tok/s=16503.4]
Inference: 92%|█████████▏| 5465/5920 [57:10<01:48, 4.19it/s, tok/s=16503.4]
Inference: 92%|█████████▏| 5465/5920 [57:10<01:48, 4.19it/s, tok/s=16503.7]
Inference: 92%|█████████▏| 5466/5920 [57:10<01:42, 4.43it/s, tok/s=16503.7]
Inference: 92%|█████████▏| 5466/5920 [57:10<01:42, 4.43it/s, tok/s=16505.1]
Inference: 92%|█████████▏| 5467/5920 [57:12<04:34, 1.65it/s, tok/s=16505.1]
Inference: 92%|█████████▏| 5467/5920 [57:12<04:34, 1.65it/s, tok/s=16497.8]
Inference: 92%|█████████▏| 5468/5920 [57:12<04:18, 1.75it/s, tok/s=16497.8]
Inference: 92%|█████████▏| 5468/5920 [57:12<04:18, 1.75it/s, tok/s=16500.8]
Inference: 92%|█████████▏| 5469/5920 [57:12<04:17, 1.75it/s, tok/s=16502.6]
Inference: 92%|█████████▏| 5470/5920 [57:12<04:17, 1.75it/s, tok/s=16506.2]
Inference: 92%|█████████▏| 5471/5920 [57:13<03:35, 2.08it/s, tok/s=16506.2]
Inference: 92%|█████████▏| 5471/5920 [57:13<03:35, 2.08it/s, tok/s=16502.3]
Inference: 92%|█████████▏| 5472/5920 [57:14<03:25, 2.18it/s, tok/s=16502.3]
Inference: 92%|█████████▏| 5472/5920 [57:14<03:25, 2.18it/s, tok/s=16503.2]
Inference: 92%|█████████▏| 5473/5920 [57:14<03:17, 2.26it/s, tok/s=16503.2]
Inference: 92%|█████████▏| 5473/5920 [57:14<03:17, 2.26it/s, tok/s=16504.3]
Inference: 92%|█████████▏| 5474/5920 [57:14<03:16, 2.27it/s, tok/s=16504.3]
Inference: 92%|█████████▏| 5474/5920 [57:14<03:16, 2.27it/s, tok/s=16502.9]
Inference: 92%|█████████▏| 5475/5920 [57:15<04:17, 1.73it/s, tok/s=16502.9]
Inference: 92%|█████████▏| 5475/5920 [57:15<04:17, 1.73it/s, tok/s=16499.2]
Inference: 92%|█████████▎| 5476/5920 [57:16<03:32, 2.09it/s, tok/s=16499.2]
Inference: 92%|█████████▎| 5476/5920 [57:16<03:32, 2.09it/s, tok/s=16501.2]
Inference: 93%|█████████▎| 5477/5920 [57:16<03:02, 2.43it/s, tok/s=16501.2]
Inference: 93%|█████████▎| 5477/5920 [57:16<03:02, 2.43it/s, tok/s=16502.6]
Inference: 93%|█████████▎| 5478/5920 [57:16<03:01, 2.43it/s, tok/s=16502.6]
Inference: 93%|█████████▎| 5478/5920 [57:16<03:01, 2.43it/s, tok/s=16503.3]
Inference: 93%|█████████▎| 5479/5920 [57:16<02:22, 3.10it/s, tok/s=16503.3]
Inference: 93%|█████████▎| 5479/5920 [57:16<02:22, 3.10it/s, tok/s=16504.2]
Inference: 93%|█████████▎| 5480/5920 [57:17<02:19, 3.16it/s, tok/s=16504.2]
Inference: 93%|█████████▎| 5480/5920 [57:17<02:19, 3.16it/s, tok/s=16505.5]
Inference: 93%|█████████▎| 5481/5920 [57:17<02:18, 3.16it/s, tok/s=16507.3]
Inference: 93%|█████████▎| 5482/5920 [57:17<01:35, 4.58it/s, tok/s=16507.3]
Inference: 93%|█████████▎| 5482/5920 [57:17<01:35, 4.58it/s, tok/s=16509.0]
Inference: 93%|█████████▎| 5483/5920 [57:17<01:32, 4.74it/s, tok/s=16509.0]
Inference: 93%|█████████▎| 5483/5920 [57:17<01:32, 4.74it/s, tok/s=16515.3]
Inference: 93%|█████████▎| 5484/5920 [57:17<01:32, 4.74it/s, tok/s=16516.1]
Inference: 93%|█████████▎| 5485/5920 [57:18<01:50, 3.95it/s, tok/s=16516.1]
Inference: 93%|█████████▎| 5485/5920 [57:18<01:50, 3.95it/s, tok/s=16514.9]
Inference: 93%|█████████▎| 5486/5920 [57:18<02:23, 3.03it/s, tok/s=16514.9]
Inference: 93%|█████████▎| 5486/5920 [57:18<02:23, 3.03it/s, tok/s=16513.4]
Inference: 93%|█████████▎| 5487/5920 [57:19<02:11, 3.30it/s, tok/s=16513.4]
Inference: 93%|█████████▎| 5487/5920 [57:19<02:11, 3.30it/s, tok/s=16512.9]
Inference: 93%|█████████▎| 5488/5920 [57:19<02:11, 3.30it/s, tok/s=16513.3]
Inference: 93%|█████████▎| 5489/5920 [57:19<02:29, 2.88it/s, tok/s=16513.3]
Inference: 93%|█████████▎| 5489/5920 [57:19<02:29, 2.88it/s, tok/s=16514.0]
Inference: 93%|█████████▎| 5490/5920 [57:20<03:02, 2.36it/s, tok/s=16514.0]
Inference: 93%|█████████▎| 5490/5920 [57:20<03:02, 2.36it/s, tok/s=16513.0]
Inference: 93%|█████████▎| 5491/5920 [57:20<03:02, 2.36it/s, tok/s=16515.9]
Inference: 93%|█████████▎| 5492/5920 [57:20<02:06, 3.39it/s, tok/s=16515.9]
Inference: 93%|█████████▎| 5492/5920 [57:20<02:06, 3.39it/s, tok/s=16515.8]
Inference: 93%|█████████▎| 5493/5920 [57:22<03:38, 1.96it/s, tok/s=16515.8]
Inference: 93%|█████████▎| 5493/5920 [57:22<03:38, 1.96it/s, tok/s=16512.7]
Inference: 93%|█████████▎| 5494/5920 [57:22<02:58, 2.38it/s, tok/s=16512.7]
Inference: 93%|█████████▎| 5494/5920 [57:22<02:58, 2.38it/s, tok/s=16512.7]
Inference: 93%|█████████▎| 5495/5920 [57:22<02:28, 2.87it/s, tok/s=16512.7]
Inference: 93%|█████████▎| 5495/5920 [57:22<02:28, 2.87it/s, tok/s=16515.3]
Inference: 93%|█████████▎| 5496/5920 [57:22<02:33, 2.76it/s, tok/s=16515.3]
Inference: 93%|█████████▎| 5496/5920 [57:22<02:33, 2.76it/s, tok/s=16516.5]
Inference: 93%|█████████▎| 5497/5920 [57:23<03:36, 1.96it/s, tok/s=16516.5]
Inference: 93%|█████████▎| 5497/5920 [57:23<03:36, 1.96it/s, tok/s=16515.7]
Inference: 93%|█████████▎| 5498/5920 [57:24<05:16, 1.33it/s, tok/s=16515.7]
Inference: 93%|█████████▎| 5498/5920 [57:24<05:16, 1.33it/s, tok/s=16511.3]
Inference: 93%|█████████▎| 5499/5920 [57:25<05:15, 1.33it/s, tok/s=16515.8]
Inference: 93%|█████████▎| 5500/5920 [57:25<03:07, 2.24it/s, tok/s=16515.8]
Inference: 93%|█████████▎| 5500/5920 [57:25<03:07, 2.24it/s, tok/s=16515.8]
Inference: 93%|█████████▎| 5501/5920 [57:25<03:44, 1.86it/s, tok/s=16515.8]
Inference: 93%|█████████▎| 5501/5920 [57:25<03:44, 1.86it/s, tok/s=16512.6]
Inference: 93%|█████████▎| 5502/5920 [57:26<03:44, 1.86it/s, tok/s=16516.0]
Inference: 93%|█████████▎| 5503/5920 [57:26<02:23, 2.91it/s, tok/s=16516.0]
Inference: 93%|█████████▎| 5503/5920 [57:26<02:23, 2.91it/s, tok/s=16517.5]
Inference: 93%|█████████▎| 5504/5920 [57:26<02:22, 2.91it/s, tok/s=16519.4]
Inference: 93%|█████████▎| 5505/5920 [57:26<02:14, 3.08it/s, tok/s=16519.4]
Inference: 93%|█████████▎| 5505/5920 [57:26<02:14, 3.08it/s, tok/s=16519.4]
Inference: 93%|█████████▎| 5506/5920 [57:27<02:45, 2.50it/s, tok/s=16519.4]
Inference: 93%|█████████▎| 5506/5920 [57:27<02:45, 2.50it/s, tok/s=16519.2]
Inference: 93%|█████████▎| 5507/5920 [57:27<02:26, 2.82it/s, tok/s=16519.2]
Inference: 93%|█████████▎| 5507/5920 [57:27<02:26, 2.82it/s, tok/s=16519.1]
Inference: 93%|█████████▎| 5508/5920 [57:27<02:26, 2.82it/s, tok/s=16521.5]
Inference: 93%|█████████▎| 5509/5920 [57:28<02:49, 2.43it/s, tok/s=16521.5]
Inference: 93%|█████████▎| 5509/5920 [57:28<02:49, 2.43it/s, tok/s=16521.8]
Inference: 93%|█████████▎| 5510/5920 [57:28<02:24, 2.85it/s, tok/s=16521.8]
Inference: 93%|█████████▎| 5510/5920 [57:28<02:24, 2.85it/s, tok/s=16521.6]
Inference: 93%|█████████▎| 5511/5920 [57:28<02:11, 3.11it/s, tok/s=16521.6]
Inference: 93%|█████████▎| 5511/5920 [57:28<02:11, 3.11it/s, tok/s=16523.4]
Inference: 93%|█████████▎| 5512/5920 [57:29<03:10, 2.15it/s, tok/s=16523.4]
Inference: 93%|█████████▎| 5512/5920 [57:29<03:10, 2.15it/s, tok/s=16520.4]
Inference: 93%|█████████▎| 5513/5920 [57:29<02:30, 2.71it/s, tok/s=16520.4]
Inference: 93%|█████████▎| 5513/5920 [57:29<02:30, 2.71it/s, tok/s=16524.0]
Inference: 93%|█████████▎| 5514/5920 [57:29<02:29, 2.71it/s, tok/s=16525.5]
Inference: 93%|█████████▎| 5515/5920 [57:30<02:25, 2.78it/s, tok/s=16525.5]
Inference: 93%|█████████▎| 5515/5920 [57:30<02:25, 2.78it/s, tok/s=16524.5]
Inference: 93%|█████████▎| 5516/5920 [57:30<02:25, 2.78it/s, tok/s=16527.5]
Inference: 93%|█████████▎| 5517/5920 [57:31<02:29, 2.69it/s, tok/s=16527.5]
Inference: 93%|█████████▎| 5517/5920 [57:31<02:29, 2.69it/s, tok/s=16524.7]
Inference: 93%|█████████▎| 5518/5920 [57:31<02:22, 2.83it/s, tok/s=16524.7]
Inference: 93%|█████████▎| 5518/5920 [57:31<02:22, 2.83it/s, tok/s=16529.0]
Inference: 93%|█████████▎| 5519/5920 [57:31<02:18, 2.89it/s, tok/s=16529.0]
Inference: 93%|█████████▎| 5519/5920 [57:31<02:18, 2.89it/s, tok/s=16529.5]
Inference: 93%|█████████▎| 5520/5920 [57:32<01:55, 3.47it/s, tok/s=16529.5]
Inference: 93%|█████████▎| 5520/5920 [57:32<01:55, 3.47it/s, tok/s=16529.6]
Inference: 93%|█████████▎| 5521/5920 [57:32<01:45, 3.79it/s, tok/s=16529.6]
Inference: 93%|█████████▎| 5521/5920 [57:32<01:45, 3.79it/s, tok/s=16533.5]
Inference: 93%|█████████▎| 5522/5920 [57:32<01:44, 3.79it/s, tok/s=16534.2]
Inference: 93%|█████████▎| 5523/5920 [57:32<01:09, 5.74it/s, tok/s=16534.2]
Inference: 93%|█████████▎| 5523/5920 [57:32<01:09, 5.74it/s, tok/s=16536.4]
Inference: 93%|█████████▎| 5524/5920 [57:32<01:05, 6.03it/s, tok/s=16536.4]
Inference: 93%|█████████▎| 5524/5920 [57:32<01:05, 6.03it/s, tok/s=16538.1]
Inference: 93%|█████████▎| 5525/5920 [57:33<01:51, 3.54it/s, tok/s=16538.1]
Inference: 93%|█████████▎| 5525/5920 [57:33<01:51, 3.54it/s, tok/s=16537.3]
Inference: 93%|█████████▎| 5526/5920 [57:33<01:48, 3.64it/s, tok/s=16537.3]
Inference: 93%|█████████▎| 5526/5920 [57:33<01:48, 3.64it/s, tok/s=16536.8]
Inference: 93%|█████████▎| 5527/5920 [57:33<01:50, 3.57it/s, tok/s=16536.8]
Inference: 93%|█████████▎| 5527/5920 [57:33<01:50, 3.57it/s, tok/s=16536.6]
Inference: 93%|█████████▎| 5528/5920 [57:35<03:43, 1.75it/s, tok/s=16536.6]
Inference: 93%|█████████▎| 5528/5920 [57:35<03:43, 1.75it/s, tok/s=16533.3]
Inference: 93%|█████████▎| 5529/5920 [57:35<03:51, 1.69it/s, tok/s=16533.3]
Inference: 93%|█████████▎| 5529/5920 [57:35<03:51, 1.69it/s, tok/s=16531.2]
Inference: 93%|█████████▎| 5530/5920 [57:35<03:02, 2.14it/s, tok/s=16531.2]
Inference: 93%|█████████▎| 5530/5920 [57:35<03:02, 2.14it/s, tok/s=16534.9]
Inference: 93%|█████████▎| 5531/5920 [57:36<02:40, 2.42it/s, tok/s=16534.9]
Inference: 93%|█████████▎| 5531/5920 [57:36<02:40, 2.42it/s, tok/s=16535.6]
Inference: 93%|█████████▎| 5532/5920 [57:36<02:15, 2.86it/s, tok/s=16535.6]
Inference: 93%|█████████▎| 5532/5920 [57:36<02:15, 2.86it/s, tok/s=16535.8]
Inference: 93%|█████████▎| 5533/5920 [57:36<01:50, 3.52it/s, tok/s=16535.8]
Inference: 93%|█████████▎| 5533/5920 [57:36<01:50, 3.52it/s, tok/s=16536.1]
Inference: 93%|█████████▎| 5534/5920 [57:36<01:41, 3.82it/s, tok/s=16536.1]
Inference: 93%|█████████▎| 5534/5920 [57:36<01:41, 3.82it/s, tok/s=16537.1]
Inference: 93%|█████████▎| 5535/5920 [57:36<01:40, 3.82it/s, tok/s=16538.9]
Inference: 94%|█████████▎| 5536/5920 [57:38<02:57, 2.16it/s, tok/s=16538.9]
Inference: 94%|█████████▎| 5536/5920 [57:38<02:57, 2.16it/s, tok/s=16536.1]
Inference: 94%|█████████▎| 5537/5920 [57:40<05:25, 1.17it/s, tok/s=16536.1]
Inference: 94%|█████████▎| 5537/5920 [57:40<05:25, 1.17it/s, tok/s=16527.1]
Inference: 94%|█████████▎| 5538/5920 [57:40<04:53, 1.30it/s, tok/s=16527.1]
Inference: 94%|█████████▎| 5538/5920 [57:40<04:53, 1.30it/s, tok/s=16526.5]
Inference: 94%|█████████▎| 5539/5920 [57:41<04:30, 1.41it/s, tok/s=16526.5]
Inference: 94%|█████████▎| 5539/5920 [57:41<04:30, 1.41it/s, tok/s=16525.1]
Inference: 94%|█████████▎| 5540/5920 [57:41<04:29, 1.41it/s, tok/s=16532.5]
Inference: 94%|█████████▎| 5541/5920 [57:41<04:28, 1.41it/s, tok/s=16533.0]
Inference: 94%|█████████▎| 5542/5920 [57:42<03:57, 1.59it/s, tok/s=16533.0]
Inference: 94%|█████████▎| 5542/5920 [57:42<03:57, 1.59it/s, tok/s=16529.0]
Inference: 94%|█████████▎| 5543/5920 [57:42<03:17, 1.91it/s, tok/s=16529.0]
Inference: 94%|█████████▎| 5543/5920 [57:42<03:17, 1.91it/s, tok/s=16531.2]
Inference: 94%|█████████▎| 5544/5920 [57:43<03:22, 1.86it/s, tok/s=16531.2]
Inference: 94%|█████████▎| 5544/5920 [57:43<03:22, 1.86it/s, tok/s=16530.6]
Inference: 94%|█████████▎| 5545/5920 [57:44<03:38, 1.72it/s, tok/s=16530.6]
Inference: 94%|█████████▎| 5545/5920 [57:44<03:38, 1.72it/s, tok/s=16532.5]
Inference: 94%|█████████▎| 5546/5920 [57:44<03:37, 1.72it/s, tok/s=16532.5]
Inference: 94%|█████████▎| 5546/5920 [57:44<03:37, 1.72it/s, tok/s=16537.4]
Inference: 94%|█████████▎| 5547/5920 [57:44<03:36, 1.72it/s, tok/s=16538.0]
Inference: 94%|█████████▎| 5548/5920 [57:45<02:23, 2.58it/s, tok/s=16538.0]
Inference: 94%|█████████▎| 5548/5920 [57:45<02:23, 2.58it/s, tok/s=16538.8]
Inference: 94%|█████████▎| 5549/5920 [57:45<03:05, 2.00it/s, tok/s=16538.8]
Inference: 94%|█████████▎| 5549/5920 [57:45<03:05, 2.00it/s, tok/s=16535.3]
Inference: 94%|█████████▍| 5550/5920 [57:46<02:40, 2.31it/s, tok/s=16535.3]
Inference: 94%|█████████▍| 5550/5920 [57:46<02:40, 2.31it/s, tok/s=16535.9]
Inference: 94%|█████████▍| 5551/5920 [57:46<02:11, 2.81it/s, tok/s=16535.9]
Inference: 94%|█████████▍| 5551/5920 [57:46<02:11, 2.81it/s, tok/s=16536.7]
Inference: 94%|█████████▍| 5552/5920 [57:46<02:11, 2.81it/s, tok/s=16539.5]
Inference: 94%|█████████▍| 5553/5920 [57:47<02:07, 2.87it/s, tok/s=16539.5]
Inference: 94%|█████████▍| 5553/5920 [57:47<02:07, 2.87it/s, tok/s=16539.2]
Inference: 94%|█████████▍| 5554/5920 [57:47<02:03, 2.97it/s, tok/s=16539.2]
Inference: 94%|█████████▍| 5554/5920 [57:47<02:03, 2.97it/s, tok/s=16542.0]
Inference: 94%|█████████▍| 5555/5920 [57:47<02:03, 2.97it/s, tok/s=16547.0]
Inference: 94%|█████████▍| 5556/5920 [57:47<02:02, 2.97it/s, tok/s=16553.4]
Inference: 94%|█████████▍| 5557/5920 [57:47<01:29, 4.04it/s, tok/s=16553.4]
Inference: 94%|█████████▍| 5557/5920 [57:47<01:29, 4.04it/s, tok/s=16552.5]
Inference: 94%|█████████▍| 5558/5920 [57:48<01:38, 3.66it/s, tok/s=16552.5]
Inference: 94%|█████████▍| 5558/5920 [57:48<01:38, 3.66it/s, tok/s=16553.4]
Inference: 94%|█████████▍| 5559/5920 [57:48<01:40, 3.58it/s, tok/s=16553.4]
Inference: 94%|█████████▍| 5559/5920 [57:48<01:40, 3.58it/s, tok/s=16553.6]
Inference: 94%|█████████▍| 5560/5920 [57:48<01:48, 3.33it/s, tok/s=16553.6]
Inference: 94%|█████████▍| 5560/5920 [57:48<01:48, 3.33it/s, tok/s=16553.5]
Inference: 94%|█████████▍| 5561/5920 [57:49<02:33, 2.34it/s, tok/s=16553.5]
Inference: 94%|█████████▍| 5561/5920 [57:49<02:33, 2.34it/s, tok/s=16553.1]
Inference: 94%|█████████▍| 5562/5920 [57:49<02:32, 2.34it/s, tok/s=16557.2]
Inference: 94%|█████████▍| 5563/5920 [57:49<02:32, 2.34it/s, tok/s=16558.8]
Inference: 94%|█████████▍| 5564/5920 [57:50<01:37, 3.64it/s, tok/s=16558.8]
Inference: 94%|█████████▍| 5564/5920 [57:50<01:37, 3.64it/s, tok/s=16563.2]
Inference: 94%|█████████▍| 5565/5920 [57:50<02:05, 2.83it/s, tok/s=16563.2]
Inference: 94%|█████████▍| 5565/5920 [57:50<02:05, 2.83it/s, tok/s=16562.0]
Inference: 94%|█████████▍| 5566/5920 [57:50<02:04, 2.83it/s, tok/s=16564.5]
Inference: 94%|█████████▍| 5567/5920 [57:51<01:42, 3.44it/s, tok/s=16564.5]
Inference: 94%|█████████▍| 5567/5920 [57:51<01:42, 3.44it/s, tok/s=16564.7]
Inference: 94%|█████████▍| 5568/5920 [57:51<01:38, 3.57it/s, tok/s=16564.7]
Inference: 94%|█████████▍| 5568/5920 [57:51<01:38, 3.57it/s, tok/s=16565.2]
Inference: 94%|█████████▍| 5569/5920 [57:51<01:33, 3.74it/s, tok/s=16565.2]
Inference: 94%|█████████▍| 5569/5920 [57:51<01:33, 3.74it/s, tok/s=16566.0]
Inference: 94%|█████████▍| 5570/5920 [57:51<01:40, 3.50it/s, tok/s=16566.0]
Inference: 94%|█████████▍| 5570/5920 [57:51<01:40, 3.50it/s, tok/s=16567.8]
Inference: 94%|█████████▍| 5571/5920 [57:52<01:38, 3.53it/s, tok/s=16567.8]
Inference: 94%|█████████▍| 5571/5920 [57:52<01:38, 3.53it/s, tok/s=16569.5]
Inference: 94%|█████████▍| 5572/5920 [57:52<01:27, 3.97it/s, tok/s=16569.5]
Inference: 94%|█████████▍| 5572/5920 [57:52<01:27, 3.97it/s, tok/s=16571.9]
Inference: 94%|█████████▍| 5573/5920 [57:53<02:27, 2.35it/s, tok/s=16571.9]
Inference: 94%|█████████▍| 5573/5920 [57:53<02:27, 2.35it/s, tok/s=16570.7]
Inference: 94%|█████████▍| 5574/5920 [57:53<02:06, 2.73it/s, tok/s=16570.7]
Inference: 94%|█████████▍| 5574/5920 [57:53<02:06, 2.73it/s, tok/s=16570.5]
Inference: 94%|█████████▍| 5575/5920 [57:53<02:06, 2.73it/s, tok/s=16574.5]
Inference: 94%|█████████▍| 5576/5920 [57:54<01:55, 2.97it/s, tok/s=16574.5]
Inference: 94%|█████████▍| 5576/5920 [57:54<01:55, 2.97it/s, tok/s=16572.9]
Inference: 94%|█████████▍| 5577/5920 [57:55<02:54, 1.96it/s, tok/s=16572.9]
Inference: 94%|█████████▍| 5577/5920 [57:55<02:54, 1.96it/s, tok/s=16574.5]
Inference: 94%|█████████▍| 5578/5920 [57:55<03:14, 1.76it/s, tok/s=16574.5]
Inference: 94%|█████████▍| 5578/5920 [57:55<03:14, 1.76it/s, tok/s=16573.6]
Inference: 94%|█████████▍| 5579/5920 [57:55<03:13, 1.76it/s, tok/s=16575.1]
Inference: 94%|█████████▍| 5580/5920 [57:56<02:12, 2.56it/s, tok/s=16575.1]
Inference: 94%|█████████▍| 5580/5920 [57:56<02:12, 2.56it/s, tok/s=16577.2]
Inference: 94%|█████████▍| 5581/5920 [57:56<02:12, 2.56it/s, tok/s=16585.8]
Inference: 94%|█████████▍| 5582/5920 [57:56<02:13, 2.53it/s, tok/s=16585.8]
Inference: 94%|█████████▍| 5582/5920 [57:56<02:13, 2.53it/s, tok/s=16584.0]
Inference: 94%|█████████▍| 5583/5920 [57:57<02:17, 2.44it/s, tok/s=16584.0]
Inference: 94%|█████████▍| 5583/5920 [57:57<02:17, 2.44it/s, tok/s=16585.6]
Inference: 94%|█████████▍| 5584/5920 [57:58<02:50, 1.96it/s, tok/s=16585.6]
Inference: 94%|█████████▍| 5584/5920 [57:58<02:50, 1.96it/s, tok/s=16586.8]
Inference: 94%|█████████▍| 5585/5920 [57:58<02:44, 2.03it/s, tok/s=16586.8]
Inference: 94%|█████████▍| 5585/5920 [57:58<02:44, 2.03it/s, tok/s=16586.1]
Inference: 94%|█████████▍| 5586/5920 [57:58<02:13, 2.50it/s, tok/s=16586.1]
Inference: 94%|█████████▍| 5586/5920 [57:58<02:13, 2.50it/s, tok/s=16588.0]
Inference: 94%|█████████▍| 5587/5920 [57:59<02:56, 1.89it/s, tok/s=16588.0]
Inference: 94%|█████████▍| 5587/5920 [57:59<02:56, 1.89it/s, tok/s=16587.1]
Inference: 94%|█████████▍| 5588/5920 [57:59<02:29, 2.22it/s, tok/s=16587.1]
Inference: 94%|█████████▍| 5588/5920 [57:59<02:29, 2.22it/s, tok/s=16589.3]
Inference: 94%|█████████▍| 5589/5920 [58:00<02:44, 2.02it/s, tok/s=16589.3]
Inference: 94%|█████████▍| 5589/5920 [58:00<02:44, 2.02it/s, tok/s=16590.5]
Inference: 94%|█████████▍| 5590/5920 [58:01<02:45, 2.00it/s, tok/s=16590.5]
Inference: 94%|█████████▍| 5590/5920 [58:01<02:45, 2.00it/s, tok/s=16591.8]
Inference: 94%|█████████▍| 5591/5920 [58:01<02:52, 1.91it/s, tok/s=16591.8]
Inference: 94%|█████████▍| 5591/5920 [58:01<02:52, 1.91it/s, tok/s=16592.4]
Inference: 94%|█████████▍| 5592/5920 [58:02<03:27, 1.58it/s, tok/s=16592.4]
Inference: 94%|█████████▍| 5592/5920 [58:02<03:27, 1.58it/s, tok/s=16591.9]
Inference: 94%|█████████▍| 5593/5920 [58:04<05:00, 1.09it/s, tok/s=16591.9]
Inference: 94%|█████████▍| 5593/5920 [58:04<05:00, 1.09it/s, tok/s=16587.0]
Inference: 94%|█████████▍| 5594/5920 [58:04<04:54, 1.11it/s, tok/s=16587.0]
Inference: 94%|█████████▍| 5594/5920 [58:04<04:54, 1.11it/s, tok/s=16585.7]
Inference: 95%|█████████▍| 5595/5920 [58:04<04:53, 1.11it/s, tok/s=16587.9]
Inference: 95%|█████████▍| 5596/5920 [58:05<02:46, 1.95it/s, tok/s=16587.9]
Inference: 95%|█████████▍| 5596/5920 [58:05<02:46, 1.95it/s, tok/s=16588.4]
Inference: 95%|█████████▍| 5597/5920 [58:05<02:45, 1.95it/s, tok/s=16591.3]
Inference: 95%|█████████▍| 5598/5920 [58:05<02:38, 2.03it/s, tok/s=16591.3]
Inference: 95%|█████████▍| 5598/5920 [58:05<02:38, 2.03it/s, tok/s=16591.4]
Inference: 95%|█████████▍| 5599/5920 [58:06<02:18, 2.32it/s, tok/s=16591.4]
Inference: 95%|█████████▍| 5599/5920 [58:06<02:18, 2.32it/s, tok/s=16593.6]
Inference: 95%|█████████▍| 5600/5920 [58:06<02:10, 2.45it/s, tok/s=16593.6]
Inference: 95%|█████████▍| 5600/5920 [58:06<02:10, 2.45it/s, tok/s=16594.8]
Inference: 95%|█████████▍| 5601/5920 [58:06<02:09, 2.47it/s, tok/s=16594.8]
Inference: 95%|█████████▍| 5601/5920 [58:06<02:09, 2.47it/s, tok/s=16596.5]
Inference: 95%|█████████▍| 5602/5920 [58:07<02:00, 2.63it/s, tok/s=16596.5]
Inference: 95%|█████████▍| 5602/5920 [58:07<02:00, 2.63it/s, tok/s=16598.2]
Inference: 95%|█████████▍| 5603/5920 [58:07<01:47, 2.95it/s, tok/s=16598.2]
Inference: 95%|█████████▍| 5603/5920 [58:07<01:47, 2.95it/s, tok/s=16601.2]
Inference: 95%|█████████▍| 5604/5920 [58:07<01:47, 2.95it/s, tok/s=16604.2]
Inference: 95%|█████████▍| 5605/5920 [58:07<01:20, 3.92it/s, tok/s=16604.2]
Inference: 95%|█████████▍| 5605/5920 [58:07<01:20, 3.92it/s, tok/s=16605.4]
Inference: 95%|█████████▍| 5606/5920 [58:08<01:40, 3.11it/s, tok/s=16605.4]
Inference: 95%|█████████▍| 5606/5920 [58:08<01:40, 3.11it/s, tok/s=16611.5]
Inference: 95%|█████████▍| 5607/5920 [58:08<01:40, 3.11it/s, tok/s=16614.7]
Inference: 95%|█████████▍| 5608/5920 [58:08<01:27, 3.59it/s, tok/s=16614.7]
Inference: 95%|█████████▍| 5608/5920 [58:08<01:27, 3.59it/s, tok/s=16615.7]
Inference: 95%|█████████▍| 5609/5920 [58:09<01:27, 3.54it/s, tok/s=16615.7]
Inference: 95%|█████████▍| 5609/5920 [58:09<01:27, 3.54it/s, tok/s=16618.6]
Inference: 95%|█████████▍| 5610/5920 [58:09<02:16, 2.27it/s, tok/s=16618.6]
Inference: 95%|█████████▍| 5610/5920 [58:09<02:16, 2.27it/s, tok/s=16619.1]
Inference: 95%|█████████▍| 5611/5920 [58:10<02:10, 2.37it/s, tok/s=16619.1]
Inference: 95%|█████████▍| 5611/5920 [58:10<02:10, 2.37it/s, tok/s=16619.9]
Inference: 95%|█████████▍| 5612/5920 [58:10<02:00, 2.55it/s, tok/s=16619.9]
Inference: 95%|█████████▍| 5612/5920 [58:10<02:00, 2.55it/s, tok/s=16623.2]
Inference: 95%|█████████▍| 5613/5920 [58:10<02:00, 2.55it/s, tok/s=16624.8]
Inference: 95%|█████████▍| 5614/5920 [58:12<03:37, 1.40it/s, tok/s=16624.8]
Inference: 95%|█████████▍| 5614/5920 [58:12<03:37, 1.40it/s, tok/s=16617.4]
Inference: 95%|█████████▍| 5615/5920 [58:13<03:12, 1.58it/s, tok/s=16617.4]
Inference: 95%|█████████▍| 5615/5920 [58:13<03:12, 1.58it/s, tok/s=16617.2]
Inference: 95%|█████████▍| 5616/5920 [58:13<02:39, 1.91it/s, tok/s=16617.2]
Inference: 95%|█████████▍| 5616/5920 [58:13<02:39, 1.91it/s, tok/s=16618.4]
Inference: 95%|█████████▍| 5617/5920 [58:13<02:34, 1.96it/s, tok/s=16618.4]
Inference: 95%|█████████▍| 5617/5920 [58:13<02:34, 1.96it/s, tok/s=16621.5]
Inference: 95%|█████████▍| 5618/5920 [58:14<02:22, 2.12it/s, tok/s=16621.5]
Inference: 95%|█████████▍| 5618/5920 [58:14<02:22, 2.12it/s, tok/s=16622.9]
Inference: 95%|█████████▍| 5619/5920 [58:14<01:53, 2.65it/s, tok/s=16622.9]
Inference: 95%|█████████▍| 5619/5920 [58:14<01:53, 2.65it/s, tok/s=16626.0]
Inference: 95%|█████████▍| 5620/5920 [58:15<02:11, 2.28it/s, tok/s=16626.0]
Inference: 95%|█████████▍| 5620/5920 [58:15<02:11, 2.28it/s, tok/s=16625.5]
Inference: 95%|█████████▍| 5621/5920 [58:15<02:11, 2.28it/s, tok/s=16626.7]
Inference: 95%|█████████▍| 5622/5920 [58:16<02:15, 2.19it/s, tok/s=16626.7]
Inference: 95%|█████████▍| 5622/5920 [58:16<02:15, 2.19it/s, tok/s=16629.2]
Inference: 95%|█████████▍| 5623/5920 [58:16<01:51, 2.65it/s, tok/s=16629.2]
Inference: 95%|█████████▍| 5623/5920 [58:16<01:51, 2.65it/s, tok/s=16631.0]
Inference: 95%|█████████▌| 5624/5920 [58:16<01:50, 2.68it/s, tok/s=16631.0]
Inference: 95%|█████████▌| 5624/5920 [58:16<01:50, 2.68it/s, tok/s=16631.9]
Inference: 95%|█████████▌| 5625/5920 [58:17<02:56, 1.67it/s, tok/s=16631.9]
Inference: 95%|█████████▌| 5625/5920 [58:17<02:56, 1.67it/s, tok/s=16628.2]
Inference: 95%|█████████▌| 5626/5920 [58:17<02:56, 1.67it/s, tok/s=16629.3]
Inference: 95%|█████████▌| 5627/5920 [58:18<02:00, 2.44it/s, tok/s=16629.3]
Inference: 95%|█████████▌| 5627/5920 [58:18<02:00, 2.44it/s, tok/s=16630.3]
Inference: 95%|█████████▌| 5628/5920 [58:18<02:09, 2.25it/s, tok/s=16630.3]
Inference: 95%|█████████▌| 5628/5920 [58:18<02:09, 2.25it/s, tok/s=16629.8]
Inference: 95%|█████████▌| 5629/5920 [58:18<01:55, 2.52it/s, tok/s=16629.8]
Inference: 95%|█████████▌| 5629/5920 [58:18<01:55, 2.52it/s, tok/s=16630.6]
Inference: 95%|█████████▌| 5630/5920 [58:19<01:44, 2.79it/s, tok/s=16630.6]
Inference: 95%|█████████▌| 5630/5920 [58:19<01:44, 2.79it/s, tok/s=16631.4]
Inference: 95%|█████████▌| 5631/5920 [58:19<01:50, 2.60it/s, tok/s=16631.4]
Inference: 95%|█████████▌| 5631/5920 [58:19<01:50, 2.60it/s, tok/s=16632.5]
Inference: 95%|█████████▌| 5632/5920 [58:20<02:53, 1.66it/s, tok/s=16632.5]
Inference: 95%|█████████▌| 5632/5920 [58:20<02:53, 1.66it/s, tok/s=16629.3]
Inference: 95%|█████████▌| 5633/5920 [58:20<02:16, 2.11it/s, tok/s=16629.3]
Inference: 95%|█████████▌| 5633/5920 [58:20<02:16, 2.11it/s, tok/s=16634.0]
Inference: 95%|█████████▌| 5634/5920 [58:20<02:15, 2.11it/s, tok/s=16638.0]
Inference: 95%|█████████▌| 5635/5920 [58:21<01:26, 3.31it/s, tok/s=16638.0]
Inference: 95%|█████████▌| 5635/5920 [58:21<01:26, 3.31it/s, tok/s=16638.2]
Inference: 95%|█████████▌| 5636/5920 [58:22<02:29, 1.90it/s, tok/s=16638.2]
Inference: 95%|█████████▌| 5636/5920 [58:22<02:29, 1.90it/s, tok/s=16634.2]
Inference: 95%|█████████▌| 5637/5920 [58:22<02:23, 1.98it/s, tok/s=16634.2]
Inference: 95%|█████████▌| 5637/5920 [58:22<02:23, 1.98it/s, tok/s=16638.2]
Inference: 95%|█████████▌| 5638/5920 [58:23<02:16, 2.07it/s, tok/s=16638.2]
Inference: 95%|█████████▌| 5638/5920 [58:23<02:16, 2.07it/s, tok/s=16644.8]
Inference: 95%|█████████▌| 5639/5920 [58:23<02:07, 2.20it/s, tok/s=16644.8]
Inference: 95%|█████████▌| 5639/5920 [58:23<02:07, 2.20it/s, tok/s=16647.0]
Inference: 95%|█████████▌| 5640/5920 [58:23<01:52, 2.50it/s, tok/s=16647.0]
Inference: 95%|█████████▌| 5640/5920 [58:23<01:52, 2.50it/s, tok/s=16647.8]
Inference: 95%|█████████▌| 5641/5920 [58:24<02:05, 2.23it/s, tok/s=16647.8]
Inference: 95%|█████████▌| 5641/5920 [58:24<02:05, 2.23it/s, tok/s=16646.1]
Inference: 95%|█████████▌| 5642/5920 [58:25<03:13, 1.44it/s, tok/s=16646.1]
Inference: 95%|█████████▌| 5642/5920 [58:25<03:13, 1.44it/s, tok/s=16643.6]
Inference: 95%|█████████▌| 5643/5920 [58:25<02:36, 1.77it/s, tok/s=16643.6]
Inference: 95%|█████████▌| 5643/5920 [58:25<02:36, 1.77it/s, tok/s=16643.8]
Inference: 95%|█████████▌| 5644/5920 [58:27<04:19, 1.06it/s, tok/s=16643.8]
Inference: 95%|█████████▌| 5644/5920 [58:27<04:19, 1.06it/s, tok/s=16639.9]
Inference: 95%|█████████▌| 5645/5920 [58:28<03:55, 1.17it/s, tok/s=16639.9]
Inference: 95%|█████████▌| 5645/5920 [58:28<03:55, 1.17it/s, tok/s=16639.2]
Inference: 95%|█████████▌| 5646/5920 [58:29<03:43, 1.22it/s, tok/s=16639.2]
Inference: 95%|█████████▌| 5646/5920 [58:29<03:43, 1.22it/s, tok/s=16639.3]
Inference: 95%|█████████▌| 5647/5920 [58:29<03:12, 1.42it/s, tok/s=16639.3]
Inference: 95%|█████████▌| 5647/5920 [58:29<03:12, 1.42it/s, tok/s=16639.5]
Inference: 95%|█████████▌| 5648/5920 [58:31<05:26, 1.20s/it, tok/s=16639.5]
Inference: 95%|█████████▌| 5648/5920 [58:31<05:26, 1.20s/it, tok/s=16631.0]
Inference: 95%|█████████▌| 5649/5920 [58:32<03:59, 1.13it/s, tok/s=16631.0]
Inference: 95%|█████████▌| 5649/5920 [58:32<03:59, 1.13it/s, tok/s=16633.3]
Inference: 95%|█████████▌| 5650/5920 [58:32<03:14, 1.39it/s, tok/s=16633.3]
Inference: 95%|█████████▌| 5650/5920 [58:32<03:14, 1.39it/s, tok/s=16636.6]
Inference: 95%|█████████▌| 5651/5920 [58:32<02:58, 1.51it/s, tok/s=16636.6]
Inference: 95%|█████████▌| 5651/5920 [58:32<02:58, 1.51it/s, tok/s=16636.2]
Inference: 95%|█████████▌| 5652/5920 [58:33<02:31, 1.77it/s, tok/s=16636.2]
Inference: 95%|█████████▌| 5652/5920 [58:33<02:31, 1.77it/s, tok/s=16636.0]
Inference: 95%|█████████▌| 5653/5920 [58:33<02:27, 1.81it/s, tok/s=16636.0]
Inference: 95%|█████████▌| 5653/5920 [58:33<02:27, 1.81it/s, tok/s=16635.7]
Inference: 96%|█████████▌| 5654/5920 [58:34<02:17, 1.93it/s, tok/s=16635.7]
Inference: 96%|█████████▌| 5654/5920 [58:34<02:17, 1.93it/s, tok/s=16635.0]
Inference: 96%|█████████▌| 5655/5920 [58:34<02:09, 2.05it/s, tok/s=16635.0]
Inference: 96%|█████████▌| 5655/5920 [58:34<02:09, 2.05it/s, tok/s=16635.2]
Inference: 96%|█████████▌| 5656/5920 [58:35<01:58, 2.23it/s, tok/s=16635.2]
Inference: 96%|█████████▌| 5656/5920 [58:35<01:58, 2.23it/s, tok/s=16636.9]
Inference: 96%|█████████▌| 5657/5920 [58:35<01:39, 2.63it/s, tok/s=16636.9]
Inference: 96%|█████████▌| 5657/5920 [58:35<01:39, 2.63it/s, tok/s=16638.3]
Inference: 96%|█████████▌| 5658/5920 [58:35<01:30, 2.90it/s, tok/s=16638.3]
Inference: 96%|█████████▌| 5658/5920 [58:35<01:30, 2.90it/s, tok/s=16639.2]
Inference: 96%|█████████▌| 5659/5920 [58:35<01:19, 3.27it/s, tok/s=16639.2]
Inference: 96%|█████████▌| 5659/5920 [58:35<01:19, 3.27it/s, tok/s=16641.0]
Inference: 96%|█████████▌| 5660/5920 [58:36<01:48, 2.40it/s, tok/s=16641.0]
Inference: 96%|█████████▌| 5660/5920 [58:36<01:48, 2.40it/s, tok/s=16639.0]
Inference: 96%|█████████▌| 5661/5920 [58:36<01:34, 2.74it/s, tok/s=16639.0]
Inference: 96%|█████████▌| 5661/5920 [58:36<01:34, 2.74it/s, tok/s=16641.8]
Inference: 96%|█████████▌| 5662/5920 [58:37<01:49, 2.35it/s, tok/s=16641.8]
Inference: 96%|█████████▌| 5662/5920 [58:37<01:49, 2.35it/s, tok/s=16642.3]
Inference: 96%|█████████▌| 5663/5920 [58:38<03:22, 1.27it/s, tok/s=16642.3]
Inference: 96%|█████████▌| 5663/5920 [58:38<03:22, 1.27it/s, tok/s=16638.1]
Inference: 96%|█████████▌| 5664/5920 [58:39<02:41, 1.59it/s, tok/s=16638.1]
Inference: 96%|█████████▌| 5664/5920 [58:39<02:41, 1.59it/s, tok/s=16640.2]
Inference: 96%|█████████▌| 5665/5920 [58:39<02:04, 2.05it/s, tok/s=16640.2]
Inference: 96%|█████████▌| 5665/5920 [58:39<02:04, 2.05it/s, tok/s=16641.7]
Inference: 96%|█████████▌| 5666/5920 [58:39<01:42, 2.48it/s, tok/s=16641.7]
Inference: 96%|█████████▌| 5666/5920 [58:39<01:42, 2.48it/s, tok/s=16643.4]
Inference: 96%|█████████▌| 5667/5920 [58:39<01:35, 2.64it/s, tok/s=16643.4]
Inference: 96%|█████████▌| 5667/5920 [58:39<01:35, 2.64it/s, tok/s=16647.6]
Inference: 96%|█████████▌| 5668/5920 [58:40<01:37, 2.60it/s, tok/s=16647.6]
Inference: 96%|█████████▌| 5668/5920 [58:40<01:37, 2.60it/s, tok/s=16650.4]
Inference: 96%|█████████▌| 5669/5920 [58:40<01:36, 2.60it/s, tok/s=16652.7]
Inference: 96%|█████████▌| 5670/5920 [58:40<01:00, 4.15it/s, tok/s=16652.7]
Inference: 96%|█████████▌| 5670/5920 [58:40<01:00, 4.15it/s, tok/s=16654.6]
Inference: 96%|█████████▌| 5671/5920 [58:40<01:12, 3.44it/s, tok/s=16654.6]
Inference: 96%|█████████▌| 5671/5920 [58:40<01:12, 3.44it/s, tok/s=16658.6]
Inference: 96%|█████████▌| 5672/5920 [58:41<01:17, 3.21it/s, tok/s=16658.6]
Inference: 96%|█████████▌| 5672/5920 [58:41<01:17, 3.21it/s, tok/s=16659.5]
Inference: 96%|█████████▌| 5673/5920 [58:41<01:03, 3.91it/s, tok/s=16659.5]
Inference: 96%|█████████▌| 5673/5920 [58:41<01:03, 3.91it/s, tok/s=16661.0]
Inference: 96%|█████████▌| 5674/5920 [58:41<01:08, 3.58it/s, tok/s=16661.0]
Inference: 96%|█████████▌| 5674/5920 [58:41<01:08, 3.58it/s, tok/s=16661.8]
Inference: 96%|█████████▌| 5675/5920 [58:41<01:01, 3.98it/s, tok/s=16661.8]
Inference: 96%|█████████▌| 5675/5920 [58:41<01:01, 3.98it/s, tok/s=16663.1]
Inference: 96%|█████████▌| 5676/5920 [58:41<01:01, 3.98it/s, tok/s=16666.8]
Inference: 96%|█████████▌| 5677/5920 [58:43<01:43, 2.36it/s, tok/s=16666.8]
Inference: 96%|█████████▌| 5677/5920 [58:43<01:43, 2.36it/s, tok/s=16662.7]
Inference: 96%|█████████▌| 5678/5920 [58:43<01:41, 2.37it/s, tok/s=16662.7]
Inference: 96%|█████████▌| 5678/5920 [58:43<01:41, 2.37it/s, tok/s=16662.7]
Inference: 96%|█████████▌| 5679/5920 [58:44<02:30, 1.60it/s, tok/s=16662.7]
Inference: 96%|█████████▌| 5679/5920 [58:44<02:30, 1.60it/s, tok/s=16658.9]
Inference: 96%|█████████▌| 5680/5920 [58:45<03:08, 1.27it/s, tok/s=16658.9]
Inference: 96%|█████████▌| 5680/5920 [58:45<03:08, 1.27it/s, tok/s=16658.4]
Inference: 96%|█████████▌| 5681/5920 [58:46<02:31, 1.58it/s, tok/s=16658.4]
Inference: 96%|█████████▌| 5681/5920 [58:46<02:31, 1.58it/s, tok/s=16661.0]
Inference: 96%|█████████▌| 5682/5920 [58:46<02:10, 1.83it/s, tok/s=16661.0]
Inference: 96%|█████████▌| 5682/5920 [58:46<02:10, 1.83it/s, tok/s=16662.0]
Inference: 96%|█████████▌| 5683/5920 [58:46<02:05, 1.89it/s, tok/s=16662.0]
Inference: 96%|█████████▌| 5683/5920 [58:46<02:05, 1.89it/s, tok/s=16663.9]
Inference: 96%|█████████▌| 5684/5920 [58:47<02:05, 1.89it/s, tok/s=16668.4]
Inference: 96%|█████████▌| 5685/5920 [58:47<01:43, 2.26it/s, tok/s=16668.4]
Inference: 96%|█████████▌| 5685/5920 [58:47<01:43, 2.26it/s, tok/s=16670.1]
Inference: 96%|█████████▌| 5686/5920 [58:47<01:26, 2.70it/s, tok/s=16670.1]
Inference: 96%|█████████▌| 5686/5920 [58:47<01:26, 2.70it/s, tok/s=16672.0]
Inference: 96%|█████████▌| 5687/5920 [58:48<01:32, 2.52it/s, tok/s=16672.0]
Inference: 96%|█████████▌| 5687/5920 [58:48<01:32, 2.52it/s, tok/s=16672.0]
Inference: 96%|█████████▌| 5688/5920 [58:48<01:31, 2.52it/s, tok/s=16677.7]
Inference: 96%|█████████▌| 5689/5920 [58:48<01:31, 2.52it/s, tok/s=16678.9]
Inference: 96%|█████████▌| 5690/5920 [58:49<01:24, 2.73it/s, tok/s=16678.9]
Inference: 96%|█████████▌| 5690/5920 [58:49<01:24, 2.73it/s, tok/s=16678.2]
Inference: 96%|█████████▌| 5691/5920 [58:50<02:19, 1.65it/s, tok/s=16678.2]
Inference: 96%|█████████▌| 5691/5920 [58:50<02:19, 1.65it/s, tok/s=16675.0]
Inference: 96%|█████████▌| 5692/5920 [58:52<02:56, 1.30it/s, tok/s=16675.0]
Inference: 96%|█████████▌| 5692/5920 [58:52<02:56, 1.30it/s, tok/s=16672.5]
Inference: 96%|█████████▌| 5693/5920 [58:52<02:25, 1.56it/s, tok/s=16672.5]
Inference: 96%|█████████▌| 5693/5920 [58:52<02:25, 1.56it/s, tok/s=16675.5]
Inference: 96%|█████████▌| 5694/5920 [58:52<01:59, 1.88it/s, tok/s=16675.5]
Inference: 96%|█████████▌| 5694/5920 [58:52<01:59, 1.88it/s, tok/s=16681.8]
Inference: 96%|█████████▌| 5695/5920 [58:53<01:56, 1.94it/s, tok/s=16681.8]
Inference: 96%|█████████▌| 5695/5920 [58:53<01:56, 1.94it/s, tok/s=16688.0]
Inference: 96%|█████████▌| 5696/5920 [58:54<03:01, 1.23it/s, tok/s=16688.0]
Inference: 96%|█████████▌| 5696/5920 [58:54<03:01, 1.23it/s, tok/s=16687.7]
Inference: 96%|█████████▌| 5697/5920 [58:56<04:18, 1.16s/it, tok/s=16687.7]
Inference: 96%|█████████▌| 5697/5920 [58:56<04:18, 1.16s/it, tok/s=16683.8]
Inference: 96%|█████████▋| 5698/5920 [59:00<07:33, 2.04s/it, tok/s=16683.8]
Inference: 96%|█████████▋| 5698/5920 [59:00<07:33, 2.04s/it, tok/s=16667.3]
Inference: 96%|█████████▋| 5699/5920 [59:01<05:43, 1.55s/it, tok/s=16667.3]
Inference: 96%|█████████▋| 5699/5920 [59:01<05:43, 1.55s/it, tok/s=16667.6]
Inference: 96%|█████████▋| 5700/5920 [59:02<05:00, 1.37s/it, tok/s=16667.6]
Inference: 96%|█████████▋| 5700/5920 [59:02<05:00, 1.37s/it, tok/s=16665.7]
Inference: 96%|█████████▋| 5701/5920 [59:02<03:42, 1.02s/it, tok/s=16665.7]
Inference: 96%|█████████▋| 5701/5920 [59:02<03:42, 1.02s/it, tok/s=16667.7]
Inference: 96%|█████████▋| 5702/5920 [59:03<04:18, 1.18s/it, tok/s=16667.7]
Inference: 96%|█████████▋| 5702/5920 [59:03<04:18, 1.18s/it, tok/s=16663.0]
Inference: 96%|█████████▋| 5703/5920 [59:06<05:32, 1.53s/it, tok/s=16663.0]
Inference: 96%|█████████▋| 5703/5920 [59:06<05:32, 1.53s/it, tok/s=16654.6]
Inference: 96%|█████████▋| 5704/5920 [59:06<04:32, 1.26s/it, tok/s=16654.6]
Inference: 96%|█████████▋| 5704/5920 [59:06<04:32, 1.26s/it, tok/s=16655.2]
Inference: 96%|█████████▋| 5705/5920 [59:07<03:38, 1.02s/it, tok/s=16655.2]
Inference: 96%|█████████▋| 5705/5920 [59:07<03:38, 1.02s/it, tok/s=16655.3]
Inference: 96%|█████████▋| 5706/5920 [59:08<03:27, 1.03it/s, tok/s=16655.3]
Inference: 96%|█████████▋| 5706/5920 [59:08<03:27, 1.03it/s, tok/s=16654.8]
Inference: 96%|█████████▋| 5707/5920 [59:08<02:36, 1.36it/s, tok/s=16654.8]
Inference: 96%|█████████▋| 5707/5920 [59:08<02:36, 1.36it/s, tok/s=16658.9]
Inference: 96%|█████████▋| 5708/5920 [59:08<01:57, 1.81it/s, tok/s=16658.9]
Inference: 96%|█████████▋| 5708/5920 [59:08<01:57, 1.81it/s, tok/s=16660.7]
Inference: 96%|█████████▋| 5709/5920 [59:08<01:48, 1.95it/s, tok/s=16660.7]
Inference: 96%|█████████▋| 5709/5920 [59:08<01:48, 1.95it/s, tok/s=16663.5]
Inference: 96%|█████████▋| 5710/5920 [59:09<01:36, 2.17it/s, tok/s=16663.5]
Inference: 96%|█████████▋| 5710/5920 [59:09<01:36, 2.17it/s, tok/s=16664.2]
Inference: 96%|█████████▋| 5711/5920 [59:09<01:18, 2.65it/s, tok/s=16664.2]
Inference: 96%|█████████▋| 5711/5920 [59:09<01:18, 2.65it/s, tok/s=16667.0]
Inference: 96%|█████████▋| 5712/5920 [59:12<03:45, 1.08s/it, tok/s=16667.0]
Inference: 96%|█████████▋| 5712/5920 [59:12<03:45, 1.08s/it, tok/s=16659.1]
Inference: 97%|█████████▋| 5713/5920 [59:13<03:43, 1.08s/it, tok/s=16659.1]
Inference: 97%|█████████▋| 5713/5920 [59:13<03:43, 1.08s/it, tok/s=16656.6]
Inference: 97%|█████████▋| 5714/5920 [59:14<03:32, 1.03s/it, tok/s=16656.6]
Inference: 97%|█████████▋| 5714/5920 [59:14<03:32, 1.03s/it, tok/s=16654.5]
Inference: 97%|█████████▋| 5715/5920 [59:14<02:51, 1.19it/s, tok/s=16654.5]
Inference: 97%|█████████▋| 5715/5920 [59:14<02:51, 1.19it/s, tok/s=16655.8]
Inference: 97%|█████████▋| 5716/5920 [59:15<02:48, 1.21it/s, tok/s=16655.8]
Inference: 97%|█████████▋| 5716/5920 [59:15<02:48, 1.21it/s, tok/s=16654.1]
Inference: 97%|█████████▋| 5717/5920 [59:15<02:29, 1.35it/s, tok/s=16654.1]
Inference: 97%|█████████▋| 5717/5920 [59:15<02:29, 1.35it/s, tok/s=16654.0]
Inference: 97%|█████████▋| 5718/5920 [59:16<02:45, 1.22it/s, tok/s=16654.0]
Inference: 97%|█████████▋| 5718/5920 [59:16<02:45, 1.22it/s, tok/s=16652.9]
Inference: 97%|█████████▋| 5719/5920 [59:17<02:16, 1.47it/s, tok/s=16652.9]
Inference: 97%|█████████▋| 5719/5920 [59:17<02:16, 1.47it/s, tok/s=16653.4]
Inference: 97%|█████████▋| 5720/5920 [59:17<02:07, 1.57it/s, tok/s=16653.4]
Inference: 97%|█████████▋| 5720/5920 [59:17<02:07, 1.57it/s, tok/s=16652.4]
Inference: 97%|█████████▋| 5721/5920 [59:17<01:37, 2.04it/s, tok/s=16652.4]
Inference: 97%|█████████▋| 5721/5920 [59:17<01:37, 2.04it/s, tok/s=16654.2]
Inference: 97%|█████████▋| 5722/5920 [59:19<02:23, 1.38it/s, tok/s=16654.2]
Inference: 97%|█████████▋| 5722/5920 [59:19<02:23, 1.38it/s, tok/s=16653.0]
Inference: 97%|█████████▋| 5723/5920 [59:19<02:19, 1.41it/s, tok/s=16653.0]
Inference: 97%|█████████▋| 5723/5920 [59:19<02:19, 1.41it/s, tok/s=16654.0]
Inference: 97%|█████████▋| 5724/5920 [59:20<01:49, 1.79it/s, tok/s=16654.0]
Inference: 97%|█████████▋| 5724/5920 [59:20<01:49, 1.79it/s, tok/s=16659.3]
Inference: 97%|█████████▋| 5725/5920 [59:20<01:34, 2.07it/s, tok/s=16659.3]
Inference: 97%|█████████▋| 5725/5920 [59:20<01:34, 2.07it/s, tok/s=16660.8]
Inference: 97%|█████████▋| 5726/5920 [59:20<01:16, 2.54it/s, tok/s=16660.8]
Inference: 97%|█████████▋| 5726/5920 [59:20<01:16, 2.54it/s, tok/s=16661.9]
Inference: 97%|█████████▋| 5727/5920 [59:21<01:42, 1.89it/s, tok/s=16661.9]
Inference: 97%|█████████▋| 5727/5920 [59:21<01:42, 1.89it/s, tok/s=16661.5]
Inference: 97%|█████████▋| 5728/5920 [59:23<02:41, 1.19it/s, tok/s=16661.5]
Inference: 97%|█████████▋| 5728/5920 [59:23<02:41, 1.19it/s, tok/s=16656.8]
Inference: 97%|█████████▋| 5729/5920 [59:24<03:11, 1.00s/it, tok/s=16656.8]
Inference: 97%|█████████▋| 5729/5920 [59:24<03:11, 1.00s/it, tok/s=16652.2]
Inference: 97%|█████████▋| 5730/5920 [59:26<03:43, 1.18s/it, tok/s=16652.2]
Inference: 97%|█████████▋| 5730/5920 [59:26<03:43, 1.18s/it, tok/s=16647.5]
Inference: 97%|█████████▋| 5731/5920 [59:27<04:22, 1.39s/it, tok/s=16647.5]
Inference: 97%|█████████▋| 5731/5920 [59:27<04:22, 1.39s/it, tok/s=16642.5]
Inference: 97%|█████████▋| 5732/5920 [59:27<04:20, 1.39s/it, tok/s=16646.0]
Inference: 97%|█████████▋| 5733/5920 [59:28<02:41, 1.16it/s, tok/s=16646.0]
Inference: 97%|█████████▋| 5733/5920 [59:28<02:41, 1.16it/s, tok/s=16648.1]
Inference: 97%|█████████▋| 5734/5920 [59:29<02:50, 1.09it/s, tok/s=16648.1]
Inference: 97%|█████████▋| 5734/5920 [59:29<02:50, 1.09it/s, tok/s=16645.9]
Inference: 97%|█████████▋| 5735/5920 [59:31<03:28, 1.13s/it, tok/s=16645.9]
Inference: 97%|█████████▋| 5735/5920 [59:31<03:28, 1.13s/it, tok/s=16640.6]
Inference: 97%|█████████▋| 5736/5920 [59:31<03:27, 1.13s/it, tok/s=16642.6]
Inference: 97%|█████████▋| 5737/5920 [59:31<02:12, 1.38it/s, tok/s=16642.6]
Inference: 97%|█████████▋| 5737/5920 [59:31<02:12, 1.38it/s, tok/s=16646.4]
Inference: 97%|█████████▋| 5738/5920 [59:31<01:55, 1.58it/s, tok/s=16646.4]
Inference: 97%|█████████▋| 5738/5920 [59:31<01:55, 1.58it/s, tok/s=16649.9]
Inference: 97%|█████████▋| 5739/5920 [59:32<01:56, 1.56it/s, tok/s=16649.9]
Inference: 97%|█████████▋| 5739/5920 [59:32<01:56, 1.56it/s, tok/s=16651.1]
Inference: 97%|█████████▋| 5740/5920 [59:35<03:34, 1.19s/it, tok/s=16651.1]
Inference: 97%|█████████▋| 5740/5920 [59:35<03:34, 1.19s/it, tok/s=16643.3]
Inference: 97%|█████████▋| 5741/5920 [59:36<03:49, 1.28s/it, tok/s=16643.3]
Inference: 97%|█████████▋| 5741/5920 [59:36<03:49, 1.28s/it, tok/s=16638.4]
Inference: 97%|█████████▋| 5742/5920 [59:37<03:02, 1.03s/it, tok/s=16638.4]
Inference: 97%|█████████▋| 5742/5920 [59:37<03:02, 1.03s/it, tok/s=16640.8]
Inference: 97%|█████████▋| 5743/5920 [59:37<02:30, 1.18it/s, tok/s=16640.8]
Inference: 97%|█████████▋| 5743/5920 [59:37<02:30, 1.18it/s, tok/s=16641.3]
Inference: 97%|█████████▋| 5744/5920 [59:38<02:51, 1.02it/s, tok/s=16641.3]
Inference: 97%|█████████▋| 5744/5920 [59:38<02:51, 1.02it/s, tok/s=16640.0]
Inference: 97%|█████████▋| 5745/5920 [59:38<02:50, 1.02it/s, tok/s=16645.5]
Inference: 97%|█████████▋| 5746/5920 [59:39<02:09, 1.35it/s, tok/s=16645.5]
Inference: 97%|█████████▋| 5746/5920 [59:39<02:09, 1.35it/s, tok/s=16645.9]
Inference: 97%|█████████▋| 5747/5920 [59:42<03:49, 1.33s/it, tok/s=16645.9]
Inference: 97%|█████████▋| 5747/5920 [59:42<03:49, 1.33s/it, tok/s=16634.4]
Inference: 97%|█████████▋| 5748/5920 [59:43<03:23, 1.18s/it, tok/s=16634.4]
Inference: 97%|█████████▋| 5748/5920 [59:43<03:23, 1.18s/it, tok/s=16637.4]
Inference: 97%|█████████▋| 5749/5920 [59:44<02:58, 1.05s/it, tok/s=16637.4]
Inference: 97%|█████████▋| 5749/5920 [59:44<02:58, 1.05s/it, tok/s=16638.3]
Inference: 97%|█████████▋| 5750/5920 [59:47<04:41, 1.65s/it, tok/s=16638.3]
Inference: 97%|█████████▋| 5750/5920 [59:47<04:41, 1.65s/it, tok/s=16626.3]
Inference: 97%|█████████▋| 5751/5920 [59:47<03:29, 1.24s/it, tok/s=16626.3]
Inference: 97%|█████████▋| 5751/5920 [59:47<03:29, 1.24s/it, tok/s=16628.5]
Inference: 97%|█████████▋| 5752/5920 [59:47<02:35, 1.08it/s, tok/s=16628.5]
Inference: 97%|█████████▋| 5752/5920 [59:47<02:35, 1.08it/s, tok/s=16632.8]
Inference: 97%|█████████▋| 5753/5920 [59:49<03:24, 1.22s/it, tok/s=16632.8]
Inference: 97%|█████████▋| 5753/5920 [59:49<03:24, 1.22s/it, tok/s=16627.1]
Inference: 97%|█████████▋| 5754/5920 [59:55<06:44, 2.44s/it, tok/s=16627.1]
Inference: 97%|█████████▋| 5754/5920 [59:55<06:44, 2.44s/it, tok/s=16605.9]
Inference: 97%|█████████▋| 5755/5920 [59:58<07:03, 2.57s/it, tok/s=16605.9]
Inference: 97%|█████████▋| 5755/5920 [59:58<07:03, 2.57s/it, tok/s=16599.2]
Inference: 97%|█████████▋| 5756/5920 [59:58<05:16, 1.93s/it, tok/s=16599.2]
Inference: 97%|█████████▋| 5756/5920 [59:58<05:16, 1.93s/it, tok/s=16599.1]
Inference: 97%|█████████▋| 5757/5920 [59:59<04:19, 1.59s/it, tok/s=16599.1]
Inference: 97%|█████████▋| 5757/5920 [59:59<04:19, 1.59s/it, tok/s=16597.5]
Inference: 97%|█████████▋| 5758/5920 [59:59<03:28, 1.29s/it, tok/s=16597.5]
Inference: 97%|█████████▋| 5758/5920 [59:59<03:28, 1.29s/it, tok/s=16601.0]
Inference: 97%|█████████▋| 5759/5920 [1:00:00<02:36, 1.03it/s, tok/s=16601.0]
Inference: 97%|█████████▋| 5759/5920 [1:00:00<02:36, 1.03it/s, tok/s=16602.8]
Inference: 97%|█████████▋| 5760/5920 [1:00:00<02:27, 1.09it/s, tok/s=16602.8]
Inference: 97%|█████████▋| 5760/5920 [1:00:00<02:27, 1.09it/s, tok/s=16605.1]
Inference: 97%|█████████▋| 5761/5920 [1:00:01<02:21, 1.12it/s, tok/s=16605.1]
Inference: 97%|█████████▋| 5761/5920 [1:00:01<02:21, 1.12it/s, tok/s=16607.8]
Inference: 97%|█████████▋| 5762/5920 [1:00:02<02:34, 1.02it/s, tok/s=16607.8]
Inference: 97%|█████████▋| 5762/5920 [1:00:02<02:34, 1.02it/s, tok/s=16606.6]
Inference: 97%|█████████▋| 5763/5920 [1:00:05<03:27, 1.32s/it, tok/s=16606.6]
Inference: 97%|█████████▋| 5763/5920 [1:00:05<03:27, 1.32s/it, tok/s=16601.5]
Inference: 97%|█████████▋| 5764/5920 [1:00:05<02:29, 1.05it/s, tok/s=16601.5]
Inference: 97%|█████████▋| 5764/5920 [1:00:05<02:29, 1.05it/s, tok/s=16605.2]
Inference: 97%|█████████▋| 5765/5920 [1:00:05<02:28, 1.05it/s, tok/s=16610.8]
Inference: 97%|█████████▋| 5766/5920 [1:00:05<01:47, 1.43it/s, tok/s=16610.8]
Inference: 97%|█████████▋| 5766/5920 [1:00:05<01:47, 1.43it/s, tok/s=16611.2]
Inference: 97%|█████████▋| 5767/5920 [1:00:06<01:25, 1.79it/s, tok/s=16611.2]
Inference: 97%|█████████▋| 5767/5920 [1:00:06<01:25, 1.79it/s, tok/s=16615.6]
Inference: 97%|█████████▋| 5768/5920 [1:00:07<02:05, 1.21it/s, tok/s=16615.6]
Inference: 97%|█████████▋| 5768/5920 [1:00:07<02:05, 1.21it/s, tok/s=16611.9]
Inference: 97%|█████████▋| 5769/5920 [1:00:07<01:36, 1.56it/s, tok/s=16611.9]
Inference: 97%|█████████▋| 5769/5920 [1:00:07<01:36, 1.56it/s, tok/s=16614.5]
Inference: 97%|█████████▋| 5770/5920 [1:00:08<01:28, 1.69it/s, tok/s=16614.5]
Inference: 97%|█████████▋| 5770/5920 [1:00:08<01:28, 1.69it/s, tok/s=16614.8]
Inference: 97%|█████████▋| 5771/5920 [1:00:09<02:11, 1.14it/s, tok/s=16614.8]
Inference: 97%|█████████▋| 5771/5920 [1:00:09<02:11, 1.14it/s, tok/s=16611.3]
Inference: 98%|█████████▊| 5772/5920 [1:00:10<01:42, 1.44it/s, tok/s=16611.3]
Inference: 98%|█████████▊| 5772/5920 [1:00:10<01:42, 1.44it/s, tok/s=16614.5]
Inference: 98%|█████████▊| 5773/5920 [1:00:10<01:24, 1.73it/s, tok/s=16614.5]
Inference: 98%|█████████▊| 5773/5920 [1:00:10<01:24, 1.73it/s, tok/s=16620.5]
Inference: 98%|█████████▊| 5774/5920 [1:00:10<01:24, 1.73it/s, tok/s=16624.5]
Inference: 98%|█████████▊| 5775/5920 [1:00:10<00:58, 2.47it/s, tok/s=16624.5]
Inference: 98%|█████████▊| 5775/5920 [1:00:10<00:58, 2.47it/s, tok/s=16626.0]
Inference: 98%|█████████▊| 5776/5920 [1:00:11<01:07, 2.13it/s, tok/s=16626.0]
Inference: 98%|█████████▊| 5776/5920 [1:00:11<01:07, 2.13it/s, tok/s=16627.3]
Inference: 98%|█████████▊| 5777/5920 [1:00:12<01:12, 1.98it/s, tok/s=16627.3]
Inference: 98%|█████████▊| 5777/5920 [1:00:12<01:12, 1.98it/s, tok/s=16626.8]
Inference: 98%|█████████▊| 5778/5920 [1:00:12<01:15, 1.87it/s, tok/s=16626.8]
Inference: 98%|█████████▊| 5778/5920 [1:00:12<01:15, 1.87it/s, tok/s=16628.1]
Inference: 98%|█████████▊| 5779/5920 [1:00:12<01:15, 1.87it/s, tok/s=16632.4]
Inference: 98%|█████████▊| 5780/5920 [1:00:12<01:14, 1.87it/s, tok/s=16638.8]
Inference: 98%|█████████▊| 5781/5920 [1:00:13<00:45, 3.07it/s, tok/s=16638.8]
Inference: 98%|█████████▊| 5781/5920 [1:00:13<00:45, 3.07it/s, tok/s=16644.3]
Inference: 98%|█████████▊| 5782/5920 [1:00:13<00:46, 2.94it/s, tok/s=16644.3]
Inference: 98%|█████████▊| 5782/5920 [1:00:13<00:46, 2.94it/s, tok/s=16645.6]
Inference: 98%|█████████▊| 5783/5920 [1:00:13<00:40, 3.35it/s, tok/s=16645.6]
Inference: 98%|█████████▊| 5783/5920 [1:00:13<00:40, 3.35it/s, tok/s=16648.3]
Inference: 98%|█████████▊| 5784/5920 [1:00:14<01:00, 2.25it/s, tok/s=16648.3]
Inference: 98%|█████████▊| 5784/5920 [1:00:14<01:00, 2.25it/s, tok/s=16648.7]
Inference: 98%|█████████▊| 5785/5920 [1:00:14<00:48, 2.78it/s, tok/s=16648.7]
Inference: 98%|█████████▊| 5785/5920 [1:00:14<00:48, 2.78it/s, tok/s=16650.9]
Inference: 98%|█████████▊| 5786/5920 [1:00:15<00:48, 2.75it/s, tok/s=16650.9]
Inference: 98%|█████████▊| 5786/5920 [1:00:15<00:48, 2.75it/s, tok/s=16652.1]
Inference: 98%|█████████▊| 5787/5920 [1:00:15<00:43, 3.08it/s, tok/s=16652.1]
Inference: 98%|█████████▊| 5787/5920 [1:00:15<00:43, 3.08it/s, tok/s=16659.0]
Inference: 98%|█████████▊| 5788/5920 [1:00:17<02:01, 1.09it/s, tok/s=16659.0]
Inference: 98%|█████████▊| 5788/5920 [1:00:17<02:01, 1.09it/s, tok/s=16650.3]
Inference: 98%|█████████▊| 5789/5920 [1:00:18<02:04, 1.05it/s, tok/s=16650.3]
Inference: 98%|█████████▊| 5789/5920 [1:00:18<02:04, 1.05it/s, tok/s=16653.9]
Inference: 98%|█████████▊| 5790/5920 [1:00:18<01:31, 1.42it/s, tok/s=16653.9]
Inference: 98%|█████████▊| 5790/5920 [1:00:18<01:31, 1.42it/s, tok/s=16660.9]
Inference: 98%|█████████▊| 5791/5920 [1:00:19<01:36, 1.33it/s, tok/s=16660.9]
Inference: 98%|█████████▊| 5791/5920 [1:00:19<01:36, 1.33it/s, tok/s=16661.0]
Inference: 98%|█████████▊| 5792/5920 [1:00:20<01:25, 1.51it/s, tok/s=16661.0]
Inference: 98%|█████████▊| 5792/5920 [1:00:20<01:25, 1.51it/s, tok/s=16663.6]
Inference: 98%|█████████▊| 5793/5920 [1:00:21<01:46, 1.20it/s, tok/s=16663.6]
Inference: 98%|█████████▊| 5793/5920 [1:00:21<01:46, 1.20it/s, tok/s=16660.9]
Inference: 98%|█████████▊| 5794/5920 [1:00:23<02:40, 1.27s/it, tok/s=16660.9]
Inference: 98%|█████████▊| 5794/5920 [1:00:23<02:40, 1.27s/it, tok/s=16653.4]
Inference: 98%|█████████▊| 5795/5920 [1:00:24<02:30, 1.20s/it, tok/s=16653.4]
Inference: 98%|█████████▊| 5795/5920 [1:00:24<02:30, 1.20s/it, tok/s=16653.3]
Inference: 98%|█████████▊| 5796/5920 [1:00:27<03:12, 1.55s/it, tok/s=16653.3]
Inference: 98%|█████████▊| 5796/5920 [1:00:27<03:12, 1.55s/it, tok/s=16650.5]
Inference: 98%|█████████▊| 5797/5920 [1:00:27<03:10, 1.55s/it, tok/s=16655.1]
Inference: 98%|█████████▊| 5798/5920 [1:00:29<02:35, 1.28s/it, tok/s=16655.1]
Inference: 98%|█████████▊| 5798/5920 [1:00:29<02:35, 1.28s/it, tok/s=16648.4]
Inference: 98%|█████████▊| 5799/5920 [1:00:29<02:04, 1.03s/it, tok/s=16648.4]
Inference: 98%|█████████▊| 5799/5920 [1:00:29<02:04, 1.03s/it, tok/s=16652.1]
Inference: 98%|█████████▊| 5800/5920 [1:00:29<01:36, 1.24it/s, tok/s=16652.1]
Inference: 98%|█████████▊| 5800/5920 [1:00:29<01:36, 1.24it/s, tok/s=16655.9]
Inference: 98%|█████████▊| 5801/5920 [1:00:30<01:42, 1.16it/s, tok/s=16655.9]
Inference: 98%|█████████▊| 5801/5920 [1:00:30<01:42, 1.16it/s, tok/s=16658.4]
Inference: 98%|█████████▊| 5802/5920 [1:00:30<01:21, 1.45it/s, tok/s=16658.4]
Inference: 98%|█████████▊| 5802/5920 [1:00:30<01:21, 1.45it/s, tok/s=16665.6]
Inference: 98%|█████████▊| 5803/5920 [1:00:31<01:19, 1.48it/s, tok/s=16665.6]
Inference: 98%|█████████▊| 5803/5920 [1:00:31<01:19, 1.48it/s, tok/s=16667.8]
Inference: 98%|█████████▊| 5804/5920 [1:00:31<01:01, 1.89it/s, tok/s=16667.8]
Inference: 98%|█████████▊| 5804/5920 [1:00:31<01:01, 1.89it/s, tok/s=16670.0]
Inference: 98%|█████████▊| 5805/5920 [1:00:33<01:58, 1.03s/it, tok/s=16670.0]
Inference: 98%|█████████▊| 5805/5920 [1:00:33<01:58, 1.03s/it, tok/s=16665.1]
Inference: 98%|█████████▊| 5806/5920 [1:00:34<01:40, 1.14it/s, tok/s=16665.1]
Inference: 98%|█████████▊| 5806/5920 [1:00:34<01:40, 1.14it/s, tok/s=16667.6]
Inference: 98%|█████████▊| 5807/5920 [1:00:37<02:41, 1.43s/it, tok/s=16667.6]
Inference: 98%|█████████▊| 5807/5920 [1:00:37<02:41, 1.43s/it, tok/s=16660.7]
Inference: 98%|█████████▊| 5808/5920 [1:00:39<03:19, 1.78s/it, tok/s=16660.7]
Inference: 98%|█████████▊| 5808/5920 [1:00:39<03:19, 1.78s/it, tok/s=16653.4]
Inference: 98%|█████████▊| 5809/5920 [1:00:40<02:44, 1.48s/it, tok/s=16653.4]
Inference: 98%|█████████▊| 5809/5920 [1:00:40<02:44, 1.48s/it, tok/s=16652.7]
Inference: 98%|█████████▊| 5810/5920 [1:00:41<02:19, 1.27s/it, tok/s=16652.7]
Inference: 98%|█████████▊| 5810/5920 [1:00:41<02:19, 1.27s/it, tok/s=16655.1]
Inference: 98%|█████████▊| 5811/5920 [1:00:42<02:15, 1.25s/it, tok/s=16655.1]
Inference: 98%|█████████▊| 5811/5920 [1:00:42<02:15, 1.25s/it, tok/s=16654.9]
Inference: 98%|█████████▊| 5812/5920 [1:00:46<03:32, 1.97s/it, tok/s=16654.9]
Inference: 98%|█████████▊| 5812/5920 [1:00:46<03:32, 1.97s/it, tok/s=16640.9]
Inference: 98%|█████████▊| 5813/5920 [1:00:47<03:14, 1.82s/it, tok/s=16640.9]
Inference: 98%|█████████▊| 5813/5920 [1:00:47<03:14, 1.82s/it, tok/s=16639.3]
Inference: 98%|█████████▊| 5814/5920 [1:00:48<02:39, 1.51s/it, tok/s=16639.3]
Inference: 98%|█████████▊| 5814/5920 [1:00:48<02:39, 1.51s/it, tok/s=16638.5]
Inference: 98%|█████████▊| 5815/5920 [1:00:50<02:48, 1.60s/it, tok/s=16638.5]
Inference: 98%|█████████▊| 5815/5920 [1:00:50<02:48, 1.60s/it, tok/s=16636.8]
Inference: 98%|█████████▊| 5816/5920 [1:00:50<02:12, 1.27s/it, tok/s=16636.8]
Inference: 98%|█████████▊| 5816/5920 [1:00:50<02:12, 1.27s/it, tok/s=16640.5]
Inference: 98%|█████████▊| 5817/5920 [1:00:54<03:18, 1.93s/it, tok/s=16640.5]
Inference: 98%|█████████▊| 5817/5920 [1:00:54<03:18, 1.93s/it, tok/s=16629.4]
Inference: 98%|█████████▊| 5818/5920 [1:00:54<02:34, 1.51s/it, tok/s=16629.4]
Inference: 98%|█████████▊| 5818/5920 [1:00:54<02:34, 1.51s/it, tok/s=16629.6]
Inference: 98%|█████████▊| 5819/5920 [1:00:56<02:35, 1.54s/it, tok/s=16629.6]
Inference: 98%|█████████▊| 5819/5920 [1:00:56<02:35, 1.54s/it, tok/s=16626.4]
Inference: 98%|█████████▊| 5820/5920 [1:00:59<03:36, 2.16s/it, tok/s=16626.4]
Inference: 98%|█████████▊| 5820/5920 [1:00:59<03:36, 2.16s/it, tok/s=16614.2]
Inference: 98%|█████████▊| 5821/5920 [1:01:03<04:10, 2.53s/it, tok/s=16614.2]
Inference: 98%|█████████▊| 5821/5920 [1:01:03<04:10, 2.53s/it, tok/s=16604.0]
Inference: 98%|█████████▊| 5822/5920 [1:01:05<03:46, 2.31s/it, tok/s=16604.0]
Inference: 98%|█████████▊| 5822/5920 [1:01:05<03:46, 2.31s/it, tok/s=16598.3]
Inference: 98%|█████████▊| 5823/5920 [1:01:06<03:19, 2.06s/it, tok/s=16598.3]
Inference: 98%|█████████▊| 5823/5920 [1:01:06<03:19, 2.06s/it, tok/s=16595.7]
Inference: 98%|█████████▊| 5824/5920 [1:01:06<02:23, 1.50s/it, tok/s=16595.7]
Inference: 98%|█████████▊| 5824/5920 [1:01:06<02:23, 1.50s/it, tok/s=16600.2]
Inference: 98%|█████████▊| 5825/5920 [1:01:07<02:02, 1.29s/it, tok/s=16600.2]
Inference: 98%|█████████▊| 5825/5920 [1:01:07<02:02, 1.29s/it, tok/s=16598.9]
Inference: 98%|█████████▊| 5826/5920 [1:01:07<01:34, 1.00s/it, tok/s=16598.9]
Inference: 98%|█████████▊| 5826/5920 [1:01:07<01:34, 1.00s/it, tok/s=16601.0]
Inference: 98%|█████████▊| 5827/5920 [1:01:08<01:16, 1.22it/s, tok/s=16601.0]
Inference: 98%|█████████▊| 5827/5920 [1:01:08<01:16, 1.22it/s, tok/s=16602.6]
Inference: 98%|█████████▊| 5828/5920 [1:01:10<01:51, 1.21s/it, tok/s=16602.6]
Inference: 98%|█████████▊| 5828/5920 [1:01:10<01:51, 1.21s/it, tok/s=16600.6]
Inference: 98%|█████████▊| 5829/5920 [1:01:15<03:23, 2.24s/it, tok/s=16600.6]
Inference: 98%|█████████▊| 5829/5920 [1:01:15<03:23, 2.24s/it, tok/s=16582.8]
Inference: 98%|█████████▊| 5830/5920 [1:01:15<03:21, 2.24s/it, tok/s=16588.1]
Inference: 98%|█████████▊| 5831/5920 [1:01:16<02:14, 1.51s/it, tok/s=16588.1]
Inference: 98%|█████████▊| 5831/5920 [1:01:16<02:14, 1.51s/it, tok/s=16588.2]
Inference: 99%|█████████▊| 5832/5920 [1:01:18<02:31, 1.72s/it, tok/s=16588.2]
Inference: 99%|█████████▊| 5832/5920 [1:01:18<02:31, 1.72s/it, tok/s=16582.1]
Inference: 99%|█████████▊| 5833/5920 [1:01:20<02:40, 1.84s/it, tok/s=16582.1]
Inference: 99%|█████████▊| 5833/5920 [1:01:20<02:40, 1.84s/it, tok/s=16575.6]
Inference: 99%|█████████▊| 5834/5920 [1:01:23<02:46, 1.93s/it, tok/s=16575.6]
Inference: 99%|█████████▊| 5834/5920 [1:01:23<02:46, 1.93s/it, tok/s=16570.7]
Inference: 99%|█████████▊| 5835/5920 [1:01:23<02:12, 1.56s/it, tok/s=16570.7]
Inference: 99%|█████████▊| 5835/5920 [1:01:23<02:12, 1.56s/it, tok/s=16574.1]
Inference: 99%|█████████▊| 5836/5920 [1:01:24<01:57, 1.39s/it, tok/s=16574.1]
Inference: 99%|█████████▊| 5836/5920 [1:01:24<01:57, 1.39s/it, tok/s=16573.2]
Inference: 99%|█████████▊| 5837/5920 [1:01:25<01:37, 1.17s/it, tok/s=16573.2]
Inference: 99%|█████████▊| 5837/5920 [1:01:25<01:37, 1.17s/it, tok/s=16575.7]
Inference: 99%|█████████▊| 5838/5920 [1:01:25<01:19, 1.03it/s, tok/s=16575.7]
Inference: 99%|█████████▊| 5838/5920 [1:01:25<01:19, 1.03it/s, tok/s=16577.2]
Inference: 99%|█████████▊| 5839/5920 [1:01:27<01:31, 1.12s/it, tok/s=16577.2]
Inference: 99%|█████████▊| 5839/5920 [1:01:27<01:31, 1.12s/it, tok/s=16575.3]
Inference: 99%|█████████▊| 5840/5920 [1:01:27<01:09, 1.14it/s, tok/s=16575.3]
Inference: 99%|█████████▊| 5840/5920 [1:01:27<01:09, 1.14it/s, tok/s=16578.0]
Inference: 99%|█████████▊| 5841/5920 [1:01:31<02:19, 1.76s/it, tok/s=16578.0]
Inference: 99%|█████████▊| 5841/5920 [1:01:31<02:19, 1.76s/it, tok/s=16567.5]
Inference: 99%|█████████▊| 5842/5920 [1:01:31<01:49, 1.40s/it, tok/s=16567.5]
Inference: 99%|█████████▊| 5842/5920 [1:01:31<01:49, 1.40s/it, tok/s=16568.3]
Inference: 99%|█████████▊| 5843/5920 [1:01:33<01:48, 1.41s/it, tok/s=16568.3]
Inference: 99%|█████████▊| 5843/5920 [1:01:33<01:48, 1.41s/it, tok/s=16567.4]
Inference: 99%|█████████▊| 5844/5920 [1:01:33<01:24, 1.11s/it, tok/s=16567.4]
Inference: 99%|█████████▊| 5844/5920 [1:01:33<01:24, 1.11s/it, tok/s=16569.7]
Inference: 99%|█████████▊| 5845/5920 [1:01:34<01:21, 1.08s/it, tok/s=16569.7]
Inference: 99%|█████████▊| 5845/5920 [1:01:34<01:21, 1.08s/it, tok/s=16573.3]
Inference: 99%|█████████▉| 5846/5920 [1:01:34<00:59, 1.24it/s, tok/s=16573.3]
Inference: 99%|█████████▉| 5846/5920 [1:01:34<00:59, 1.24it/s, tok/s=16578.6]
Inference: 99%|█████████▉| 5847/5920 [1:01:37<01:37, 1.33s/it, tok/s=16578.6]
Inference: 99%|█████████▉| 5847/5920 [1:01:37<01:37, 1.33s/it, tok/s=16575.2]
Inference: 99%|█████████▉| 5848/5920 [1:01:37<01:11, 1.00it/s, tok/s=16575.2]
Inference: 99%|█████████▉| 5848/5920 [1:01:37<01:11, 1.00it/s, tok/s=16579.0]
Inference: 99%|█████████▉| 5849/5920 [1:01:39<01:18, 1.11s/it, tok/s=16579.0]
Inference: 99%|█████████▉| 5849/5920 [1:01:39<01:18, 1.11s/it, tok/s=16578.4]
Inference: 99%|█████████▉| 5850/5920 [1:01:39<01:07, 1.04it/s, tok/s=16578.4]
Inference: 99%|█████████▉| 5850/5920 [1:01:39<01:07, 1.04it/s, tok/s=16581.4]
Inference: 99%|█████████▉| 5851/5920 [1:01:39<00:50, 1.37it/s, tok/s=16581.4]
Inference: 99%|█████████▉| 5851/5920 [1:01:39<00:50, 1.37it/s, tok/s=16584.4]
Inference: 99%|█████████▉| 5852/5920 [1:01:44<02:03, 1.82s/it, tok/s=16584.4]
Inference: 99%|█████████▉| 5852/5920 [1:01:44<02:03, 1.82s/it, tok/s=16569.5]
Inference: 99%|█████████▉| 5853/5920 [1:01:46<02:12, 1.98s/it, tok/s=16569.5]
Inference: 99%|█████████▉| 5853/5920 [1:01:46<02:12, 1.98s/it, tok/s=16563.4]
Inference: 99%|█████████▉| 5854/5920 [1:01:48<02:09, 1.97s/it, tok/s=16563.4]
Inference: 99%|█████████▉| 5854/5920 [1:01:48<02:09, 1.97s/it, tok/s=16562.0]
Inference: 99%|█████████▉| 5855/5920 [1:01:48<01:33, 1.44s/it, tok/s=16562.0]
Inference: 99%|█████████▉| 5855/5920 [1:01:48<01:33, 1.44s/it, tok/s=16564.6]
Inference: 99%|█████████▉| 5856/5920 [1:01:53<02:34, 2.41s/it, tok/s=16564.6]
Inference: 99%|█████████▉| 5856/5920 [1:01:53<02:34, 2.41s/it, tok/s=16547.8]
Inference: 99%|█████████▉| 5857/5920 [1:01:53<01:49, 1.73s/it, tok/s=16547.8]
Inference: 99%|█████████▉| 5857/5920 [1:01:53<01:49, 1.73s/it, tok/s=16550.7]
Inference: 99%|█████████▉| 5858/5920 [1:01:53<01:19, 1.29s/it, tok/s=16550.7]
Inference: 99%|█████████▉| 5858/5920 [1:01:53<01:19, 1.29s/it, tok/s=16554.3]
Inference: 99%|█████████▉| 5859/5920 [1:01:57<01:56, 1.91s/it, tok/s=16554.3]
Inference: 99%|█████████▉| 5859/5920 [1:01:57<01:56, 1.91s/it, tok/s=16543.5]
Inference: 99%|█████████▉| 5860/5920 [1:02:07<04:20, 4.34s/it, tok/s=16543.5]
Inference: 99%|█████████▉| 5860/5920 [1:02:07<04:20, 4.34s/it, tok/s=16502.2]
Inference: 99%|█████████▉| 5861/5920 [1:02:10<03:56, 4.00s/it, tok/s=16502.2]
Inference: 99%|█████████▉| 5861/5920 [1:02:10<03:56, 4.00s/it, tok/s=16493.9]
Inference: 99%|█████████▉| 5862/5920 [1:02:11<02:54, 3.00s/it, tok/s=16493.9]
Inference: 99%|█████████▉| 5862/5920 [1:02:11<02:54, 3.00s/it, tok/s=16494.9]
Inference: 99%|█████████▉| 5863/5920 [1:02:11<02:11, 2.31s/it, tok/s=16494.9]
Inference: 99%|█████████▉| 5863/5920 [1:02:11<02:11, 2.31s/it, tok/s=16495.9]
Inference: 99%|█████████▉| 5864/5920 [1:02:13<01:53, 2.03s/it, tok/s=16495.9]
Inference: 99%|█████████▉| 5864/5920 [1:02:13<01:53, 2.03s/it, tok/s=16497.0]
Inference: 99%|█████████▉| 5865/5920 [1:02:13<01:51, 2.03s/it, tok/s=16499.7]
Inference: 99%|█████████▉| 5866/5920 [1:02:16<01:43, 1.91s/it, tok/s=16499.7]
Inference: 99%|█████████▉| 5866/5920 [1:02:16<01:43, 1.91s/it, tok/s=16489.1]
Inference: 99%|█████████▉| 5867/5920 [1:02:18<01:41, 1.92s/it, tok/s=16489.1]
Inference: 99%|█████████▉| 5867/5920 [1:02:18<01:41, 1.92s/it, tok/s=16487.6]
Inference: 99%|█████████▉| 5868/5920 [1:02:20<01:32, 1.78s/it, tok/s=16487.6]
Inference: 99%|█████████▉| 5868/5920 [1:02:20<01:32, 1.78s/it, tok/s=16484.9]
Inference: 99%|█████████▉| 5869/5920 [1:02:24<02:11, 2.58s/it, tok/s=16484.9]
Inference: 99%|█████████▉| 5869/5920 [1:02:24<02:11, 2.58s/it, tok/s=16468.2]
Inference: 99%|█████████▉| 5870/5920 [1:02:28<02:25, 2.91s/it, tok/s=16468.2]
Inference: 99%|█████████▉| 5870/5920 [1:02:28<02:25, 2.91s/it, tok/s=16456.7]
Inference: 99%|█████████▉| 5871/5920 [1:02:32<02:32, 3.11s/it, tok/s=16456.7]
Inference: 99%|█████████▉| 5871/5920 [1:02:32<02:32, 3.11s/it, tok/s=16447.2]
Inference: 99%|█████████▉| 5872/5920 [1:02:35<02:34, 3.23s/it, tok/s=16447.2]
Inference: 99%|█████████▉| 5872/5920 [1:02:35<02:34, 3.23s/it, tok/s=16439.9]
Inference: 99%|█████████▉| 5873/5920 [1:02:35<01:48, 2.32s/it, tok/s=16439.9]
Inference: 99%|█████████▉| 5873/5920 [1:02:35<01:48, 2.32s/it, tok/s=16445.4]
Inference: 99%|█████████▉| 5874/5920 [1:02:47<03:47, 4.94s/it, tok/s=16445.4]
Inference: 99%|█████████▉| 5874/5920 [1:02:47<03:47, 4.94s/it, tok/s=16402.6]
Inference: 99%|█████████▉| 5875/5920 [1:02:51<03:38, 4.87s/it, tok/s=16402.6]
Inference: 99%|█████████▉| 5875/5920 [1:02:51<03:38, 4.87s/it, tok/s=16389.3]
Inference: 99%|█████████▉| 5876/5920 [1:02:52<02:46, 3.78s/it, tok/s=16389.3]
Inference: 99%|█████████▉| 5876/5920 [1:02:52<02:46, 3.78s/it, tok/s=16388.1]
Inference: 99%|█████████▉| 5877/5920 [1:02:53<01:59, 2.77s/it, tok/s=16388.1]
Inference: 99%|█████████▉| 5877/5920 [1:02:53<01:59, 2.77s/it, tok/s=16391.8]
Inference: 99%|█████████▉| 5878/5920 [1:02:54<01:39, 2.36s/it, tok/s=16391.8]
Inference: 99%|█████████▉| 5878/5920 [1:02:54<01:39, 2.36s/it, tok/s=16391.8]
Inference: 99%|█████████▉| 5879/5920 [1:02:54<01:10, 1.71s/it, tok/s=16391.8]
Inference: 99%|█████████▉| 5879/5920 [1:02:54<01:10, 1.71s/it, tok/s=16396.3]
Inference: 99%|█████████▉| 5880/5920 [1:02:59<01:38, 2.46s/it, tok/s=16396.3]
Inference: 99%|█████████▉| 5880/5920 [1:02:59<01:38, 2.46s/it, tok/s=16384.1]
Inference: 99%|█████████▉| 5881/5920 [1:03:06<02:35, 3.98s/it, tok/s=16384.1]
Inference: 99%|█████████▉| 5881/5920 [1:03:06<02:35, 3.98s/it, tok/s=16356.4]
Inference: 99%|█████████▉| 5882/5920 [1:03:14<03:20, 5.29s/it, tok/s=16356.4]
Inference: 99%|█████████▉| 5882/5920 [1:03:14<03:20, 5.29s/it, tok/s=16326.2]
Inference: 99%|█████████▉| 5883/5920 [1:03:16<02:28, 4.01s/it, tok/s=16326.2]
Inference: 99%|█████████▉| 5883/5920 [1:03:16<02:28, 4.01s/it, tok/s=16326.7]
Inference: 99%|█████████▉| 5884/5920 [1:03:16<01:47, 2.98s/it, tok/s=16326.7]
Inference: 99%|█████████▉| 5884/5920 [1:03:16<01:47, 2.98s/it, tok/s=16328.9]
Inference: 99%|█████████▉| 5885/5920 [1:03:18<01:31, 2.62s/it, tok/s=16328.9]
Inference: 99%|█████████▉| 5885/5920 [1:03:18<01:31, 2.62s/it, tok/s=16326.8]
Inference: 99%|█████████▉| 5886/5920 [1:03:19<01:10, 2.09s/it, tok/s=16326.8]
Inference: 99%|█████████▉| 5886/5920 [1:03:19<01:10, 2.09s/it, tok/s=16328.1]
Inference: 99%|█████████▉| 5887/5920 [1:03:27<02:13, 4.05s/it, tok/s=16328.1]
Inference: 99%|█████████▉| 5887/5920 [1:03:27<02:13, 4.05s/it, tok/s=16296.5]
Inference: 99%|█████████▉| 5888/5920 [1:03:28<01:40, 3.15s/it, tok/s=16296.5]
Inference: 99%|█████████▉| 5888/5920 [1:03:28<01:40, 3.15s/it, tok/s=16296.4]
Inference: 99%|█████████▉| 5889/5920 [1:03:29<01:11, 2.31s/it, tok/s=16296.4]
Inference: 99%|█████████▉| 5889/5920 [1:03:29<01:11, 2.31s/it, tok/s=16301.8]
Inference: 99%|█████████▉| 5890/5920 [1:03:30<00:58, 1.95s/it, tok/s=16301.8]
Inference: 99%|█████████▉| 5890/5920 [1:03:30<00:58, 1.95s/it, tok/s=16301.5]
Inference: 100%|█████████▉| 5891/5920 [1:03:37<01:37, 3.36s/it, tok/s=16301.5]
Inference: 100%|█████████▉| 5891/5920 [1:03:37<01:37, 3.36s/it, tok/s=16277.2]
Inference: 100%|█████████▉| 5892/5920 [1:03:38<01:20, 2.86s/it, tok/s=16277.2]
Inference: 100%|█████████▉| 5892/5920 [1:03:38<01:20, 2.86s/it, tok/s=16277.1]
Inference: 100%|█████████▉| 5893/5920 [1:03:41<01:13, 2.73s/it, tok/s=16277.1]
Inference: 100%|█████████▉| 5893/5920 [1:03:41<01:13, 2.73s/it, tok/s=16273.9]
Inference: 100%|█████████▉| 5894/5920 [1:03:48<01:49, 4.19s/it, tok/s=16273.9]
Inference: 100%|█████████▉| 5894/5920 [1:03:48<01:49, 4.19s/it, tok/s=16247.3]
Inference: 100%|█████████▉| 5895/5920 [1:03:54<01:54, 4.57s/it, tok/s=16247.3]
Inference: 100%|█████████▉| 5895/5920 [1:03:54<01:54, 4.57s/it, tok/s=16229.9]
Inference: 100%|█████████▉| 5896/5920 [1:03:59<01:51, 4.65s/it, tok/s=16229.9]
Inference: 100%|█████████▉| 5896/5920 [1:03:59<01:51, 4.65s/it, tok/s=16216.6]
Inference: 100%|█████████▉| 5897/5920 [1:04:00<01:24, 3.67s/it, tok/s=16216.6]
Inference: 100%|█████████▉| 5897/5920 [1:04:00<01:24, 3.67s/it, tok/s=16214.9]
Inference: 100%|█████████▉| 5898/5920 [1:04:02<01:10, 3.22s/it, tok/s=16214.9]
Inference: 100%|█████████▉| 5898/5920 [1:04:02<01:10, 3.22s/it, tok/s=16213.6]
Inference: 100%|█████████▉| 5899/5920 [1:04:08<01:23, 3.97s/it, tok/s=16213.6]
Inference: 100%|█████████▉| 5899/5920 [1:04:08<01:23, 3.97s/it, tok/s=16194.4]
Inference: 100%|█████████▉| 5900/5920 [1:04:10<01:08, 3.42s/it, tok/s=16194.4]
Inference: 100%|█████████▉| 5900/5920 [1:04:10<01:08, 3.42s/it, tok/s=16190.1]
Inference: 100%|█████████▉| 5901/5920 [1:04:12<00:56, 2.95s/it, tok/s=16190.1]
Inference: 100%|█████████▉| 5901/5920 [1:04:12<00:56, 2.95s/it, tok/s=16190.2]
Inference: 100%|█████████▉| 5902/5920 [1:04:12<00:38, 2.13s/it, tok/s=16190.2]
Inference: 100%|█████████▉| 5902/5920 [1:04:12<00:38, 2.13s/it, tok/s=16195.8]
Inference: 100%|█████████▉| 5903/5920 [1:04:20<01:05, 3.86s/it, tok/s=16195.8]
Inference: 100%|█████████▉| 5903/5920 [1:04:20<01:05, 3.86s/it, tok/s=16168.5]
Inference: 100%|█████████▉| 5904/5920 [1:04:21<00:48, 3.05s/it, tok/s=16168.5]
Inference: 100%|█████████▉| 5904/5920 [1:04:21<00:48, 3.05s/it, tok/s=16168.2]
Inference: 100%|█████████▉| 5905/5920 [1:04:24<00:43, 2.90s/it, tok/s=16168.2]
Inference: 100%|█████████▉| 5905/5920 [1:04:24<00:43, 2.90s/it, tok/s=16164.5]
Inference: 100%|█████████▉| 5906/5920 [1:04:25<00:36, 2.58s/it, tok/s=16164.5]
Inference: 100%|█████████▉| 5906/5920 [1:04:25<00:36, 2.58s/it, tok/s=16162.0]
Inference: 100%|█████████▉| 5907/5920 [1:04:28<00:32, 2.49s/it, tok/s=16162.0]
Inference: 100%|█████████▉| 5907/5920 [1:04:28<00:32, 2.49s/it, tok/s=16158.6]
Inference: 100%|█████████▉| 5908/5920 [1:04:35<00:47, 3.95s/it, tok/s=16158.6]
Inference: 100%|█████████▉| 5908/5920 [1:04:35<00:47, 3.95s/it, tok/s=16134.0]
Inference: 100%|█████████▉| 5909/5920 [1:04:42<00:53, 4.90s/it, tok/s=16134.0]
Inference: 100%|█████████▉| 5909/5920 [1:04:42<00:53, 4.90s/it, tok/s=16111.6]
Inference: 100%|█████████▉| 5910/5920 [1:04:45<00:42, 4.26s/it, tok/s=16111.6]
Inference: 100%|█████████▉| 5910/5920 [1:04:45<00:42, 4.26s/it, tok/s=16107.8]
Inference: 100%|█████████▉| 5911/5920 [1:04:49<00:36, 4.04s/it, tok/s=16107.8]
Inference: 100%|█████████▉| 5911/5920 [1:04:49<00:36, 4.04s/it, tok/s=16099.6]
Inference: 100%|█████████▉| 5912/5920 [1:05:04<01:00, 7.56s/it, tok/s=16099.6]
Inference: 100%|█████████▉| 5912/5920 [1:05:04<01:00, 7.56s/it, tok/s=16039.7]
Inference: 100%|█████████▉| 5913/5920 [1:05:11<00:51, 7.34s/it, tok/s=16039.7]
Inference: 100%|█████████▉| 5913/5920 [1:05:11<00:51, 7.34s/it, tok/s=16017.3]
Inference: 100%|█████████▉| 5914/5920 [1:05:16<00:38, 6.47s/it, tok/s=16017.3]
Inference: 100%|█████████▉| 5914/5920 [1:05:16<00:38, 6.47s/it, tok/s=16006.9]
Inference: 100%|█████████▉| 5915/5920 [1:05:38<00:55, 11.18s/it, tok/s=16006.9]
Inference: 100%|█████████▉| 5915/5920 [1:05:38<00:55, 11.18s/it, tok/s=15923.6]
Inference: 100%|█████████▉| 5916/5920 [1:05:44<00:39, 9.82s/it, tok/s=15923.6]
Inference: 100%|█████████▉| 5916/5920 [1:05:44<00:39, 9.82s/it, tok/s=15904.5]
Inference: 100%|█████████▉| 5916/5920 [1:05:44<00:02, 1.50it/s, tok/s=15904.5]
{"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-muon-20260413_090005/iter_0001023-hf/eval_20260417051759/step01_prepared.jsonl"}, "step02": {"backend": "offline", "input_path": "/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-muon-20260413_090005/iter_0001023-hf/eval_20260417051759/step01_prepared.jsonl", "output_path": "/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-muon-20260413_090005/iter_0001023-hf/eval_20260417051759/step02_inference.jsonl"}, "step03": {"input_path": "/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-muon-20260413_090005/iter_0001023-hf/eval_20260417051759/step02_inference.jsonl", "score_output_path": "/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-muon-20260413_090005/iter_0001023-hf/eval_20260417051759/step03_score.jsonl", "final_eval_output_path": "/jpfs-5p/chenyanxu.9/model/Qwen3-8B-onpolicy-profiling-muon-20260413_090005/iter_0001023-hf/eval_20260417051759/step03_final_eval.jsonl", "metrics": {"gpqa_diamond": {"avg_k": 0.5774111675126904, "pass_k": 0.766497461928934, "avg_total_tokens": 10612.370558375635, "avg_thinking_tokens": 0.0, "max_thinking_tokens": 0.0, "min_thinking_tokens": 0.0}, "hmmt2025": {"avg_k": 0.31666666666666665, "pass_k": 0.43333333333333335, "avg_total_tokens": 18237.566666666666, "avg_thinking_tokens": 0.0, "max_thinking_tokens": 0.0, "min_thinking_tokens": 0.0}, "aime2024": {"avg_k": 0.6802083333333333, "pass_k": 0.9, "avg_total_tokens": 14426.277083333332, "avg_thinking_tokens": 0.0, "max_thinking_tokens": 0.0, "min_thinking_tokens": 0.0}, "aime2025": {"avg_k": 0.571875, "pass_k": 0.8666666666666667, "avg_total_tokens": 15452.192708333334, "avg_thinking_tokens": 0.0, "max_thinking_tokens": 0.0, "min_thinking_tokens": 0.0}, "math500": {"avg_k": 0.7485, "pass_k": 0.768, "avg_total_tokens": 4490.75, "avg_thinking_tokens": 0.0, "max_thinking_tokens": 0.0, "min_thinking_tokens": 0.0}, "minerva": {"avg_k": 0.3290441176470588, "pass_k": 0.38235294117647056, "avg_total_tokens": 6507.237132352941, "avg_thinking_tokens": 0.0, "max_thinking_tokens": 0.0, "min_thinking_tokens": 0.0}, "overall": {"avg_k": 0.6000676132521975, "pass_k": 0.6657223796033994, "avg_total_tokens": 9346.815584854632, "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-3hp1opqx'
Reference in New Issue View Git Blame Copy Permalink