commit 78bc4e463a63983e5a871c9ca67ee80be0de9620 Author: ModelHub XC Date: Mon Aug 24 12:09:14 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: laion/Qwen3-8B_exp_tas_trajectory_minimal_traces_save-strategy_steps Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..f835ae0 --- /dev/null +++ b/.gitattributes @@ -0,0 +1,56 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text + + +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zstandard filter=lfs diff=lfs merge=lfs -text +*.tfevents* filter=lfs diff=lfs merge=lfs -text +*.db* filter=lfs diff=lfs merge=lfs -text +*.ark* filter=lfs diff=lfs merge=lfs -text +**/*ckpt*data* filter=lfs diff=lfs merge=lfs -text +**/*ckpt*.meta filter=lfs diff=lfs merge=lfs -text +**/*ckpt*.index filter=lfs diff=lfs merge=lfs -text + +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.gguf* filter=lfs diff=lfs merge=lfs -text +*.ggml filter=lfs diff=lfs merge=lfs -text +*.llamafile* filter=lfs diff=lfs merge=lfs -text +*.pt2 filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text + +training_args.bin filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text +model-00001-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +model-00004-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +vocab.json filter=lfs diff=lfs merge=lfs -text +model-00002-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +model-00003-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +merges.txt filter=lfs diff=lfs merge=lfs -text \ No newline at end of file diff --git a/README.md b/README.md new file mode 100644 index 0000000..02a06fc --- /dev/null +++ b/README.md @@ -0,0 +1,60 @@ +--- +library_name: transformers +license: apache-2.0 +base_model: Qwen/Qwen3-8B +tags: +- llama-factory +- full +- generated_from_trainer +model-index: +- name: Qwen3-8B_exp_tas_trajectory_minimal_traces_save-strategy_steps + results: [] +--- + + + +# Qwen3-8B_exp_tas_trajectory_minimal_traces_save-strategy_steps + +This model is a fine-tuned version of [Qwen/Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B) on the DCAgent/exp_tas_trajectory_minimal_traces dataset. + +## Model description + +More information needed + +## Intended uses & limitations + +More information needed + +## Training and evaluation data + +More information needed + +## Training procedure + +### Training hyperparameters + +The following hyperparameters were used during training: +- learning_rate: 0.0001 +- train_batch_size: 1 +- eval_batch_size: 8 +- seed: 42 +- distributed_type: multi-GPU +- num_devices: 32 +- total_train_batch_size: 32 +- total_eval_batch_size: 256 +- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.87,0.99) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments +- lr_scheduler_type: cosine +- lr_scheduler_warmup_ratio: 0.005 +- num_epochs: 8.0 + +### Training results + + + +### Framework versions + +- Transformers 4.55.0 +- Pytorch 2.7.1+cu128 +- Datasets 3.6.0 +- Tokenizers 0.21.1 diff --git a/added_tokens.json b/added_tokens.json new file mode 100644 index 0000000..b54f913 --- /dev/null +++ b/added_tokens.json @@ -0,0 +1,28 @@ +{ + "": 151668, + "": 151658, + "": 151666, + "": 151667, + "": 151657, + "": 151665, + "<|box_end|>": 151649, + "<|box_start|>": 151648, + "<|endoftext|>": 151643, + "<|file_sep|>": 151664, + "<|fim_middle|>": 151660, + "<|fim_pad|>": 151662, + "<|fim_prefix|>": 151659, + "<|fim_suffix|>": 151661, + "<|im_end|>": 151645, + "<|im_start|>": 151644, + "<|image_pad|>": 151655, + "<|object_ref_end|>": 151647, + "<|object_ref_start|>": 151646, + "<|quad_end|>": 151651, + "<|quad_start|>": 151650, + "<|repo_name|>": 151663, + "<|video_pad|>": 151656, + "<|vision_end|>": 151653, + "<|vision_pad|>": 151654, + "<|vision_start|>": 151652 +} diff --git a/all_results.json b/all_results.json new file mode 100644 index 0000000..ad111b9 --- /dev/null +++ b/all_results.json @@ -0,0 +1,16 @@ +{ + "achieved_tflops_per_gpu": 0.0010303227944526512, + "achieved_tflops_per_gpu_theoretical": 243.92912852326808, + "epoch": 8.0, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.003110480960458517, + "mfu_percent": 0.0003302316648886703, + "mfu_percent_theoretical": 78.18241298822694, + "total_flos": 1155060078084096.0, + "train_loss": 0.14387717474665906, + "train_runtime": 35033.3193, + "train_samples_per_second": 2.295, + "train_steps_per_second": 0.072, + "valid_targets_mean": 2280.5, + "valid_targets_min": 275 +} \ No newline at end of file diff --git a/chat_template.jinja b/chat_template.jinja new file mode 100644 index 0000000..01be9b3 --- /dev/null +++ b/chat_template.jinja @@ -0,0 +1,89 @@ +{%- if tools %} + {{- '<|im_start|>system\n' }} + {%- if messages[0].role == 'system' %} + {{- messages[0].content + '\n\n' }} + {%- endif %} + {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within XML tags:\n" }} + {%- for tool in tools %} + {{- "\n" }} + {{- tool | tojson }} + {%- endfor %} + {{- "\n\n\nFor each function call, return a json object with function name and arguments within XML tags:\n\n{\"name\": , \"arguments\": }\n<|im_end|>\n" }} +{%- else %} + {%- if messages[0].role == 'system' %} + {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }} + {%- endif %} +{%- endif %} +{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %} +{%- for message in messages[::-1] %} + {%- set index = (messages|length - 1) - loop.index0 %} + {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('') and message.content.endswith('')) %} + {%- set ns.multi_step_tool = false %} + {%- set ns.last_query_index = index %} + {%- endif %} +{%- endfor %} +{%- for message in messages %} + {%- if message.content is string %} + {%- set content = message.content %} + {%- else %} + {%- set content = '' %} + {%- endif %} + {%- if (message.role == "user") or (message.role == "system" and not loop.first) %} + {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }} + {%- elif message.role == "assistant" %} + {%- set reasoning_content = '' %} + {%- if message.reasoning_content is string %} + {%- set reasoning_content = message.reasoning_content %} + {%- else %} + {%- if '' in content %} + {%- set reasoning_content = content.split('')[0].rstrip('\n').split('')[-1].lstrip('\n') %} + {%- set content = content.split('')[-1].lstrip('\n') %} + {%- endif %} + {%- endif %} + {%- if loop.index0 > ns.last_query_index %} + {%- if loop.last or (not loop.last and reasoning_content) %} + {{- '<|im_start|>' + message.role + '\n\n' + reasoning_content.strip('\n') + '\n\n\n' + content.lstrip('\n') }} + {%- else %} + {{- '<|im_start|>' + message.role + '\n' + content }} + {%- endif %} + {%- else %} + {{- '<|im_start|>' + message.role + '\n' + content }} + {%- endif %} + {%- if message.tool_calls %} + {%- for tool_call in message.tool_calls %} + {%- if (loop.first and content) or (not loop.first) %} + {{- '\n' }} + {%- endif %} + {%- if tool_call.function %} + {%- set tool_call = tool_call.function %} + {%- endif %} + {{- '\n{"name": "' }} + {{- tool_call.name }} + {{- '", "arguments": ' }} + {%- if tool_call.arguments is string %} + {{- tool_call.arguments }} + {%- else %} + {{- tool_call.arguments | tojson }} + {%- endif %} + {{- '}\n' }} + {%- endfor %} + {%- endif %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "tool" %} + {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %} + {{- '<|im_start|>user' }} + {%- endif %} + {{- '\n\n' }} + {{- content }} + {{- '\n' }} + {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %} + {{- '<|im_end|>\n' }} + {%- endif %} + {%- endif %} +{%- endfor %} +{%- if add_generation_prompt %} + {{- '<|im_start|>assistant\n' }} + {%- if enable_thinking is defined and enable_thinking is false %} + {{- '\n\n\n\n' }} + {%- endif %} +{%- endif %} \ No newline at end of file diff --git a/config.json b/config.json new file mode 100644 index 0000000..cac9b4a --- /dev/null +++ b/config.json @@ -0,0 +1,68 @@ +{ + "architectures": [ + "Qwen3ForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 151643, + "eos_token_id": 151645, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 12288, + "layer_types": [ + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention", + "full_attention" + ], + "max_position_embeddings": 40960, + "max_window_layers": 36, + "model_type": "qwen3", + "num_attention_heads": 32, + "num_hidden_layers": 36, + "num_key_value_heads": 8, + "rms_norm_eps": 1e-06, + "rope_scaling": null, + "rope_theta": 1000000, + "sliding_window": null, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.55.0", + "use_cache": false, + "use_sliding_window": false, + "vocab_size": 151936 +} diff --git a/configuration.json b/configuration.json new file mode 100644 index 0000000..159097f --- /dev/null +++ b/configuration.json @@ -0,0 +1 @@ +{"framework": "pytorch", "task": "others", "allow_remote": true} \ No newline at end of file diff --git a/generation_config.json b/generation_config.json new file mode 100644 index 0000000..4b23077 --- /dev/null +++ b/generation_config.json @@ -0,0 +1,13 @@ +{ + "bos_token_id": 151643, + "do_sample": true, + "eos_token_id": [ + 151645, + 151643 + ], + "pad_token_id": 151643, + "temperature": 0.6, + "top_k": 20, + "top_p": 0.95, + "transformers_version": "4.55.0" +} diff --git a/merges.txt b/merges.txt new file mode 100644 index 0000000..80c1a19 --- /dev/null +++ b/merges.txt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8831e4f1a044471340f7c0a83d7bd71306a5b867e95fd870f74d0c5308a904d5 +size 1671853 diff --git a/model-00001-of-00004.safetensors b/model-00001-of-00004.safetensors new file mode 100644 index 0000000..576be2d --- /dev/null +++ b/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a2db297eae8d5f81da570dedca74e05fd5266fa09d803fa8c1119de5edd4fa9f +size 4902257696 diff --git a/model-00002-of-00004.safetensors b/model-00002-of-00004.safetensors new file mode 100644 index 0000000..4e0fdb0 --- /dev/null +++ b/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3330e00ca95ad52dafc3575f6aa2fbb0d7d72c6504a67f6d0512fd7d42333620 +size 4915960368 diff --git a/model-00003-of-00004.safetensors b/model-00003-of-00004.safetensors new file mode 100644 index 0000000..7fdd18f --- /dev/null +++ b/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7c12e007a9654b1d69f64180f0859399576fb658be8fa6a3509101be71efb879 +size 4983068496 diff --git a/model-00004-of-00004.safetensors b/model-00004-of-00004.safetensors new file mode 100644 index 0000000..c86b599 --- /dev/null +++ b/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ed8ddab475e863e62de67cbf1428d4fbe12bb62fe6d304cc46cf70004d25b2a4 +size 1580230264 diff --git a/model.safetensors.index.json b/model.safetensors.index.json new file mode 100644 index 0000000..ba886c0 --- /dev/null +++ b/model.safetensors.index.json @@ -0,0 +1,407 @@ +{ + "metadata": { + "total_parameters": 308224, + "total_size": 16381470720 + }, + "weight_map": { + "lm_head.weight": "model-00004-of-00004.safetensors", + "model.embed_tokens.weight": "model-00001-of-00004.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.k_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.q_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.k_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.q_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.k_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.q_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.k_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.q_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.k_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.q_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.k_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.q_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.k_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.q_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.k_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.q_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.k_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.q_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.k_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.q_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.k_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.q_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.k_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.q_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.k_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.q_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.k_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.q_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.21.self_attn.k_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.self_attn.q_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.22.self_attn.k_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.22.self_attn.q_norm.weight": "model-00002-of-00004.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.k_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.q_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.k_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.q_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.k_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.q_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.k_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.q_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.k_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.q_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.k_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.q_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.k_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.q_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.k_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.q_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.k_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.q_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.k_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.q_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.32.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.32.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.32.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.32.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.32.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.32.self_attn.k_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.32.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.32.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.32.self_attn.q_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.32.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.32.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.33.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.33.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.33.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.33.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.33.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.33.self_attn.k_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.33.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.33.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.33.self_attn.q_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.33.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.33.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.34.input_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.34.mlp.down_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.34.mlp.gate_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.34.mlp.up_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.34.post_attention_layernorm.weight": "model-00003-of-00004.safetensors", + "model.layers.34.self_attn.k_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.34.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.34.self_attn.o_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.34.self_attn.q_norm.weight": "model-00003-of-00004.safetensors", + "model.layers.34.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.34.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.35.input_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.35.mlp.down_proj.weight": "model-00004-of-00004.safetensors", + "model.layers.35.mlp.gate_proj.weight": "model-00004-of-00004.safetensors", + "model.layers.35.mlp.up_proj.weight": "model-00004-of-00004.safetensors", + "model.layers.35.post_attention_layernorm.weight": "model-00004-of-00004.safetensors", + "model.layers.35.self_attn.k_norm.weight": "model-00004-of-00004.safetensors", + "model.layers.35.self_attn.k_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.35.self_attn.o_proj.weight": "model-00004-of-00004.safetensors", + "model.layers.35.self_attn.q_norm.weight": "model-00004-of-00004.safetensors", + "model.layers.35.self_attn.q_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.35.self_attn.v_proj.weight": "model-00003-of-00004.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.k_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.q_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.k_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.q_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.k_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.q_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.k_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.q_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.k_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.q_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors", + "model.layers.9.self_attn.k_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.self_attn.q_norm.weight": "model-00001-of-00004.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00004.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00004.safetensors", + "model.norm.weight": "model-00004-of-00004.safetensors" + } +} diff --git a/run_summary.json b/run_summary.json new file mode 100644 index 0000000..3a3d240 --- /dev/null +++ b/run_summary.json @@ -0,0 +1,12 @@ +{ + "agent_name": null, + "training_start": null, + "training_end": null, + "created_by": "DCAgent", + "base_model_name": "Qwen/Qwen3-8B", + "dataset_name": "DCAgent/exp_tas_trajectory_minimal_traces", + "training_type": "SFT", + "training_parameters": "https://huggingface.co/laion/Qwen3-8B_exp_tas_trajectory_minimal_traces_save-strategy_steps/blob/main/config.json", + "wandb_link": "https://wandb.ai/dogml/dc-agent/runs/Qwen3-8B_exp_tas_trajectory_minimal_traces_save-strategy_steps", + "traces_location_s3": null +} \ No newline at end of file diff --git a/special_tokens_map.json b/special_tokens_map.json new file mode 100644 index 0000000..ac23c0a --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1,31 @@ +{ + "additional_special_tokens": [ + "<|im_start|>", + "<|im_end|>", + "<|object_ref_start|>", + "<|object_ref_end|>", + "<|box_start|>", + "<|box_end|>", + "<|quad_start|>", + "<|quad_end|>", + "<|vision_start|>", + "<|vision_end|>", + "<|vision_pad|>", + "<|image_pad|>", + "<|video_pad|>" + ], + "eos_token": { + "content": "<|im_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|endoftext|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000..cd71f61 --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4 +size 11422654 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000..e9dc937 --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,240 @@ +{ + "add_bos_token": false, + "add_prefix_space": false, + "added_tokens_decoder": { + "151643": { + "content": "<|endoftext|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151644": { + "content": "<|im_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151645": { + "content": "<|im_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151646": { + "content": "<|object_ref_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151647": { + "content": "<|object_ref_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151648": { + "content": "<|box_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151649": { + "content": "<|box_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151650": { + "content": "<|quad_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151651": { + "content": "<|quad_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151652": { + "content": "<|vision_start|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151653": { + "content": "<|vision_end|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151654": { + "content": "<|vision_pad|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151655": { + "content": "<|image_pad|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151656": { + "content": "<|video_pad|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "151657": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151658": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151659": { + "content": "<|fim_prefix|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151660": { + "content": "<|fim_middle|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151661": { + "content": "<|fim_suffix|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151662": { + "content": "<|fim_pad|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151663": { + "content": "<|repo_name|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151664": { + "content": "<|file_sep|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151665": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151666": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151667": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + }, + "151668": { + "content": "", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": false + } + }, + "additional_special_tokens": [ + "<|im_start|>", + "<|im_end|>", + "<|object_ref_start|>", + "<|object_ref_end|>", + "<|box_start|>", + "<|box_end|>", + "<|quad_start|>", + "<|quad_end|>", + "<|vision_start|>", + "<|vision_end|>", + "<|vision_pad|>", + "<|image_pad|>", + "<|video_pad|>" + ], + "bos_token": null, + "clean_up_tokenization_spaces": false, + "eos_token": "<|im_end|>", + "errors": "replace", + "extra_special_tokens": {}, + "model_max_length": 32768, + "pad_token": "<|endoftext|>", + "padding_side": "right", + "split_special_tokens": false, + "tokenizer_class": "Qwen2Tokenizer", + "unk_token": null +} diff --git a/train_results.json b/train_results.json new file mode 100644 index 0000000..ad111b9 --- /dev/null +++ b/train_results.json @@ -0,0 +1,16 @@ +{ + "achieved_tflops_per_gpu": 0.0010303227944526512, + "achieved_tflops_per_gpu_theoretical": 243.92912852326808, + "epoch": 8.0, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.003110480960458517, + "mfu_percent": 0.0003302316648886703, + "mfu_percent_theoretical": 78.18241298822694, + "total_flos": 1155060078084096.0, + "train_loss": 0.14387717474665906, + "train_runtime": 35033.3193, + "train_samples_per_second": 2.295, + "train_steps_per_second": 0.072, + "valid_targets_mean": 2280.5, + "valid_targets_min": 275 +} \ No newline at end of file diff --git a/trainer_log.jsonl b/trainer_log.jsonl new file mode 100644 index 0000000..c3fddb4 --- /dev/null +++ b/trainer_log.jsonl @@ -0,0 +1,525 @@ +{"current_steps": 5, "total_steps": 2520, "loss": 0.9183, "lr": 3.0769230769230774e-05, "epoch": 0.015873015873015872, "percentage": 0.2, "elapsed_time": "0:01:49", "remaining_time": "15:21:21"} +{"current_steps": 10, "total_steps": 2520, "loss": 0.677, "lr": 6.923076923076924e-05, "epoch": 0.031746031746031744, "percentage": 0.4, "elapsed_time": "0:02:58", "remaining_time": "12:24:44"} +{"current_steps": 15, "total_steps": 2520, "loss": 0.5679, "lr": 9.99999607417416e-05, "epoch": 0.047619047619047616, "percentage": 0.6, "elapsed_time": "0:04:02", "remaining_time": "11:15:26"} +{"current_steps": 20, "total_steps": 2520, "loss": 0.5439, "lr": 9.999858670917045e-05, "epoch": 0.06349206349206349, "percentage": 0.79, "elapsed_time": "0:05:08", "remaining_time": "10:42:39"} +{"current_steps": 25, "total_steps": 2520, "loss": 0.5406, "lr": 9.999524982532699e-05, "epoch": 0.07936507936507936, "percentage": 0.99, "elapsed_time": "0:06:14", "remaining_time": "10:22:56"} +{"current_steps": 30, "total_steps": 2520, "loss": 0.5202, "lr": 9.998995022121104e-05, "epoch": 0.09523809523809523, "percentage": 1.19, "elapsed_time": "0:07:14", "remaining_time": "10:01:10"} +{"current_steps": 35, "total_steps": 2520, "loss": 0.5274, "lr": 9.998268810487518e-05, "epoch": 0.1111111111111111, "percentage": 1.39, "elapsed_time": "0:08:20", "remaining_time": "9:52:35"} +{"current_steps": 40, "total_steps": 2520, "loss": 0.5177, "lr": 9.997346376141656e-05, "epoch": 0.12698412698412698, "percentage": 1.59, "elapsed_time": "0:09:22", "remaining_time": "9:41:14"} +{"current_steps": 45, "total_steps": 2520, "loss": 0.5007, "lr": 9.99622775529657e-05, "epoch": 0.14285714285714285, "percentage": 1.79, "elapsed_time": "0:10:25", "remaining_time": "9:33:45"} +{"current_steps": 50, "total_steps": 2520, "loss": 0.4833, "lr": 9.994912991867228e-05, "epoch": 0.15873015873015872, "percentage": 1.98, "elapsed_time": "0:11:32", "remaining_time": "9:30:15"} +{"current_steps": 55, "total_steps": 2520, "loss": 0.4872, "lr": 9.99340213746879e-05, "epoch": 0.1746031746031746, "percentage": 2.18, "elapsed_time": "0:12:35", "remaining_time": "9:24:32"} +{"current_steps": 60, "total_steps": 2520, "loss": 0.5027, "lr": 9.991695251414583e-05, "epoch": 0.19047619047619047, "percentage": 2.38, "elapsed_time": "0:13:40", "remaining_time": "9:20:44"} +{"current_steps": 65, "total_steps": 2520, "loss": 0.4943, "lr": 9.989792400713771e-05, "epoch": 0.20634920634920634, "percentage": 2.58, "elapsed_time": "0:14:43", "remaining_time": "9:16:05"} +{"current_steps": 70, "total_steps": 2520, "loss": 0.4771, "lr": 9.987693660068722e-05, "epoch": 0.2222222222222222, "percentage": 2.78, "elapsed_time": "0:15:49", "remaining_time": "9:14:04"} +{"current_steps": 75, "total_steps": 2520, "loss": 0.4757, "lr": 9.985399111872081e-05, "epoch": 0.23809523809523808, "percentage": 2.98, "elapsed_time": "0:16:56", "remaining_time": "9:12:27"} +{"current_steps": 80, "total_steps": 2520, "loss": 0.4658, "lr": 9.982908846203529e-05, "epoch": 0.25396825396825395, "percentage": 3.17, "elapsed_time": "0:18:02", "remaining_time": "9:10:09"} +{"current_steps": 85, "total_steps": 2520, "loss": 0.4686, "lr": 9.980222960826254e-05, "epoch": 0.2698412698412698, "percentage": 3.37, "elapsed_time": "0:19:07", "remaining_time": "9:07:53"} +{"current_steps": 90, "total_steps": 2520, "loss": 0.484, "lr": 9.977341561183109e-05, "epoch": 0.2857142857142857, "percentage": 3.57, "elapsed_time": "0:20:11", "remaining_time": "9:05:17"} +{"current_steps": 95, "total_steps": 2520, "loss": 0.4722, "lr": 9.974264760392466e-05, "epoch": 0.30158730158730157, "percentage": 3.77, "elapsed_time": "0:21:11", "remaining_time": "9:00:50"} +{"current_steps": 100, "total_steps": 2520, "loss": 0.4854, "lr": 9.97099267924379e-05, "epoch": 0.31746031746031744, "percentage": 3.97, "elapsed_time": "0:22:15", "remaining_time": "8:58:34"} +{"current_steps": 105, "total_steps": 2520, "loss": 0.4829, "lr": 9.967525446192882e-05, "epoch": 0.3333333333333333, "percentage": 4.17, "elapsed_time": "0:47:51", "remaining_time": "18:20:35"} +{"current_steps": 110, "total_steps": 2520, "loss": 0.4595, "lr": 9.963863197356849e-05, "epoch": 0.3492063492063492, "percentage": 4.37, "elapsed_time": "0:49:01", "remaining_time": "17:54:02"} +{"current_steps": 115, "total_steps": 2520, "loss": 0.4953, "lr": 9.960006076508747e-05, "epoch": 0.36507936507936506, "percentage": 4.56, "elapsed_time": "0:50:07", "remaining_time": "17:28:12"} +{"current_steps": 120, "total_steps": 2520, "loss": 0.4536, "lr": 9.95595423507195e-05, "epoch": 0.38095238095238093, "percentage": 4.76, "elapsed_time": "0:51:10", "remaining_time": "17:03:24"} +{"current_steps": 125, "total_steps": 2520, "loss": 0.4567, "lr": 9.951707832114193e-05, "epoch": 0.3968253968253968, "percentage": 4.96, "elapsed_time": "0:52:15", "remaining_time": "16:41:15"} +{"current_steps": 130, "total_steps": 2520, "loss": 0.4516, "lr": 9.947267034341341e-05, "epoch": 0.4126984126984127, "percentage": 5.16, "elapsed_time": "0:53:22", "remaining_time": "16:21:09"} +{"current_steps": 135, "total_steps": 2520, "loss": 0.453, "lr": 9.942632016090832e-05, "epoch": 0.42857142857142855, "percentage": 5.36, "elapsed_time": "0:54:25", "remaining_time": "16:01:28"} +{"current_steps": 140, "total_steps": 2520, "loss": 0.4376, "lr": 9.937802959324838e-05, "epoch": 0.4444444444444444, "percentage": 5.56, "elapsed_time": "0:55:30", "remaining_time": "15:43:38"} +{"current_steps": 145, "total_steps": 2520, "loss": 0.4697, "lr": 9.932780053623121e-05, "epoch": 0.4603174603174603, "percentage": 5.75, "elapsed_time": "0:56:34", "remaining_time": "15:26:42"} +{"current_steps": 150, "total_steps": 2520, "loss": 0.4609, "lr": 9.927563496175593e-05, "epoch": 0.47619047619047616, "percentage": 5.95, "elapsed_time": "0:57:33", "remaining_time": "15:09:24"} +{"current_steps": 155, "total_steps": 2520, "loss": 0.4845, "lr": 9.922153491774572e-05, "epoch": 0.49206349206349204, "percentage": 6.15, "elapsed_time": "0:58:33", "remaining_time": "14:53:25"} +{"current_steps": 160, "total_steps": 2520, "loss": 0.4779, "lr": 9.91655025280674e-05, "epoch": 0.5079365079365079, "percentage": 6.35, "elapsed_time": "0:59:35", "remaining_time": "14:38:57"} +{"current_steps": 165, "total_steps": 2520, "loss": 0.4592, "lr": 9.910753999244811e-05, "epoch": 0.5238095238095238, "percentage": 6.55, "elapsed_time": "1:00:35", "remaining_time": "14:24:54"} +{"current_steps": 170, "total_steps": 2520, "loss": 0.4724, "lr": 9.904764958638889e-05, "epoch": 0.5396825396825397, "percentage": 6.75, "elapsed_time": "1:01:35", "remaining_time": "14:11:21"} +{"current_steps": 175, "total_steps": 2520, "loss": 0.4631, "lr": 9.898583366107538e-05, "epoch": 0.5555555555555556, "percentage": 6.94, "elapsed_time": "1:02:35", "remaining_time": "13:58:49"} +{"current_steps": 180, "total_steps": 2520, "loss": 0.4667, "lr": 9.892209464328556e-05, "epoch": 0.5714285714285714, "percentage": 7.14, "elapsed_time": "1:03:39", "remaining_time": "13:47:37"} +{"current_steps": 185, "total_steps": 2520, "loss": 0.4667, "lr": 9.885643503529439e-05, "epoch": 0.5873015873015873, "percentage": 7.34, "elapsed_time": "1:04:46", "remaining_time": "13:37:37"} +{"current_steps": 190, "total_steps": 2520, "loss": 0.4646, "lr": 9.878885741477563e-05, "epoch": 0.6031746031746031, "percentage": 7.54, "elapsed_time": "1:05:45", "remaining_time": "13:26:22"} +{"current_steps": 195, "total_steps": 2520, "loss": 0.474, "lr": 9.871936443470063e-05, "epoch": 0.6190476190476191, "percentage": 7.74, "elapsed_time": "1:06:51", "remaining_time": "13:17:13"} +{"current_steps": 200, "total_steps": 2520, "loss": 0.4619, "lr": 9.864795882323421e-05, "epoch": 0.6349206349206349, "percentage": 7.94, "elapsed_time": "1:07:54", "remaining_time": "13:07:47"} +{"current_steps": 105, "total_steps": 2520, "loss": 0.4829, "lr": 9.967525446192882e-05, "epoch": 0.3333333333333333, "percentage": 4.17, "elapsed_time": "0:01:52", "remaining_time": "0:43:00"} +{"current_steps": 110, "total_steps": 2520, "loss": 0.4593, "lr": 9.963863197356849e-05, "epoch": 0.3492063492063492, "percentage": 4.37, "elapsed_time": "0:03:07", "remaining_time": "1:08:19"} +{"current_steps": 115, "total_steps": 2520, "loss": 0.4956, "lr": 9.960006076508747e-05, "epoch": 0.36507936507936506, "percentage": 4.56, "elapsed_time": "0:04:15", "remaining_time": "1:29:05"} +{"current_steps": 120, "total_steps": 2520, "loss": 0.454, "lr": 9.95595423507195e-05, "epoch": 0.38095238095238093, "percentage": 4.76, "elapsed_time": "0:05:19", "remaining_time": "1:46:33"} +{"current_steps": 125, "total_steps": 2520, "loss": 0.4566, "lr": 9.951707832114193e-05, "epoch": 0.3968253968253968, "percentage": 4.96, "elapsed_time": "0:06:27", "remaining_time": "2:03:35"} +{"current_steps": 130, "total_steps": 2520, "loss": 0.4509, "lr": 9.947267034341341e-05, "epoch": 0.4126984126984127, "percentage": 5.16, "elapsed_time": "0:07:34", "remaining_time": "2:19:23"} +{"current_steps": 135, "total_steps": 2520, "loss": 0.4527, "lr": 9.942632016090832e-05, "epoch": 0.42857142857142855, "percentage": 5.36, "elapsed_time": "0:08:39", "remaining_time": "2:32:54"} +{"current_steps": 140, "total_steps": 2520, "loss": 0.4381, "lr": 9.937802959324838e-05, "epoch": 0.4444444444444444, "percentage": 5.56, "elapsed_time": "0:09:45", "remaining_time": "2:45:54"} +{"current_steps": 145, "total_steps": 2520, "loss": 0.4693, "lr": 9.932780053623121e-05, "epoch": 0.4603174603174603, "percentage": 5.75, "elapsed_time": "0:10:50", "remaining_time": "2:57:35"} +{"current_steps": 150, "total_steps": 2520, "loss": 0.4604, "lr": 9.927563496175593e-05, "epoch": 0.47619047619047616, "percentage": 5.95, "elapsed_time": "0:11:50", "remaining_time": "3:07:08"} +{"current_steps": 155, "total_steps": 2520, "loss": 0.4843, "lr": 9.922153491774572e-05, "epoch": 0.49206349206349204, "percentage": 6.15, "elapsed_time": "0:12:50", "remaining_time": "3:15:56"} +{"current_steps": 160, "total_steps": 2520, "loss": 0.4776, "lr": 9.91655025280674e-05, "epoch": 0.5079365079365079, "percentage": 6.35, "elapsed_time": "0:13:53", "remaining_time": "3:24:49"} +{"current_steps": 165, "total_steps": 2520, "loss": 0.4586, "lr": 9.910753999244811e-05, "epoch": 0.5238095238095238, "percentage": 6.55, "elapsed_time": "0:14:54", "remaining_time": "3:32:41"} +{"current_steps": 170, "total_steps": 2520, "loss": 0.4729, "lr": 9.904764958638889e-05, "epoch": 0.5396825396825397, "percentage": 6.75, "elapsed_time": "0:15:53", "remaining_time": "3:39:43"} +{"current_steps": 175, "total_steps": 2520, "loss": 0.4635, "lr": 9.898583366107538e-05, "epoch": 0.5555555555555556, "percentage": 6.94, "elapsed_time": "0:16:54", "remaining_time": "3:46:36"} +{"current_steps": 180, "total_steps": 2520, "loss": 0.4664, "lr": 9.892209464328556e-05, "epoch": 0.5714285714285714, "percentage": 7.14, "elapsed_time": "0:17:58", "remaining_time": "3:53:45"} +{"current_steps": 185, "total_steps": 2520, "loss": 0.4663, "lr": 9.885643503529439e-05, "epoch": 0.5873015873015873, "percentage": 7.34, "elapsed_time": "0:19:06", "remaining_time": "4:01:09"} +{"current_steps": 190, "total_steps": 2520, "loss": 0.4642, "lr": 9.878885741477563e-05, "epoch": 0.6031746031746031, "percentage": 7.54, "elapsed_time": "0:20:05", "remaining_time": "4:06:27"} +{"current_steps": 195, "total_steps": 2520, "loss": 0.4744, "lr": 9.871936443470063e-05, "epoch": 0.6190476190476191, "percentage": 7.74, "elapsed_time": "0:21:12", "remaining_time": "4:12:51"} +{"current_steps": 200, "total_steps": 2520, "loss": 0.4621, "lr": 9.864795882323421e-05, "epoch": 0.6349206349206349, "percentage": 7.94, "elapsed_time": "0:22:16", "remaining_time": "4:18:19"} +{"current_steps": 205, "total_steps": 2520, "loss": 0.4526, "lr": 9.857464338362748e-05, "epoch": 0.6507936507936508, "percentage": 8.13, "elapsed_time": "0:24:32", "remaining_time": "4:37:05"} +{"current_steps": 210, "total_steps": 2520, "loss": 0.4395, "lr": 9.849942099410792e-05, "epoch": 0.6666666666666666, "percentage": 8.33, "elapsed_time": "0:25:32", "remaining_time": "4:40:53"} +{"current_steps": 215, "total_steps": 2520, "loss": 0.4503, "lr": 9.842229460776622e-05, "epoch": 0.6825396825396826, "percentage": 8.53, "elapsed_time": "0:26:36", "remaining_time": "4:45:12"} +{"current_steps": 220, "total_steps": 2520, "loss": 0.4555, "lr": 9.834326725244049e-05, "epoch": 0.6984126984126984, "percentage": 8.73, "elapsed_time": "0:27:39", "remaining_time": "4:49:04"} +{"current_steps": 225, "total_steps": 2520, "loss": 0.4473, "lr": 9.826234203059731e-05, "epoch": 0.7142857142857143, "percentage": 8.93, "elapsed_time": "0:28:45", "remaining_time": "4:53:19"} +{"current_steps": 230, "total_steps": 2520, "loss": 0.4442, "lr": 9.817952211921e-05, "epoch": 0.7301587301587301, "percentage": 9.13, "elapsed_time": "0:29:42", "remaining_time": "4:55:46"} +{"current_steps": 235, "total_steps": 2520, "loss": 0.4461, "lr": 9.809481076963383e-05, "epoch": 0.746031746031746, "percentage": 9.33, "elapsed_time": "0:30:46", "remaining_time": "4:59:18"} +{"current_steps": 240, "total_steps": 2520, "loss": 0.4432, "lr": 9.800821130747837e-05, "epoch": 0.7619047619047619, "percentage": 9.52, "elapsed_time": "0:31:45", "remaining_time": "5:01:43"} +{"current_steps": 245, "total_steps": 2520, "loss": 0.4509, "lr": 9.791972713247704e-05, "epoch": 0.7777777777777778, "percentage": 9.72, "elapsed_time": "0:32:46", "remaining_time": "5:04:16"} +{"current_steps": 250, "total_steps": 2520, "loss": 0.4416, "lr": 9.782936171835353e-05, "epoch": 0.7936507936507936, "percentage": 9.92, "elapsed_time": "0:33:49", "remaining_time": "5:07:05"} +{"current_steps": 255, "total_steps": 2520, "loss": 0.451, "lr": 9.773711861268549e-05, "epoch": 0.8095238095238095, "percentage": 10.12, "elapsed_time": "0:34:49", "remaining_time": "5:09:21"} +{"current_steps": 260, "total_steps": 2520, "loss": 0.446, "lr": 9.764300143676518e-05, "epoch": 0.8253968253968254, "percentage": 10.32, "elapsed_time": "0:35:54", "remaining_time": "5:12:06"} +{"current_steps": 265, "total_steps": 2520, "loss": 0.4865, "lr": 9.754701388545745e-05, "epoch": 0.8412698412698413, "percentage": 10.52, "elapsed_time": "0:36:57", "remaining_time": "5:14:25"} +{"current_steps": 270, "total_steps": 2520, "loss": 0.4571, "lr": 9.744915972705453e-05, "epoch": 0.8571428571428571, "percentage": 10.71, "elapsed_time": "0:37:56", "remaining_time": "5:16:13"} +{"current_steps": 275, "total_steps": 2520, "loss": 0.4779, "lr": 9.734944280312824e-05, "epoch": 0.873015873015873, "percentage": 10.91, "elapsed_time": "0:38:59", "remaining_time": "5:18:19"} +{"current_steps": 280, "total_steps": 2520, "loss": 0.4495, "lr": 9.7247867028379e-05, "epoch": 0.8888888888888888, "percentage": 11.11, "elapsed_time": "0:40:00", "remaining_time": "5:20:03"} +{"current_steps": 285, "total_steps": 2520, "loss": 0.4394, "lr": 9.714443639048232e-05, "epoch": 0.9047619047619048, "percentage": 11.31, "elapsed_time": "0:41:02", "remaining_time": "5:21:51"} +{"current_steps": 290, "total_steps": 2520, "loss": 0.4622, "lr": 9.703915494993215e-05, "epoch": 0.9206349206349206, "percentage": 11.51, "elapsed_time": "0:42:04", "remaining_time": "5:23:34"} +{"current_steps": 295, "total_steps": 2520, "loss": 0.4422, "lr": 9.693202683988151e-05, "epoch": 0.9365079365079365, "percentage": 11.71, "elapsed_time": "0:43:11", "remaining_time": "5:25:46"} +{"current_steps": 300, "total_steps": 2520, "loss": 0.4536, "lr": 9.682305626598023e-05, "epoch": 0.9523809523809523, "percentage": 11.9, "elapsed_time": "0:44:11", "remaining_time": "5:27:00"} +{"current_steps": 305, "total_steps": 2520, "loss": 0.4463, "lr": 9.671224750620981e-05, "epoch": 0.9682539682539683, "percentage": 12.1, "elapsed_time": "0:46:35", "remaining_time": "5:38:22"} +{"current_steps": 310, "total_steps": 2520, "loss": 0.4605, "lr": 9.659960491071554e-05, "epoch": 0.9841269841269841, "percentage": 12.3, "elapsed_time": "0:47:36", "remaining_time": "5:39:25"} +{"current_steps": 315, "total_steps": 2520, "loss": 0.4564, "lr": 9.64851329016356e-05, "epoch": 1.0, "percentage": 12.5, "elapsed_time": "0:48:39", "remaining_time": "5:40:39"} +{"current_steps": 320, "total_steps": 2520, "loss": 0.3484, "lr": 9.636883597292762e-05, "epoch": 1.0158730158730158, "percentage": 12.7, "elapsed_time": "0:49:46", "remaining_time": "5:42:14"} +{"current_steps": 325, "total_steps": 2520, "loss": 0.3353, "lr": 9.625071869019215e-05, "epoch": 1.0317460317460316, "percentage": 12.9, "elapsed_time": "0:50:48", "remaining_time": "5:43:08"} +{"current_steps": 330, "total_steps": 2520, "loss": 0.3332, "lr": 9.613078569049344e-05, "epoch": 1.0476190476190477, "percentage": 13.1, "elapsed_time": "0:51:47", "remaining_time": "5:43:45"} +{"current_steps": 335, "total_steps": 2520, "loss": 0.3401, "lr": 9.600904168217734e-05, "epoch": 1.0634920634920635, "percentage": 13.29, "elapsed_time": "0:52:43", "remaining_time": "5:43:55"} +{"current_steps": 340, "total_steps": 2520, "loss": 0.3486, "lr": 9.588549144468664e-05, "epoch": 1.0793650793650793, "percentage": 13.49, "elapsed_time": "0:53:47", "remaining_time": "5:44:53"} +{"current_steps": 345, "total_steps": 2520, "loss": 0.3307, "lr": 9.576013982837324e-05, "epoch": 1.0952380952380953, "percentage": 13.69, "elapsed_time": "0:54:43", "remaining_time": "5:45:01"} +{"current_steps": 350, "total_steps": 2520, "loss": 0.3211, "lr": 9.563299175430782e-05, "epoch": 1.1111111111111112, "percentage": 13.89, "elapsed_time": "0:55:48", "remaining_time": "5:45:59"} +{"current_steps": 355, "total_steps": 2520, "loss": 0.3229, "lr": 9.550405221408664e-05, "epoch": 1.126984126984127, "percentage": 14.09, "elapsed_time": "0:56:43", "remaining_time": "5:45:55"} +{"current_steps": 360, "total_steps": 2520, "loss": 0.3419, "lr": 9.537332626963561e-05, "epoch": 1.1428571428571428, "percentage": 14.29, "elapsed_time": "0:57:37", "remaining_time": "5:45:45"} +{"current_steps": 365, "total_steps": 2520, "loss": 0.334, "lr": 9.524081905301152e-05, "epoch": 1.1587301587301586, "percentage": 14.48, "elapsed_time": "0:58:42", "remaining_time": "5:46:37"} +{"current_steps": 370, "total_steps": 2520, "loss": 0.3577, "lr": 9.510653576620056e-05, "epoch": 1.1746031746031746, "percentage": 14.68, "elapsed_time": "0:59:47", "remaining_time": "5:47:27"} +{"current_steps": 375, "total_steps": 2520, "loss": 0.354, "lr": 9.497048168091418e-05, "epoch": 1.1904761904761905, "percentage": 14.88, "elapsed_time": "1:00:48", "remaining_time": "5:47:48"} +{"current_steps": 380, "total_steps": 2520, "loss": 0.3391, "lr": 9.483266213838202e-05, "epoch": 1.2063492063492063, "percentage": 15.08, "elapsed_time": "1:01:42", "remaining_time": "5:47:32"} +{"current_steps": 385, "total_steps": 2520, "loss": 0.3836, "lr": 9.469308254914231e-05, "epoch": 1.2222222222222223, "percentage": 15.28, "elapsed_time": "1:02:39", "remaining_time": "5:47:27"} +{"current_steps": 390, "total_steps": 2520, "loss": 0.3491, "lr": 9.455174839282941e-05, "epoch": 1.2380952380952381, "percentage": 15.48, "elapsed_time": "1:03:41", "remaining_time": "5:47:52"} +{"current_steps": 395, "total_steps": 2520, "loss": 0.3491, "lr": 9.440866521795874e-05, "epoch": 1.253968253968254, "percentage": 15.67, "elapsed_time": "1:04:46", "remaining_time": "5:48:28"} +{"current_steps": 400, "total_steps": 2520, "loss": 0.3533, "lr": 9.426383864170891e-05, "epoch": 1.2698412698412698, "percentage": 15.87, "elapsed_time": "1:05:49", "remaining_time": "5:48:51"} +{"current_steps": 405, "total_steps": 2520, "loss": 0.3356, "lr": 9.411727434970121e-05, "epoch": 1.2857142857142856, "percentage": 16.07, "elapsed_time": "1:08:06", "remaining_time": "5:55:38"} +{"current_steps": 410, "total_steps": 2520, "loss": 0.3658, "lr": 9.396897809577643e-05, "epoch": 1.3015873015873016, "percentage": 16.27, "elapsed_time": "1:09:06", "remaining_time": "5:55:36"} +{"current_steps": 415, "total_steps": 2520, "loss": 0.3729, "lr": 9.381895570176893e-05, "epoch": 1.3174603174603174, "percentage": 16.47, "elapsed_time": "1:10:07", "remaining_time": "5:55:40"} +{"current_steps": 420, "total_steps": 2520, "loss": 0.3503, "lr": 9.366721305727813e-05, "epoch": 1.3333333333333333, "percentage": 16.67, "elapsed_time": "1:11:06", "remaining_time": "5:55:31"} +{"current_steps": 425, "total_steps": 2520, "loss": 0.3474, "lr": 9.351375611943724e-05, "epoch": 1.3492063492063493, "percentage": 16.87, "elapsed_time": "1:12:09", "remaining_time": "5:55:41"} +{"current_steps": 430, "total_steps": 2520, "loss": 0.346, "lr": 9.335859091267952e-05, "epoch": 1.3650793650793651, "percentage": 17.06, "elapsed_time": "1:13:06", "remaining_time": "5:55:18"} +{"current_steps": 435, "total_steps": 2520, "loss": 0.3528, "lr": 9.320172352850156e-05, "epoch": 1.380952380952381, "percentage": 17.26, "elapsed_time": "1:14:03", "remaining_time": "5:54:55"} +{"current_steps": 440, "total_steps": 2520, "loss": 0.3361, "lr": 9.304316012522437e-05, "epoch": 1.3968253968253967, "percentage": 17.46, "elapsed_time": "1:15:05", "remaining_time": "5:54:59"} +{"current_steps": 445, "total_steps": 2520, "loss": 0.3616, "lr": 9.288290692775143e-05, "epoch": 1.4126984126984126, "percentage": 17.66, "elapsed_time": "1:16:09", "remaining_time": "5:55:07"} +{"current_steps": 450, "total_steps": 2520, "loss": 0.3754, "lr": 9.272097022732443e-05, "epoch": 1.4285714285714286, "percentage": 17.86, "elapsed_time": "1:17:11", "remaining_time": "5:55:06"} +{"current_steps": 455, "total_steps": 2520, "loss": 0.3625, "lr": 9.255735638127623e-05, "epoch": 1.4444444444444444, "percentage": 18.06, "elapsed_time": "1:18:17", "remaining_time": "5:55:20"} +{"current_steps": 460, "total_steps": 2520, "loss": 0.3439, "lr": 9.239207181278131e-05, "epoch": 1.4603174603174602, "percentage": 18.25, "elapsed_time": "1:19:19", "remaining_time": "5:55:15"} +{"current_steps": 465, "total_steps": 2520, "loss": 0.3335, "lr": 9.222512301060358e-05, "epoch": 1.4761904761904763, "percentage": 18.45, "elapsed_time": "1:20:20", "remaining_time": "5:55:04"} +{"current_steps": 470, "total_steps": 2520, "loss": 0.3371, "lr": 9.205651652884169e-05, "epoch": 1.492063492063492, "percentage": 18.65, "elapsed_time": "1:21:20", "remaining_time": "5:54:46"} +{"current_steps": 475, "total_steps": 2520, "loss": 0.3785, "lr": 9.188625898667165e-05, "epoch": 1.507936507936508, "percentage": 18.85, "elapsed_time": "1:22:21", "remaining_time": "5:54:36"} +{"current_steps": 480, "total_steps": 2520, "loss": 0.3298, "lr": 9.171435706808709e-05, "epoch": 1.5238095238095237, "percentage": 19.05, "elapsed_time": "1:23:24", "remaining_time": "5:54:28"} +{"current_steps": 485, "total_steps": 2520, "loss": 0.3699, "lr": 9.154081752163675e-05, "epoch": 1.5396825396825395, "percentage": 19.25, "elapsed_time": "1:24:26", "remaining_time": "5:54:17"} +{"current_steps": 490, "total_steps": 2520, "loss": 0.3417, "lr": 9.136564716015956e-05, "epoch": 1.5555555555555556, "percentage": 19.44, "elapsed_time": "1:25:23", "remaining_time": "5:53:45"} +{"current_steps": 495, "total_steps": 2520, "loss": 0.3814, "lr": 9.118885286051726e-05, "epoch": 1.5714285714285714, "percentage": 19.64, "elapsed_time": "1:26:25", "remaining_time": "5:53:34"} +{"current_steps": 500, "total_steps": 2520, "loss": 0.3483, "lr": 9.101044156332437e-05, "epoch": 1.5873015873015874, "percentage": 19.84, "elapsed_time": "1:27:27", "remaining_time": "5:53:18"} +{"current_steps": 505, "total_steps": 2520, "loss": 0.3624, "lr": 9.083042027267567e-05, "epoch": 1.6031746031746033, "percentage": 20.04, "elapsed_time": "1:32:06", "remaining_time": "6:07:31"} +{"current_steps": 510, "total_steps": 2520, "loss": 0.3363, "lr": 9.064879605587132e-05, "epoch": 1.619047619047619, "percentage": 20.24, "elapsed_time": "1:33:12", "remaining_time": "6:07:22"} +{"current_steps": 515, "total_steps": 2520, "loss": 0.3535, "lr": 9.046557604313937e-05, "epoch": 1.6349206349206349, "percentage": 20.44, "elapsed_time": "1:34:13", "remaining_time": "6:06:50"} +{"current_steps": 520, "total_steps": 2520, "loss": 0.374, "lr": 9.028076742735583e-05, "epoch": 1.6507936507936507, "percentage": 20.63, "elapsed_time": "1:35:15", "remaining_time": "6:06:23"} +{"current_steps": 525, "total_steps": 2520, "loss": 0.3767, "lr": 9.009437746376231e-05, "epoch": 1.6666666666666665, "percentage": 20.83, "elapsed_time": "1:36:15", "remaining_time": "6:05:48"} +{"current_steps": 530, "total_steps": 2520, "loss": 0.3348, "lr": 8.990641346968117e-05, "epoch": 1.6825396825396826, "percentage": 21.03, "elapsed_time": "1:37:20", "remaining_time": "6:05:28"} +{"current_steps": 535, "total_steps": 2520, "loss": 0.3716, "lr": 8.97168828242283e-05, "epoch": 1.6984126984126984, "percentage": 21.23, "elapsed_time": "1:38:21", "remaining_time": "6:04:56"} +{"current_steps": 540, "total_steps": 2520, "loss": 0.3516, "lr": 8.952579296802339e-05, "epoch": 1.7142857142857144, "percentage": 21.43, "elapsed_time": "1:39:13", "remaining_time": "6:03:48"} +{"current_steps": 545, "total_steps": 2520, "loss": 0.3683, "lr": 8.933315140289782e-05, "epoch": 1.7301587301587302, "percentage": 21.63, "elapsed_time": "1:40:10", "remaining_time": "6:03:01"} +{"current_steps": 550, "total_steps": 2520, "loss": 0.352, "lr": 8.91389656916002e-05, "epoch": 1.746031746031746, "percentage": 21.83, "elapsed_time": "1:41:14", "remaining_time": "6:02:36"} +{"current_steps": 555, "total_steps": 2520, "loss": 0.3603, "lr": 8.894324345749939e-05, "epoch": 1.7619047619047619, "percentage": 22.02, "elapsed_time": "1:42:18", "remaining_time": "6:02:15"} +{"current_steps": 560, "total_steps": 2520, "loss": 0.3726, "lr": 8.874599238428533e-05, "epoch": 1.7777777777777777, "percentage": 22.22, "elapsed_time": "1:43:20", "remaining_time": "6:01:40"} +{"current_steps": 565, "total_steps": 2520, "loss": 0.3506, "lr": 8.85472202156673e-05, "epoch": 1.7936507936507935, "percentage": 22.42, "elapsed_time": "1:44:18", "remaining_time": "6:00:55"} +{"current_steps": 570, "total_steps": 2520, "loss": 0.3483, "lr": 8.834693475506992e-05, "epoch": 1.8095238095238095, "percentage": 22.62, "elapsed_time": "1:45:23", "remaining_time": "6:00:33"} +{"current_steps": 575, "total_steps": 2520, "loss": 0.3578, "lr": 8.814514386532691e-05, "epoch": 1.8253968253968254, "percentage": 22.82, "elapsed_time": "1:46:26", "remaining_time": "6:00:02"} +{"current_steps": 580, "total_steps": 2520, "loss": 0.328, "lr": 8.794185546837224e-05, "epoch": 1.8412698412698414, "percentage": 23.02, "elapsed_time": "1:47:29", "remaining_time": "5:59:31"} +{"current_steps": 585, "total_steps": 2520, "loss": 0.361, "lr": 8.773707754492928e-05, "epoch": 1.8571428571428572, "percentage": 23.21, "elapsed_time": "1:48:33", "remaining_time": "5:59:05"} +{"current_steps": 590, "total_steps": 2520, "loss": 0.3505, "lr": 8.753081813419743e-05, "epoch": 1.873015873015873, "percentage": 23.41, "elapsed_time": "1:49:35", "remaining_time": "5:58:31"} +{"current_steps": 595, "total_steps": 2520, "loss": 0.3555, "lr": 8.73230853335365e-05, "epoch": 1.8888888888888888, "percentage": 23.61, "elapsed_time": "1:50:36", "remaining_time": "5:57:51"} +{"current_steps": 600, "total_steps": 2520, "loss": 0.3519, "lr": 8.711388729814882e-05, "epoch": 1.9047619047619047, "percentage": 23.81, "elapsed_time": "1:51:38", "remaining_time": "5:57:16"} +{"current_steps": 605, "total_steps": 2520, "loss": 0.3807, "lr": 8.690323224075917e-05, "epoch": 1.9206349206349205, "percentage": 24.01, "elapsed_time": "1:56:34", "remaining_time": "6:08:59"} +{"current_steps": 610, "total_steps": 2520, "loss": 0.3537, "lr": 8.669112843129221e-05, "epoch": 1.9365079365079365, "percentage": 24.21, "elapsed_time": "1:57:41", "remaining_time": "6:08:29"} +{"current_steps": 615, "total_steps": 2520, "loss": 0.3664, "lr": 8.647758419654798e-05, "epoch": 1.9523809523809523, "percentage": 24.4, "elapsed_time": "1:58:35", "remaining_time": "6:07:19"} +{"current_steps": 620, "total_steps": 2520, "loss": 0.3745, "lr": 8.626260791987488e-05, "epoch": 1.9682539682539684, "percentage": 24.6, "elapsed_time": "1:59:35", "remaining_time": "6:06:28"} +{"current_steps": 625, "total_steps": 2520, "loss": 0.3551, "lr": 8.604620804084065e-05, "epoch": 1.9841269841269842, "percentage": 24.8, "elapsed_time": "2:00:31", "remaining_time": "6:05:26"} +{"current_steps": 630, "total_steps": 2520, "loss": 0.3453, "lr": 8.582839305490094e-05, "epoch": 2.0, "percentage": 25.0, "elapsed_time": "2:01:23", "remaining_time": "6:04:10"} +{"current_steps": 635, "total_steps": 2520, "loss": 0.23, "lr": 8.560917151306593e-05, "epoch": 2.015873015873016, "percentage": 25.2, "elapsed_time": "2:02:30", "remaining_time": "6:03:41"} +{"current_steps": 640, "total_steps": 2520, "loss": 0.2316, "lr": 8.538855202156453e-05, "epoch": 2.0317460317460316, "percentage": 25.4, "elapsed_time": "2:03:31", "remaining_time": "6:02:51"} +{"current_steps": 645, "total_steps": 2520, "loss": 0.2271, "lr": 8.516654324150652e-05, "epoch": 2.0476190476190474, "percentage": 25.6, "elapsed_time": "2:04:34", "remaining_time": "6:02:09"} +{"current_steps": 650, "total_steps": 2520, "loss": 0.2251, "lr": 8.49431538885426e-05, "epoch": 2.0634920634920633, "percentage": 25.79, "elapsed_time": "2:05:38", "remaining_time": "6:01:26"} +{"current_steps": 655, "total_steps": 2520, "loss": 0.2359, "lr": 8.471839273252217e-05, "epoch": 2.0793650793650795, "percentage": 25.99, "elapsed_time": "2:06:37", "remaining_time": "6:00:33"} +{"current_steps": 660, "total_steps": 2520, "loss": 0.2322, "lr": 8.449226859714907e-05, "epoch": 2.0952380952380953, "percentage": 26.19, "elapsed_time": "2:07:40", "remaining_time": "5:59:49"} +{"current_steps": 665, "total_steps": 2520, "loss": 0.2417, "lr": 8.426479035963513e-05, "epoch": 2.111111111111111, "percentage": 26.39, "elapsed_time": "2:08:42", "remaining_time": "5:59:00"} +{"current_steps": 670, "total_steps": 2520, "loss": 0.2185, "lr": 8.403596695035174e-05, "epoch": 2.126984126984127, "percentage": 26.59, "elapsed_time": "2:09:44", "remaining_time": "5:58:14"} +{"current_steps": 675, "total_steps": 2520, "loss": 0.2272, "lr": 8.380580735247925e-05, "epoch": 2.142857142857143, "percentage": 26.79, "elapsed_time": "2:10:44", "remaining_time": "5:57:22"} +{"current_steps": 680, "total_steps": 2520, "loss": 0.2251, "lr": 8.35743206016542e-05, "epoch": 2.1587301587301586, "percentage": 26.98, "elapsed_time": "2:11:48", "remaining_time": "5:56:40"} +{"current_steps": 685, "total_steps": 2520, "loss": 0.2446, "lr": 8.334151578561478e-05, "epoch": 2.1746031746031744, "percentage": 27.18, "elapsed_time": "2:12:46", "remaining_time": "5:55:41"} +{"current_steps": 690, "total_steps": 2520, "loss": 0.2381, "lr": 8.310740204384387e-05, "epoch": 2.1904761904761907, "percentage": 27.38, "elapsed_time": "2:13:51", "remaining_time": "5:55:01"} +{"current_steps": 695, "total_steps": 2520, "loss": 0.2379, "lr": 8.287198856721042e-05, "epoch": 2.2063492063492065, "percentage": 27.58, "elapsed_time": "2:14:48", "remaining_time": "5:53:59"} +{"current_steps": 700, "total_steps": 2520, "loss": 0.2215, "lr": 8.263528459760844e-05, "epoch": 2.2222222222222223, "percentage": 27.78, "elapsed_time": "2:15:51", "remaining_time": "5:53:12"} +{"current_steps": 705, "total_steps": 2520, "loss": 0.2371, "lr": 8.23972994275944e-05, "epoch": 2.238095238095238, "percentage": 27.98, "elapsed_time": "2:23:04", "remaining_time": "6:08:19"} +{"current_steps": 710, "total_steps": 2520, "loss": 0.229, "lr": 8.215804240002225e-05, "epoch": 2.253968253968254, "percentage": 28.17, "elapsed_time": "2:24:05", "remaining_time": "6:07:18"} +{"current_steps": 715, "total_steps": 2520, "loss": 0.2232, "lr": 8.191752290767671e-05, "epoch": 2.2698412698412698, "percentage": 28.37, "elapsed_time": "2:25:04", "remaining_time": "6:06:13"} +{"current_steps": 720, "total_steps": 2520, "loss": 0.2407, "lr": 8.167575039290448e-05, "epoch": 2.2857142857142856, "percentage": 28.57, "elapsed_time": "2:26:06", "remaining_time": "6:05:15"} +{"current_steps": 725, "total_steps": 2520, "loss": 0.2522, "lr": 8.143273434724363e-05, "epoch": 2.3015873015873014, "percentage": 28.77, "elapsed_time": "2:27:12", "remaining_time": "6:04:26"} +{"current_steps": 730, "total_steps": 2520, "loss": 0.2386, "lr": 8.118848431105091e-05, "epoch": 2.317460317460317, "percentage": 28.97, "elapsed_time": "2:28:18", "remaining_time": "6:03:40"} +{"current_steps": 735, "total_steps": 2520, "loss": 0.2225, "lr": 8.094300987312725e-05, "epoch": 2.3333333333333335, "percentage": 29.17, "elapsed_time": "2:29:16", "remaining_time": "6:02:32"} +{"current_steps": 740, "total_steps": 2520, "loss": 0.239, "lr": 8.069632067034129e-05, "epoch": 2.3492063492063493, "percentage": 29.37, "elapsed_time": "2:30:20", "remaining_time": "6:01:37"} +{"current_steps": 745, "total_steps": 2520, "loss": 0.239, "lr": 8.044842638725107e-05, "epoch": 2.365079365079365, "percentage": 29.56, "elapsed_time": "2:31:20", "remaining_time": "6:00:33"} +{"current_steps": 750, "total_steps": 2520, "loss": 0.2396, "lr": 8.019933675572389e-05, "epoch": 2.380952380952381, "percentage": 29.76, "elapsed_time": "2:32:20", "remaining_time": "5:59:32"} +{"current_steps": 755, "total_steps": 2520, "loss": 0.2406, "lr": 7.994906155455411e-05, "epoch": 2.3968253968253967, "percentage": 29.96, "elapsed_time": "2:33:23", "remaining_time": "5:58:35"} +{"current_steps": 760, "total_steps": 2520, "loss": 0.2544, "lr": 7.969761060907943e-05, "epoch": 2.4126984126984126, "percentage": 30.16, "elapsed_time": "2:34:24", "remaining_time": "5:57:34"} +{"current_steps": 765, "total_steps": 2520, "loss": 0.2502, "lr": 7.944499379079502e-05, "epoch": 2.4285714285714284, "percentage": 30.36, "elapsed_time": "2:35:25", "remaining_time": "5:56:33"} +{"current_steps": 770, "total_steps": 2520, "loss": 0.2309, "lr": 7.919122101696606e-05, "epoch": 2.4444444444444446, "percentage": 30.56, "elapsed_time": "2:36:31", "remaining_time": "5:55:43"} +{"current_steps": 775, "total_steps": 2520, "loss": 0.2198, "lr": 7.893630225023842e-05, "epoch": 2.4603174603174605, "percentage": 30.75, "elapsed_time": "2:37:31", "remaining_time": "5:54:40"} +{"current_steps": 780, "total_steps": 2520, "loss": 0.2328, "lr": 7.868024749824745e-05, "epoch": 2.4761904761904763, "percentage": 30.95, "elapsed_time": "2:38:39", "remaining_time": "5:53:56"} +{"current_steps": 785, "total_steps": 2520, "loss": 0.2377, "lr": 7.842306681322522e-05, "epoch": 2.492063492063492, "percentage": 31.15, "elapsed_time": "2:39:32", "remaining_time": "5:52:36"} +{"current_steps": 790, "total_steps": 2520, "loss": 0.2401, "lr": 7.816477029160582e-05, "epoch": 2.507936507936508, "percentage": 31.35, "elapsed_time": "2:40:36", "remaining_time": "5:51:42"} +{"current_steps": 795, "total_steps": 2520, "loss": 0.2321, "lr": 7.790536807362896e-05, "epoch": 2.5238095238095237, "percentage": 31.55, "elapsed_time": "2:41:38", "remaining_time": "5:50:44"} +{"current_steps": 800, "total_steps": 2520, "loss": 0.2363, "lr": 7.7644870342942e-05, "epoch": 2.5396825396825395, "percentage": 31.75, "elapsed_time": "2:42:33", "remaining_time": "5:49:30"} +{"current_steps": 805, "total_steps": 2520, "loss": 0.2393, "lr": 7.738328732620001e-05, "epoch": 2.5555555555555554, "percentage": 31.94, "elapsed_time": "2:48:20", "remaining_time": "5:58:37"} +{"current_steps": 810, "total_steps": 2520, "loss": 0.2455, "lr": 7.712062929266444e-05, "epoch": 2.571428571428571, "percentage": 32.14, "elapsed_time": "2:49:20", "remaining_time": "5:57:29"} +{"current_steps": 815, "total_steps": 2520, "loss": 0.2324, "lr": 7.685690655379984e-05, "epoch": 2.5873015873015874, "percentage": 32.34, "elapsed_time": "2:50:18", "remaining_time": "5:56:17"} +{"current_steps": 820, "total_steps": 2520, "loss": 0.2429, "lr": 7.659212946286912e-05, "epoch": 2.6031746031746033, "percentage": 32.54, "elapsed_time": "2:51:21", "remaining_time": "5:55:15"} +{"current_steps": 825, "total_steps": 2520, "loss": 0.2598, "lr": 7.632630841452709e-05, "epoch": 2.619047619047619, "percentage": 32.74, "elapsed_time": "2:52:24", "remaining_time": "5:54:13"} +{"current_steps": 830, "total_steps": 2520, "loss": 0.2586, "lr": 7.605945384441238e-05, "epoch": 2.634920634920635, "percentage": 32.94, "elapsed_time": "2:53:26", "remaining_time": "5:53:09"} +{"current_steps": 835, "total_steps": 2520, "loss": 0.262, "lr": 7.579157622873779e-05, "epoch": 2.6507936507936507, "percentage": 33.13, "elapsed_time": "2:54:20", "remaining_time": "5:51:49"} +{"current_steps": 840, "total_steps": 2520, "loss": 0.2446, "lr": 7.552268608387889e-05, "epoch": 2.6666666666666665, "percentage": 33.33, "elapsed_time": "2:55:16", "remaining_time": "5:50:32"} +{"current_steps": 845, "total_steps": 2520, "loss": 0.2279, "lr": 7.525279396596137e-05, "epoch": 2.682539682539683, "percentage": 33.53, "elapsed_time": "2:56:21", "remaining_time": "5:49:34"} +{"current_steps": 850, "total_steps": 2520, "loss": 0.2368, "lr": 7.498191047044641e-05, "epoch": 2.6984126984126986, "percentage": 33.73, "elapsed_time": "2:57:22", "remaining_time": "5:48:29"} +{"current_steps": 855, "total_steps": 2520, "loss": 0.2369, "lr": 7.471004623171493e-05, "epoch": 2.7142857142857144, "percentage": 33.93, "elapsed_time": "2:58:25", "remaining_time": "5:47:27"} +{"current_steps": 860, "total_steps": 2520, "loss": 0.2422, "lr": 7.443721192264991e-05, "epoch": 2.7301587301587302, "percentage": 34.13, "elapsed_time": "2:59:29", "remaining_time": "5:46:27"} +{"current_steps": 865, "total_steps": 2520, "loss": 0.2565, "lr": 7.416341825421754e-05, "epoch": 2.746031746031746, "percentage": 34.33, "elapsed_time": "3:00:28", "remaining_time": "5:45:18"} +{"current_steps": 870, "total_steps": 2520, "loss": 0.2452, "lr": 7.388867597504664e-05, "epoch": 2.761904761904762, "percentage": 34.52, "elapsed_time": "3:01:30", "remaining_time": "5:44:14"} +{"current_steps": 875, "total_steps": 2520, "loss": 0.2451, "lr": 7.361299587100671e-05, "epoch": 2.7777777777777777, "percentage": 34.72, "elapsed_time": "3:02:30", "remaining_time": "5:43:07"} +{"current_steps": 880, "total_steps": 2520, "loss": 0.252, "lr": 7.333638876478453e-05, "epoch": 2.7936507936507935, "percentage": 34.92, "elapsed_time": "3:03:30", "remaining_time": "5:42:00"} +{"current_steps": 885, "total_steps": 2520, "loss": 0.238, "lr": 7.305886551545926e-05, "epoch": 2.8095238095238093, "percentage": 35.12, "elapsed_time": "3:04:28", "remaining_time": "5:40:49"} +{"current_steps": 890, "total_steps": 2520, "loss": 0.24, "lr": 7.27804370180761e-05, "epoch": 2.825396825396825, "percentage": 35.32, "elapsed_time": "3:05:30", "remaining_time": "5:39:44"} +{"current_steps": 895, "total_steps": 2520, "loss": 0.2462, "lr": 7.250111420321863e-05, "epoch": 2.8412698412698414, "percentage": 35.52, "elapsed_time": "3:06:26", "remaining_time": "5:38:30"} +{"current_steps": 900, "total_steps": 2520, "loss": 0.2519, "lr": 7.222090803657965e-05, "epoch": 2.857142857142857, "percentage": 35.71, "elapsed_time": "3:07:23", "remaining_time": "5:37:18"} +{"current_steps": 905, "total_steps": 2520, "loss": 0.2562, "lr": 7.193982951853072e-05, "epoch": 2.873015873015873, "percentage": 35.91, "elapsed_time": "3:11:02", "remaining_time": "5:40:54"} +{"current_steps": 910, "total_steps": 2520, "loss": 0.263, "lr": 7.165788968369027e-05, "epoch": 2.888888888888889, "percentage": 36.11, "elapsed_time": "3:12:04", "remaining_time": "5:39:49"} +{"current_steps": 915, "total_steps": 2520, "loss": 0.2613, "lr": 7.137509960049043e-05, "epoch": 2.9047619047619047, "percentage": 36.31, "elapsed_time": "3:13:03", "remaining_time": "5:38:38"} +{"current_steps": 920, "total_steps": 2520, "loss": 0.2456, "lr": 7.109147037074249e-05, "epoch": 2.9206349206349205, "percentage": 36.51, "elapsed_time": "3:13:58", "remaining_time": "5:37:21"} +{"current_steps": 925, "total_steps": 2520, "loss": 0.2474, "lr": 7.080701312920106e-05, "epoch": 2.9365079365079367, "percentage": 36.71, "elapsed_time": "3:15:01", "remaining_time": "5:36:17"} +{"current_steps": 930, "total_steps": 2520, "loss": 0.2417, "lr": 7.052173904312699e-05, "epoch": 2.9523809523809526, "percentage": 36.9, "elapsed_time": "3:16:00", "remaining_time": "5:35:07"} +{"current_steps": 935, "total_steps": 2520, "loss": 0.2409, "lr": 7.023565931184888e-05, "epoch": 2.9682539682539684, "percentage": 37.1, "elapsed_time": "3:17:00", "remaining_time": "5:33:58"} +{"current_steps": 940, "total_steps": 2520, "loss": 0.2392, "lr": 6.994878516632347e-05, "epoch": 2.984126984126984, "percentage": 37.3, "elapsed_time": "3:17:58", "remaining_time": "5:32:45"} +{"current_steps": 945, "total_steps": 2520, "loss": 0.2302, "lr": 6.966112786869471e-05, "epoch": 3.0, "percentage": 37.5, "elapsed_time": "3:18:58", "remaining_time": "5:31:38"} +{"current_steps": 950, "total_steps": 2520, "loss": 0.1329, "lr": 6.937269871185171e-05, "epoch": 3.015873015873016, "percentage": 37.7, "elapsed_time": "3:20:01", "remaining_time": "5:30:34"} +{"current_steps": 955, "total_steps": 2520, "loss": 0.1295, "lr": 6.908350901898522e-05, "epoch": 3.0317460317460316, "percentage": 37.9, "elapsed_time": "3:21:05", "remaining_time": "5:29:31"} +{"current_steps": 960, "total_steps": 2520, "loss": 0.1263, "lr": 6.87935701431433e-05, "epoch": 3.0476190476190474, "percentage": 38.1, "elapsed_time": "3:22:03", "remaining_time": "5:28:20"} +{"current_steps": 965, "total_steps": 2520, "loss": 0.123, "lr": 6.850289346678552e-05, "epoch": 3.0634920634920633, "percentage": 38.29, "elapsed_time": "3:23:06", "remaining_time": "5:27:16"} +{"current_steps": 970, "total_steps": 2520, "loss": 0.1207, "lr": 6.821149040133608e-05, "epoch": 3.0793650793650795, "percentage": 38.49, "elapsed_time": "3:23:59", "remaining_time": "5:25:57"} +{"current_steps": 975, "total_steps": 2520, "loss": 0.1311, "lr": 6.791937238673592e-05, "epoch": 3.0952380952380953, "percentage": 38.69, "elapsed_time": "3:24:55", "remaining_time": "5:24:44"} +{"current_steps": 980, "total_steps": 2520, "loss": 0.126, "lr": 6.762655089099353e-05, "epoch": 3.111111111111111, "percentage": 38.89, "elapsed_time": "3:25:51", "remaining_time": "5:23:30"} +{"current_steps": 985, "total_steps": 2520, "loss": 0.1395, "lr": 6.733303740973476e-05, "epoch": 3.126984126984127, "percentage": 39.09, "elapsed_time": "3:26:53", "remaining_time": "5:22:24"} +{"current_steps": 990, "total_steps": 2520, "loss": 0.141, "lr": 6.703884346575147e-05, "epoch": 3.142857142857143, "percentage": 39.29, "elapsed_time": "3:27:55", "remaining_time": "5:21:19"} +{"current_steps": 995, "total_steps": 2520, "loss": 0.1329, "lr": 6.674398060854931e-05, "epoch": 3.1587301587301586, "percentage": 39.48, "elapsed_time": "3:28:56", "remaining_time": "5:20:14"} +{"current_steps": 1000, "total_steps": 2520, "loss": 0.1301, "lr": 6.644846041389414e-05, "epoch": 3.1746031746031744, "percentage": 39.68, "elapsed_time": "3:29:53", "remaining_time": "5:19:01"} +{"current_steps": 1005, "total_steps": 2520, "loss": 0.1255, "lr": 6.615229448335769e-05, "epoch": 3.1904761904761907, "percentage": 39.88, "elapsed_time": "3:36:47", "remaining_time": "5:26:48"} +{"current_steps": 1010, "total_steps": 2520, "loss": 0.1283, "lr": 6.58554944438621e-05, "epoch": 3.2063492063492065, "percentage": 40.08, "elapsed_time": "3:37:50", "remaining_time": "5:25:40"} +{"current_steps": 1015, "total_steps": 2520, "loss": 0.13, "lr": 6.555807194722339e-05, "epoch": 3.2222222222222223, "percentage": 40.28, "elapsed_time": "3:38:49", "remaining_time": "5:24:27"} +{"current_steps": 1020, "total_steps": 2520, "loss": 0.1297, "lr": 6.526003866969412e-05, "epoch": 3.238095238095238, "percentage": 40.48, "elapsed_time": "3:39:48", "remaining_time": "5:23:14"} +{"current_steps": 1025, "total_steps": 2520, "loss": 0.1324, "lr": 6.4961406311505e-05, "epoch": 3.253968253968254, "percentage": 40.67, "elapsed_time": "3:40:52", "remaining_time": "5:22:08"} +{"current_steps": 1030, "total_steps": 2520, "loss": 0.1397, "lr": 6.466218659640545e-05, "epoch": 3.2698412698412698, "percentage": 40.87, "elapsed_time": "3:41:49", "remaining_time": "5:20:52"} +{"current_steps": 1035, "total_steps": 2520, "loss": 0.1309, "lr": 6.436239127120351e-05, "epoch": 3.2857142857142856, "percentage": 41.07, "elapsed_time": "3:42:48", "remaining_time": "5:19:41"} +{"current_steps": 1040, "total_steps": 2520, "loss": 0.1356, "lr": 6.406203210530455e-05, "epoch": 3.3015873015873014, "percentage": 41.27, "elapsed_time": "3:43:47", "remaining_time": "5:18:28"} +{"current_steps": 1045, "total_steps": 2520, "loss": 0.1362, "lr": 6.376112089024928e-05, "epoch": 3.317460317460317, "percentage": 41.47, "elapsed_time": "3:44:47", "remaining_time": "5:17:16"} +{"current_steps": 1050, "total_steps": 2520, "loss": 0.1413, "lr": 6.345966943925085e-05, "epoch": 3.3333333333333335, "percentage": 41.67, "elapsed_time": "3:45:48", "remaining_time": "5:16:07"} +{"current_steps": 1055, "total_steps": 2520, "loss": 0.1285, "lr": 6.315768958673103e-05, "epoch": 3.3492063492063493, "percentage": 41.87, "elapsed_time": "3:46:54", "remaining_time": "5:15:05"} +{"current_steps": 1060, "total_steps": 2520, "loss": 0.1284, "lr": 6.285519318785568e-05, "epoch": 3.365079365079365, "percentage": 42.06, "elapsed_time": "3:47:53", "remaining_time": "5:13:53"} +{"current_steps": 1065, "total_steps": 2520, "loss": 0.133, "lr": 6.25521921180693e-05, "epoch": 3.380952380952381, "percentage": 42.26, "elapsed_time": "3:48:55", "remaining_time": "5:12:45"} +{"current_steps": 1070, "total_steps": 2520, "loss": 0.1358, "lr": 6.224869827262885e-05, "epoch": 3.3968253968253967, "percentage": 42.46, "elapsed_time": "3:49:58", "remaining_time": "5:11:39"} +{"current_steps": 1075, "total_steps": 2520, "loss": 0.1465, "lr": 6.194472356613667e-05, "epoch": 3.4126984126984126, "percentage": 42.66, "elapsed_time": "3:51:01", "remaining_time": "5:10:32"} +{"current_steps": 1080, "total_steps": 2520, "loss": 0.1364, "lr": 6.16402799320729e-05, "epoch": 3.4285714285714284, "percentage": 42.86, "elapsed_time": "3:52:06", "remaining_time": "5:09:28"} +{"current_steps": 1085, "total_steps": 2520, "loss": 0.1353, "lr": 6.133537932232684e-05, "epoch": 3.4444444444444446, "percentage": 43.06, "elapsed_time": "3:53:02", "remaining_time": "5:08:13"} +{"current_steps": 1090, "total_steps": 2520, "loss": 0.1376, "lr": 6.1030033706727815e-05, "epoch": 3.4603174603174605, "percentage": 43.25, "elapsed_time": "3:54:00", "remaining_time": "5:06:59"} +{"current_steps": 1095, "total_steps": 2520, "loss": 0.134, "lr": 6.0724255072575275e-05, "epoch": 3.4761904761904763, "percentage": 43.45, "elapsed_time": "3:55:01", "remaining_time": "5:05:51"} +{"current_steps": 1100, "total_steps": 2520, "loss": 0.1382, "lr": 6.0418055424168154e-05, "epoch": 3.492063492063492, "percentage": 43.65, "elapsed_time": "3:55:59", "remaining_time": "5:04:38"} +{"current_steps": 1105, "total_steps": 2520, "loss": 0.1342, "lr": 6.011144678233359e-05, "epoch": 3.507936507936508, "percentage": 43.85, "elapsed_time": "4:02:24", "remaining_time": "5:10:24"} +{"current_steps": 1110, "total_steps": 2520, "loss": 0.1371, "lr": 5.9804441183955104e-05, "epoch": 3.5238095238095237, "percentage": 44.05, "elapsed_time": "4:03:27", "remaining_time": "5:09:15"} +{"current_steps": 1115, "total_steps": 2520, "loss": 0.1303, "lr": 5.9497050681499955e-05, "epoch": 3.5396825396825395, "percentage": 44.25, "elapsed_time": "4:04:29", "remaining_time": "5:08:04"} +{"current_steps": 1120, "total_steps": 2520, "loss": 0.1328, "lr": 5.9189287342545996e-05, "epoch": 3.5555555555555554, "percentage": 44.44, "elapsed_time": "4:05:23", "remaining_time": "5:06:44"} +{"current_steps": 1125, "total_steps": 2520, "loss": 0.1414, "lr": 5.888116324930798e-05, "epoch": 3.571428571428571, "percentage": 44.64, "elapsed_time": "4:06:25", "remaining_time": "5:05:34"} +{"current_steps": 1130, "total_steps": 2520, "loss": 0.1381, "lr": 5.8572690498163205e-05, "epoch": 3.5873015873015874, "percentage": 44.84, "elapsed_time": "4:07:24", "remaining_time": "5:04:20"} +{"current_steps": 1135, "total_steps": 2520, "loss": 0.1384, "lr": 5.826388119917658e-05, "epoch": 3.6031746031746033, "percentage": 45.04, "elapsed_time": "4:08:16", "remaining_time": "5:02:57"} +{"current_steps": 1140, "total_steps": 2520, "loss": 0.1305, "lr": 5.795474747562533e-05, "epoch": 3.619047619047619, "percentage": 45.24, "elapsed_time": "4:09:20", "remaining_time": "5:01:49"} +{"current_steps": 1145, "total_steps": 2520, "loss": 0.136, "lr": 5.7645301463522895e-05, "epoch": 3.634920634920635, "percentage": 45.44, "elapsed_time": "4:10:22", "remaining_time": "5:00:40"} +{"current_steps": 1150, "total_steps": 2520, "loss": 0.1355, "lr": 5.7335555311142677e-05, "epoch": 3.6507936507936507, "percentage": 45.63, "elapsed_time": "4:11:23", "remaining_time": "4:59:28"} +{"current_steps": 1155, "total_steps": 2520, "loss": 0.1348, "lr": 5.702552117854093e-05, "epoch": 3.6666666666666665, "percentage": 45.83, "elapsed_time": "4:12:20", "remaining_time": "4:58:12"} +{"current_steps": 1160, "total_steps": 2520, "loss": 0.1312, "lr": 5.671521123707955e-05, "epoch": 3.682539682539683, "percentage": 46.03, "elapsed_time": "4:13:17", "remaining_time": "4:56:57"} +{"current_steps": 1165, "total_steps": 2520, "loss": 0.1402, "lr": 5.640463766894813e-05, "epoch": 3.6984126984126986, "percentage": 46.23, "elapsed_time": "4:14:16", "remaining_time": "4:55:44"} +{"current_steps": 1170, "total_steps": 2520, "loss": 0.1351, "lr": 5.609381266668578e-05, "epoch": 3.7142857142857144, "percentage": 46.43, "elapsed_time": "4:15:21", "remaining_time": "4:54:39"} +{"current_steps": 1175, "total_steps": 2520, "loss": 0.1408, "lr": 5.5782748432702426e-05, "epoch": 3.7301587301587302, "percentage": 46.63, "elapsed_time": "4:16:23", "remaining_time": "4:53:28"} +{"current_steps": 1180, "total_steps": 2520, "loss": 0.1413, "lr": 5.54714571787998e-05, "epoch": 3.746031746031746, "percentage": 46.83, "elapsed_time": "4:17:24", "remaining_time": "4:52:19"} +{"current_steps": 1185, "total_steps": 2520, "loss": 0.1337, "lr": 5.5159951125692e-05, "epoch": 3.761904761904762, "percentage": 47.02, "elapsed_time": "4:18:17", "remaining_time": "4:50:59"} +{"current_steps": 1190, "total_steps": 2520, "loss": 0.1473, "lr": 5.484824250252574e-05, "epoch": 3.7777777777777777, "percentage": 47.22, "elapsed_time": "4:19:18", "remaining_time": "4:49:49"} +{"current_steps": 1195, "total_steps": 2520, "loss": 0.1431, "lr": 5.453634354640028e-05, "epoch": 3.7936507936507935, "percentage": 47.42, "elapsed_time": "4:20:14", "remaining_time": "4:48:33"} +{"current_steps": 1200, "total_steps": 2520, "loss": 0.1432, "lr": 5.422426650188698e-05, "epoch": 3.8095238095238093, "percentage": 47.62, "elapsed_time": "4:21:12", "remaining_time": "4:47:20"} +{"current_steps": 1205, "total_steps": 2520, "loss": 0.1369, "lr": 5.391202362054859e-05, "epoch": 3.825396825396825, "percentage": 47.82, "elapsed_time": "4:24:45", "remaining_time": "4:48:55"} +{"current_steps": 1210, "total_steps": 2520, "loss": 0.137, "lr": 5.359962716045835e-05, "epoch": 3.8412698412698414, "percentage": 48.02, "elapsed_time": "4:25:48", "remaining_time": "4:47:46"} +{"current_steps": 1215, "total_steps": 2520, "loss": 0.1536, "lr": 5.328708938571872e-05, "epoch": 3.857142857142857, "percentage": 48.21, "elapsed_time": "4:26:46", "remaining_time": "4:46:32"} +{"current_steps": 1220, "total_steps": 2520, "loss": 0.1391, "lr": 5.2974422565979866e-05, "epoch": 3.873015873015873, "percentage": 48.41, "elapsed_time": "4:27:47", "remaining_time": "4:45:21"} +{"current_steps": 1225, "total_steps": 2520, "loss": 0.1364, "lr": 5.266163897595804e-05, "epoch": 3.888888888888889, "percentage": 48.61, "elapsed_time": "4:28:42", "remaining_time": "4:44:03"} +{"current_steps": 1230, "total_steps": 2520, "loss": 0.1401, "lr": 5.234875089495368e-05, "epoch": 3.9047619047619047, "percentage": 48.81, "elapsed_time": "4:29:45", "remaining_time": "4:42:54"} +{"current_steps": 1235, "total_steps": 2520, "loss": 0.139, "lr": 5.203577060636935e-05, "epoch": 3.9206349206349205, "percentage": 49.01, "elapsed_time": "4:30:43", "remaining_time": "4:41:40"} +{"current_steps": 1240, "total_steps": 2520, "loss": 0.1283, "lr": 5.172271039722749e-05, "epoch": 3.9365079365079367, "percentage": 49.21, "elapsed_time": "4:31:45", "remaining_time": "4:40:31"} +{"current_steps": 1245, "total_steps": 2520, "loss": 0.134, "lr": 5.140958255768812e-05, "epoch": 3.9523809523809526, "percentage": 49.4, "elapsed_time": "4:32:40", "remaining_time": "4:39:14"} +{"current_steps": 1250, "total_steps": 2520, "loss": 0.1367, "lr": 5.109639938056625e-05, "epoch": 3.9682539682539684, "percentage": 49.6, "elapsed_time": "4:33:42", "remaining_time": "4:38:05"} +{"current_steps": 1255, "total_steps": 2520, "loss": 0.1324, "lr": 5.078317316084935e-05, "epoch": 3.984126984126984, "percentage": 49.8, "elapsed_time": "4:34:43", "remaining_time": "4:36:54"} +{"current_steps": 1260, "total_steps": 2520, "loss": 0.1324, "lr": 5.0469916195214694e-05, "epoch": 4.0, "percentage": 50.0, "elapsed_time": "4:35:46", "remaining_time": "4:35:46"} +{"current_steps": 1265, "total_steps": 2520, "loss": 0.0613, "lr": 5.015664078154655e-05, "epoch": 4.015873015873016, "percentage": 50.2, "elapsed_time": "4:36:47", "remaining_time": "4:34:36"} +{"current_steps": 1270, "total_steps": 2520, "loss": 0.0607, "lr": 4.9843359218453455e-05, "epoch": 4.031746031746032, "percentage": 50.4, "elapsed_time": "4:37:51", "remaining_time": "4:33:29"} +{"current_steps": 1275, "total_steps": 2520, "loss": 0.0681, "lr": 4.953008380478532e-05, "epoch": 4.0476190476190474, "percentage": 50.6, "elapsed_time": "4:38:52", "remaining_time": "4:32:18"} +{"current_steps": 1280, "total_steps": 2520, "loss": 0.0605, "lr": 4.921682683915066e-05, "epoch": 4.063492063492063, "percentage": 50.79, "elapsed_time": "4:39:56", "remaining_time": "4:31:12"} +{"current_steps": 1285, "total_steps": 2520, "loss": 0.0651, "lr": 4.8903600619433775e-05, "epoch": 4.079365079365079, "percentage": 50.99, "elapsed_time": "4:40:57", "remaining_time": "4:30:01"} +{"current_steps": 1290, "total_steps": 2520, "loss": 0.0636, "lr": 4.85904174423119e-05, "epoch": 4.095238095238095, "percentage": 51.19, "elapsed_time": "4:41:59", "remaining_time": "4:28:52"} +{"current_steps": 1295, "total_steps": 2520, "loss": 0.058, "lr": 4.8277289602772514e-05, "epoch": 4.111111111111111, "percentage": 51.39, "elapsed_time": "4:42:58", "remaining_time": "4:27:40"} +{"current_steps": 1300, "total_steps": 2520, "loss": 0.0675, "lr": 4.796422939363066e-05, "epoch": 4.1269841269841265, "percentage": 51.59, "elapsed_time": "4:44:00", "remaining_time": "4:26:32"} +{"current_steps": 1305, "total_steps": 2520, "loss": 0.0687, "lr": 4.7651249105046325e-05, "epoch": 4.142857142857143, "percentage": 51.79, "elapsed_time": "4:48:48", "remaining_time": "4:28:53"} +{"current_steps": 1310, "total_steps": 2520, "loss": 0.0606, "lr": 4.733836102404197e-05, "epoch": 4.158730158730159, "percentage": 51.98, "elapsed_time": "4:49:46", "remaining_time": "4:27:39"} +{"current_steps": 1315, "total_steps": 2520, "loss": 0.0716, "lr": 4.7025577434020146e-05, "epoch": 4.174603174603175, "percentage": 52.18, "elapsed_time": "4:50:39", "remaining_time": "4:26:20"} +{"current_steps": 1320, "total_steps": 2520, "loss": 0.0694, "lr": 4.671291061428129e-05, "epoch": 4.190476190476191, "percentage": 52.38, "elapsed_time": "4:51:39", "remaining_time": "4:25:08"} +{"current_steps": 1325, "total_steps": 2520, "loss": 0.0688, "lr": 4.640037283954165e-05, "epoch": 4.2063492063492065, "percentage": 52.58, "elapsed_time": "4:52:43", "remaining_time": "4:24:00"} +{"current_steps": 1330, "total_steps": 2520, "loss": 0.065, "lr": 4.608797637945142e-05, "epoch": 4.222222222222222, "percentage": 52.78, "elapsed_time": "4:53:46", "remaining_time": "4:22:50"} +{"current_steps": 1335, "total_steps": 2520, "loss": 0.0662, "lr": 4.577573349811304e-05, "epoch": 4.238095238095238, "percentage": 52.98, "elapsed_time": "4:54:49", "remaining_time": "4:21:41"} +{"current_steps": 1340, "total_steps": 2520, "loss": 0.074, "lr": 4.5463656453599726e-05, "epoch": 4.253968253968254, "percentage": 53.17, "elapsed_time": "4:55:39", "remaining_time": "4:20:20"} +{"current_steps": 1345, "total_steps": 2520, "loss": 0.063, "lr": 4.515175749747426e-05, "epoch": 4.26984126984127, "percentage": 53.37, "elapsed_time": "4:56:39", "remaining_time": "4:19:09"} +{"current_steps": 1350, "total_steps": 2520, "loss": 0.0657, "lr": 4.484004887430803e-05, "epoch": 4.285714285714286, "percentage": 53.57, "elapsed_time": "4:57:40", "remaining_time": "4:17:58"} +{"current_steps": 1355, "total_steps": 2520, "loss": 0.0673, "lr": 4.452854282120022e-05, "epoch": 4.301587301587301, "percentage": 53.77, "elapsed_time": "4:58:41", "remaining_time": "4:16:48"} +{"current_steps": 1360, "total_steps": 2520, "loss": 0.0655, "lr": 4.4217251567297586e-05, "epoch": 4.317460317460317, "percentage": 53.97, "elapsed_time": "4:59:47", "remaining_time": "4:15:41"} +{"current_steps": 1365, "total_steps": 2520, "loss": 0.0635, "lr": 4.390618733331423e-05, "epoch": 4.333333333333333, "percentage": 54.17, "elapsed_time": "5:00:46", "remaining_time": "4:14:29"} +{"current_steps": 1370, "total_steps": 2520, "loss": 0.0648, "lr": 4.359536233105187e-05, "epoch": 4.349206349206349, "percentage": 54.37, "elapsed_time": "5:01:48", "remaining_time": "4:13:20"} +{"current_steps": 1375, "total_steps": 2520, "loss": 0.0681, "lr": 4.328478876292045e-05, "epoch": 4.365079365079365, "percentage": 54.56, "elapsed_time": "5:02:43", "remaining_time": "4:12:04"} +{"current_steps": 1380, "total_steps": 2520, "loss": 0.0675, "lr": 4.297447882145907e-05, "epoch": 4.380952380952381, "percentage": 54.76, "elapsed_time": "5:03:44", "remaining_time": "4:10:54"} +{"current_steps": 1385, "total_steps": 2520, "loss": 0.0696, "lr": 4.266444468885735e-05, "epoch": 4.396825396825397, "percentage": 54.96, "elapsed_time": "5:04:45", "remaining_time": "4:09:44"} +{"current_steps": 1390, "total_steps": 2520, "loss": 0.0664, "lr": 4.235469853647711e-05, "epoch": 4.412698412698413, "percentage": 55.16, "elapsed_time": "5:05:44", "remaining_time": "4:08:33"} +{"current_steps": 1395, "total_steps": 2520, "loss": 0.0742, "lr": 4.204525252437469e-05, "epoch": 4.428571428571429, "percentage": 55.36, "elapsed_time": "5:06:45", "remaining_time": "4:07:23"} +{"current_steps": 1400, "total_steps": 2520, "loss": 0.0663, "lr": 4.173611880082342e-05, "epoch": 4.444444444444445, "percentage": 55.56, "elapsed_time": "5:07:45", "remaining_time": "4:06:12"} +{"current_steps": 1405, "total_steps": 2520, "loss": 0.0634, "lr": 4.1427309501836806e-05, "epoch": 4.4603174603174605, "percentage": 55.75, "elapsed_time": "5:13:20", "remaining_time": "4:08:40"} +{"current_steps": 1410, "total_steps": 2520, "loss": 0.065, "lr": 4.111883675069202e-05, "epoch": 4.476190476190476, "percentage": 55.95, "elapsed_time": "5:14:24", "remaining_time": "4:07:30"} +{"current_steps": 1415, "total_steps": 2520, "loss": 0.0714, "lr": 4.081071265745402e-05, "epoch": 4.492063492063492, "percentage": 56.15, "elapsed_time": "5:15:22", "remaining_time": "4:06:16"} +{"current_steps": 1420, "total_steps": 2520, "loss": 0.0679, "lr": 4.0502949318500064e-05, "epoch": 4.507936507936508, "percentage": 56.35, "elapsed_time": "5:16:20", "remaining_time": "4:05:03"} +{"current_steps": 1425, "total_steps": 2520, "loss": 0.0689, "lr": 4.01955588160449e-05, "epoch": 4.523809523809524, "percentage": 56.55, "elapsed_time": "5:17:18", "remaining_time": "4:03:49"} +{"current_steps": 1430, "total_steps": 2520, "loss": 0.0618, "lr": 3.9888553217666415e-05, "epoch": 4.5396825396825395, "percentage": 56.75, "elapsed_time": "5:18:19", "remaining_time": "4:02:38"} +{"current_steps": 1435, "total_steps": 2520, "loss": 0.0683, "lr": 3.9581944575831864e-05, "epoch": 4.555555555555555, "percentage": 56.94, "elapsed_time": "5:19:23", "remaining_time": "4:01:29"} +{"current_steps": 1440, "total_steps": 2520, "loss": 0.0679, "lr": 3.927574492742473e-05, "epoch": 4.571428571428571, "percentage": 57.14, "elapsed_time": "5:20:23", "remaining_time": "4:00:17"} +{"current_steps": 1445, "total_steps": 2520, "loss": 0.0641, "lr": 3.896996629327219e-05, "epoch": 4.587301587301587, "percentage": 57.34, "elapsed_time": "5:21:24", "remaining_time": "3:59:06"} +{"current_steps": 1450, "total_steps": 2520, "loss": 0.0734, "lr": 3.8664620677673186e-05, "epoch": 4.603174603174603, "percentage": 57.54, "elapsed_time": "5:22:23", "remaining_time": "3:57:54"} +{"current_steps": 1455, "total_steps": 2520, "loss": 0.0688, "lr": 3.8359720067927115e-05, "epoch": 4.619047619047619, "percentage": 57.74, "elapsed_time": "5:23:21", "remaining_time": "3:56:40"} +{"current_steps": 1460, "total_steps": 2520, "loss": 0.062, "lr": 3.805527643386334e-05, "epoch": 4.634920634920634, "percentage": 57.94, "elapsed_time": "5:24:22", "remaining_time": "3:55:30"} +{"current_steps": 1465, "total_steps": 2520, "loss": 0.0672, "lr": 3.775130172737117e-05, "epoch": 4.650793650793651, "percentage": 58.13, "elapsed_time": "5:25:24", "remaining_time": "3:54:19"} +{"current_steps": 1470, "total_steps": 2520, "loss": 0.0614, "lr": 3.7447807881930694e-05, "epoch": 4.666666666666667, "percentage": 58.33, "elapsed_time": "5:26:24", "remaining_time": "3:53:09"} +{"current_steps": 1475, "total_steps": 2520, "loss": 0.0636, "lr": 3.7144806812144324e-05, "epoch": 4.682539682539683, "percentage": 58.53, "elapsed_time": "5:27:27", "remaining_time": "3:51:59"} +{"current_steps": 1480, "total_steps": 2520, "loss": 0.0682, "lr": 3.6842310413269e-05, "epoch": 4.698412698412699, "percentage": 58.73, "elapsed_time": "5:28:31", "remaining_time": "3:50:51"} +{"current_steps": 1485, "total_steps": 2520, "loss": 0.0638, "lr": 3.654033056074918e-05, "epoch": 4.714285714285714, "percentage": 58.93, "elapsed_time": "5:29:31", "remaining_time": "3:49:40"} +{"current_steps": 1490, "total_steps": 2520, "loss": 0.0721, "lr": 3.6238879109750735e-05, "epoch": 4.73015873015873, "percentage": 59.13, "elapsed_time": "5:30:33", "remaining_time": "3:48:30"} +{"current_steps": 1495, "total_steps": 2520, "loss": 0.0668, "lr": 3.593796789469546e-05, "epoch": 4.746031746031746, "percentage": 59.33, "elapsed_time": "5:31:33", "remaining_time": "3:47:19"} +{"current_steps": 1500, "total_steps": 2520, "loss": 0.0718, "lr": 3.563760872879649e-05, "epoch": 4.761904761904762, "percentage": 59.52, "elapsed_time": "5:32:38", "remaining_time": "3:46:11"} +{"current_steps": 1505, "total_steps": 2520, "loss": 0.068, "lr": 3.5337813403594545e-05, "epoch": 4.777777777777778, "percentage": 59.72, "elapsed_time": "5:43:37", "remaining_time": "3:51:44"} +{"current_steps": 1510, "total_steps": 2520, "loss": 0.0658, "lr": 3.5038593688495005e-05, "epoch": 4.7936507936507935, "percentage": 59.92, "elapsed_time": "5:44:39", "remaining_time": "3:50:31"} +{"current_steps": 1515, "total_steps": 2520, "loss": 0.0677, "lr": 3.4739961330305894e-05, "epoch": 4.809523809523809, "percentage": 60.12, "elapsed_time": "5:45:42", "remaining_time": "3:49:19"} +{"current_steps": 1520, "total_steps": 2520, "loss": 0.0656, "lr": 3.444192805277663e-05, "epoch": 4.825396825396825, "percentage": 60.32, "elapsed_time": "5:46:42", "remaining_time": "3:48:05"} +{"current_steps": 1525, "total_steps": 2520, "loss": 0.0674, "lr": 3.414450555613792e-05, "epoch": 4.841269841269841, "percentage": 60.52, "elapsed_time": "5:47:48", "remaining_time": "3:46:56"} +{"current_steps": 1530, "total_steps": 2520, "loss": 0.0675, "lr": 3.3847705516642317e-05, "epoch": 4.857142857142857, "percentage": 60.71, "elapsed_time": "5:48:43", "remaining_time": "3:45:38"} +{"current_steps": 1535, "total_steps": 2520, "loss": 0.0645, "lr": 3.355153958610587e-05, "epoch": 4.8730158730158735, "percentage": 60.91, "elapsed_time": "5:49:48", "remaining_time": "3:44:27"} +{"current_steps": 1540, "total_steps": 2520, "loss": 0.0648, "lr": 3.325601939145069e-05, "epoch": 4.888888888888889, "percentage": 61.11, "elapsed_time": "5:50:50", "remaining_time": "3:43:15"} +{"current_steps": 1545, "total_steps": 2520, "loss": 0.0678, "lr": 3.296115653424854e-05, "epoch": 4.904761904761905, "percentage": 61.31, "elapsed_time": "5:51:53", "remaining_time": "3:42:04"} +{"current_steps": 1550, "total_steps": 2520, "loss": 0.0655, "lr": 3.266696259026526e-05, "epoch": 4.920634920634921, "percentage": 61.51, "elapsed_time": "5:52:51", "remaining_time": "3:40:49"} +{"current_steps": 1555, "total_steps": 2520, "loss": 0.0637, "lr": 3.237344910900648e-05, "epoch": 4.936507936507937, "percentage": 61.71, "elapsed_time": "5:53:52", "remaining_time": "3:39:36"} +{"current_steps": 1560, "total_steps": 2520, "loss": 0.0638, "lr": 3.208062761326408e-05, "epoch": 4.9523809523809526, "percentage": 61.9, "elapsed_time": "5:54:53", "remaining_time": "3:38:23"} +{"current_steps": 1565, "total_steps": 2520, "loss": 0.0681, "lr": 3.178850959866393e-05, "epoch": 4.968253968253968, "percentage": 62.1, "elapsed_time": "5:55:54", "remaining_time": "3:37:10"} +{"current_steps": 1570, "total_steps": 2520, "loss": 0.0673, "lr": 3.14971065332145e-05, "epoch": 4.984126984126984, "percentage": 62.3, "elapsed_time": "5:56:53", "remaining_time": "3:35:57"} +{"current_steps": 1575, "total_steps": 2520, "loss": 0.0616, "lr": 3.1206429856856706e-05, "epoch": 5.0, "percentage": 62.5, "elapsed_time": "5:57:56", "remaining_time": "3:34:46"} +{"current_steps": 1580, "total_steps": 2520, "loss": 0.0284, "lr": 3.091649098101479e-05, "epoch": 5.015873015873016, "percentage": 62.7, "elapsed_time": "5:59:00", "remaining_time": "3:33:35"} +{"current_steps": 1585, "total_steps": 2520, "loss": 0.0264, "lr": 3.06273012881483e-05, "epoch": 5.031746031746032, "percentage": 62.9, "elapsed_time": "6:00:00", "remaining_time": "3:32:22"} +{"current_steps": 1590, "total_steps": 2520, "loss": 0.0308, "lr": 3.0338872131305284e-05, "epoch": 5.0476190476190474, "percentage": 63.1, "elapsed_time": "6:01:00", "remaining_time": "3:31:09"} +{"current_steps": 1595, "total_steps": 2520, "loss": 0.0283, "lr": 3.0051214833676545e-05, "epoch": 5.063492063492063, "percentage": 63.29, "elapsed_time": "6:02:03", "remaining_time": "3:29:58"} +{"current_steps": 1600, "total_steps": 2520, "loss": 0.0291, "lr": 2.9764340688151137e-05, "epoch": 5.079365079365079, "percentage": 63.49, "elapsed_time": "6:03:05", "remaining_time": "3:28:46"} +{"current_steps": 1605, "total_steps": 2520, "loss": 0.0292, "lr": 2.9478260956873028e-05, "epoch": 5.095238095238095, "percentage": 63.69, "elapsed_time": "6:06:22", "remaining_time": "3:28:52"} +{"current_steps": 1610, "total_steps": 2520, "loss": 0.0281, "lr": 2.919298687079895e-05, "epoch": 5.111111111111111, "percentage": 63.89, "elapsed_time": "6:07:23", "remaining_time": "3:27:39"} +{"current_steps": 1615, "total_steps": 2520, "loss": 0.0279, "lr": 2.8908529629257543e-05, "epoch": 5.1269841269841265, "percentage": 64.09, "elapsed_time": "6:08:20", "remaining_time": "3:26:24"} +{"current_steps": 1620, "total_steps": 2520, "loss": 0.031, "lr": 2.8624900399509603e-05, "epoch": 5.142857142857143, "percentage": 64.29, "elapsed_time": "6:09:14", "remaining_time": "3:25:08"} +{"current_steps": 1625, "total_steps": 2520, "loss": 0.0261, "lr": 2.8342110316309745e-05, "epoch": 5.158730158730159, "percentage": 64.48, "elapsed_time": "6:10:13", "remaining_time": "3:23:54"} +{"current_steps": 1630, "total_steps": 2520, "loss": 0.029, "lr": 2.8060170481469293e-05, "epoch": 5.174603174603175, "percentage": 64.68, "elapsed_time": "6:11:17", "remaining_time": "3:22:43"} +{"current_steps": 1635, "total_steps": 2520, "loss": 0.027, "lr": 2.777909196342035e-05, "epoch": 5.190476190476191, "percentage": 64.88, "elapsed_time": "6:12:16", "remaining_time": "3:21:30"} +{"current_steps": 1640, "total_steps": 2520, "loss": 0.0309, "lr": 2.749888579678138e-05, "epoch": 5.2063492063492065, "percentage": 65.08, "elapsed_time": "6:13:18", "remaining_time": "3:20:18"} +{"current_steps": 1645, "total_steps": 2520, "loss": 0.0267, "lr": 2.721956298192392e-05, "epoch": 5.222222222222222, "percentage": 65.28, "elapsed_time": "6:14:15", "remaining_time": "3:19:04"} +{"current_steps": 1650, "total_steps": 2520, "loss": 0.0287, "lr": 2.6941134484540774e-05, "epoch": 5.238095238095238, "percentage": 65.48, "elapsed_time": "6:15:18", "remaining_time": "3:17:53"} +{"current_steps": 1655, "total_steps": 2520, "loss": 0.0276, "lr": 2.6663611235215486e-05, "epoch": 5.253968253968254, "percentage": 65.67, "elapsed_time": "6:16:15", "remaining_time": "3:16:39"} +{"current_steps": 1660, "total_steps": 2520, "loss": 0.0293, "lr": 2.6387004128993314e-05, "epoch": 5.26984126984127, "percentage": 65.87, "elapsed_time": "6:17:15", "remaining_time": "3:15:26"} +{"current_steps": 1665, "total_steps": 2520, "loss": 0.0287, "lr": 2.6111324024953378e-05, "epoch": 5.285714285714286, "percentage": 66.07, "elapsed_time": "6:18:22", "remaining_time": "3:14:17"} +{"current_steps": 1670, "total_steps": 2520, "loss": 0.0291, "lr": 2.5836581745782475e-05, "epoch": 5.301587301587301, "percentage": 66.27, "elapsed_time": "6:19:17", "remaining_time": "3:13:03"} +{"current_steps": 1675, "total_steps": 2520, "loss": 0.028, "lr": 2.556278807735008e-05, "epoch": 5.317460317460317, "percentage": 66.47, "elapsed_time": "6:20:17", "remaining_time": "3:11:51"} +{"current_steps": 1680, "total_steps": 2520, "loss": 0.0331, "lr": 2.5289953768285092e-05, "epoch": 5.333333333333333, "percentage": 66.67, "elapsed_time": "6:21:21", "remaining_time": "3:10:40"} +{"current_steps": 1685, "total_steps": 2520, "loss": 0.0303, "lr": 2.501808952955359e-05, "epoch": 5.349206349206349, "percentage": 66.87, "elapsed_time": "6:22:25", "remaining_time": "3:09:30"} +{"current_steps": 1690, "total_steps": 2520, "loss": 0.0298, "lr": 2.474720603403866e-05, "epoch": 5.365079365079365, "percentage": 67.06, "elapsed_time": "6:23:28", "remaining_time": "3:08:19"} +{"current_steps": 1695, "total_steps": 2520, "loss": 0.0266, "lr": 2.447731391612112e-05, "epoch": 5.380952380952381, "percentage": 67.26, "elapsed_time": "6:24:28", "remaining_time": "3:07:07"} +{"current_steps": 1700, "total_steps": 2520, "loss": 0.0309, "lr": 2.4208423771262238e-05, "epoch": 5.396825396825397, "percentage": 67.46, "elapsed_time": "6:25:32", "remaining_time": "3:05:58"} +{"current_steps": 1705, "total_steps": 2520, "loss": 0.0274, "lr": 2.3940546155587618e-05, "epoch": 5.412698412698413, "percentage": 67.66, "elapsed_time": "6:28:45", "remaining_time": "3:05:49"} +{"current_steps": 1710, "total_steps": 2520, "loss": 0.0285, "lr": 2.367369158547292e-05, "epoch": 5.428571428571429, "percentage": 67.86, "elapsed_time": "6:29:42", "remaining_time": "3:04:35"} +{"current_steps": 1715, "total_steps": 2520, "loss": 0.0287, "lr": 2.3407870537130898e-05, "epoch": 5.444444444444445, "percentage": 68.06, "elapsed_time": "6:30:47", "remaining_time": "3:03:26"} +{"current_steps": 1720, "total_steps": 2520, "loss": 0.0321, "lr": 2.314309344620019e-05, "epoch": 5.4603174603174605, "percentage": 68.25, "elapsed_time": "6:31:46", "remaining_time": "3:02:13"} +{"current_steps": 1725, "total_steps": 2520, "loss": 0.0301, "lr": 2.287937070733557e-05, "epoch": 5.476190476190476, "percentage": 68.45, "elapsed_time": "6:32:46", "remaining_time": "3:01:01"} +{"current_steps": 1730, "total_steps": 2520, "loss": 0.0299, "lr": 2.2616712673799994e-05, "epoch": 5.492063492063492, "percentage": 68.65, "elapsed_time": "6:33:48", "remaining_time": "2:59:49"} +{"current_steps": 1735, "total_steps": 2520, "loss": 0.0299, "lr": 2.2355129657058004e-05, "epoch": 5.507936507936508, "percentage": 68.85, "elapsed_time": "6:34:50", "remaining_time": "2:58:38"} +{"current_steps": 1740, "total_steps": 2520, "loss": 0.0267, "lr": 2.2094631926371045e-05, "epoch": 5.523809523809524, "percentage": 69.05, "elapsed_time": "6:35:51", "remaining_time": "2:57:27"} +{"current_steps": 1745, "total_steps": 2520, "loss": 0.03, "lr": 2.18352297083942e-05, "epoch": 5.5396825396825395, "percentage": 69.25, "elapsed_time": "6:36:49", "remaining_time": "2:56:14"} +{"current_steps": 1750, "total_steps": 2520, "loss": 0.0311, "lr": 2.1576933186774777e-05, "epoch": 5.555555555555555, "percentage": 69.44, "elapsed_time": "6:37:46", "remaining_time": "2:55:01"} +{"current_steps": 1755, "total_steps": 2520, "loss": 0.0293, "lr": 2.131975250175256e-05, "epoch": 5.571428571428571, "percentage": 69.64, "elapsed_time": "6:38:44", "remaining_time": "2:53:48"} +{"current_steps": 1760, "total_steps": 2520, "loss": 0.0309, "lr": 2.1063697749761603e-05, "epoch": 5.587301587301587, "percentage": 69.84, "elapsed_time": "6:39:46", "remaining_time": "2:52:37"} +{"current_steps": 1765, "total_steps": 2520, "loss": 0.0297, "lr": 2.080877898303394e-05, "epoch": 5.603174603174603, "percentage": 70.04, "elapsed_time": "6:40:51", "remaining_time": "2:51:28"} +{"current_steps": 1770, "total_steps": 2520, "loss": 0.0282, "lr": 2.0555006209204997e-05, "epoch": 5.619047619047619, "percentage": 70.24, "elapsed_time": "6:41:50", "remaining_time": "2:50:16"} +{"current_steps": 1775, "total_steps": 2520, "loss": 0.029, "lr": 2.030238939092059e-05, "epoch": 5.634920634920634, "percentage": 70.44, "elapsed_time": "6:42:52", "remaining_time": "2:49:05"} +{"current_steps": 1780, "total_steps": 2520, "loss": 0.0282, "lr": 2.0050938445445894e-05, "epoch": 5.650793650793651, "percentage": 70.63, "elapsed_time": "6:43:52", "remaining_time": "2:47:54"} +{"current_steps": 1785, "total_steps": 2520, "loss": 0.0279, "lr": 1.980066324427613e-05, "epoch": 5.666666666666667, "percentage": 70.83, "elapsed_time": "6:44:50", "remaining_time": "2:46:41"} +{"current_steps": 1790, "total_steps": 2520, "loss": 0.032, "lr": 1.9551573612748923e-05, "epoch": 5.682539682539683, "percentage": 71.03, "elapsed_time": "6:45:48", "remaining_time": "2:45:29"} +{"current_steps": 1795, "total_steps": 2520, "loss": 0.027, "lr": 1.9303679329658723e-05, "epoch": 5.698412698412699, "percentage": 71.23, "elapsed_time": "6:46:46", "remaining_time": "2:44:17"} +{"current_steps": 1800, "total_steps": 2520, "loss": 0.0282, "lr": 1.905699012687275e-05, "epoch": 5.714285714285714, "percentage": 71.43, "elapsed_time": "6:47:47", "remaining_time": "2:43:07"} +{"current_steps": 1805, "total_steps": 2520, "loss": 0.0274, "lr": 1.881151568894909e-05, "epoch": 5.73015873015873, "percentage": 71.63, "elapsed_time": "6:51:55", "remaining_time": "2:43:10"} +{"current_steps": 1810, "total_steps": 2520, "loss": 0.0292, "lr": 1.8567265652756378e-05, "epoch": 5.746031746031746, "percentage": 71.83, "elapsed_time": "6:52:56", "remaining_time": "2:41:58"} +{"current_steps": 1815, "total_steps": 2520, "loss": 0.0253, "lr": 1.8324249607095534e-05, "epoch": 5.761904761904762, "percentage": 72.02, "elapsed_time": "6:53:55", "remaining_time": "2:40:47"} +{"current_steps": 1820, "total_steps": 2520, "loss": 0.0275, "lr": 1.8082477092323297e-05, "epoch": 5.777777777777778, "percentage": 72.22, "elapsed_time": "6:54:56", "remaining_time": "2:39:35"} +{"current_steps": 1825, "total_steps": 2520, "loss": 0.028, "lr": 1.7841957599977755e-05, "epoch": 5.7936507936507935, "percentage": 72.42, "elapsed_time": "6:55:53", "remaining_time": "2:38:22"} +{"current_steps": 1830, "total_steps": 2520, "loss": 0.0274, "lr": 1.760270057240559e-05, "epoch": 5.809523809523809, "percentage": 72.62, "elapsed_time": "6:56:56", "remaining_time": "2:37:12"} +{"current_steps": 1835, "total_steps": 2520, "loss": 0.0275, "lr": 1.736471540239156e-05, "epoch": 5.825396825396825, "percentage": 72.82, "elapsed_time": "6:57:57", "remaining_time": "2:36:01"} +{"current_steps": 1840, "total_steps": 2520, "loss": 0.0299, "lr": 1.712801143278961e-05, "epoch": 5.841269841269841, "percentage": 73.02, "elapsed_time": "6:59:01", "remaining_time": "2:34:51"} +{"current_steps": 1845, "total_steps": 2520, "loss": 0.0258, "lr": 1.6892597956156148e-05, "epoch": 5.857142857142857, "percentage": 73.21, "elapsed_time": "7:00:04", "remaining_time": "2:33:41"} +{"current_steps": 1850, "total_steps": 2520, "loss": 0.0257, "lr": 1.6658484214385234e-05, "epoch": 5.8730158730158735, "percentage": 73.41, "elapsed_time": "7:01:04", "remaining_time": "2:32:30"} +{"current_steps": 1855, "total_steps": 2520, "loss": 0.0294, "lr": 1.6425679398345812e-05, "epoch": 5.888888888888889, "percentage": 73.61, "elapsed_time": "7:02:08", "remaining_time": "2:31:19"} +{"current_steps": 1860, "total_steps": 2520, "loss": 0.0267, "lr": 1.619419264752076e-05, "epoch": 5.904761904761905, "percentage": 73.81, "elapsed_time": "7:03:09", "remaining_time": "2:30:09"} +{"current_steps": 1865, "total_steps": 2520, "loss": 0.0264, "lr": 1.5964033049648262e-05, "epoch": 5.920634920634921, "percentage": 74.01, "elapsed_time": "7:04:07", "remaining_time": "2:28:57"} +{"current_steps": 1870, "total_steps": 2520, "loss": 0.03, "lr": 1.5735209640364873e-05, "epoch": 5.936507936507937, "percentage": 74.21, "elapsed_time": "7:05:03", "remaining_time": "2:27:44"} +{"current_steps": 1875, "total_steps": 2520, "loss": 0.0282, "lr": 1.5507731402850956e-05, "epoch": 5.9523809523809526, "percentage": 74.4, "elapsed_time": "7:06:05", "remaining_time": "2:26:34"} +{"current_steps": 1880, "total_steps": 2520, "loss": 0.0263, "lr": 1.528160726747783e-05, "epoch": 5.968253968253968, "percentage": 74.6, "elapsed_time": "7:07:07", "remaining_time": "2:25:24"} +{"current_steps": 1885, "total_steps": 2520, "loss": 0.0278, "lr": 1.5056846111457407e-05, "epoch": 5.984126984126984, "percentage": 74.8, "elapsed_time": "7:08:09", "remaining_time": "2:24:13"} +{"current_steps": 1890, "total_steps": 2520, "loss": 0.0249, "lr": 1.483345675849348e-05, "epoch": 6.0, "percentage": 75.0, "elapsed_time": "7:09:11", "remaining_time": "2:23:03"} +{"current_steps": 1895, "total_steps": 2520, "loss": 0.0117, "lr": 1.4611447978435478e-05, "epoch": 6.015873015873016, "percentage": 75.2, "elapsed_time": "7:10:11", "remaining_time": "2:21:53"} +{"current_steps": 1900, "total_steps": 2520, "loss": 0.0111, "lr": 1.439082848693406e-05, "epoch": 6.031746031746032, "percentage": 75.4, "elapsed_time": "7:11:10", "remaining_time": "2:20:41"} +{"current_steps": 1905, "total_steps": 2520, "loss": 0.0106, "lr": 1.4171606945099076e-05, "epoch": 6.0476190476190474, "percentage": 75.6, "elapsed_time": "7:17:48", "remaining_time": "2:21:20"} +{"current_steps": 1910, "total_steps": 2520, "loss": 0.0108, "lr": 1.3953791959159368e-05, "epoch": 6.063492063492063, "percentage": 75.79, "elapsed_time": "7:18:52", "remaining_time": "2:20:09"} +{"current_steps": 1915, "total_steps": 2520, "loss": 0.0113, "lr": 1.3737392080125134e-05, "epoch": 6.079365079365079, "percentage": 75.99, "elapsed_time": "7:19:53", "remaining_time": "2:18:58"} +{"current_steps": 1920, "total_steps": 2520, "loss": 0.0111, "lr": 1.3522415803452027e-05, "epoch": 6.095238095238095, "percentage": 76.19, "elapsed_time": "7:20:49", "remaining_time": "2:17:45"} +{"current_steps": 1925, "total_steps": 2520, "loss": 0.0117, "lr": 1.3308871568707798e-05, "epoch": 6.111111111111111, "percentage": 76.39, "elapsed_time": "7:21:50", "remaining_time": "2:16:34"} +{"current_steps": 1930, "total_steps": 2520, "loss": 0.0109, "lr": 1.3096767759240836e-05, "epoch": 6.1269841269841265, "percentage": 76.59, "elapsed_time": "7:22:53", "remaining_time": "2:15:23"} +{"current_steps": 1935, "total_steps": 2520, "loss": 0.0109, "lr": 1.2886112701851178e-05, "epoch": 6.142857142857143, "percentage": 76.79, "elapsed_time": "7:23:53", "remaining_time": "2:14:11"} +{"current_steps": 1940, "total_steps": 2520, "loss": 0.0118, "lr": 1.2676914666463508e-05, "epoch": 6.158730158730159, "percentage": 76.98, "elapsed_time": "7:24:53", "remaining_time": "2:13:00"} +{"current_steps": 1945, "total_steps": 2520, "loss": 0.0119, "lr": 1.2469181865802576e-05, "epoch": 6.174603174603175, "percentage": 77.18, "elapsed_time": "7:25:58", "remaining_time": "2:11:50"} +{"current_steps": 1950, "total_steps": 2520, "loss": 0.0119, "lr": 1.2262922455070719e-05, "epoch": 6.190476190476191, "percentage": 77.38, "elapsed_time": "7:26:55", "remaining_time": "2:10:38"} +{"current_steps": 1955, "total_steps": 2520, "loss": 0.0119, "lr": 1.2058144531627774e-05, "epoch": 6.2063492063492065, "percentage": 77.58, "elapsed_time": "7:27:56", "remaining_time": "2:09:27"} +{"current_steps": 1960, "total_steps": 2520, "loss": 0.0108, "lr": 1.1854856134673097e-05, "epoch": 6.222222222222222, "percentage": 77.78, "elapsed_time": "7:29:00", "remaining_time": "2:08:17"} +{"current_steps": 1965, "total_steps": 2520, "loss": 0.0113, "lr": 1.1653065244930083e-05, "epoch": 6.238095238095238, "percentage": 77.98, "elapsed_time": "7:29:59", "remaining_time": "2:07:05"} +{"current_steps": 1970, "total_steps": 2520, "loss": 0.0126, "lr": 1.1452779784332718e-05, "epoch": 6.253968253968254, "percentage": 78.17, "elapsed_time": "7:31:00", "remaining_time": "2:05:54"} +{"current_steps": 1975, "total_steps": 2520, "loss": 0.0106, "lr": 1.1254007615714685e-05, "epoch": 6.26984126984127, "percentage": 78.37, "elapsed_time": "7:32:01", "remaining_time": "2:04:44"} +{"current_steps": 1980, "total_steps": 2520, "loss": 0.0112, "lr": 1.1056756542500613e-05, "epoch": 6.285714285714286, "percentage": 78.57, "elapsed_time": "7:32:59", "remaining_time": "2:03:32"} +{"current_steps": 1985, "total_steps": 2520, "loss": 0.0113, "lr": 1.086103430839982e-05, "epoch": 6.301587301587301, "percentage": 78.77, "elapsed_time": "7:34:02", "remaining_time": "2:02:22"} +{"current_steps": 1990, "total_steps": 2520, "loss": 0.0136, "lr": 1.066684859710218e-05, "epoch": 6.317460317460317, "percentage": 78.97, "elapsed_time": "7:35:07", "remaining_time": "2:01:12"} +{"current_steps": 1995, "total_steps": 2520, "loss": 0.0122, "lr": 1.0474207031976618e-05, "epoch": 6.333333333333333, "percentage": 79.17, "elapsed_time": "7:36:09", "remaining_time": "2:00:02"} +{"current_steps": 2000, "total_steps": 2520, "loss": 0.0115, "lr": 1.0283117175771701e-05, "epoch": 6.349206349206349, "percentage": 79.37, "elapsed_time": "7:37:11", "remaining_time": "1:58:52"} +{"current_steps": 2005, "total_steps": 2520, "loss": 0.011, "lr": 1.0093586530318849e-05, "epoch": 6.365079365079365, "percentage": 79.56, "elapsed_time": "7:39:59", "remaining_time": "1:58:09"} +{"current_steps": 2010, "total_steps": 2520, "loss": 0.012, "lr": 9.905622536237708e-06, "epoch": 6.380952380952381, "percentage": 79.76, "elapsed_time": "7:40:59", "remaining_time": "1:56:58"} +{"current_steps": 2015, "total_steps": 2520, "loss": 0.0115, "lr": 9.719232572644187e-06, "epoch": 6.396825396825397, "percentage": 79.96, "elapsed_time": "7:42:00", "remaining_time": "1:55:47"} +{"current_steps": 2020, "total_steps": 2520, "loss": 0.0115, "lr": 9.534423956860638e-06, "epoch": 6.412698412698413, "percentage": 80.16, "elapsed_time": "7:42:58", "remaining_time": "1:54:35"} +{"current_steps": 2025, "total_steps": 2520, "loss": 0.0112, "lr": 9.351203944128694e-06, "epoch": 6.428571428571429, "percentage": 80.36, "elapsed_time": "7:43:58", "remaining_time": "1:53:24"} +{"current_steps": 2030, "total_steps": 2520, "loss": 0.0126, "lr": 9.16957972732434e-06, "epoch": 6.444444444444445, "percentage": 80.56, "elapsed_time": "7:44:56", "remaining_time": "1:52:13"} +{"current_steps": 2035, "total_steps": 2520, "loss": 0.0112, "lr": 8.989558436675643e-06, "epoch": 6.4603174603174605, "percentage": 80.75, "elapsed_time": "7:45:55", "remaining_time": "1:51:02"} +{"current_steps": 2040, "total_steps": 2520, "loss": 0.01, "lr": 8.811147139482745e-06, "epoch": 6.476190476190476, "percentage": 80.95, "elapsed_time": "7:46:58", "remaining_time": "1:49:52"} +{"current_steps": 2045, "total_steps": 2520, "loss": 0.0117, "lr": 8.634352839840459e-06, "epoch": 6.492063492063492, "percentage": 81.15, "elapsed_time": "7:47:58", "remaining_time": "1:48:41"} +{"current_steps": 2050, "total_steps": 2520, "loss": 0.0106, "lr": 8.45918247836327e-06, "epoch": 6.507936507936508, "percentage": 81.35, "elapsed_time": "7:48:56", "remaining_time": "1:47:30"} +{"current_steps": 2055, "total_steps": 2520, "loss": 0.0109, "lr": 8.285642931912918e-06, "epoch": 6.523809523809524, "percentage": 81.55, "elapsed_time": "7:49:57", "remaining_time": "1:46:20"} +{"current_steps": 2060, "total_steps": 2520, "loss": 0.0114, "lr": 8.113741013328352e-06, "epoch": 6.5396825396825395, "percentage": 81.75, "elapsed_time": "7:50:58", "remaining_time": "1:45:10"} +{"current_steps": 2065, "total_steps": 2520, "loss": 0.0131, "lr": 7.943483471158326e-06, "epoch": 6.555555555555555, "percentage": 81.94, "elapsed_time": "7:51:57", "remaining_time": "1:43:59"} +{"current_steps": 2070, "total_steps": 2520, "loss": 0.0133, "lr": 7.774876989396434e-06, "epoch": 6.571428571428571, "percentage": 82.14, "elapsed_time": "7:52:54", "remaining_time": "1:42:48"} +{"current_steps": 2075, "total_steps": 2520, "loss": 0.0106, "lr": 7.607928187218699e-06, "epoch": 6.587301587301587, "percentage": 82.34, "elapsed_time": "7:53:51", "remaining_time": "1:41:37"} +{"current_steps": 2080, "total_steps": 2520, "loss": 0.011, "lr": 7.442643618723777e-06, "epoch": 6.603174603174603, "percentage": 82.54, "elapsed_time": "7:54:54", "remaining_time": "1:40:27"} +{"current_steps": 2085, "total_steps": 2520, "loss": 0.0111, "lr": 7.2790297726755716e-06, "epoch": 6.619047619047619, "percentage": 82.74, "elapsed_time": "7:55:58", "remaining_time": "1:39:18"} +{"current_steps": 2090, "total_steps": 2520, "loss": 0.01, "lr": 7.117093072248571e-06, "epoch": 6.634920634920634, "percentage": 82.94, "elapsed_time": "7:56:58", "remaining_time": "1:38:07"} +{"current_steps": 2095, "total_steps": 2520, "loss": 0.0109, "lr": 6.9568398747756396e-06, "epoch": 6.650793650793651, "percentage": 83.13, "elapsed_time": "7:57:58", "remaining_time": "1:36:57"} +{"current_steps": 2100, "total_steps": 2520, "loss": 0.0117, "lr": 6.798276471498444e-06, "epoch": 6.666666666666667, "percentage": 83.33, "elapsed_time": "7:59:00", "remaining_time": "1:35:48"} +{"current_steps": 2105, "total_steps": 2520, "loss": 0.0109, "lr": 6.6414090873204886e-06, "epoch": 6.682539682539683, "percentage": 83.53, "elapsed_time": "8:01:59", "remaining_time": "1:35:01"} +{"current_steps": 2110, "total_steps": 2520, "loss": 0.0114, "lr": 6.486243880562759e-06, "epoch": 6.698412698412699, "percentage": 83.73, "elapsed_time": "8:02:56", "remaining_time": "1:33:50"} +{"current_steps": 2115, "total_steps": 2520, "loss": 0.0115, "lr": 6.3327869427218855e-06, "epoch": 6.714285714285714, "percentage": 83.93, "elapsed_time": "8:03:57", "remaining_time": "1:32:40"} +{"current_steps": 2120, "total_steps": 2520, "loss": 0.0113, "lr": 6.181044298231081e-06, "epoch": 6.73015873015873, "percentage": 84.13, "elapsed_time": "8:04:54", "remaining_time": "1:31:29"} +{"current_steps": 2125, "total_steps": 2520, "loss": 0.0109, "lr": 6.031021904223572e-06, "epoch": 6.746031746031746, "percentage": 84.33, "elapsed_time": "8:05:54", "remaining_time": "1:30:19"} +{"current_steps": 2130, "total_steps": 2520, "loss": 0.0118, "lr": 5.882725650298787e-06, "epoch": 6.761904761904762, "percentage": 84.52, "elapsed_time": "8:06:53", "remaining_time": "1:29:08"} +{"current_steps": 2135, "total_steps": 2520, "loss": 0.0122, "lr": 5.736161358291092e-06, "epoch": 6.777777777777778, "percentage": 84.72, "elapsed_time": "8:07:56", "remaining_time": "1:27:59"} +{"current_steps": 2140, "total_steps": 2520, "loss": 0.0102, "lr": 5.5913347820412635e-06, "epoch": 6.7936507936507935, "percentage": 84.92, "elapsed_time": "8:09:02", "remaining_time": "1:26:50"} +{"current_steps": 2145, "total_steps": 2520, "loss": 0.0112, "lr": 5.44825160717059e-06, "epoch": 6.809523809523809, "percentage": 85.12, "elapsed_time": "8:10:01", "remaining_time": "1:25:40"} +{"current_steps": 2150, "total_steps": 2520, "loss": 0.0121, "lr": 5.306917450857702e-06, "epoch": 6.825396825396825, "percentage": 85.32, "elapsed_time": "8:11:04", "remaining_time": "1:24:30"} +{"current_steps": 2155, "total_steps": 2520, "loss": 0.0106, "lr": 5.167337861617982e-06, "epoch": 6.841269841269841, "percentage": 85.52, "elapsed_time": "8:12:05", "remaining_time": "1:23:20"} +{"current_steps": 2160, "total_steps": 2520, "loss": 0.0107, "lr": 5.029518319085824e-06, "epoch": 6.857142857142857, "percentage": 85.71, "elapsed_time": "8:12:59", "remaining_time": "1:22:09"} +{"current_steps": 2165, "total_steps": 2520, "loss": 0.011, "lr": 4.893464233799433e-06, "epoch": 6.8730158730158735, "percentage": 85.91, "elapsed_time": "8:14:01", "remaining_time": "1:21:00"} +{"current_steps": 2170, "total_steps": 2520, "loss": 0.0104, "lr": 4.759180946988495e-06, "epoch": 6.888888888888889, "percentage": 86.11, "elapsed_time": "8:15:00", "remaining_time": "1:19:50"} +{"current_steps": 2175, "total_steps": 2520, "loss": 0.0105, "lr": 4.626673730364395e-06, "epoch": 6.904761904761905, "percentage": 86.31, "elapsed_time": "8:16:01", "remaining_time": "1:18:40"} +{"current_steps": 2180, "total_steps": 2520, "loss": 0.0109, "lr": 4.495947785913368e-06, "epoch": 6.920634920634921, "percentage": 86.51, "elapsed_time": "8:17:08", "remaining_time": "1:17:32"} +{"current_steps": 2185, "total_steps": 2520, "loss": 0.0107, "lr": 4.367008245692189e-06, "epoch": 6.936507936507937, "percentage": 86.71, "elapsed_time": "8:18:10", "remaining_time": "1:16:22"} +{"current_steps": 2190, "total_steps": 2520, "loss": 0.0097, "lr": 4.239860171626769e-06, "epoch": 6.9523809523809526, "percentage": 86.9, "elapsed_time": "8:19:13", "remaining_time": "1:15:13"} +{"current_steps": 2195, "total_steps": 2520, "loss": 0.0095, "lr": 4.114508555313351e-06, "epoch": 6.968253968253968, "percentage": 87.1, "elapsed_time": "8:20:13", "remaining_time": "1:14:03"} +{"current_steps": 2200, "total_steps": 2520, "loss": 0.011, "lr": 3.990958317822663e-06, "epoch": 6.984126984126984, "percentage": 87.3, "elapsed_time": "8:21:16", "remaining_time": "1:12:54"} +{"current_steps": 2205, "total_steps": 2520, "loss": 0.0094, "lr": 3.86921430950658e-06, "epoch": 7.0, "percentage": 87.5, "elapsed_time": "8:24:00", "remaining_time": "1:12:00"} +{"current_steps": 2210, "total_steps": 2520, "loss": 0.0051, "lr": 3.7492813098078506e-06, "epoch": 7.015873015873016, "percentage": 87.7, "elapsed_time": "8:25:01", "remaining_time": "1:10:50"} +{"current_steps": 2215, "total_steps": 2520, "loss": 0.0052, "lr": 3.6311640270723757e-06, "epoch": 7.031746031746032, "percentage": 87.9, "elapsed_time": "8:26:07", "remaining_time": "1:09:41"} +{"current_steps": 2220, "total_steps": 2520, "loss": 0.0048, "lr": 3.5148670983644104e-06, "epoch": 7.0476190476190474, "percentage": 88.1, "elapsed_time": "8:27:09", "remaining_time": "1:08:32"} +{"current_steps": 2225, "total_steps": 2520, "loss": 0.0049, "lr": 3.400395089284475e-06, "epoch": 7.063492063492063, "percentage": 88.29, "elapsed_time": "8:28:10", "remaining_time": "1:07:22"} +{"current_steps": 2230, "total_steps": 2520, "loss": 0.0047, "lr": 3.287752493790186e-06, "epoch": 7.079365079365079, "percentage": 88.49, "elapsed_time": "8:29:05", "remaining_time": "1:06:12"} +{"current_steps": 2235, "total_steps": 2520, "loss": 0.0047, "lr": 3.1769437340197715e-06, "epoch": 7.095238095238095, "percentage": 88.69, "elapsed_time": "8:30:06", "remaining_time": "1:05:02"} +{"current_steps": 2240, "total_steps": 2520, "loss": 0.005, "lr": 3.067973160118498e-06, "epoch": 7.111111111111111, "percentage": 88.89, "elapsed_time": "8:31:07", "remaining_time": "1:03:53"} +{"current_steps": 2245, "total_steps": 2520, "loss": 0.0049, "lr": 2.9608450500678565e-06, "epoch": 7.1269841269841265, "percentage": 89.09, "elapsed_time": "8:32:03", "remaining_time": "1:02:43"} +{"current_steps": 2250, "total_steps": 2520, "loss": 0.0043, "lr": 2.8555636095176975e-06, "epoch": 7.142857142857143, "percentage": 89.29, "elapsed_time": "8:33:05", "remaining_time": "1:01:34"} +{"current_steps": 2255, "total_steps": 2520, "loss": 0.0046, "lr": 2.7521329716210074e-06, "epoch": 7.158730158730159, "percentage": 89.48, "elapsed_time": "8:34:04", "remaining_time": "1:00:24"} +{"current_steps": 2260, "total_steps": 2520, "loss": 0.0049, "lr": 2.6505571968717725e-06, "epoch": 7.174603174603175, "percentage": 89.68, "elapsed_time": "8:35:06", "remaining_time": "0:59:15"} +{"current_steps": 2265, "total_steps": 2520, "loss": 0.005, "lr": 2.550840272945465e-06, "epoch": 7.190476190476191, "percentage": 89.88, "elapsed_time": "8:36:00", "remaining_time": "0:58:05"} +{"current_steps": 2270, "total_steps": 2520, "loss": 0.0045, "lr": 2.4529861145425605e-06, "epoch": 7.2063492063492065, "percentage": 90.08, "elapsed_time": "8:36:53", "remaining_time": "0:56:55"} +{"current_steps": 2275, "total_steps": 2520, "loss": 0.0048, "lr": 2.3569985632348247e-06, "epoch": 7.222222222222222, "percentage": 90.28, "elapsed_time": "8:37:51", "remaining_time": "0:55:46"} +{"current_steps": 2280, "total_steps": 2520, "loss": 0.0042, "lr": 2.2628813873145303e-06, "epoch": 7.238095238095238, "percentage": 90.48, "elapsed_time": "8:38:57", "remaining_time": "0:54:37"} +{"current_steps": 2285, "total_steps": 2520, "loss": 0.0048, "lr": 2.17063828164647e-06, "epoch": 7.253968253968254, "percentage": 90.67, "elapsed_time": "8:39:55", "remaining_time": "0:53:28"} +{"current_steps": 2290, "total_steps": 2520, "loss": 0.0052, "lr": 2.0802728675229587e-06, "epoch": 7.26984126984127, "percentage": 90.87, "elapsed_time": "8:40:57", "remaining_time": "0:52:19"} +{"current_steps": 2295, "total_steps": 2520, "loss": 0.0054, "lr": 1.9917886925216234e-06, "epoch": 7.285714285714286, "percentage": 91.07, "elapsed_time": "8:41:56", "remaining_time": "0:51:10"} +{"current_steps": 2300, "total_steps": 2520, "loss": 0.0049, "lr": 1.9051892303661834e-06, "epoch": 7.301587301587301, "percentage": 91.27, "elapsed_time": "8:42:56", "remaining_time": "0:50:01"} +{"current_steps": 2305, "total_steps": 2520, "loss": 0.0054, "lr": 1.8204778807900003e-06, "epoch": 7.317460317460317, "percentage": 91.47, "elapsed_time": "8:48:42", "remaining_time": "0:49:18"} +{"current_steps": 2310, "total_steps": 2520, "loss": 0.0053, "lr": 1.7376579694026896e-06, "epoch": 7.333333333333333, "percentage": 91.67, "elapsed_time": "8:49:42", "remaining_time": "0:48:09"} +{"current_steps": 2315, "total_steps": 2520, "loss": 0.0046, "lr": 1.6567327475595195e-06, "epoch": 7.349206349206349, "percentage": 91.87, "elapsed_time": "8:50:42", "remaining_time": "0:46:59"} +{"current_steps": 2320, "total_steps": 2520, "loss": 0.0044, "lr": 1.577705392233797e-06, "epoch": 7.365079365079365, "percentage": 92.06, "elapsed_time": "8:51:44", "remaining_time": "0:45:50"} +{"current_steps": 2325, "total_steps": 2520, "loss": 0.0046, "lr": 1.5005790058920943e-06, "epoch": 7.380952380952381, "percentage": 92.26, "elapsed_time": "8:52:48", "remaining_time": "0:44:41"} +{"current_steps": 2330, "total_steps": 2520, "loss": 0.0054, "lr": 1.4253566163725252e-06, "epoch": 7.396825396825397, "percentage": 92.46, "elapsed_time": "8:53:46", "remaining_time": "0:43:31"} +{"current_steps": 2335, "total_steps": 2520, "loss": 0.005, "lr": 1.3520411767658059e-06, "epoch": 7.412698412698413, "percentage": 92.66, "elapsed_time": "8:54:48", "remaining_time": "0:42:22"} +{"current_steps": 2340, "total_steps": 2520, "loss": 0.0043, "lr": 1.2806355652993762e-06, "epoch": 7.428571428571429, "percentage": 92.86, "elapsed_time": "8:55:47", "remaining_time": "0:41:12"} +{"current_steps": 2345, "total_steps": 2520, "loss": 0.0046, "lr": 1.2111425852243785e-06, "epoch": 7.444444444444445, "percentage": 93.06, "elapsed_time": "8:56:52", "remaining_time": "0:40:03"} +{"current_steps": 2350, "total_steps": 2520, "loss": 0.0052, "lr": 1.143564964705618e-06, "epoch": 7.4603174603174605, "percentage": 93.25, "elapsed_time": "8:57:46", "remaining_time": "0:38:54"} +{"current_steps": 2355, "total_steps": 2520, "loss": 0.0047, "lr": 1.0779053567144427e-06, "epoch": 7.476190476190476, "percentage": 93.45, "elapsed_time": "8:58:49", "remaining_time": "0:37:45"} +{"current_steps": 2360, "total_steps": 2520, "loss": 0.0052, "lr": 1.014166338924627e-06, "epoch": 7.492063492063492, "percentage": 93.65, "elapsed_time": "8:59:48", "remaining_time": "0:36:35"} +{"current_steps": 2365, "total_steps": 2520, "loss": 0.0045, "lr": 9.523504136111306e-07, "epoch": 7.507936507936508, "percentage": 93.85, "elapsed_time": "9:00:49", "remaining_time": "0:35:26"} +{"current_steps": 2370, "total_steps": 2520, "loss": 0.0057, "lr": 8.92460007551904e-07, "epoch": 7.523809523809524, "percentage": 94.05, "elapsed_time": "9:01:45", "remaining_time": "0:34:17"} +{"current_steps": 2375, "total_steps": 2520, "loss": 0.0051, "lr": 8.344974719326104e-07, "epoch": 7.5396825396825395, "percentage": 94.25, "elapsed_time": "9:02:43", "remaining_time": "0:33:08"} +{"current_steps": 2380, "total_steps": 2520, "loss": 0.0045, "lr": 7.784650822542871e-07, "epoch": 7.555555555555555, "percentage": 94.44, "elapsed_time": "9:03:47", "remaining_time": "0:31:59"} +{"current_steps": 2385, "total_steps": 2520, "loss": 0.0049, "lr": 7.243650382440736e-07, "epoch": 7.571428571428571, "percentage": 94.64, "elapsed_time": "9:04:43", "remaining_time": "0:30:50"} +{"current_steps": 2390, "total_steps": 2520, "loss": 0.0046, "lr": 6.721994637687967e-07, "epoch": 7.587301587301587, "percentage": 94.84, "elapsed_time": "9:05:44", "remaining_time": "0:29:41"} +{"current_steps": 2395, "total_steps": 2520, "loss": 0.0054, "lr": 6.219704067516374e-07, "epoch": 7.603174603174603, "percentage": 95.04, "elapsed_time": "9:06:50", "remaining_time": "0:28:32"} +{"current_steps": 2400, "total_steps": 2520, "loss": 0.0055, "lr": 5.736798390916898e-07, "epoch": 7.619047619047619, "percentage": 95.24, "elapsed_time": "9:07:42", "remaining_time": "0:27:23"} +{"current_steps": 2405, "total_steps": 2520, "loss": 0.005, "lr": 5.273296565865948e-07, "epoch": 7.634920634920634, "percentage": 95.44, "elapsed_time": "9:10:41", "remaining_time": "0:26:19"} +{"current_steps": 2410, "total_steps": 2520, "loss": 0.005, "lr": 4.829216788580726e-07, "epoch": 7.650793650793651, "percentage": 95.63, "elapsed_time": "9:11:33", "remaining_time": "0:25:10"} +{"current_steps": 2415, "total_steps": 2520, "loss": 0.0057, "lr": 4.404576492805179e-07, "epoch": 7.666666666666667, "percentage": 95.83, "elapsed_time": "9:12:35", "remaining_time": "0:24:01"} +{"current_steps": 2420, "total_steps": 2520, "loss": 0.0044, "lr": 3.999392349125386e-07, "epoch": 7.682539682539683, "percentage": 96.03, "elapsed_time": "9:13:36", "remaining_time": "0:22:52"} +{"current_steps": 2425, "total_steps": 2520, "loss": 0.0044, "lr": 3.613680264315189e-07, "epoch": 7.698412698412699, "percentage": 96.23, "elapsed_time": "9:14:35", "remaining_time": "0:21:43"} +{"current_steps": 2430, "total_steps": 2520, "loss": 0.0045, "lr": 3.247455380711806e-07, "epoch": 7.714285714285714, "percentage": 96.43, "elapsed_time": "9:15:35", "remaining_time": "0:20:34"} +{"current_steps": 2435, "total_steps": 2520, "loss": 0.0053, "lr": 2.900732075621082e-07, "epoch": 7.73015873015873, "percentage": 96.63, "elapsed_time": "9:16:35", "remaining_time": "0:19:25"} +{"current_steps": 2440, "total_steps": 2520, "loss": 0.0045, "lr": 2.5735239607534434e-07, "epoch": 7.746031746031746, "percentage": 96.83, "elapsed_time": "9:17:30", "remaining_time": "0:18:16"} +{"current_steps": 2445, "total_steps": 2520, "loss": 0.0047, "lr": 2.2658438816892112e-07, "epoch": 7.761904761904762, "percentage": 97.02, "elapsed_time": "9:18:27", "remaining_time": "0:17:07"} +{"current_steps": 2450, "total_steps": 2520, "loss": 0.0047, "lr": 1.9777039173746182e-07, "epoch": 7.777777777777778, "percentage": 97.22, "elapsed_time": "9:19:29", "remaining_time": "0:15:59"} +{"current_steps": 2455, "total_steps": 2520, "loss": 0.0045, "lr": 1.709115379647186e-07, "epoch": 7.7936507936507935, "percentage": 97.42, "elapsed_time": "9:20:25", "remaining_time": "0:14:50"} +{"current_steps": 2460, "total_steps": 2520, "loss": 0.0048, "lr": 1.460088812792082e-07, "epoch": 7.809523809523809, "percentage": 97.62, "elapsed_time": "9:21:17", "remaining_time": "0:13:41"} +{"current_steps": 2465, "total_steps": 2520, "loss": 0.0051, "lr": 1.2306339931279499e-07, "epoch": 7.825396825396825, "percentage": 97.82, "elapsed_time": "9:22:17", "remaining_time": "0:12:32"} +{"current_steps": 2470, "total_steps": 2520, "loss": 0.0048, "lr": 1.0207599286229941e-07, "epoch": 7.841269841269841, "percentage": 98.02, "elapsed_time": "9:23:22", "remaining_time": "0:11:24"} +{"current_steps": 2475, "total_steps": 2520, "loss": 0.0046, "lr": 8.304748585417078e-08, "epoch": 7.857142857142857, "percentage": 98.21, "elapsed_time": "9:24:21", "remaining_time": "0:10:15"} +{"current_steps": 2480, "total_steps": 2520, "loss": 0.005, "lr": 6.597862531210197e-08, "epoch": 7.8730158730158735, "percentage": 98.41, "elapsed_time": "9:25:21", "remaining_time": "0:09:07"} +{"current_steps": 2485, "total_steps": 2520, "loss": 0.0045, "lr": 5.0870081327725194e-08, "epoch": 7.888888888888889, "percentage": 98.61, "elapsed_time": "9:26:29", "remaining_time": "0:07:58"} +{"current_steps": 2490, "total_steps": 2520, "loss": 0.0048, "lr": 3.7722447034305164e-08, "epoch": 7.904761904761905, "percentage": 98.81, "elapsed_time": "9:27:31", "remaining_time": "0:06:50"} +{"current_steps": 2495, "total_steps": 2520, "loss": 0.0045, "lr": 2.653623858344667e-08, "epoch": 7.920634920634921, "percentage": 99.01, "elapsed_time": "9:28:34", "remaining_time": "0:05:41"} +{"current_steps": 2500, "total_steps": 2520, "loss": 0.0042, "lr": 1.731189512482745e-08, "epoch": 7.936507936507937, "percentage": 99.21, "elapsed_time": "9:29:38", "remaining_time": "0:04:33"} +{"current_steps": 2505, "total_steps": 2520, "loss": 0.0049, "lr": 1.0049778788967513e-08, "epoch": 7.9523809523809526, "percentage": 99.4, "elapsed_time": "9:39:40", "remaining_time": "0:03:28"} +{"current_steps": 2510, "total_steps": 2520, "loss": 0.0049, "lr": 4.750174673018304e-09, "epoch": 7.968253968253968, "percentage": 99.6, "elapsed_time": "9:40:42", "remaining_time": "0:02:18"} +{"current_steps": 2515, "total_steps": 2520, "loss": 0.0048, "lr": 1.4132908295549919e-09, "epoch": 7.984126984126984, "percentage": 99.8, "elapsed_time": "9:41:46", "remaining_time": "0:01:09"} +{"current_steps": 2520, "total_steps": 2520, "loss": 0.0044, "lr": 3.925825840522812e-11, "epoch": 8.0, "percentage": 100.0, "elapsed_time": "9:42:49", "remaining_time": "0:00:00"} +{"current_steps": 2520, "total_steps": 2520, "epoch": 8.0, "percentage": 100.0, "elapsed_time": "9:43:50", "remaining_time": "0:00:00"} diff --git a/trainer_state.json b/trainer_state.json new file mode 100644 index 0000000..1be7539 --- /dev/null +++ b/trainer_state.json @@ -0,0 +1,5591 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 8.0, + "eval_steps": 500, + "global_step": 2520, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.015873015873015872, + "grad_norm": 2.243458566621128, + "learning_rate": 3.0769230769230774e-05, + "loss": 0.9183, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.7241313457489014, + "step": 5, + "valid_targets_mean": 2840.1, + "valid_targets_min": 232 + }, + { + "epoch": 0.031746031746031744, + "grad_norm": 0.875951419213133, + "learning_rate": 6.923076923076924e-05, + "loss": 0.677, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.597373902797699, + "step": 10, + "valid_targets_mean": 2251.9, + "valid_targets_min": 194 + }, + { + "epoch": 0.047619047619047616, + "grad_norm": 0.7203240004314375, + "learning_rate": 9.99999607417416e-05, + "loss": 0.5679, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.5559351444244385, + "step": 15, + "valid_targets_mean": 2512.3, + "valid_targets_min": 223 + }, + { + "epoch": 0.06349206349206349, + "grad_norm": 0.6991343357457058, + "learning_rate": 9.999858670917045e-05, + "loss": 0.5439, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.5753902196884155, + "step": 20, + "valid_targets_mean": 2404.1, + "valid_targets_min": 124 + }, + { + "epoch": 0.07936507936507936, + "grad_norm": 0.5454838524513616, + "learning_rate": 9.999524982532699e-05, + "loss": 0.5406, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.5201640129089355, + "step": 25, + "valid_targets_mean": 2456.9, + "valid_targets_min": 494 + }, + { + "epoch": 0.09523809523809523, + "grad_norm": 0.5084862570840438, + "learning_rate": 9.998995022121104e-05, + "loss": 0.5202, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4681701064109802, + "step": 30, + "valid_targets_mean": 2273.3, + "valid_targets_min": 335 + }, + { + "epoch": 0.1111111111111111, + "grad_norm": 0.541337715607852, + "learning_rate": 9.998268810487518e-05, + "loss": 0.5274, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.530846118927002, + "step": 35, + "valid_targets_mean": 2673.7, + "valid_targets_min": 260 + }, + { + "epoch": 0.12698412698412698, + "grad_norm": 0.5493266092196112, + "learning_rate": 9.997346376141656e-05, + "loss": 0.5177, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.462390661239624, + "step": 40, + "valid_targets_mean": 2173.5, + "valid_targets_min": 209 + }, + { + "epoch": 0.14285714285714285, + "grad_norm": 0.4949153352257213, + "learning_rate": 9.99622775529657e-05, + "loss": 0.5007, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4589081108570099, + "step": 45, + "valid_targets_mean": 2497.8, + "valid_targets_min": 551 + }, + { + "epoch": 0.15873015873015872, + "grad_norm": 0.4925469902178904, + "learning_rate": 9.994912991867228e-05, + "loss": 0.4833, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4633740782737732, + "step": 50, + "valid_targets_mean": 2515.8, + "valid_targets_min": 263 + }, + { + "epoch": 0.1746031746031746, + "grad_norm": 0.4977083044642213, + "learning_rate": 9.99340213746879e-05, + "loss": 0.4872, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.472017377614975, + "step": 55, + "valid_targets_mean": 2695.8, + "valid_targets_min": 176 + }, + { + "epoch": 0.19047619047619047, + "grad_norm": 0.473046164620766, + "learning_rate": 9.991695251414583e-05, + "loss": 0.5027, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.5453957319259644, + "step": 60, + "valid_targets_mean": 2641.6, + "valid_targets_min": 191 + }, + { + "epoch": 0.20634920634920634, + "grad_norm": 0.5110503803534091, + "learning_rate": 9.989792400713771e-05, + "loss": 0.4943, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.46123188734054565, + "step": 65, + "valid_targets_mean": 2060.5, + "valid_targets_min": 172 + }, + { + "epoch": 0.2222222222222222, + "grad_norm": 0.5317141576348992, + "learning_rate": 9.987693660068722e-05, + "loss": 0.4771, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4513200521469116, + "step": 70, + "valid_targets_mean": 2187.9, + "valid_targets_min": 450 + }, + { + "epoch": 0.23809523809523808, + "grad_norm": 0.4576199397482037, + "learning_rate": 9.985399111872081e-05, + "loss": 0.4757, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.47312021255493164, + "step": 75, + "valid_targets_mean": 2376.3, + "valid_targets_min": 228 + }, + { + "epoch": 0.25396825396825395, + "grad_norm": 0.4452149596927674, + "learning_rate": 9.982908846203529e-05, + "loss": 0.4658, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.47530248761177063, + "step": 80, + "valid_targets_mean": 2791.6, + "valid_targets_min": 189 + }, + { + "epoch": 0.2698412698412698, + "grad_norm": 0.46636375439030336, + "learning_rate": 9.980222960826254e-05, + "loss": 0.4686, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.41578030586242676, + "step": 85, + "valid_targets_mean": 2515.5, + "valid_targets_min": 246 + }, + { + "epoch": 0.2857142857142857, + "grad_norm": 0.42460312226406843, + "learning_rate": 9.977341561183109e-05, + "loss": 0.484, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4327179789543152, + "step": 90, + "valid_targets_mean": 3151.5, + "valid_targets_min": 597 + }, + { + "epoch": 0.30158730158730157, + "grad_norm": 0.5344446278063606, + "learning_rate": 9.974264760392466e-05, + "loss": 0.4722, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.5050898790359497, + "step": 95, + "valid_targets_mean": 2412.5, + "valid_targets_min": 234 + }, + { + "epoch": 0.31746031746031744, + "grad_norm": 0.499374507337321, + "learning_rate": 9.97099267924379e-05, + "loss": 0.4854, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.5286675691604614, + "step": 100, + "valid_targets_mean": 2194.5, + "valid_targets_min": 220 + }, + { + "epoch": 0.3333333333333333, + "grad_norm": 0.4644821565286184, + "learning_rate": 9.967525446192882e-05, + "loss": 0.4829, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.46753236651420593, + "step": 105, + "valid_targets_mean": 3076.4, + "valid_targets_min": 346 + }, + { + "epoch": 0.3492063492063492, + "grad_norm": 0.45875011863782966, + "learning_rate": 9.963863197356849e-05, + "loss": 0.4593, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4288153648376465, + "step": 110, + "valid_targets_mean": 2591.6, + "valid_targets_min": 205 + }, + { + "epoch": 0.36507936507936506, + "grad_norm": 0.4241575076803944, + "learning_rate": 9.960006076508747e-05, + "loss": 0.4956, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.5102588534355164, + "step": 115, + "valid_targets_mean": 2851.6, + "valid_targets_min": 301 + }, + { + "epoch": 0.38095238095238093, + "grad_norm": 0.4669849374601501, + "learning_rate": 9.95595423507195e-05, + "loss": 0.454, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4462718963623047, + "step": 120, + "valid_targets_mean": 2187.5, + "valid_targets_min": 218 + }, + { + "epoch": 0.3968253968253968, + "grad_norm": 0.42730150506373527, + "learning_rate": 9.951707832114193e-05, + "loss": 0.4566, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4672713279724121, + "step": 125, + "valid_targets_mean": 2710.9, + "valid_targets_min": 384 + }, + { + "epoch": 0.4126984126984127, + "grad_norm": 0.4452659913895723, + "learning_rate": 9.947267034341341e-05, + "loss": 0.4509, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.43273213505744934, + "step": 130, + "valid_targets_mean": 2510.0, + "valid_targets_min": 196 + }, + { + "epoch": 0.42857142857142855, + "grad_norm": 0.5230814946594939, + "learning_rate": 9.942632016090832e-05, + "loss": 0.4527, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.484851211309433, + "step": 135, + "valid_targets_mean": 3063.4, + "valid_targets_min": 376 + }, + { + "epoch": 0.4444444444444444, + "grad_norm": 0.48972021367909374, + "learning_rate": 9.937802959324838e-05, + "loss": 0.4381, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.45229780673980713, + "step": 140, + "valid_targets_mean": 2292.6, + "valid_targets_min": 198 + }, + { + "epoch": 0.4603174603174603, + "grad_norm": 0.4817977691355018, + "learning_rate": 9.932780053623121e-05, + "loss": 0.4693, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.46955937147140503, + "step": 145, + "valid_targets_mean": 2205.5, + "valid_targets_min": 312 + }, + { + "epoch": 0.47619047619047616, + "grad_norm": 0.40496345573574893, + "learning_rate": 9.927563496175593e-05, + "loss": 0.4604, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4356059432029724, + "step": 150, + "valid_targets_mean": 2779.3, + "valid_targets_min": 239 + }, + { + "epoch": 0.49206349206349204, + "grad_norm": 0.41383480740789197, + "learning_rate": 9.922153491774572e-05, + "loss": 0.4843, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4460809826850891, + "step": 155, + "valid_targets_mean": 2414.3, + "valid_targets_min": 232 + }, + { + "epoch": 0.5079365079365079, + "grad_norm": 0.41895286288742994, + "learning_rate": 9.91655025280674e-05, + "loss": 0.4776, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4513111710548401, + "step": 160, + "valid_targets_mean": 2400.3, + "valid_targets_min": 237 + }, + { + "epoch": 0.5238095238095238, + "grad_norm": 0.40903129533307225, + "learning_rate": 9.910753999244811e-05, + "loss": 0.4586, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.48089268803596497, + "step": 165, + "valid_targets_mean": 2635.8, + "valid_targets_min": 200 + }, + { + "epoch": 0.5396825396825397, + "grad_norm": 0.49411761138340926, + "learning_rate": 9.904764958638889e-05, + "loss": 0.4729, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.5136486291885376, + "step": 170, + "valid_targets_mean": 1939.4, + "valid_targets_min": 261 + }, + { + "epoch": 0.5555555555555556, + "grad_norm": 0.4316830149626285, + "learning_rate": 9.898583366107538e-05, + "loss": 0.4635, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4691350758075714, + "step": 175, + "valid_targets_mean": 2472.9, + "valid_targets_min": 273 + }, + { + "epoch": 0.5714285714285714, + "grad_norm": 0.37205019952750906, + "learning_rate": 9.892209464328556e-05, + "loss": 0.4664, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.43540042638778687, + "step": 180, + "valid_targets_mean": 3117.2, + "valid_targets_min": 440 + }, + { + "epoch": 0.5873015873015873, + "grad_norm": 0.35432966215877393, + "learning_rate": 9.885643503529439e-05, + "loss": 0.4663, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4152136445045471, + "step": 185, + "valid_targets_mean": 2810.6, + "valid_targets_min": 597 + }, + { + "epoch": 0.6031746031746031, + "grad_norm": 0.404018236827967, + "learning_rate": 9.878885741477563e-05, + "loss": 0.4642, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.43308812379837036, + "step": 190, + "valid_targets_mean": 2800.1, + "valid_targets_min": 323 + }, + { + "epoch": 0.6190476190476191, + "grad_norm": 0.4623038031782474, + "learning_rate": 9.871936443470063e-05, + "loss": 0.4744, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.44736790657043457, + "step": 195, + "valid_targets_mean": 2348.9, + "valid_targets_min": 286 + }, + { + "epoch": 0.6349206349206349, + "grad_norm": 0.42732679372391985, + "learning_rate": 9.864795882323421e-05, + "loss": 0.4621, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.45076048374176025, + "step": 200, + "valid_targets_mean": 2237.4, + "valid_targets_min": 638 + }, + { + "epoch": 0.6507936507936508, + "grad_norm": 0.4070469918060331, + "learning_rate": 9.857464338362748e-05, + "loss": 0.4526, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.44417518377304077, + "step": 205, + "valid_targets_mean": 2662.0, + "valid_targets_min": 185 + }, + { + "epoch": 0.6666666666666666, + "grad_norm": 0.40086844994258647, + "learning_rate": 9.849942099410792e-05, + "loss": 0.4395, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4628518223762512, + "step": 210, + "valid_targets_mean": 3103.4, + "valid_targets_min": 379 + }, + { + "epoch": 0.6825396825396826, + "grad_norm": 0.44803489338529057, + "learning_rate": 9.842229460776622e-05, + "loss": 0.4503, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4513833224773407, + "step": 215, + "valid_targets_mean": 2300.3, + "valid_targets_min": 202 + }, + { + "epoch": 0.6984126984126984, + "grad_norm": 0.4245178577427608, + "learning_rate": 9.834326725244049e-05, + "loss": 0.4555, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4850739538669586, + "step": 220, + "valid_targets_mean": 2528.8, + "valid_targets_min": 218 + }, + { + "epoch": 0.7142857142857143, + "grad_norm": 0.42640017439283534, + "learning_rate": 9.826234203059731e-05, + "loss": 0.4473, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4509563446044922, + "step": 225, + "valid_targets_mean": 2254.6, + "valid_targets_min": 469 + }, + { + "epoch": 0.7301587301587301, + "grad_norm": 0.4060256385753192, + "learning_rate": 9.817952211921e-05, + "loss": 0.4442, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.44804853200912476, + "step": 230, + "valid_targets_mean": 2336.8, + "valid_targets_min": 348 + }, + { + "epoch": 0.746031746031746, + "grad_norm": 0.39415634507460307, + "learning_rate": 9.809481076963383e-05, + "loss": 0.4461, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4222237467765808, + "step": 235, + "valid_targets_mean": 2616.1, + "valid_targets_min": 158 + }, + { + "epoch": 0.7619047619047619, + "grad_norm": 0.3793209216220147, + "learning_rate": 9.800821130747837e-05, + "loss": 0.4432, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.417599081993103, + "step": 240, + "valid_targets_mean": 2518.8, + "valid_targets_min": 244 + }, + { + "epoch": 0.7777777777777778, + "grad_norm": 0.36302483079818443, + "learning_rate": 9.791972713247704e-05, + "loss": 0.4509, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.42642349004745483, + "step": 245, + "valid_targets_mean": 2760.2, + "valid_targets_min": 359 + }, + { + "epoch": 0.7936507936507936, + "grad_norm": 0.4504642601762292, + "learning_rate": 9.782936171835353e-05, + "loss": 0.4416, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.46558064222335815, + "step": 250, + "valid_targets_mean": 2025.4, + "valid_targets_min": 120 + }, + { + "epoch": 0.8095238095238095, + "grad_norm": 0.4060999422939176, + "learning_rate": 9.773711861268549e-05, + "loss": 0.451, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4156520962715149, + "step": 255, + "valid_targets_mean": 2295.9, + "valid_targets_min": 189 + }, + { + "epoch": 0.8253968253968254, + "grad_norm": 0.3824798659443881, + "learning_rate": 9.764300143676518e-05, + "loss": 0.446, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4083912968635559, + "step": 260, + "valid_targets_mean": 2399.7, + "valid_targets_min": 196 + }, + { + "epoch": 0.8412698412698413, + "grad_norm": 0.4121574137558252, + "learning_rate": 9.754701388545745e-05, + "loss": 0.4865, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4813106656074524, + "step": 265, + "valid_targets_mean": 2399.8, + "valid_targets_min": 441 + }, + { + "epoch": 0.8571428571428571, + "grad_norm": 0.4108251966958842, + "learning_rate": 9.744915972705453e-05, + "loss": 0.4571, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.46524688601493835, + "step": 270, + "valid_targets_mean": 2407.6, + "valid_targets_min": 265 + }, + { + "epoch": 0.873015873015873, + "grad_norm": 0.43265442758622796, + "learning_rate": 9.734944280312824e-05, + "loss": 0.4779, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4791699945926666, + "step": 275, + "valid_targets_mean": 2258.2, + "valid_targets_min": 249 + }, + { + "epoch": 0.8888888888888888, + "grad_norm": 0.4328678190185699, + "learning_rate": 9.7247867028379e-05, + "loss": 0.4495, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4616977274417877, + "step": 280, + "valid_targets_mean": 2313.3, + "valid_targets_min": 185 + }, + { + "epoch": 0.9047619047619048, + "grad_norm": 0.3739881069982373, + "learning_rate": 9.714443639048232e-05, + "loss": 0.4394, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4122163653373718, + "step": 285, + "valid_targets_mean": 2567.9, + "valid_targets_min": 217 + }, + { + "epoch": 0.9206349206349206, + "grad_norm": 0.39552374005588, + "learning_rate": 9.703915494993215e-05, + "loss": 0.4622, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.45014989376068115, + "step": 290, + "valid_targets_mean": 2683.9, + "valid_targets_min": 266 + }, + { + "epoch": 0.9365079365079365, + "grad_norm": 0.3884148391603768, + "learning_rate": 9.693202683988151e-05, + "loss": 0.4422, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4628767967224121, + "step": 295, + "valid_targets_mean": 2285.8, + "valid_targets_min": 434 + }, + { + "epoch": 0.9523809523809523, + "grad_norm": 0.43310195986149597, + "learning_rate": 9.682305626598023e-05, + "loss": 0.4536, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.531997561454773, + "step": 300, + "valid_targets_mean": 2283.9, + "valid_targets_min": 264 + }, + { + "epoch": 0.9682539682539683, + "grad_norm": 0.4042349297889921, + "learning_rate": 9.671224750620981e-05, + "loss": 0.4463, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4553319811820984, + "step": 305, + "valid_targets_mean": 2139.8, + "valid_targets_min": 554 + }, + { + "epoch": 0.9841269841269841, + "grad_norm": 0.39570309574981977, + "learning_rate": 9.659960491071554e-05, + "loss": 0.4605, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4491097629070282, + "step": 310, + "valid_targets_mean": 2278.8, + "valid_targets_min": 219 + }, + { + "epoch": 1.0, + "grad_norm": 0.39547083187490994, + "learning_rate": 9.64851329016356e-05, + "loss": 0.4564, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.48211491107940674, + "step": 315, + "valid_targets_mean": 2471.2, + "valid_targets_min": 195 + }, + { + "epoch": 1.0158730158730158, + "grad_norm": 0.44266851326090795, + "learning_rate": 9.636883597292762e-05, + "loss": 0.3484, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.34535109996795654, + "step": 320, + "valid_targets_mean": 2675.8, + "valid_targets_min": 180 + }, + { + "epoch": 1.0317460317460316, + "grad_norm": 0.40690207113226806, + "learning_rate": 9.625071869019215e-05, + "loss": 0.3353, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.367917001247406, + "step": 325, + "valid_targets_mean": 2409.7, + "valid_targets_min": 241 + }, + { + "epoch": 1.0476190476190477, + "grad_norm": 0.4015590216927649, + "learning_rate": 9.613078569049344e-05, + "loss": 0.3332, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.35667768120765686, + "step": 330, + "valid_targets_mean": 2619.7, + "valid_targets_min": 963 + }, + { + "epoch": 1.0634920634920635, + "grad_norm": 0.4596693540802222, + "learning_rate": 9.600904168217734e-05, + "loss": 0.3401, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.39760622382164, + "step": 335, + "valid_targets_mean": 2627.2, + "valid_targets_min": 476 + }, + { + "epoch": 1.0793650793650793, + "grad_norm": 0.35473676844478375, + "learning_rate": 9.588549144468664e-05, + "loss": 0.3486, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.31826117634773254, + "step": 340, + "valid_targets_mean": 2848.5, + "valid_targets_min": 426 + }, + { + "epoch": 1.0952380952380953, + "grad_norm": 0.37339727714745113, + "learning_rate": 9.576013982837324e-05, + "loss": 0.3307, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3318113684654236, + "step": 345, + "valid_targets_mean": 2772.2, + "valid_targets_min": 346 + }, + { + "epoch": 1.1111111111111112, + "grad_norm": 0.3817083727667357, + "learning_rate": 9.563299175430782e-05, + "loss": 0.3211, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3441043794155121, + "step": 350, + "valid_targets_mean": 2942.0, + "valid_targets_min": 438 + }, + { + "epoch": 1.126984126984127, + "grad_norm": 0.4429926293637699, + "learning_rate": 9.550405221408664e-05, + "loss": 0.3229, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3329640030860901, + "step": 355, + "valid_targets_mean": 1941.2, + "valid_targets_min": 584 + }, + { + "epoch": 1.1428571428571428, + "grad_norm": 0.3930329982535386, + "learning_rate": 9.537332626963561e-05, + "loss": 0.3419, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3196127712726593, + "step": 360, + "valid_targets_mean": 2259.1, + "valid_targets_min": 124 + }, + { + "epoch": 1.1587301587301586, + "grad_norm": 0.34861697345236586, + "learning_rate": 9.524081905301152e-05, + "loss": 0.334, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3146742582321167, + "step": 365, + "valid_targets_mean": 2784.4, + "valid_targets_min": 145 + }, + { + "epoch": 1.1746031746031746, + "grad_norm": 0.4239796844440749, + "learning_rate": 9.510653576620056e-05, + "loss": 0.3577, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.35452699661254883, + "step": 370, + "valid_targets_mean": 2033.4, + "valid_targets_min": 215 + }, + { + "epoch": 1.1904761904761905, + "grad_norm": 0.40060655993111693, + "learning_rate": 9.497048168091418e-05, + "loss": 0.354, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3834336996078491, + "step": 375, + "valid_targets_mean": 2535.3, + "valid_targets_min": 232 + }, + { + "epoch": 1.2063492063492063, + "grad_norm": 0.3795109430124761, + "learning_rate": 9.483266213838202e-05, + "loss": 0.3391, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2972096800804138, + "step": 380, + "valid_targets_mean": 2523.3, + "valid_targets_min": 451 + }, + { + "epoch": 1.2222222222222223, + "grad_norm": 0.39191590831287, + "learning_rate": 9.469308254914231e-05, + "loss": 0.3836, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3700210452079773, + "step": 385, + "valid_targets_mean": 2706.9, + "valid_targets_min": 267 + }, + { + "epoch": 1.2380952380952381, + "grad_norm": 0.3659998026203586, + "learning_rate": 9.455174839282941e-05, + "loss": 0.3491, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.33657610416412354, + "step": 390, + "valid_targets_mean": 2722.0, + "valid_targets_min": 194 + }, + { + "epoch": 1.253968253968254, + "grad_norm": 0.37418168008907987, + "learning_rate": 9.440866521795874e-05, + "loss": 0.3491, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.35256943106651306, + "step": 395, + "valid_targets_mean": 2740.6, + "valid_targets_min": 224 + }, + { + "epoch": 1.2698412698412698, + "grad_norm": 0.35309965836220497, + "learning_rate": 9.426383864170891e-05, + "loss": 0.3533, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3299960196018219, + "step": 400, + "valid_targets_mean": 2572.2, + "valid_targets_min": 163 + }, + { + "epoch": 1.2857142857142856, + "grad_norm": 0.34519207131687596, + "learning_rate": 9.411727434970121e-05, + "loss": 0.3356, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.32315561175346375, + "step": 405, + "valid_targets_mean": 2863.9, + "valid_targets_min": 274 + }, + { + "epoch": 1.3015873015873016, + "grad_norm": 0.3685286433615304, + "learning_rate": 9.396897809577643e-05, + "loss": 0.3658, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3561249375343323, + "step": 410, + "valid_targets_mean": 2906.5, + "valid_targets_min": 359 + }, + { + "epoch": 1.3174603174603174, + "grad_norm": 0.4113265718110088, + "learning_rate": 9.381895570176893e-05, + "loss": 0.3729, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.37798845767974854, + "step": 415, + "valid_targets_mean": 2214.1, + "valid_targets_min": 278 + }, + { + "epoch": 1.3333333333333333, + "grad_norm": 0.3853929919058591, + "learning_rate": 9.366721305727813e-05, + "loss": 0.3503, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3528241515159607, + "step": 420, + "valid_targets_mean": 2844.3, + "valid_targets_min": 171 + }, + { + "epoch": 1.3492063492063493, + "grad_norm": 0.38070551456783475, + "learning_rate": 9.351375611943724e-05, + "loss": 0.3474, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3346797227859497, + "step": 425, + "valid_targets_mean": 2769.2, + "valid_targets_min": 201 + }, + { + "epoch": 1.3650793650793651, + "grad_norm": 0.37631713069814143, + "learning_rate": 9.335859091267952e-05, + "loss": 0.346, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3484334349632263, + "step": 430, + "valid_targets_mean": 2739.4, + "valid_targets_min": 214 + }, + { + "epoch": 1.380952380952381, + "grad_norm": 0.3751155959444023, + "learning_rate": 9.320172352850156e-05, + "loss": 0.3528, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.35233962535858154, + "step": 435, + "valid_targets_mean": 2489.8, + "valid_targets_min": 175 + }, + { + "epoch": 1.3968253968253967, + "grad_norm": 0.3817044233424013, + "learning_rate": 9.304316012522437e-05, + "loss": 0.3361, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.32587337493896484, + "step": 440, + "valid_targets_mean": 2374.7, + "valid_targets_min": 199 + }, + { + "epoch": 1.4126984126984126, + "grad_norm": 0.34414484705070847, + "learning_rate": 9.288290692775143e-05, + "loss": 0.3616, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.35864099860191345, + "step": 445, + "valid_targets_mean": 3126.6, + "valid_targets_min": 183 + }, + { + "epoch": 1.4285714285714286, + "grad_norm": 0.39454981089408503, + "learning_rate": 9.272097022732443e-05, + "loss": 0.3754, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3884466290473938, + "step": 450, + "valid_targets_mean": 2727.4, + "valid_targets_min": 250 + }, + { + "epoch": 1.4444444444444444, + "grad_norm": 0.34048023100820024, + "learning_rate": 9.255735638127623e-05, + "loss": 0.3625, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.342367947101593, + "step": 455, + "valid_targets_mean": 2992.0, + "valid_targets_min": 263 + }, + { + "epoch": 1.4603174603174602, + "grad_norm": 0.40481943376113955, + "learning_rate": 9.239207181278131e-05, + "loss": 0.3439, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3563992977142334, + "step": 460, + "valid_targets_mean": 2321.1, + "valid_targets_min": 206 + }, + { + "epoch": 1.4761904761904763, + "grad_norm": 0.3222626598906936, + "learning_rate": 9.222512301060358e-05, + "loss": 0.3335, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3124883770942688, + "step": 465, + "valid_targets_mean": 2704.8, + "valid_targets_min": 437 + }, + { + "epoch": 1.492063492063492, + "grad_norm": 0.36741237060725396, + "learning_rate": 9.205651652884169e-05, + "loss": 0.3371, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3461281657218933, + "step": 470, + "valid_targets_mean": 2850.3, + "valid_targets_min": 446 + }, + { + "epoch": 1.507936507936508, + "grad_norm": 0.41977956031028407, + "learning_rate": 9.188625898667165e-05, + "loss": 0.3785, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.40994706749916077, + "step": 475, + "valid_targets_mean": 2107.3, + "valid_targets_min": 207 + }, + { + "epoch": 1.5238095238095237, + "grad_norm": 0.3378625048872071, + "learning_rate": 9.171435706808709e-05, + "loss": 0.3298, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.32193559408187866, + "step": 480, + "valid_targets_mean": 3214.6, + "valid_targets_min": 220 + }, + { + "epoch": 1.5396825396825395, + "grad_norm": 0.4059575372698797, + "learning_rate": 9.154081752163675e-05, + "loss": 0.3699, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3578077554702759, + "step": 485, + "valid_targets_mean": 2024.2, + "valid_targets_min": 203 + }, + { + "epoch": 1.5555555555555556, + "grad_norm": 0.36398042351787546, + "learning_rate": 9.136564716015956e-05, + "loss": 0.3417, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.33834683895111084, + "step": 490, + "valid_targets_mean": 2358.5, + "valid_targets_min": 261 + }, + { + "epoch": 1.5714285714285714, + "grad_norm": 0.3433357187699167, + "learning_rate": 9.118885286051726e-05, + "loss": 0.3814, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3285355865955353, + "step": 495, + "valid_targets_mean": 2969.2, + "valid_targets_min": 144 + }, + { + "epoch": 1.5873015873015874, + "grad_norm": 0.3653356208371385, + "learning_rate": 9.101044156332437e-05, + "loss": 0.3483, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.36525341868400574, + "step": 500, + "valid_targets_mean": 2578.8, + "valid_targets_min": 678 + }, + { + "epoch": 1.6031746031746033, + "grad_norm": 0.35624343925463897, + "learning_rate": 9.083042027267567e-05, + "loss": 0.3624, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3465462327003479, + "step": 505, + "valid_targets_mean": 3138.6, + "valid_targets_min": 223 + }, + { + "epoch": 1.619047619047619, + "grad_norm": 0.4001082968087676, + "learning_rate": 9.064879605587132e-05, + "loss": 0.3363, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.35884204506874084, + "step": 510, + "valid_targets_mean": 2125.3, + "valid_targets_min": 167 + }, + { + "epoch": 1.6349206349206349, + "grad_norm": 0.35115647038426234, + "learning_rate": 9.046557604313937e-05, + "loss": 0.3535, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3274953067302704, + "step": 515, + "valid_targets_mean": 3062.6, + "valid_targets_min": 187 + }, + { + "epoch": 1.6507936507936507, + "grad_norm": 0.38386897998469693, + "learning_rate": 9.028076742735583e-05, + "loss": 0.374, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3895227909088135, + "step": 520, + "valid_targets_mean": 2499.5, + "valid_targets_min": 378 + }, + { + "epoch": 1.6666666666666665, + "grad_norm": 0.38046511406876304, + "learning_rate": 9.009437746376231e-05, + "loss": 0.3767, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.32906001806259155, + "step": 525, + "valid_targets_mean": 2462.9, + "valid_targets_min": 195 + }, + { + "epoch": 1.6825396825396826, + "grad_norm": 0.3174081597031139, + "learning_rate": 8.990641346968117e-05, + "loss": 0.3348, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.30704906582832336, + "step": 530, + "valid_targets_mean": 2873.2, + "valid_targets_min": 195 + }, + { + "epoch": 1.6984126984126984, + "grad_norm": 0.3843546825896823, + "learning_rate": 8.97168828242283e-05, + "loss": 0.3716, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.38304901123046875, + "step": 535, + "valid_targets_mean": 2569.9, + "valid_targets_min": 154 + }, + { + "epoch": 1.7142857142857144, + "grad_norm": 0.3642936094633957, + "learning_rate": 8.952579296802339e-05, + "loss": 0.3516, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3422582745552063, + "step": 540, + "valid_targets_mean": 2326.2, + "valid_targets_min": 421 + }, + { + "epoch": 1.7301587301587302, + "grad_norm": 0.35372441258435955, + "learning_rate": 8.933315140289782e-05, + "loss": 0.3683, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3545994758605957, + "step": 545, + "valid_targets_mean": 2525.8, + "valid_targets_min": 571 + }, + { + "epoch": 1.746031746031746, + "grad_norm": 0.3488760670304193, + "learning_rate": 8.91389656916002e-05, + "loss": 0.352, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.33395880460739136, + "step": 550, + "valid_targets_mean": 2334.3, + "valid_targets_min": 286 + }, + { + "epoch": 1.7619047619047619, + "grad_norm": 0.3449515653785241, + "learning_rate": 8.894324345749939e-05, + "loss": 0.3603, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3752964437007904, + "step": 555, + "valid_targets_mean": 2698.6, + "valid_targets_min": 334 + }, + { + "epoch": 1.7777777777777777, + "grad_norm": 0.44582537065228367, + "learning_rate": 8.874599238428533e-05, + "loss": 0.3726, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4217669367790222, + "step": 560, + "valid_targets_mean": 2369.1, + "valid_targets_min": 208 + }, + { + "epoch": 1.7936507936507935, + "grad_norm": 0.38499792532948907, + "learning_rate": 8.85472202156673e-05, + "loss": 0.3506, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3763793110847473, + "step": 565, + "valid_targets_mean": 2362.7, + "valid_targets_min": 413 + }, + { + "epoch": 1.8095238095238095, + "grad_norm": 0.3693504938924601, + "learning_rate": 8.834693475506992e-05, + "loss": 0.3483, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.32478874921798706, + "step": 570, + "valid_targets_mean": 2128.4, + "valid_targets_min": 230 + }, + { + "epoch": 1.8253968253968254, + "grad_norm": 0.29665875947923503, + "learning_rate": 8.814514386532691e-05, + "loss": 0.3578, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3050738573074341, + "step": 575, + "valid_targets_mean": 3005.0, + "valid_targets_min": 316 + }, + { + "epoch": 1.8412698412698414, + "grad_norm": 0.3171821826733686, + "learning_rate": 8.794185546837224e-05, + "loss": 0.328, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2995409071445465, + "step": 580, + "valid_targets_mean": 2781.0, + "valid_targets_min": 185 + }, + { + "epoch": 1.8571428571428572, + "grad_norm": 0.3476177271627817, + "learning_rate": 8.773707754492928e-05, + "loss": 0.361, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.35048457980155945, + "step": 585, + "valid_targets_mean": 2753.2, + "valid_targets_min": 530 + }, + { + "epoch": 1.873015873015873, + "grad_norm": 0.3558076242469748, + "learning_rate": 8.753081813419743e-05, + "loss": 0.3505, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.33344173431396484, + "step": 590, + "valid_targets_mean": 2813.1, + "valid_targets_min": 277 + }, + { + "epoch": 1.8888888888888888, + "grad_norm": 0.3903780835141171, + "learning_rate": 8.73230853335365e-05, + "loss": 0.3555, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.39296555519104004, + "step": 595, + "valid_targets_mean": 2520.8, + "valid_targets_min": 226 + }, + { + "epoch": 1.9047619047619047, + "grad_norm": 0.34998722992075165, + "learning_rate": 8.711388729814882e-05, + "loss": 0.3519, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3555290102958679, + "step": 600, + "valid_targets_mean": 2505.5, + "valid_targets_min": 104 + }, + { + "epoch": 1.9206349206349205, + "grad_norm": 0.35332584534974654, + "learning_rate": 8.690323224075917e-05, + "loss": 0.3807, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3768095374107361, + "step": 605, + "valid_targets_mean": 2586.9, + "valid_targets_min": 621 + }, + { + "epoch": 1.9365079365079365, + "grad_norm": 0.3272031935125674, + "learning_rate": 8.669112843129221e-05, + "loss": 0.3537, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.31432756781578064, + "step": 610, + "valid_targets_mean": 2751.9, + "valid_targets_min": 453 + }, + { + "epoch": 1.9523809523809523, + "grad_norm": 0.32415179706650094, + "learning_rate": 8.647758419654798e-05, + "loss": 0.3664, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.33924877643585205, + "step": 615, + "valid_targets_mean": 2780.2, + "valid_targets_min": 165 + }, + { + "epoch": 1.9682539682539684, + "grad_norm": 0.33717210666067793, + "learning_rate": 8.626260791987488e-05, + "loss": 0.3745, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.36188873648643494, + "step": 620, + "valid_targets_mean": 2830.4, + "valid_targets_min": 247 + }, + { + "epoch": 1.9841269841269842, + "grad_norm": 0.3677591957217286, + "learning_rate": 8.604620804084065e-05, + "loss": 0.3551, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.35614416003227234, + "step": 625, + "valid_targets_mean": 2520.5, + "valid_targets_min": 202 + }, + { + "epoch": 2.0, + "grad_norm": 0.38311285270433343, + "learning_rate": 8.582839305490094e-05, + "loss": 0.3453, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.28444457054138184, + "step": 630, + "valid_targets_mean": 2149.9, + "valid_targets_min": 424 + }, + { + "epoch": 2.015873015873016, + "grad_norm": 0.48301794004636245, + "learning_rate": 8.560917151306593e-05, + "loss": 0.23, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2242504507303238, + "step": 635, + "valid_targets_mean": 2896.3, + "valid_targets_min": 258 + }, + { + "epoch": 2.0317460317460316, + "grad_norm": 0.4040551655412978, + "learning_rate": 8.538855202156453e-05, + "loss": 0.2316, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21489444375038147, + "step": 640, + "valid_targets_mean": 2087.8, + "valid_targets_min": 193 + }, + { + "epoch": 2.0476190476190474, + "grad_norm": 0.4009196407760802, + "learning_rate": 8.516654324150652e-05, + "loss": 0.2271, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24874655902385712, + "step": 645, + "valid_targets_mean": 2557.5, + "valid_targets_min": 259 + }, + { + "epoch": 2.0634920634920633, + "grad_norm": 0.3866484634250721, + "learning_rate": 8.49431538885426e-05, + "loss": 0.2251, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22177374362945557, + "step": 650, + "valid_targets_mean": 2882.0, + "valid_targets_min": 426 + }, + { + "epoch": 2.0793650793650795, + "grad_norm": 0.3886427299302076, + "learning_rate": 8.471839273252217e-05, + "loss": 0.2359, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2031370997428894, + "step": 655, + "valid_targets_mean": 2485.6, + "valid_targets_min": 273 + }, + { + "epoch": 2.0952380952380953, + "grad_norm": 0.4239975962571399, + "learning_rate": 8.449226859714907e-05, + "loss": 0.2322, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23341144621372223, + "step": 660, + "valid_targets_mean": 2194.4, + "valid_targets_min": 203 + }, + { + "epoch": 2.111111111111111, + "grad_norm": 0.4281422340604251, + "learning_rate": 8.426479035963513e-05, + "loss": 0.2417, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2638523280620575, + "step": 665, + "valid_targets_mean": 2667.8, + "valid_targets_min": 945 + }, + { + "epoch": 2.126984126984127, + "grad_norm": 0.3776281695920411, + "learning_rate": 8.403596695035174e-05, + "loss": 0.2185, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2292603850364685, + "step": 670, + "valid_targets_mean": 2797.4, + "valid_targets_min": 398 + }, + { + "epoch": 2.142857142857143, + "grad_norm": 0.37212049930540597, + "learning_rate": 8.380580735247925e-05, + "loss": 0.2272, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23629912734031677, + "step": 675, + "valid_targets_mean": 2955.2, + "valid_targets_min": 269 + }, + { + "epoch": 2.1587301587301586, + "grad_norm": 0.421959896165175, + "learning_rate": 8.35743206016542e-05, + "loss": 0.2251, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2337156981229782, + "step": 680, + "valid_targets_mean": 2348.4, + "valid_targets_min": 250 + }, + { + "epoch": 2.1746031746031744, + "grad_norm": 0.372920260054872, + "learning_rate": 8.334151578561478e-05, + "loss": 0.2446, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24325747787952423, + "step": 685, + "valid_targets_mean": 2694.3, + "valid_targets_min": 283 + }, + { + "epoch": 2.1904761904761907, + "grad_norm": 0.39791139142167553, + "learning_rate": 8.310740204384387e-05, + "loss": 0.2381, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.25120609998703003, + "step": 690, + "valid_targets_mean": 2783.2, + "valid_targets_min": 426 + }, + { + "epoch": 2.2063492063492065, + "grad_norm": 0.4436539263165052, + "learning_rate": 8.287198856721042e-05, + "loss": 0.2379, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2446824014186859, + "step": 695, + "valid_targets_mean": 2084.3, + "valid_targets_min": 310 + }, + { + "epoch": 2.2222222222222223, + "grad_norm": 0.40522796147508056, + "learning_rate": 8.263528459760844e-05, + "loss": 0.2215, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21620988845825195, + "step": 700, + "valid_targets_mean": 2285.0, + "valid_targets_min": 419 + }, + { + "epoch": 2.238095238095238, + "grad_norm": 0.34858776335166425, + "learning_rate": 8.23972994275944e-05, + "loss": 0.2371, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22805678844451904, + "step": 705, + "valid_targets_mean": 2930.4, + "valid_targets_min": 266 + }, + { + "epoch": 2.253968253968254, + "grad_norm": 0.38174605875388073, + "learning_rate": 8.215804240002225e-05, + "loss": 0.229, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2130299210548401, + "step": 710, + "valid_targets_mean": 2618.8, + "valid_targets_min": 204 + }, + { + "epoch": 2.2698412698412698, + "grad_norm": 0.4443345512276022, + "learning_rate": 8.191752290767671e-05, + "loss": 0.2232, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23588870465755463, + "step": 715, + "valid_targets_mean": 1828.8, + "valid_targets_min": 211 + }, + { + "epoch": 2.2857142857142856, + "grad_norm": 0.3972504110088561, + "learning_rate": 8.167575039290448e-05, + "loss": 0.2407, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2563818395137787, + "step": 720, + "valid_targets_mean": 2860.3, + "valid_targets_min": 265 + }, + { + "epoch": 2.3015873015873014, + "grad_norm": 0.4156399667141594, + "learning_rate": 8.143273434724363e-05, + "loss": 0.2522, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.25209513306617737, + "step": 725, + "valid_targets_mean": 2486.2, + "valid_targets_min": 212 + }, + { + "epoch": 2.317460317460317, + "grad_norm": 0.3244609773168615, + "learning_rate": 8.118848431105091e-05, + "loss": 0.2386, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19948849081993103, + "step": 730, + "valid_targets_mean": 3095.0, + "valid_targets_min": 546 + }, + { + "epoch": 2.3333333333333335, + "grad_norm": 0.3415245420136681, + "learning_rate": 8.094300987312725e-05, + "loss": 0.2225, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1969720721244812, + "step": 735, + "valid_targets_mean": 2681.8, + "valid_targets_min": 756 + }, + { + "epoch": 2.3492063492063493, + "grad_norm": 0.4370336400721154, + "learning_rate": 8.069632067034129e-05, + "loss": 0.239, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24939577281475067, + "step": 740, + "valid_targets_mean": 1994.1, + "valid_targets_min": 260 + }, + { + "epoch": 2.365079365079365, + "grad_norm": 0.4120106407197172, + "learning_rate": 8.044842638725107e-05, + "loss": 0.239, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2335195690393448, + "step": 745, + "valid_targets_mean": 2264.1, + "valid_targets_min": 168 + }, + { + "epoch": 2.380952380952381, + "grad_norm": 0.42070813918631883, + "learning_rate": 8.019933675572389e-05, + "loss": 0.2396, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24204698204994202, + "step": 750, + "valid_targets_mean": 2170.7, + "valid_targets_min": 199 + }, + { + "epoch": 2.3968253968253967, + "grad_norm": 0.419212858920235, + "learning_rate": 7.994906155455411e-05, + "loss": 0.2406, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24102634191513062, + "step": 755, + "valid_targets_mean": 2129.0, + "valid_targets_min": 207 + }, + { + "epoch": 2.4126984126984126, + "grad_norm": 0.42357924115475587, + "learning_rate": 7.969761060907943e-05, + "loss": 0.2544, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2568979263305664, + "step": 760, + "valid_targets_mean": 2368.4, + "valid_targets_min": 401 + }, + { + "epoch": 2.4285714285714284, + "grad_norm": 0.4076313045109732, + "learning_rate": 7.944499379079502e-05, + "loss": 0.2502, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2677553594112396, + "step": 765, + "valid_targets_mean": 2660.7, + "valid_targets_min": 591 + }, + { + "epoch": 2.4444444444444446, + "grad_norm": 0.4196500894497793, + "learning_rate": 7.919122101696606e-05, + "loss": 0.2309, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24167239665985107, + "step": 770, + "valid_targets_mean": 2116.9, + "valid_targets_min": 478 + }, + { + "epoch": 2.4603174603174605, + "grad_norm": 0.3542307896844505, + "learning_rate": 7.893630225023842e-05, + "loss": 0.2198, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2239854782819748, + "step": 775, + "valid_targets_mean": 2851.4, + "valid_targets_min": 185 + }, + { + "epoch": 2.4761904761904763, + "grad_norm": 0.36876931222254006, + "learning_rate": 7.868024749824745e-05, + "loss": 0.2328, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20965051651000977, + "step": 780, + "valid_targets_mean": 2660.9, + "valid_targets_min": 488 + }, + { + "epoch": 2.492063492063492, + "grad_norm": 0.43447654920868173, + "learning_rate": 7.842306681322522e-05, + "loss": 0.2377, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.26887047290802, + "step": 785, + "valid_targets_mean": 2632.7, + "valid_targets_min": 257 + }, + { + "epoch": 2.507936507936508, + "grad_norm": 0.36050528366191936, + "learning_rate": 7.816477029160582e-05, + "loss": 0.2401, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21294507384300232, + "step": 790, + "valid_targets_mean": 2962.2, + "valid_targets_min": 144 + }, + { + "epoch": 2.5238095238095237, + "grad_norm": 0.3880288424105226, + "learning_rate": 7.790536807362896e-05, + "loss": 0.2321, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22806823253631592, + "step": 795, + "valid_targets_mean": 2448.3, + "valid_targets_min": 374 + }, + { + "epoch": 2.5396825396825395, + "grad_norm": 0.3506793231312844, + "learning_rate": 7.7644870342942e-05, + "loss": 0.2363, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2214076817035675, + "step": 800, + "valid_targets_mean": 2864.0, + "valid_targets_min": 509 + }, + { + "epoch": 2.5555555555555554, + "grad_norm": 0.379471431759816, + "learning_rate": 7.738328732620001e-05, + "loss": 0.2393, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.25795531272888184, + "step": 805, + "valid_targets_mean": 2795.9, + "valid_targets_min": 151 + }, + { + "epoch": 2.571428571428571, + "grad_norm": 0.4451012775555021, + "learning_rate": 7.712062929266444e-05, + "loss": 0.2455, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2695578634738922, + "step": 810, + "valid_targets_mean": 1996.1, + "valid_targets_min": 254 + }, + { + "epoch": 2.5873015873015874, + "grad_norm": 0.4149362223372113, + "learning_rate": 7.685690655379984e-05, + "loss": 0.2324, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23884569108486176, + "step": 815, + "valid_targets_mean": 2139.7, + "valid_targets_min": 391 + }, + { + "epoch": 2.6031746031746033, + "grad_norm": 0.33851723695257135, + "learning_rate": 7.659212946286912e-05, + "loss": 0.2429, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19433313608169556, + "step": 820, + "valid_targets_mean": 2543.6, + "valid_targets_min": 511 + }, + { + "epoch": 2.619047619047619, + "grad_norm": 0.35815135403445764, + "learning_rate": 7.632630841452709e-05, + "loss": 0.2598, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2559223771095276, + "step": 825, + "valid_targets_mean": 2778.4, + "valid_targets_min": 517 + }, + { + "epoch": 2.634920634920635, + "grad_norm": 0.3737784747271872, + "learning_rate": 7.605945384441238e-05, + "loss": 0.2586, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.26184919476509094, + "step": 830, + "valid_targets_mean": 2810.3, + "valid_targets_min": 247 + }, + { + "epoch": 2.6507936507936507, + "grad_norm": 0.4368069127718209, + "learning_rate": 7.579157622873779e-05, + "loss": 0.262, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2675510346889496, + "step": 835, + "valid_targets_mean": 2133.8, + "valid_targets_min": 287 + }, + { + "epoch": 2.6666666666666665, + "grad_norm": 0.4607291757869953, + "learning_rate": 7.552268608387889e-05, + "loss": 0.2446, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24789302051067352, + "step": 840, + "valid_targets_mean": 1905.4, + "valid_targets_min": 182 + }, + { + "epoch": 2.682539682539683, + "grad_norm": 0.34870228881154824, + "learning_rate": 7.525279396596137e-05, + "loss": 0.2279, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24618202447891235, + "step": 845, + "valid_targets_mean": 3167.5, + "valid_targets_min": 602 + }, + { + "epoch": 2.6984126984126986, + "grad_norm": 0.35465017069371135, + "learning_rate": 7.498191047044641e-05, + "loss": 0.2368, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2216503918170929, + "step": 850, + "valid_targets_mean": 2710.4, + "valid_targets_min": 210 + }, + { + "epoch": 2.7142857142857144, + "grad_norm": 0.3694078809115807, + "learning_rate": 7.471004623171493e-05, + "loss": 0.2369, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22898633778095245, + "step": 855, + "valid_targets_mean": 2611.0, + "valid_targets_min": 439 + }, + { + "epoch": 2.7301587301587302, + "grad_norm": 0.4063948486616755, + "learning_rate": 7.443721192264991e-05, + "loss": 0.2422, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2550041973590851, + "step": 860, + "valid_targets_mean": 2341.5, + "valid_targets_min": 191 + }, + { + "epoch": 2.746031746031746, + "grad_norm": 0.4288530527451921, + "learning_rate": 7.416341825421754e-05, + "loss": 0.2565, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24957627058029175, + "step": 865, + "valid_targets_mean": 1987.2, + "valid_targets_min": 193 + }, + { + "epoch": 2.761904761904762, + "grad_norm": 0.40661468646129983, + "learning_rate": 7.388867597504664e-05, + "loss": 0.2452, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.25630831718444824, + "step": 870, + "valid_targets_mean": 2386.3, + "valid_targets_min": 211 + }, + { + "epoch": 2.7777777777777777, + "grad_norm": 0.3702684154765434, + "learning_rate": 7.361299587100671e-05, + "loss": 0.2451, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24962452054023743, + "step": 875, + "valid_targets_mean": 2979.6, + "valid_targets_min": 207 + }, + { + "epoch": 2.7936507936507935, + "grad_norm": 0.3810506190477333, + "learning_rate": 7.333638876478453e-05, + "loss": 0.252, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2484670877456665, + "step": 880, + "valid_targets_mean": 2572.2, + "valid_targets_min": 154 + }, + { + "epoch": 2.8095238095238093, + "grad_norm": 0.39071678551885725, + "learning_rate": 7.305886551545926e-05, + "loss": 0.238, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23982301354408264, + "step": 885, + "valid_targets_mean": 2476.7, + "valid_targets_min": 291 + }, + { + "epoch": 2.825396825396825, + "grad_norm": 0.3975478513650243, + "learning_rate": 7.27804370180761e-05, + "loss": 0.24, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24334388971328735, + "step": 890, + "valid_targets_mean": 2065.4, + "valid_targets_min": 435 + }, + { + "epoch": 2.8412698412698414, + "grad_norm": 0.4016247640667921, + "learning_rate": 7.250111420321863e-05, + "loss": 0.2462, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2949928045272827, + "step": 895, + "valid_targets_mean": 2715.7, + "valid_targets_min": 393 + }, + { + "epoch": 2.857142857142857, + "grad_norm": 0.36382334192185584, + "learning_rate": 7.222090803657965e-05, + "loss": 0.2519, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20590443909168243, + "step": 900, + "valid_targets_mean": 2803.6, + "valid_targets_min": 634 + }, + { + "epoch": 2.873015873015873, + "grad_norm": 0.4065528042062189, + "learning_rate": 7.193982951853072e-05, + "loss": 0.2562, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2717926800251007, + "step": 905, + "valid_targets_mean": 2288.0, + "valid_targets_min": 271 + }, + { + "epoch": 2.888888888888889, + "grad_norm": 0.4234959220579958, + "learning_rate": 7.165788968369027e-05, + "loss": 0.263, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.28883954882621765, + "step": 910, + "valid_targets_mean": 2644.8, + "valid_targets_min": 249 + }, + { + "epoch": 2.9047619047619047, + "grad_norm": 0.4070144412051948, + "learning_rate": 7.137509960049043e-05, + "loss": 0.2613, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2804872989654541, + "step": 915, + "valid_targets_mean": 2510.8, + "valid_targets_min": 237 + }, + { + "epoch": 2.9206349206349205, + "grad_norm": 0.38610071592310075, + "learning_rate": 7.109147037074249e-05, + "loss": 0.2456, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21869128942489624, + "step": 920, + "valid_targets_mean": 2177.8, + "valid_targets_min": 484 + }, + { + "epoch": 2.9365079365079367, + "grad_norm": 0.40345040317953745, + "learning_rate": 7.080701312920106e-05, + "loss": 0.2474, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2644454836845398, + "step": 925, + "valid_targets_mean": 2261.0, + "valid_targets_min": 238 + }, + { + "epoch": 2.9523809523809526, + "grad_norm": 0.3583946683422614, + "learning_rate": 7.052173904312699e-05, + "loss": 0.2417, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22916853427886963, + "step": 930, + "valid_targets_mean": 2698.3, + "valid_targets_min": 165 + }, + { + "epoch": 2.9682539682539684, + "grad_norm": 0.4358104870135561, + "learning_rate": 7.023565931184888e-05, + "loss": 0.2409, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2504892945289612, + "step": 935, + "valid_targets_mean": 2044.6, + "valid_targets_min": 187 + }, + { + "epoch": 2.984126984126984, + "grad_norm": 0.35861624630391126, + "learning_rate": 6.994878516632347e-05, + "loss": 0.2392, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21946647763252258, + "step": 940, + "valid_targets_mean": 2683.1, + "valid_targets_min": 221 + }, + { + "epoch": 3.0, + "grad_norm": 0.3621902860532973, + "learning_rate": 6.966112786869471e-05, + "loss": 0.2302, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17726458609104156, + "step": 945, + "valid_targets_mean": 2966.3, + "valid_targets_min": 195 + }, + { + "epoch": 3.015873015873016, + "grad_norm": 0.7698595898965136, + "learning_rate": 6.937269871185171e-05, + "loss": 0.1329, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14550477266311646, + "step": 950, + "valid_targets_mean": 3079.0, + "valid_targets_min": 491 + }, + { + "epoch": 3.0317460317460316, + "grad_norm": 0.39771873905282246, + "learning_rate": 6.908350901898522e-05, + "loss": 0.1295, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11025917530059814, + "step": 955, + "valid_targets_mean": 2584.4, + "valid_targets_min": 414 + }, + { + "epoch": 3.0476190476190474, + "grad_norm": 0.4769825627649299, + "learning_rate": 6.87935701431433e-05, + "loss": 0.1263, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12527763843536377, + "step": 960, + "valid_targets_mean": 2373.9, + "valid_targets_min": 265 + }, + { + "epoch": 3.0634920634920633, + "grad_norm": 0.38699023640284025, + "learning_rate": 6.850289346678552e-05, + "loss": 0.123, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11964291334152222, + "step": 965, + "valid_targets_mean": 2580.1, + "valid_targets_min": 237 + }, + { + "epoch": 3.0793650793650795, + "grad_norm": 0.4187109262081189, + "learning_rate": 6.821149040133608e-05, + "loss": 0.1207, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1253824681043625, + "step": 970, + "valid_targets_mean": 2780.5, + "valid_targets_min": 270 + }, + { + "epoch": 3.0952380952380953, + "grad_norm": 0.4039308347954258, + "learning_rate": 6.791937238673592e-05, + "loss": 0.1311, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13876624405384064, + "step": 975, + "valid_targets_mean": 2746.4, + "valid_targets_min": 184 + }, + { + "epoch": 3.111111111111111, + "grad_norm": 0.4119149153525771, + "learning_rate": 6.762655089099353e-05, + "loss": 0.126, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12426751852035522, + "step": 980, + "valid_targets_mean": 2511.7, + "valid_targets_min": 207 + }, + { + "epoch": 3.126984126984127, + "grad_norm": 0.46008210222911483, + "learning_rate": 6.733303740973476e-05, + "loss": 0.1395, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16604162752628326, + "step": 985, + "valid_targets_mean": 2552.5, + "valid_targets_min": 367 + }, + { + "epoch": 3.142857142857143, + "grad_norm": 0.4647484378514777, + "learning_rate": 6.703884346575147e-05, + "loss": 0.141, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16042569279670715, + "step": 990, + "valid_targets_mean": 2469.8, + "valid_targets_min": 415 + }, + { + "epoch": 3.1587301587301586, + "grad_norm": 0.47755267977982285, + "learning_rate": 6.674398060854931e-05, + "loss": 0.1329, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1621413379907608, + "step": 995, + "valid_targets_mean": 2650.8, + "valid_targets_min": 281 + }, + { + "epoch": 3.1746031746031744, + "grad_norm": 0.4386333898681944, + "learning_rate": 6.644846041389414e-05, + "loss": 0.1301, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13621389865875244, + "step": 1000, + "valid_targets_mean": 2533.6, + "valid_targets_min": 198 + }, + { + "epoch": 3.1904761904761907, + "grad_norm": 0.3833701449823746, + "learning_rate": 6.615229448335769e-05, + "loss": 0.1255, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11770902574062347, + "step": 1005, + "valid_targets_mean": 2939.8, + "valid_targets_min": 174 + }, + { + "epoch": 3.2063492063492065, + "grad_norm": 0.3796820044476812, + "learning_rate": 6.58554944438621e-05, + "loss": 0.1283, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11332163214683533, + "step": 1010, + "valid_targets_mean": 2567.3, + "valid_targets_min": 260 + }, + { + "epoch": 3.2222222222222223, + "grad_norm": 0.5075170962580158, + "learning_rate": 6.555807194722339e-05, + "loss": 0.13, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14096391201019287, + "step": 1015, + "valid_targets_mean": 1829.0, + "valid_targets_min": 151 + }, + { + "epoch": 3.238095238095238, + "grad_norm": 0.3999084211320186, + "learning_rate": 6.526003866969412e-05, + "loss": 0.1297, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12311086058616638, + "step": 1020, + "valid_targets_mean": 2487.8, + "valid_targets_min": 198 + }, + { + "epoch": 3.253968253968254, + "grad_norm": 0.391206393557319, + "learning_rate": 6.4961406311505e-05, + "loss": 0.1324, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13543544709682465, + "step": 1025, + "valid_targets_mean": 3050.8, + "valid_targets_min": 690 + }, + { + "epoch": 3.2698412698412698, + "grad_norm": 0.4436044982466343, + "learning_rate": 6.466218659640545e-05, + "loss": 0.1397, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1440407782793045, + "step": 1030, + "valid_targets_mean": 2539.8, + "valid_targets_min": 770 + }, + { + "epoch": 3.2857142857142856, + "grad_norm": 0.38687968913920323, + "learning_rate": 6.436239127120351e-05, + "loss": 0.1309, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11935325711965561, + "step": 1035, + "valid_targets_mean": 3138.8, + "valid_targets_min": 252 + }, + { + "epoch": 3.3015873015873014, + "grad_norm": 0.42123320878362913, + "learning_rate": 6.406203210530455e-05, + "loss": 0.1356, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13043540716171265, + "step": 1040, + "valid_targets_mean": 2836.7, + "valid_targets_min": 371 + }, + { + "epoch": 3.317460317460317, + "grad_norm": 0.4798447649837025, + "learning_rate": 6.376112089024928e-05, + "loss": 0.1362, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15823490917682648, + "step": 1045, + "valid_targets_mean": 2094.2, + "valid_targets_min": 361 + }, + { + "epoch": 3.3333333333333335, + "grad_norm": 0.37072830531708845, + "learning_rate": 6.345966943925085e-05, + "loss": 0.1413, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.138170063495636, + "step": 1050, + "valid_targets_mean": 3093.5, + "valid_targets_min": 580 + }, + { + "epoch": 3.3492063492063493, + "grad_norm": 0.41743261394490494, + "learning_rate": 6.315768958673103e-05, + "loss": 0.1285, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13960057497024536, + "step": 1055, + "valid_targets_mean": 2723.5, + "valid_targets_min": 888 + }, + { + "epoch": 3.365079365079365, + "grad_norm": 0.43708682445854935, + "learning_rate": 6.285519318785568e-05, + "loss": 0.1284, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12192855030298233, + "step": 1060, + "valid_targets_mean": 2301.8, + "valid_targets_min": 314 + }, + { + "epoch": 3.380952380952381, + "grad_norm": 0.4538344385944368, + "learning_rate": 6.25521921180693e-05, + "loss": 0.133, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1502968668937683, + "step": 1065, + "valid_targets_mean": 2469.1, + "valid_targets_min": 136 + }, + { + "epoch": 3.3968253968253967, + "grad_norm": 0.36421773467170515, + "learning_rate": 6.224869827262885e-05, + "loss": 0.1358, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11965043097734451, + "step": 1070, + "valid_targets_mean": 3127.5, + "valid_targets_min": 235 + }, + { + "epoch": 3.4126984126984126, + "grad_norm": 0.45855586988281494, + "learning_rate": 6.194472356613667e-05, + "loss": 0.1465, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14145061373710632, + "step": 1075, + "valid_targets_mean": 2312.7, + "valid_targets_min": 209 + }, + { + "epoch": 3.4285714285714284, + "grad_norm": 0.4278180099766449, + "learning_rate": 6.16402799320729e-05, + "loss": 0.1364, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15281164646148682, + "step": 1080, + "valid_targets_mean": 2915.1, + "valid_targets_min": 793 + }, + { + "epoch": 3.4444444444444446, + "grad_norm": 0.4612738913924034, + "learning_rate": 6.133537932232684e-05, + "loss": 0.1353, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15492600202560425, + "step": 1085, + "valid_targets_mean": 2412.3, + "valid_targets_min": 472 + }, + { + "epoch": 3.4603174603174605, + "grad_norm": 0.4489452836999345, + "learning_rate": 6.1030033706727815e-05, + "loss": 0.1376, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1348612904548645, + "step": 1090, + "valid_targets_mean": 2547.2, + "valid_targets_min": 154 + }, + { + "epoch": 3.4761904761904763, + "grad_norm": 0.4688428889546676, + "learning_rate": 6.0724255072575275e-05, + "loss": 0.134, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14454475045204163, + "step": 1095, + "valid_targets_mean": 2236.1, + "valid_targets_min": 168 + }, + { + "epoch": 3.492063492063492, + "grad_norm": 0.488004478053982, + "learning_rate": 6.0418055424168154e-05, + "loss": 0.1382, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1340714991092682, + "step": 1100, + "valid_targets_mean": 2013.9, + "valid_targets_min": 191 + }, + { + "epoch": 3.507936507936508, + "grad_norm": 0.3613516138219136, + "learning_rate": 6.011144678233359e-05, + "loss": 0.1342, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11541695147752762, + "step": 1105, + "valid_targets_mean": 3018.2, + "valid_targets_min": 501 + }, + { + "epoch": 3.5238095238095237, + "grad_norm": 0.4547154945941303, + "learning_rate": 5.9804441183955104e-05, + "loss": 0.1371, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13645395636558533, + "step": 1110, + "valid_targets_mean": 2076.8, + "valid_targets_min": 189 + }, + { + "epoch": 3.5396825396825395, + "grad_norm": 0.4360025006403801, + "learning_rate": 5.9497050681499955e-05, + "loss": 0.1303, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12478509545326233, + "step": 1115, + "valid_targets_mean": 2308.9, + "valid_targets_min": 281 + }, + { + "epoch": 3.5555555555555554, + "grad_norm": 0.3691582477837132, + "learning_rate": 5.9189287342545996e-05, + "loss": 0.1328, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11068733036518097, + "step": 1120, + "valid_targets_mean": 2764.3, + "valid_targets_min": 216 + }, + { + "epoch": 3.571428571428571, + "grad_norm": 0.46179185274509543, + "learning_rate": 5.888116324930798e-05, + "loss": 0.1414, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14813870191574097, + "step": 1125, + "valid_targets_mean": 2459.1, + "valid_targets_min": 206 + }, + { + "epoch": 3.5873015873015874, + "grad_norm": 0.42985954111311864, + "learning_rate": 5.8572690498163205e-05, + "loss": 0.1381, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1348619908094406, + "step": 1130, + "valid_targets_mean": 2376.4, + "valid_targets_min": 262 + }, + { + "epoch": 3.6031746031746033, + "grad_norm": 0.45907224067724306, + "learning_rate": 5.826388119917658e-05, + "loss": 0.1384, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14189521968364716, + "step": 1135, + "valid_targets_mean": 2123.6, + "valid_targets_min": 223 + }, + { + "epoch": 3.619047619047619, + "grad_norm": 0.36688284831423773, + "learning_rate": 5.795474747562533e-05, + "loss": 0.1305, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11347411572933197, + "step": 1140, + "valid_targets_mean": 2830.4, + "valid_targets_min": 207 + }, + { + "epoch": 3.634920634920635, + "grad_norm": 0.4886171236368861, + "learning_rate": 5.7645301463522895e-05, + "loss": 0.136, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15021659433841705, + "step": 1145, + "valid_targets_mean": 1987.8, + "valid_targets_min": 217 + }, + { + "epoch": 3.6507936507936507, + "grad_norm": 0.41917429825731817, + "learning_rate": 5.7335555311142677e-05, + "loss": 0.1355, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14572878181934357, + "step": 1150, + "valid_targets_mean": 2584.3, + "valid_targets_min": 210 + }, + { + "epoch": 3.6666666666666665, + "grad_norm": 0.3735229838969736, + "learning_rate": 5.702552117854093e-05, + "loss": 0.1348, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12277483940124512, + "step": 1155, + "valid_targets_mean": 3142.4, + "valid_targets_min": 207 + }, + { + "epoch": 3.682539682539683, + "grad_norm": 0.4255484024803778, + "learning_rate": 5.671521123707955e-05, + "loss": 0.1312, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1331755518913269, + "step": 1160, + "valid_targets_mean": 2552.5, + "valid_targets_min": 251 + }, + { + "epoch": 3.6984126984126986, + "grad_norm": 0.45191642366787854, + "learning_rate": 5.640463766894813e-05, + "loss": 0.1402, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15837690234184265, + "step": 1165, + "valid_targets_mean": 2426.0, + "valid_targets_min": 267 + }, + { + "epoch": 3.7142857142857144, + "grad_norm": 0.39497444187563313, + "learning_rate": 5.609381266668578e-05, + "loss": 0.1351, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1302325278520584, + "step": 1170, + "valid_targets_mean": 2751.1, + "valid_targets_min": 249 + }, + { + "epoch": 3.7301587301587302, + "grad_norm": 0.42025552602910415, + "learning_rate": 5.5782748432702426e-05, + "loss": 0.1408, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13248401880264282, + "step": 1175, + "valid_targets_mean": 2381.6, + "valid_targets_min": 200 + }, + { + "epoch": 3.746031746031746, + "grad_norm": 0.39047726482208395, + "learning_rate": 5.54714571787998e-05, + "loss": 0.1413, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1318662315607071, + "step": 1180, + "valid_targets_mean": 2843.3, + "valid_targets_min": 238 + }, + { + "epoch": 3.761904761904762, + "grad_norm": 0.43055282656428, + "learning_rate": 5.5159951125692e-05, + "loss": 0.1337, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15757383406162262, + "step": 1185, + "valid_targets_mean": 2592.0, + "valid_targets_min": 227 + }, + { + "epoch": 3.7777777777777777, + "grad_norm": 0.49369028945670396, + "learning_rate": 5.484824250252574e-05, + "loss": 0.1473, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14920111000537872, + "step": 1190, + "valid_targets_mean": 2094.6, + "valid_targets_min": 165 + }, + { + "epoch": 3.7936507936507935, + "grad_norm": 0.4173137104628635, + "learning_rate": 5.453634354640028e-05, + "loss": 0.1431, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1300673484802246, + "step": 1195, + "valid_targets_mean": 2440.9, + "valid_targets_min": 124 + }, + { + "epoch": 3.8095238095238093, + "grad_norm": 0.40542581437194386, + "learning_rate": 5.422426650188698e-05, + "loss": 0.1432, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1281275749206543, + "step": 1200, + "valid_targets_mean": 2444.2, + "valid_targets_min": 258 + }, + { + "epoch": 3.825396825396825, + "grad_norm": 0.42799016053358296, + "learning_rate": 5.391202362054859e-05, + "loss": 0.1369, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13898375630378723, + "step": 1205, + "valid_targets_mean": 2447.6, + "valid_targets_min": 241 + }, + { + "epoch": 3.8412698412698414, + "grad_norm": 0.45521585053534835, + "learning_rate": 5.359962716045835e-05, + "loss": 0.137, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13215912878513336, + "step": 1210, + "valid_targets_mean": 2011.7, + "valid_targets_min": 176 + }, + { + "epoch": 3.857142857142857, + "grad_norm": 0.4817423800748524, + "learning_rate": 5.328708938571872e-05, + "loss": 0.1536, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18796545267105103, + "step": 1215, + "valid_targets_mean": 2458.4, + "valid_targets_min": 245 + }, + { + "epoch": 3.873015873015873, + "grad_norm": 0.44566851409235214, + "learning_rate": 5.2974422565979866e-05, + "loss": 0.1391, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13472338020801544, + "step": 1220, + "valid_targets_mean": 2213.6, + "valid_targets_min": 430 + }, + { + "epoch": 3.888888888888889, + "grad_norm": 0.43356223036890074, + "learning_rate": 5.266163897595804e-05, + "loss": 0.1364, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13793230056762695, + "step": 1225, + "valid_targets_mean": 2253.7, + "valid_targets_min": 263 + }, + { + "epoch": 3.9047619047619047, + "grad_norm": 0.4454928717598988, + "learning_rate": 5.234875089495368e-05, + "loss": 0.1401, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14466333389282227, + "step": 1230, + "valid_targets_mean": 2334.3, + "valid_targets_min": 214 + }, + { + "epoch": 3.9206349206349205, + "grad_norm": 0.4258312369515503, + "learning_rate": 5.203577060636935e-05, + "loss": 0.139, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12505504488945007, + "step": 1235, + "valid_targets_mean": 2209.3, + "valid_targets_min": 504 + }, + { + "epoch": 3.9365079365079367, + "grad_norm": 0.43629400676014396, + "learning_rate": 5.172271039722749e-05, + "loss": 0.1283, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11868347227573395, + "step": 1240, + "valid_targets_mean": 2271.5, + "valid_targets_min": 245 + }, + { + "epoch": 3.9523809523809526, + "grad_norm": 0.4468590506575495, + "learning_rate": 5.140958255768812e-05, + "loss": 0.134, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14880114793777466, + "step": 1245, + "valid_targets_mean": 2210.5, + "valid_targets_min": 192 + }, + { + "epoch": 3.9682539682539684, + "grad_norm": 0.4376822825879029, + "learning_rate": 5.109639938056625e-05, + "loss": 0.1367, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12730690836906433, + "step": 1250, + "valid_targets_mean": 2174.6, + "valid_targets_min": 391 + }, + { + "epoch": 3.984126984126984, + "grad_norm": 0.3845553258041374, + "learning_rate": 5.078317316084935e-05, + "loss": 0.1324, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12826502323150635, + "step": 1255, + "valid_targets_mean": 2918.4, + "valid_targets_min": 215 + }, + { + "epoch": 4.0, + "grad_norm": 0.34735966713139915, + "learning_rate": 5.0469916195214694e-05, + "loss": 0.1324, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08627285808324814, + "step": 1260, + "valid_targets_mean": 2630.5, + "valid_targets_min": 250 + }, + { + "epoch": 4.015873015873016, + "grad_norm": 0.5374115601003832, + "learning_rate": 5.015664078154655e-05, + "loss": 0.0613, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06476876139640808, + "step": 1265, + "valid_targets_mean": 2223.2, + "valid_targets_min": 216 + }, + { + "epoch": 4.031746031746032, + "grad_norm": 0.37171288854799694, + "learning_rate": 4.9843359218453455e-05, + "loss": 0.0607, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.058441609144210815, + "step": 1270, + "valid_targets_mean": 2688.8, + "valid_targets_min": 187 + }, + { + "epoch": 4.0476190476190474, + "grad_norm": 0.3909296567173623, + "learning_rate": 4.953008380478532e-05, + "loss": 0.0681, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07275949418544769, + "step": 1275, + "valid_targets_mean": 2998.8, + "valid_targets_min": 217 + }, + { + "epoch": 4.063492063492063, + "grad_norm": 0.4230556479711462, + "learning_rate": 4.921682683915066e-05, + "loss": 0.0605, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06004776805639267, + "step": 1280, + "valid_targets_mean": 2773.4, + "valid_targets_min": 435 + }, + { + "epoch": 4.079365079365079, + "grad_norm": 0.3870274648997481, + "learning_rate": 4.8903600619433775e-05, + "loss": 0.0651, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06690061092376709, + "step": 1285, + "valid_targets_mean": 2666.8, + "valid_targets_min": 468 + }, + { + "epoch": 4.095238095238095, + "grad_norm": 0.4325377419880736, + "learning_rate": 4.85904174423119e-05, + "loss": 0.0636, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0695696473121643, + "step": 1290, + "valid_targets_mean": 2134.7, + "valid_targets_min": 450 + }, + { + "epoch": 4.111111111111111, + "grad_norm": 0.37005628889724534, + "learning_rate": 4.8277289602772514e-05, + "loss": 0.058, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.05767609924077988, + "step": 1295, + "valid_targets_mean": 2438.5, + "valid_targets_min": 292 + }, + { + "epoch": 4.1269841269841265, + "grad_norm": 0.40358201236024654, + "learning_rate": 4.796422939363066e-05, + "loss": 0.0675, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07652418315410614, + "step": 1300, + "valid_targets_mean": 2461.6, + "valid_targets_min": 230 + }, + { + "epoch": 4.142857142857143, + "grad_norm": 0.43971941205097786, + "learning_rate": 4.7651249105046325e-05, + "loss": 0.0687, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07158204913139343, + "step": 1305, + "valid_targets_mean": 2383.7, + "valid_targets_min": 310 + }, + { + "epoch": 4.158730158730159, + "grad_norm": 0.3926552899119832, + "learning_rate": 4.733836102404197e-05, + "loss": 0.0606, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.05940824747085571, + "step": 1310, + "valid_targets_mean": 2287.8, + "valid_targets_min": 264 + }, + { + "epoch": 4.174603174603175, + "grad_norm": 0.3790650550304819, + "learning_rate": 4.7025577434020146e-05, + "loss": 0.0716, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06187836080789566, + "step": 1315, + "valid_targets_mean": 2826.3, + "valid_targets_min": 258 + }, + { + "epoch": 4.190476190476191, + "grad_norm": 0.3915816866267985, + "learning_rate": 4.671291061428129e-05, + "loss": 0.0694, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06489279866218567, + "step": 1320, + "valid_targets_mean": 2592.1, + "valid_targets_min": 439 + }, + { + "epoch": 4.2063492063492065, + "grad_norm": 0.3717790834369351, + "learning_rate": 4.640037283954165e-05, + "loss": 0.0688, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06313011050224304, + "step": 1325, + "valid_targets_mean": 2698.1, + "valid_targets_min": 533 + }, + { + "epoch": 4.222222222222222, + "grad_norm": 0.3992801441632957, + "learning_rate": 4.608797637945142e-05, + "loss": 0.065, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06738118827342987, + "step": 1330, + "valid_targets_mean": 2594.3, + "valid_targets_min": 432 + }, + { + "epoch": 4.238095238095238, + "grad_norm": 0.34131670018999777, + "learning_rate": 4.577573349811304e-05, + "loss": 0.0662, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.05766981095075607, + "step": 1335, + "valid_targets_mean": 3023.3, + "valid_targets_min": 489 + }, + { + "epoch": 4.253968253968254, + "grad_norm": 0.4706294687244321, + "learning_rate": 4.5463656453599726e-05, + "loss": 0.074, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08102245628833771, + "step": 1340, + "valid_targets_mean": 2146.8, + "valid_targets_min": 331 + }, + { + "epoch": 4.26984126984127, + "grad_norm": 0.3960464746053695, + "learning_rate": 4.515175749747426e-05, + "loss": 0.063, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.05950373038649559, + "step": 1345, + "valid_targets_mean": 2232.8, + "valid_targets_min": 237 + }, + { + "epoch": 4.285714285714286, + "grad_norm": 0.3782901423765049, + "learning_rate": 4.484004887430803e-05, + "loss": 0.0657, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.05941256508231163, + "step": 1350, + "valid_targets_mean": 2419.9, + "valid_targets_min": 180 + }, + { + "epoch": 4.301587301587301, + "grad_norm": 0.41633599303996827, + "learning_rate": 4.452854282120022e-05, + "loss": 0.0673, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06611457467079163, + "step": 1355, + "valid_targets_mean": 2522.3, + "valid_targets_min": 234 + }, + { + "epoch": 4.317460317460317, + "grad_norm": 0.4631547618724549, + "learning_rate": 4.4217251567297586e-05, + "loss": 0.0655, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07109738886356354, + "step": 1360, + "valid_targets_mean": 1865.5, + "valid_targets_min": 447 + }, + { + "epoch": 4.333333333333333, + "grad_norm": 0.3693506597023311, + "learning_rate": 4.390618733331423e-05, + "loss": 0.0635, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06450463831424713, + "step": 1365, + "valid_targets_mean": 2924.4, + "valid_targets_min": 689 + }, + { + "epoch": 4.349206349206349, + "grad_norm": 0.3585683238273711, + "learning_rate": 4.359536233105187e-05, + "loss": 0.0648, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06006067991256714, + "step": 1370, + "valid_targets_mean": 2858.0, + "valid_targets_min": 267 + }, + { + "epoch": 4.365079365079365, + "grad_norm": 0.38248289687511855, + "learning_rate": 4.328478876292045e-05, + "loss": 0.0681, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06233789771795273, + "step": 1375, + "valid_targets_mean": 2424.9, + "valid_targets_min": 442 + }, + { + "epoch": 4.380952380952381, + "grad_norm": 0.3925232026286688, + "learning_rate": 4.297447882145907e-05, + "loss": 0.0675, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06120011582970619, + "step": 1380, + "valid_targets_mean": 2372.5, + "valid_targets_min": 600 + }, + { + "epoch": 4.396825396825397, + "grad_norm": 0.4002033892280286, + "learning_rate": 4.266444468885735e-05, + "loss": 0.0696, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06046672910451889, + "step": 1385, + "valid_targets_mean": 2145.0, + "valid_targets_min": 120 + }, + { + "epoch": 4.412698412698413, + "grad_norm": 0.36759837626169345, + "learning_rate": 4.235469853647711e-05, + "loss": 0.0664, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06483858823776245, + "step": 1390, + "valid_targets_mean": 2900.9, + "valid_targets_min": 185 + }, + { + "epoch": 4.428571428571429, + "grad_norm": 0.3960802055929344, + "learning_rate": 4.204525252437469e-05, + "loss": 0.0742, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06655968725681305, + "step": 1395, + "valid_targets_mean": 2346.7, + "valid_targets_min": 191 + }, + { + "epoch": 4.444444444444445, + "grad_norm": 0.41451064024370987, + "learning_rate": 4.173611880082342e-05, + "loss": 0.0663, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07234956324100494, + "step": 1400, + "valid_targets_mean": 2242.2, + "valid_targets_min": 221 + }, + { + "epoch": 4.4603174603174605, + "grad_norm": 0.40324026907543736, + "learning_rate": 4.1427309501836806e-05, + "loss": 0.0634, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06758803129196167, + "step": 1405, + "valid_targets_mean": 2321.1, + "valid_targets_min": 454 + }, + { + "epoch": 4.476190476190476, + "grad_norm": 0.38961850241575535, + "learning_rate": 4.111883675069202e-05, + "loss": 0.065, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.05995681509375572, + "step": 1410, + "valid_targets_mean": 2332.3, + "valid_targets_min": 351 + }, + { + "epoch": 4.492063492063492, + "grad_norm": 0.42775760610838676, + "learning_rate": 4.081071265745402e-05, + "loss": 0.0714, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06675523519515991, + "step": 1415, + "valid_targets_mean": 2074.2, + "valid_targets_min": 196 + }, + { + "epoch": 4.507936507936508, + "grad_norm": 0.45128508953458496, + "learning_rate": 4.0502949318500064e-05, + "loss": 0.0679, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07593533396720886, + "step": 1420, + "valid_targets_mean": 2228.8, + "valid_targets_min": 203 + }, + { + "epoch": 4.523809523809524, + "grad_norm": 0.38554706258344384, + "learning_rate": 4.01955588160449e-05, + "loss": 0.0689, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06949351727962494, + "step": 1425, + "valid_targets_mean": 2503.9, + "valid_targets_min": 264 + }, + { + "epoch": 4.5396825396825395, + "grad_norm": 0.3567891771028985, + "learning_rate": 3.9888553217666415e-05, + "loss": 0.0618, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.05722368508577347, + "step": 1430, + "valid_targets_mean": 2989.6, + "valid_targets_min": 266 + }, + { + "epoch": 4.555555555555555, + "grad_norm": 0.3711416546957095, + "learning_rate": 3.9581944575831864e-05, + "loss": 0.0683, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06568397581577301, + "step": 1435, + "valid_targets_mean": 2932.8, + "valid_targets_min": 276 + }, + { + "epoch": 4.571428571428571, + "grad_norm": 0.3850335347043359, + "learning_rate": 3.927574492742473e-05, + "loss": 0.0679, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06129177659749985, + "step": 1440, + "valid_targets_mean": 2762.5, + "valid_targets_min": 150 + }, + { + "epoch": 4.587301587301587, + "grad_norm": 0.43630428680129985, + "learning_rate": 3.896996629327219e-05, + "loss": 0.0641, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07695893198251724, + "step": 1445, + "valid_targets_mean": 2231.8, + "valid_targets_min": 231 + }, + { + "epoch": 4.603174603174603, + "grad_norm": 0.3662894498091143, + "learning_rate": 3.8664620677673186e-05, + "loss": 0.0734, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06687343120574951, + "step": 1450, + "valid_targets_mean": 2910.8, + "valid_targets_min": 551 + }, + { + "epoch": 4.619047619047619, + "grad_norm": 0.3910920911962809, + "learning_rate": 3.8359720067927115e-05, + "loss": 0.0688, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07100112736225128, + "step": 1455, + "valid_targets_mean": 2753.8, + "valid_targets_min": 448 + }, + { + "epoch": 4.634920634920634, + "grad_norm": 0.3826085616974412, + "learning_rate": 3.805527643386334e-05, + "loss": 0.062, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.05772186815738678, + "step": 1460, + "valid_targets_mean": 2533.5, + "valid_targets_min": 245 + }, + { + "epoch": 4.650793650793651, + "grad_norm": 0.39418709977327815, + "learning_rate": 3.775130172737117e-05, + "loss": 0.0672, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06985107064247131, + "step": 1465, + "valid_targets_mean": 2884.3, + "valid_targets_min": 175 + }, + { + "epoch": 4.666666666666667, + "grad_norm": 0.37324369415203607, + "learning_rate": 3.7447807881930694e-05, + "loss": 0.0614, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.05916617065668106, + "step": 1470, + "valid_targets_mean": 2385.8, + "valid_targets_min": 265 + }, + { + "epoch": 4.682539682539683, + "grad_norm": 0.39845163689842195, + "learning_rate": 3.7144806812144324e-05, + "loss": 0.0636, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06261011958122253, + "step": 1475, + "valid_targets_mean": 2408.0, + "valid_targets_min": 215 + }, + { + "epoch": 4.698412698412699, + "grad_norm": 0.4434386199829386, + "learning_rate": 3.6842310413269e-05, + "loss": 0.0682, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0695788711309433, + "step": 1480, + "valid_targets_mean": 2167.2, + "valid_targets_min": 124 + }, + { + "epoch": 4.714285714285714, + "grad_norm": 0.34456051372968866, + "learning_rate": 3.654033056074918e-05, + "loss": 0.0638, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.05945037305355072, + "step": 1485, + "valid_targets_mean": 2710.4, + "valid_targets_min": 235 + }, + { + "epoch": 4.73015873015873, + "grad_norm": 0.4358333891609388, + "learning_rate": 3.6238879109750735e-05, + "loss": 0.0721, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06855715811252594, + "step": 1490, + "valid_targets_mean": 1982.1, + "valid_targets_min": 346 + }, + { + "epoch": 4.746031746031746, + "grad_norm": 0.4113887728007023, + "learning_rate": 3.593796789469546e-05, + "loss": 0.0668, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0679166167974472, + "step": 1495, + "valid_targets_mean": 2409.2, + "valid_targets_min": 130 + }, + { + "epoch": 4.761904761904762, + "grad_norm": 0.420164613845667, + "learning_rate": 3.563760872879649e-05, + "loss": 0.0718, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07911844551563263, + "step": 1500, + "valid_targets_mean": 2697.5, + "valid_targets_min": 218 + }, + { + "epoch": 4.777777777777778, + "grad_norm": 0.3987720363846426, + "learning_rate": 3.5337813403594545e-05, + "loss": 0.068, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06510649621486664, + "step": 1505, + "valid_targets_mean": 2443.7, + "valid_targets_min": 339 + }, + { + "epoch": 4.7936507936507935, + "grad_norm": 0.4447564398919035, + "learning_rate": 3.5038593688495005e-05, + "loss": 0.0658, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07375577837228775, + "step": 1510, + "valid_targets_mean": 2388.2, + "valid_targets_min": 277 + }, + { + "epoch": 4.809523809523809, + "grad_norm": 0.3852359661038981, + "learning_rate": 3.4739961330305894e-05, + "loss": 0.0677, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0663665384054184, + "step": 1515, + "valid_targets_mean": 2791.4, + "valid_targets_min": 370 + }, + { + "epoch": 4.825396825396825, + "grad_norm": 0.4317035639814621, + "learning_rate": 3.444192805277663e-05, + "loss": 0.0656, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0794079527258873, + "step": 1520, + "valid_targets_mean": 2863.4, + "valid_targets_min": 175 + }, + { + "epoch": 4.841269841269841, + "grad_norm": 0.3850412450525805, + "learning_rate": 3.414450555613792e-05, + "loss": 0.0674, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.05959293246269226, + "step": 1525, + "valid_targets_mean": 2372.6, + "valid_targets_min": 273 + }, + { + "epoch": 4.857142857142857, + "grad_norm": 0.42723501366552197, + "learning_rate": 3.3847705516642317e-05, + "loss": 0.0675, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07407239079475403, + "step": 1530, + "valid_targets_mean": 2309.7, + "valid_targets_min": 484 + }, + { + "epoch": 4.8730158730158735, + "grad_norm": 0.35088568584020413, + "learning_rate": 3.355153958610587e-05, + "loss": 0.0645, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06096992269158363, + "step": 1535, + "valid_targets_mean": 2741.9, + "valid_targets_min": 419 + }, + { + "epoch": 4.888888888888889, + "grad_norm": 0.33566703317803154, + "learning_rate": 3.325601939145069e-05, + "loss": 0.0648, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.05687577277421951, + "step": 1540, + "valid_targets_mean": 3076.0, + "valid_targets_min": 464 + }, + { + "epoch": 4.904761904761905, + "grad_norm": 0.4155768779104891, + "learning_rate": 3.296115653424854e-05, + "loss": 0.0678, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06845413148403168, + "step": 1545, + "valid_targets_mean": 2274.1, + "valid_targets_min": 223 + }, + { + "epoch": 4.920634920634921, + "grad_norm": 0.3597919070757417, + "learning_rate": 3.266696259026526e-05, + "loss": 0.0655, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.060910664498806, + "step": 1550, + "valid_targets_mean": 2485.0, + "valid_targets_min": 251 + }, + { + "epoch": 4.936507936507937, + "grad_norm": 0.4077971945262483, + "learning_rate": 3.237344910900648e-05, + "loss": 0.0637, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06713047623634338, + "step": 1555, + "valid_targets_mean": 2498.1, + "valid_targets_min": 497 + }, + { + "epoch": 4.9523809523809526, + "grad_norm": 0.30607354031992073, + "learning_rate": 3.208062761326408e-05, + "loss": 0.0638, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.049679048359394073, + "step": 1560, + "valid_targets_mean": 3139.7, + "valid_targets_min": 180 + }, + { + "epoch": 4.968253968253968, + "grad_norm": 0.4613110134349876, + "learning_rate": 3.178850959866393e-05, + "loss": 0.0681, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06852590292692184, + "step": 1565, + "valid_targets_mean": 2160.6, + "valid_targets_min": 218 + }, + { + "epoch": 4.984126984126984, + "grad_norm": 0.4641619026606205, + "learning_rate": 3.14971065332145e-05, + "loss": 0.0673, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08678565919399261, + "step": 1570, + "valid_targets_mean": 2183.3, + "valid_targets_min": 293 + }, + { + "epoch": 5.0, + "grad_norm": 0.2686689369236618, + "learning_rate": 3.1206429856856706e-05, + "loss": 0.0616, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.032804906368255615, + "step": 1575, + "valid_targets_mean": 2775.0, + "valid_targets_min": 221 + }, + { + "epoch": 5.015873015873016, + "grad_norm": 0.3140840660926085, + "learning_rate": 3.091649098101479e-05, + "loss": 0.0284, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.025703761726617813, + "step": 1580, + "valid_targets_mean": 2469.7, + "valid_targets_min": 278 + }, + { + "epoch": 5.031746031746032, + "grad_norm": 0.36404468154835595, + "learning_rate": 3.06273012881483e-05, + "loss": 0.0264, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02385505847632885, + "step": 1585, + "valid_targets_mean": 2425.8, + "valid_targets_min": 214 + }, + { + "epoch": 5.0476190476190474, + "grad_norm": 0.3298280719616682, + "learning_rate": 3.0338872131305284e-05, + "loss": 0.0308, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03341896831989288, + "step": 1590, + "valid_targets_mean": 2314.9, + "valid_targets_min": 523 + }, + { + "epoch": 5.063492063492063, + "grad_norm": 0.3111368932460093, + "learning_rate": 3.0051214833676545e-05, + "loss": 0.0283, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02781091071665287, + "step": 1595, + "valid_targets_mean": 2255.2, + "valid_targets_min": 202 + }, + { + "epoch": 5.079365079365079, + "grad_norm": 0.30646339884975926, + "learning_rate": 2.9764340688151137e-05, + "loss": 0.0291, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0294380821287632, + "step": 1600, + "valid_targets_mean": 2579.6, + "valid_targets_min": 183 + }, + { + "epoch": 5.095238095238095, + "grad_norm": 0.2988190205497455, + "learning_rate": 2.9478260956873028e-05, + "loss": 0.0292, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.027986258268356323, + "step": 1605, + "valid_targets_mean": 2644.2, + "valid_targets_min": 284 + }, + { + "epoch": 5.111111111111111, + "grad_norm": 0.3063066366941292, + "learning_rate": 2.919298687079895e-05, + "loss": 0.0281, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.027005605399608612, + "step": 1610, + "valid_targets_mean": 2636.3, + "valid_targets_min": 590 + }, + { + "epoch": 5.1269841269841265, + "grad_norm": 0.3555511057401375, + "learning_rate": 2.8908529629257543e-05, + "loss": 0.0279, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02809615060687065, + "step": 1615, + "valid_targets_mean": 2290.8, + "valid_targets_min": 762 + }, + { + "epoch": 5.142857142857143, + "grad_norm": 0.35316659192638694, + "learning_rate": 2.8624900399509603e-05, + "loss": 0.031, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.030622273683547974, + "step": 1620, + "valid_targets_mean": 2129.7, + "valid_targets_min": 198 + }, + { + "epoch": 5.158730158730159, + "grad_norm": 0.2841525620885731, + "learning_rate": 2.8342110316309745e-05, + "loss": 0.0261, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02445690706372261, + "step": 1625, + "valid_targets_mean": 3017.4, + "valid_targets_min": 212 + }, + { + "epoch": 5.174603174603175, + "grad_norm": 0.3555017485792108, + "learning_rate": 2.8060170481469293e-05, + "loss": 0.029, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.030888624489307404, + "step": 1630, + "valid_targets_mean": 2071.3, + "valid_targets_min": 375 + }, + { + "epoch": 5.190476190476191, + "grad_norm": 0.3779371514152101, + "learning_rate": 2.777909196342035e-05, + "loss": 0.027, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02919599786400795, + "step": 1635, + "valid_targets_mean": 2321.6, + "valid_targets_min": 247 + }, + { + "epoch": 5.2063492063492065, + "grad_norm": 0.34150563364580006, + "learning_rate": 2.749888579678138e-05, + "loss": 0.0309, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03463444113731384, + "step": 1640, + "valid_targets_mean": 2960.9, + "valid_targets_min": 229 + }, + { + "epoch": 5.222222222222222, + "grad_norm": 0.31574265117381844, + "learning_rate": 2.721956298192392e-05, + "loss": 0.0267, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.026511382311582565, + "step": 1645, + "valid_targets_mean": 2438.4, + "valid_targets_min": 434 + }, + { + "epoch": 5.238095238095238, + "grad_norm": 0.3279059079372722, + "learning_rate": 2.6941134484540774e-05, + "loss": 0.0287, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.026271160691976547, + "step": 1650, + "valid_targets_mean": 2547.0, + "valid_targets_min": 252 + }, + { + "epoch": 5.253968253968254, + "grad_norm": 0.33281387486584085, + "learning_rate": 2.6663611235215486e-05, + "loss": 0.0276, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.027903877198696136, + "step": 1655, + "valid_targets_mean": 2587.2, + "valid_targets_min": 210 + }, + { + "epoch": 5.26984126984127, + "grad_norm": 0.36940256710470704, + "learning_rate": 2.6387004128993314e-05, + "loss": 0.0293, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03158968314528465, + "step": 1660, + "valid_targets_mean": 2010.4, + "valid_targets_min": 229 + }, + { + "epoch": 5.285714285714286, + "grad_norm": 0.3458373057736632, + "learning_rate": 2.6111324024953378e-05, + "loss": 0.0287, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.030663050711154938, + "step": 1665, + "valid_targets_mean": 2221.1, + "valid_targets_min": 219 + }, + { + "epoch": 5.301587301587301, + "grad_norm": 0.3048745438757399, + "learning_rate": 2.5836581745782475e-05, + "loss": 0.0291, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02931578829884529, + "step": 1670, + "valid_targets_mean": 2349.8, + "valid_targets_min": 145 + }, + { + "epoch": 5.317460317460317, + "grad_norm": 0.3728417222091228, + "learning_rate": 2.556278807735008e-05, + "loss": 0.028, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.030653629451990128, + "step": 1675, + "valid_targets_mean": 2158.1, + "valid_targets_min": 656 + }, + { + "epoch": 5.333333333333333, + "grad_norm": 0.3408278109617429, + "learning_rate": 2.5289953768285092e-05, + "loss": 0.0331, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03102138079702854, + "step": 1680, + "valid_targets_mean": 2369.7, + "valid_targets_min": 217 + }, + { + "epoch": 5.349206349206349, + "grad_norm": 0.3687487969039094, + "learning_rate": 2.501808952955359e-05, + "loss": 0.0303, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03258369863033295, + "step": 1685, + "valid_targets_mean": 2109.6, + "valid_targets_min": 174 + }, + { + "epoch": 5.365079365079365, + "grad_norm": 0.2861073201990759, + "learning_rate": 2.474720603403866e-05, + "loss": 0.0298, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.023980431258678436, + "step": 1690, + "valid_targets_mean": 2762.9, + "valid_targets_min": 557 + }, + { + "epoch": 5.380952380952381, + "grad_norm": 0.2712864184341076, + "learning_rate": 2.447731391612112e-05, + "loss": 0.0266, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022755764424800873, + "step": 1695, + "valid_targets_mean": 3119.6, + "valid_targets_min": 561 + }, + { + "epoch": 5.396825396825397, + "grad_norm": 0.3073556507271004, + "learning_rate": 2.4208423771262238e-05, + "loss": 0.0309, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.026933487504720688, + "step": 1700, + "valid_targets_mean": 2353.7, + "valid_targets_min": 207 + }, + { + "epoch": 5.412698412698413, + "grad_norm": 0.2900894076563369, + "learning_rate": 2.3940546155587618e-05, + "loss": 0.0274, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.025398556143045425, + "step": 1705, + "valid_targets_mean": 2726.5, + "valid_targets_min": 238 + }, + { + "epoch": 5.428571428571429, + "grad_norm": 0.35055271949659134, + "learning_rate": 2.367369158547292e-05, + "loss": 0.0285, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.032015394419431686, + "step": 1710, + "valid_targets_mean": 2754.1, + "valid_targets_min": 270 + }, + { + "epoch": 5.444444444444445, + "grad_norm": 0.2946884140949086, + "learning_rate": 2.3407870537130898e-05, + "loss": 0.0287, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.027290914207696915, + "step": 1715, + "valid_targets_mean": 2720.0, + "valid_targets_min": 588 + }, + { + "epoch": 5.4603174603174605, + "grad_norm": 0.3177677204890575, + "learning_rate": 2.314309344620019e-05, + "loss": 0.0321, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03454054147005081, + "step": 1720, + "valid_targets_mean": 2578.7, + "valid_targets_min": 423 + }, + { + "epoch": 5.476190476190476, + "grad_norm": 0.3140651103378621, + "learning_rate": 2.287937070733557e-05, + "loss": 0.0301, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02843102067708969, + "step": 1725, + "valid_targets_mean": 2753.6, + "valid_targets_min": 232 + }, + { + "epoch": 5.492063492063492, + "grad_norm": 0.2852110492542093, + "learning_rate": 2.2616712673799994e-05, + "loss": 0.0299, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.032928645610809326, + "step": 1730, + "valid_targets_mean": 3194.0, + "valid_targets_min": 381 + }, + { + "epoch": 5.507936507936508, + "grad_norm": 0.3152007569381328, + "learning_rate": 2.2355129657058004e-05, + "loss": 0.0299, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.028028175234794617, + "step": 1735, + "valid_targets_mean": 2532.1, + "valid_targets_min": 215 + }, + { + "epoch": 5.523809523809524, + "grad_norm": 0.3264565583825829, + "learning_rate": 2.2094631926371045e-05, + "loss": 0.0267, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.026344938203692436, + "step": 1740, + "valid_targets_mean": 2423.4, + "valid_targets_min": 263 + }, + { + "epoch": 5.5396825396825395, + "grad_norm": 0.29962953113617385, + "learning_rate": 2.18352297083942e-05, + "loss": 0.03, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.027588889002799988, + "step": 1745, + "valid_targets_mean": 2647.0, + "valid_targets_min": 281 + }, + { + "epoch": 5.555555555555555, + "grad_norm": 0.31674225156212477, + "learning_rate": 2.1576933186774777e-05, + "loss": 0.0311, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03141658380627632, + "step": 1750, + "valid_targets_mean": 2540.5, + "valid_targets_min": 211 + }, + { + "epoch": 5.571428571428571, + "grad_norm": 0.3459407719001157, + "learning_rate": 2.131975250175256e-05, + "loss": 0.0293, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03366846591234207, + "step": 1755, + "valid_targets_mean": 2596.6, + "valid_targets_min": 462 + }, + { + "epoch": 5.587301587301587, + "grad_norm": 0.3596219329813502, + "learning_rate": 2.1063697749761603e-05, + "loss": 0.0309, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03215242177248001, + "step": 1760, + "valid_targets_mean": 2139.4, + "valid_targets_min": 235 + }, + { + "epoch": 5.603174603174603, + "grad_norm": 0.28505873368180973, + "learning_rate": 2.080877898303394e-05, + "loss": 0.0297, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03251330554485321, + "step": 1765, + "valid_targets_mean": 3298.3, + "valid_targets_min": 175 + }, + { + "epoch": 5.619047619047619, + "grad_norm": 0.35027375323684673, + "learning_rate": 2.0555006209204997e-05, + "loss": 0.0282, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.029667403548955917, + "step": 1770, + "valid_targets_mean": 2061.9, + "valid_targets_min": 217 + }, + { + "epoch": 5.634920634920634, + "grad_norm": 0.3172312959769426, + "learning_rate": 2.030238939092059e-05, + "loss": 0.029, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02634127251803875, + "step": 1775, + "valid_targets_mean": 2561.7, + "valid_targets_min": 260 + }, + { + "epoch": 5.650793650793651, + "grad_norm": 0.3019391651691695, + "learning_rate": 2.0050938445445894e-05, + "loss": 0.0282, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03007202222943306, + "step": 1780, + "valid_targets_mean": 3156.0, + "valid_targets_min": 237 + }, + { + "epoch": 5.666666666666667, + "grad_norm": 0.2898602516397339, + "learning_rate": 1.980066324427613e-05, + "loss": 0.0279, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.027274347841739655, + "step": 1785, + "valid_targets_mean": 2647.9, + "valid_targets_min": 490 + }, + { + "epoch": 5.682539682539683, + "grad_norm": 0.3105613596517962, + "learning_rate": 1.9551573612748923e-05, + "loss": 0.032, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02809463068842888, + "step": 1790, + "valid_targets_mean": 2389.3, + "valid_targets_min": 250 + }, + { + "epoch": 5.698412698412699, + "grad_norm": 0.31806935929368724, + "learning_rate": 1.9303679329658723e-05, + "loss": 0.027, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.028683535754680634, + "step": 1795, + "valid_targets_mean": 2275.5, + "valid_targets_min": 228 + }, + { + "epoch": 5.714285714285714, + "grad_norm": 0.2704723825915686, + "learning_rate": 1.905699012687275e-05, + "loss": 0.0282, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.027444101870059967, + "step": 1800, + "valid_targets_mean": 3014.2, + "valid_targets_min": 243 + }, + { + "epoch": 5.73015873015873, + "grad_norm": 0.31924182052190525, + "learning_rate": 1.881151568894909e-05, + "loss": 0.0274, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.027215439826250076, + "step": 1805, + "valid_targets_mean": 2514.9, + "valid_targets_min": 276 + }, + { + "epoch": 5.746031746031746, + "grad_norm": 0.36877991034090674, + "learning_rate": 1.8567265652756378e-05, + "loss": 0.0292, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.034090861678123474, + "step": 1810, + "valid_targets_mean": 2154.2, + "valid_targets_min": 257 + }, + { + "epoch": 5.761904761904762, + "grad_norm": 0.25835434642966726, + "learning_rate": 1.8324249607095534e-05, + "loss": 0.0253, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.023702170699834824, + "step": 1815, + "valid_targets_mean": 3114.4, + "valid_targets_min": 252 + }, + { + "epoch": 5.777777777777778, + "grad_norm": 0.2803461472574647, + "learning_rate": 1.8082477092323297e-05, + "loss": 0.0275, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02496173046529293, + "step": 1820, + "valid_targets_mean": 3060.1, + "valid_targets_min": 375 + }, + { + "epoch": 5.7936507936507935, + "grad_norm": 0.3443429591531802, + "learning_rate": 1.7841957599977755e-05, + "loss": 0.028, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02840775065124035, + "step": 1825, + "valid_targets_mean": 2287.5, + "valid_targets_min": 249 + }, + { + "epoch": 5.809523809523809, + "grad_norm": 0.2888312905070555, + "learning_rate": 1.760270057240559e-05, + "loss": 0.0274, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.026194632053375244, + "step": 1830, + "valid_targets_mean": 2947.8, + "valid_targets_min": 605 + }, + { + "epoch": 5.825396825396825, + "grad_norm": 0.3152453852181073, + "learning_rate": 1.736471540239156e-05, + "loss": 0.0275, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.027604350820183754, + "step": 1835, + "valid_targets_mean": 2781.1, + "valid_targets_min": 380 + }, + { + "epoch": 5.841269841269841, + "grad_norm": 0.3090033027048837, + "learning_rate": 1.712801143278961e-05, + "loss": 0.0299, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.027250243350863457, + "step": 1840, + "valid_targets_mean": 2713.5, + "valid_targets_min": 787 + }, + { + "epoch": 5.857142857142857, + "grad_norm": 0.2748939349164237, + "learning_rate": 1.6892597956156148e-05, + "loss": 0.0258, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02464447170495987, + "step": 1845, + "valid_targets_mean": 2945.8, + "valid_targets_min": 203 + }, + { + "epoch": 5.8730158730158735, + "grad_norm": 0.3114345688110383, + "learning_rate": 1.6658484214385234e-05, + "loss": 0.0257, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02766934409737587, + "step": 1850, + "valid_targets_mean": 2806.9, + "valid_targets_min": 472 + }, + { + "epoch": 5.888888888888889, + "grad_norm": 0.319505523382478, + "learning_rate": 1.6425679398345812e-05, + "loss": 0.0294, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.032601065933704376, + "step": 1855, + "valid_targets_mean": 2578.5, + "valid_targets_min": 689 + }, + { + "epoch": 5.904761904761905, + "grad_norm": 0.3090494248771289, + "learning_rate": 1.619419264752076e-05, + "loss": 0.0267, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.025886746123433113, + "step": 1860, + "valid_targets_mean": 2698.6, + "valid_targets_min": 230 + }, + { + "epoch": 5.920634920634921, + "grad_norm": 0.2757504664867964, + "learning_rate": 1.5964033049648262e-05, + "loss": 0.0264, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.027177568525075912, + "step": 1865, + "valid_targets_mean": 3127.8, + "valid_targets_min": 273 + }, + { + "epoch": 5.936507936507937, + "grad_norm": 0.30787880147040375, + "learning_rate": 1.5735209640364873e-05, + "loss": 0.03, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02610929310321808, + "step": 1870, + "valid_targets_mean": 2715.7, + "valid_targets_min": 367 + }, + { + "epoch": 5.9523809523809526, + "grad_norm": 0.317018977783097, + "learning_rate": 1.5507731402850956e-05, + "loss": 0.0282, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.028238074854016304, + "step": 1875, + "valid_targets_mean": 2175.2, + "valid_targets_min": 333 + }, + { + "epoch": 5.968253968253968, + "grad_norm": 0.29786683112923923, + "learning_rate": 1.528160726747783e-05, + "loss": 0.0263, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02505050227046013, + "step": 1880, + "valid_targets_mean": 2525.8, + "valid_targets_min": 690 + }, + { + "epoch": 5.984126984126984, + "grad_norm": 0.3068808318441955, + "learning_rate": 1.5056846111457407e-05, + "loss": 0.0278, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.028928183019161224, + "step": 1885, + "valid_targets_mean": 2658.9, + "valid_targets_min": 457 + }, + { + "epoch": 6.0, + "grad_norm": 0.18936901520173108, + "learning_rate": 1.483345675849348e-05, + "loss": 0.0249, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014033470302820206, + "step": 1890, + "valid_targets_mean": 2321.6, + "valid_targets_min": 432 + }, + { + "epoch": 6.015873015873016, + "grad_norm": 0.17605673343329853, + "learning_rate": 1.4611447978435478e-05, + "loss": 0.0117, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01046522706747055, + "step": 1895, + "valid_targets_mean": 2653.3, + "valid_targets_min": 265 + }, + { + "epoch": 6.031746031746032, + "grad_norm": 0.23563684210167124, + "learning_rate": 1.439082848693406e-05, + "loss": 0.0111, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01042887382209301, + "step": 1900, + "valid_targets_mean": 3026.1, + "valid_targets_min": 694 + }, + { + "epoch": 6.0476190476190474, + "grad_norm": 0.2712138322874176, + "learning_rate": 1.4171606945099076e-05, + "loss": 0.0106, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.009310530498623848, + "step": 1905, + "valid_targets_mean": 2889.7, + "valid_targets_min": 451 + }, + { + "epoch": 6.063492063492063, + "grad_norm": 0.23207053810076525, + "learning_rate": 1.3953791959159368e-05, + "loss": 0.0108, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.011318333446979523, + "step": 1910, + "valid_targets_mean": 2127.6, + "valid_targets_min": 209 + }, + { + "epoch": 6.079365079365079, + "grad_norm": 0.1734686821369126, + "learning_rate": 1.3737392080125134e-05, + "loss": 0.0113, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.008601278066635132, + "step": 1915, + "valid_targets_mean": 3053.5, + "valid_targets_min": 186 + }, + { + "epoch": 6.095238095238095, + "grad_norm": 0.2216713095036703, + "learning_rate": 1.3522415803452027e-05, + "loss": 0.0111, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01175172533839941, + "step": 1920, + "valid_targets_mean": 2333.3, + "valid_targets_min": 414 + }, + { + "epoch": 6.111111111111111, + "grad_norm": 0.2506028867818434, + "learning_rate": 1.3308871568707798e-05, + "loss": 0.0117, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.012707768939435482, + "step": 1925, + "valid_targets_mean": 2495.1, + "valid_targets_min": 226 + }, + { + "epoch": 6.1269841269841265, + "grad_norm": 0.25064581468073244, + "learning_rate": 1.3096767759240836e-05, + "loss": 0.0109, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.011413747444748878, + "step": 1930, + "valid_targets_mean": 2463.1, + "valid_targets_min": 246 + }, + { + "epoch": 6.142857142857143, + "grad_norm": 0.19975571855872182, + "learning_rate": 1.2886112701851178e-05, + "loss": 0.0109, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.009960673749446869, + "step": 1935, + "valid_targets_mean": 2510.2, + "valid_targets_min": 306 + }, + { + "epoch": 6.158730158730159, + "grad_norm": 0.1929795157959996, + "learning_rate": 1.2676914666463508e-05, + "loss": 0.0118, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01092456467449665, + "step": 1940, + "valid_targets_mean": 2655.2, + "valid_targets_min": 260 + }, + { + "epoch": 6.174603174603175, + "grad_norm": 0.2269843416014114, + "learning_rate": 1.2469181865802576e-05, + "loss": 0.0119, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.011776273138821125, + "step": 1945, + "valid_targets_mean": 2677.8, + "valid_targets_min": 246 + }, + { + "epoch": 6.190476190476191, + "grad_norm": 0.21333472136547466, + "learning_rate": 1.2262922455070719e-05, + "loss": 0.0119, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.011073673143982887, + "step": 1950, + "valid_targets_mean": 2493.4, + "valid_targets_min": 247 + }, + { + "epoch": 6.2063492063492065, + "grad_norm": 0.2103991869212945, + "learning_rate": 1.2058144531627774e-05, + "loss": 0.0119, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.012004425749182701, + "step": 1955, + "valid_targets_mean": 2612.9, + "valid_targets_min": 208 + }, + { + "epoch": 6.222222222222222, + "grad_norm": 0.22518360255990125, + "learning_rate": 1.1854856134673097e-05, + "loss": 0.0108, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.012036746367812157, + "step": 1960, + "valid_targets_mean": 2784.2, + "valid_targets_min": 144 + }, + { + "epoch": 6.238095238095238, + "grad_norm": 0.21486740182774958, + "learning_rate": 1.1653065244930083e-05, + "loss": 0.0113, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.012022515758872032, + "step": 1965, + "valid_targets_mean": 2496.6, + "valid_targets_min": 255 + }, + { + "epoch": 6.253968253968254, + "grad_norm": 0.21498623474744866, + "learning_rate": 1.1452779784332718e-05, + "loss": 0.0126, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.011240034364163876, + "step": 1970, + "valid_targets_mean": 2364.9, + "valid_targets_min": 200 + }, + { + "epoch": 6.26984126984127, + "grad_norm": 0.2205800417367191, + "learning_rate": 1.1254007615714685e-05, + "loss": 0.0106, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013215800747275352, + "step": 1975, + "valid_targets_mean": 2768.8, + "valid_targets_min": 226 + }, + { + "epoch": 6.285714285714286, + "grad_norm": 0.22923678150772725, + "learning_rate": 1.1056756542500613e-05, + "loss": 0.0112, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.011833291500806808, + "step": 1980, + "valid_targets_mean": 2483.2, + "valid_targets_min": 218 + }, + { + "epoch": 6.301587301587301, + "grad_norm": 0.22514925346955136, + "learning_rate": 1.086103430839982e-05, + "loss": 0.0113, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.010705234482884407, + "step": 1985, + "valid_targets_mean": 2367.8, + "valid_targets_min": 549 + }, + { + "epoch": 6.317460317460317, + "grad_norm": 0.24127891040490734, + "learning_rate": 1.066684859710218e-05, + "loss": 0.0136, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.016934359446167946, + "step": 1990, + "valid_targets_mean": 2301.2, + "valid_targets_min": 591 + }, + { + "epoch": 6.333333333333333, + "grad_norm": 0.27521641942693204, + "learning_rate": 1.0474207031976618e-05, + "loss": 0.0122, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01429375633597374, + "step": 1995, + "valid_targets_mean": 1941.5, + "valid_targets_min": 441 + }, + { + "epoch": 6.349206349206349, + "grad_norm": 0.20858326267887373, + "learning_rate": 1.0283117175771701e-05, + "loss": 0.0115, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.011972412467002869, + "step": 2000, + "valid_targets_mean": 2809.1, + "valid_targets_min": 504 + }, + { + "epoch": 6.365079365079365, + "grad_norm": 0.22987155146831525, + "learning_rate": 1.0093586530318849e-05, + "loss": 0.011, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.012161318212747574, + "step": 2005, + "valid_targets_mean": 2442.9, + "valid_targets_min": 423 + }, + { + "epoch": 6.380952380952381, + "grad_norm": 0.24368730006163833, + "learning_rate": 9.905622536237708e-06, + "loss": 0.012, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01644803211092949, + "step": 2010, + "valid_targets_mean": 2265.5, + "valid_targets_min": 448 + }, + { + "epoch": 6.396825396825397, + "grad_norm": 0.238657485161343, + "learning_rate": 9.719232572644187e-06, + "loss": 0.0115, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.00996311567723751, + "step": 2015, + "valid_targets_mean": 2595.9, + "valid_targets_min": 215 + }, + { + "epoch": 6.412698412698413, + "grad_norm": 0.23030608422164262, + "learning_rate": 9.534423956860638e-06, + "loss": 0.0115, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01195448823273182, + "step": 2020, + "valid_targets_mean": 2286.6, + "valid_targets_min": 455 + }, + { + "epoch": 6.428571428571429, + "grad_norm": 0.2044931359054829, + "learning_rate": 9.351203944128694e-06, + "loss": 0.0112, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.011333751492202282, + "step": 2025, + "valid_targets_mean": 2688.3, + "valid_targets_min": 221 + }, + { + "epoch": 6.444444444444445, + "grad_norm": 0.25870237735148743, + "learning_rate": 9.16957972732434e-06, + "loss": 0.0126, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01933901384472847, + "step": 2030, + "valid_targets_mean": 2576.9, + "valid_targets_min": 238 + }, + { + "epoch": 6.4603174603174605, + "grad_norm": 0.20917162837329759, + "learning_rate": 8.989558436675643e-06, + "loss": 0.0112, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.009921345859766006, + "step": 2035, + "valid_targets_mean": 2167.6, + "valid_targets_min": 167 + }, + { + "epoch": 6.476190476190476, + "grad_norm": 0.19509927211956227, + "learning_rate": 8.811147139482745e-06, + "loss": 0.01, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.009080744348466396, + "step": 2040, + "valid_targets_mean": 3169.3, + "valid_targets_min": 443 + }, + { + "epoch": 6.492063492063492, + "grad_norm": 0.28022972028797744, + "learning_rate": 8.634352839840459e-06, + "loss": 0.0117, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013772659935057163, + "step": 2045, + "valid_targets_mean": 1938.8, + "valid_targets_min": 221 + }, + { + "epoch": 6.507936507936508, + "grad_norm": 0.22163844210241243, + "learning_rate": 8.45918247836327e-06, + "loss": 0.0106, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01351403072476387, + "step": 2050, + "valid_targets_mean": 2655.9, + "valid_targets_min": 247 + }, + { + "epoch": 6.523809523809524, + "grad_norm": 0.20650863397246225, + "learning_rate": 8.285642931912918e-06, + "loss": 0.0109, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.012477180920541286, + "step": 2055, + "valid_targets_mean": 2294.6, + "valid_targets_min": 194 + }, + { + "epoch": 6.5396825396825395, + "grad_norm": 0.17302998267802144, + "learning_rate": 8.113741013328352e-06, + "loss": 0.0114, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.008748478256165981, + "step": 2060, + "valid_targets_mean": 3267.9, + "valid_targets_min": 286 + }, + { + "epoch": 6.555555555555555, + "grad_norm": 0.20466694810165378, + "learning_rate": 7.943483471158326e-06, + "loss": 0.0131, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.011773807927966118, + "step": 2065, + "valid_targets_mean": 3319.4, + "valid_targets_min": 223 + }, + { + "epoch": 6.571428571428571, + "grad_norm": 0.2934663087671904, + "learning_rate": 7.774876989396434e-06, + "loss": 0.0133, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.012026914395391941, + "step": 2070, + "valid_targets_mean": 1829.4, + "valid_targets_min": 175 + }, + { + "epoch": 6.587301587301587, + "grad_norm": 0.20031635292275735, + "learning_rate": 7.607928187218699e-06, + "loss": 0.0106, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.010328968986868858, + "step": 2075, + "valid_targets_mean": 2685.2, + "valid_targets_min": 159 + }, + { + "epoch": 6.603174603174603, + "grad_norm": 0.24134839237872796, + "learning_rate": 7.442643618723777e-06, + "loss": 0.011, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.010953256860375404, + "step": 2080, + "valid_targets_mean": 2626.5, + "valid_targets_min": 291 + }, + { + "epoch": 6.619047619047619, + "grad_norm": 0.2025863859066666, + "learning_rate": 7.2790297726755716e-06, + "loss": 0.0111, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.010732797905802727, + "step": 2085, + "valid_targets_mean": 2580.7, + "valid_targets_min": 471 + }, + { + "epoch": 6.634920634920634, + "grad_norm": 0.23553935825230604, + "learning_rate": 7.117093072248571e-06, + "loss": 0.01, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.010928872972726822, + "step": 2090, + "valid_targets_mean": 2736.0, + "valid_targets_min": 408 + }, + { + "epoch": 6.650793650793651, + "grad_norm": 0.19960886664906605, + "learning_rate": 6.9568398747756396e-06, + "loss": 0.0109, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.010096733458340168, + "step": 2095, + "valid_targets_mean": 2788.3, + "valid_targets_min": 370 + }, + { + "epoch": 6.666666666666667, + "grad_norm": 0.1914465717799404, + "learning_rate": 6.798276471498444e-06, + "loss": 0.0117, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.010824847966432571, + "step": 2100, + "valid_targets_mean": 2939.1, + "valid_targets_min": 174 + }, + { + "epoch": 6.682539682539683, + "grad_norm": 0.19170659016162409, + "learning_rate": 6.6414090873204886e-06, + "loss": 0.0109, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.010004542768001556, + "step": 2105, + "valid_targets_mean": 2455.1, + "valid_targets_min": 211 + }, + { + "epoch": 6.698412698412699, + "grad_norm": 0.1805363319621809, + "learning_rate": 6.486243880562759e-06, + "loss": 0.0114, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.008439445868134499, + "step": 2110, + "valid_targets_mean": 2589.2, + "valid_targets_min": 391 + }, + { + "epoch": 6.714285714285714, + "grad_norm": 0.2715647431366819, + "learning_rate": 6.3327869427218855e-06, + "loss": 0.0115, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.011712701991200447, + "step": 2115, + "valid_targets_mean": 2013.0, + "valid_targets_min": 258 + }, + { + "epoch": 6.73015873015873, + "grad_norm": 0.24288701193316264, + "learning_rate": 6.181044298231081e-06, + "loss": 0.0113, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.011673448607325554, + "step": 2120, + "valid_targets_mean": 2099.5, + "valid_targets_min": 210 + }, + { + "epoch": 6.746031746031746, + "grad_norm": 0.21691591263634646, + "learning_rate": 6.031021904223572e-06, + "loss": 0.0109, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.009856434538960457, + "step": 2125, + "valid_targets_mean": 2200.2, + "valid_targets_min": 223 + }, + { + "epoch": 6.761904761904762, + "grad_norm": 0.20222435286504, + "learning_rate": 5.882725650298787e-06, + "loss": 0.0118, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.011087803170084953, + "step": 2130, + "valid_targets_mean": 2714.8, + "valid_targets_min": 314 + }, + { + "epoch": 6.777777777777778, + "grad_norm": 0.17959613252458959, + "learning_rate": 5.736161358291092e-06, + "loss": 0.0122, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.009143758565187454, + "step": 2135, + "valid_targets_mean": 2677.0, + "valid_targets_min": 220 + }, + { + "epoch": 6.7936507936507935, + "grad_norm": 0.15191164674188265, + "learning_rate": 5.5913347820412635e-06, + "loss": 0.0102, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.00925508700311184, + "step": 2140, + "valid_targets_mean": 3537.6, + "valid_targets_min": 411 + }, + { + "epoch": 6.809523809523809, + "grad_norm": 0.20980074648261804, + "learning_rate": 5.44825160717059e-06, + "loss": 0.0112, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.010102735832333565, + "step": 2145, + "valid_targets_mean": 2676.0, + "valid_targets_min": 229 + }, + { + "epoch": 6.825396825396825, + "grad_norm": 0.20241979048342373, + "learning_rate": 5.306917450857702e-06, + "loss": 0.0121, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.009532380849123001, + "step": 2150, + "valid_targets_mean": 2785.3, + "valid_targets_min": 359 + }, + { + "epoch": 6.841269841269841, + "grad_norm": 0.22345477410694015, + "learning_rate": 5.167337861617982e-06, + "loss": 0.0106, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01100766658782959, + "step": 2155, + "valid_targets_mean": 2360.5, + "valid_targets_min": 440 + }, + { + "epoch": 6.857142857142857, + "grad_norm": 0.24458864041009826, + "learning_rate": 5.029518319085824e-06, + "loss": 0.0107, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013142355717718601, + "step": 2160, + "valid_targets_mean": 2065.7, + "valid_targets_min": 205 + }, + { + "epoch": 6.8730158730158735, + "grad_norm": 0.1874370840823665, + "learning_rate": 4.893464233799433e-06, + "loss": 0.011, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.008828729391098022, + "step": 2165, + "valid_targets_mean": 2836.4, + "valid_targets_min": 129 + }, + { + "epoch": 6.888888888888889, + "grad_norm": 0.199563988301068, + "learning_rate": 4.759180946988495e-06, + "loss": 0.0104, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.009349266067147255, + "step": 2170, + "valid_targets_mean": 2558.2, + "valid_targets_min": 219 + }, + { + "epoch": 6.904761904761905, + "grad_norm": 0.16160227366006275, + "learning_rate": 4.626673730364395e-06, + "loss": 0.0105, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.009358936920762062, + "step": 2175, + "valid_targets_mean": 3867.7, + "valid_targets_min": 263 + }, + { + "epoch": 6.920634920634921, + "grad_norm": 0.15244700440592213, + "learning_rate": 4.495947785913368e-06, + "loss": 0.0109, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.007771562784910202, + "step": 2180, + "valid_targets_mean": 3237.5, + "valid_targets_min": 560 + }, + { + "epoch": 6.936507936507937, + "grad_norm": 0.26478339000384865, + "learning_rate": 4.367008245692189e-06, + "loss": 0.0107, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.012884551659226418, + "step": 2185, + "valid_targets_mean": 2510.4, + "valid_targets_min": 414 + }, + { + "epoch": 6.9523809523809526, + "grad_norm": 0.18984677055490956, + "learning_rate": 4.239860171626769e-06, + "loss": 0.0097, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.008139668963849545, + "step": 2190, + "valid_targets_mean": 2885.7, + "valid_targets_min": 426 + }, + { + "epoch": 6.968253968253968, + "grad_norm": 0.16937372787465557, + "learning_rate": 4.114508555313351e-06, + "loss": 0.0095, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.00903971865773201, + "step": 2195, + "valid_targets_mean": 3300.9, + "valid_targets_min": 639 + }, + { + "epoch": 6.984126984126984, + "grad_norm": 0.2391388370052485, + "learning_rate": 3.990958317822663e-06, + "loss": 0.011, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.011284902691841125, + "step": 2200, + "valid_targets_mean": 2528.2, + "valid_targets_min": 741 + }, + { + "epoch": 7.0, + "grad_norm": 0.15573025226233533, + "learning_rate": 3.86921430950658e-06, + "loss": 0.0094, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.007273214869201183, + "step": 2205, + "valid_targets_mean": 1767.8, + "valid_targets_min": 195 + }, + { + "epoch": 7.015873015873016, + "grad_norm": 0.08459819359822927, + "learning_rate": 3.7492813098078506e-06, + "loss": 0.0051, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004210666287690401, + "step": 2210, + "valid_targets_mean": 2978.2, + "valid_targets_min": 176 + }, + { + "epoch": 7.031746031746032, + "grad_norm": 0.1214039436488733, + "learning_rate": 3.6311640270723757e-06, + "loss": 0.0052, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0047843605279922485, + "step": 2215, + "valid_targets_mean": 2487.0, + "valid_targets_min": 273 + }, + { + "epoch": 7.0476190476190474, + "grad_norm": 0.11601095671466669, + "learning_rate": 3.5148670983644104e-06, + "loss": 0.0048, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004126162733882666, + "step": 2220, + "valid_targets_mean": 2622.6, + "valid_targets_min": 227 + }, + { + "epoch": 7.063492063492063, + "grad_norm": 0.12003046421848582, + "learning_rate": 3.400395089284475e-06, + "loss": 0.0049, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004347827285528183, + "step": 2225, + "valid_targets_mean": 3081.9, + "valid_targets_min": 713 + }, + { + "epoch": 7.079365079365079, + "grad_norm": 0.14924750288287142, + "learning_rate": 3.287752493790186e-06, + "loss": 0.0047, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.005834147799760103, + "step": 2230, + "valid_targets_mean": 2144.2, + "valid_targets_min": 348 + }, + { + "epoch": 7.095238095238095, + "grad_norm": 0.09952057508796572, + "learning_rate": 3.1769437340197715e-06, + "loss": 0.0047, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004201963543891907, + "step": 2235, + "valid_targets_mean": 3046.9, + "valid_targets_min": 245 + }, + { + "epoch": 7.111111111111111, + "grad_norm": 0.15615482637774525, + "learning_rate": 3.067973160118498e-06, + "loss": 0.005, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0062007359229028225, + "step": 2240, + "valid_targets_mean": 2015.1, + "valid_targets_min": 142 + }, + { + "epoch": 7.1269841269841265, + "grad_norm": 0.1635109175720187, + "learning_rate": 2.9608450500678565e-06, + "loss": 0.0049, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0050523485988378525, + "step": 2245, + "valid_targets_mean": 2320.6, + "valid_targets_min": 272 + }, + { + "epoch": 7.142857142857143, + "grad_norm": 0.10760721467766973, + "learning_rate": 2.8555636095176975e-06, + "loss": 0.0043, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0043999748304486275, + "step": 2250, + "valid_targets_mean": 2599.4, + "valid_targets_min": 588 + }, + { + "epoch": 7.158730158730159, + "grad_norm": 0.11385015871950585, + "learning_rate": 2.7521329716210074e-06, + "loss": 0.0046, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0040392447263002396, + "step": 2255, + "valid_targets_mean": 2960.9, + "valid_targets_min": 252 + }, + { + "epoch": 7.174603174603175, + "grad_norm": 0.13181090938872578, + "learning_rate": 2.6505571968717725e-06, + "loss": 0.0049, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004630320239812136, + "step": 2260, + "valid_targets_mean": 2799.0, + "valid_targets_min": 733 + }, + { + "epoch": 7.190476190476191, + "grad_norm": 0.11249118288491969, + "learning_rate": 2.550840272945465e-06, + "loss": 0.005, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004065143410116434, + "step": 2265, + "valid_targets_mean": 2601.0, + "valid_targets_min": 195 + }, + { + "epoch": 7.2063492063492065, + "grad_norm": 0.13271491386896409, + "learning_rate": 2.4529861145425605e-06, + "loss": 0.0045, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.00510437274351716, + "step": 2270, + "valid_targets_mean": 2068.3, + "valid_targets_min": 375 + }, + { + "epoch": 7.222222222222222, + "grad_norm": 0.13083058562272745, + "learning_rate": 2.3569985632348247e-06, + "loss": 0.0048, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004860528279095888, + "step": 2275, + "valid_targets_mean": 2681.0, + "valid_targets_min": 129 + }, + { + "epoch": 7.238095238095238, + "grad_norm": 0.09133491502001839, + "learning_rate": 2.2628813873145303e-06, + "loss": 0.0042, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0034794225357472897, + "step": 2280, + "valid_targets_mean": 2977.8, + "valid_targets_min": 283 + }, + { + "epoch": 7.253968253968254, + "grad_norm": 0.10260359167143555, + "learning_rate": 2.17063828164647e-06, + "loss": 0.0048, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004080589860677719, + "step": 2285, + "valid_targets_mean": 2689.7, + "valid_targets_min": 180 + }, + { + "epoch": 7.26984126984127, + "grad_norm": 0.1277466824779668, + "learning_rate": 2.0802728675229587e-06, + "loss": 0.0052, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.005505088251084089, + "step": 2290, + "valid_targets_mean": 2520.8, + "valid_targets_min": 192 + }, + { + "epoch": 7.285714285714286, + "grad_norm": 0.13414799604012986, + "learning_rate": 1.9917886925216234e-06, + "loss": 0.0054, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004918585065752268, + "step": 2295, + "valid_targets_mean": 2286.5, + "valid_targets_min": 174 + }, + { + "epoch": 7.301587301587301, + "grad_norm": 0.14341288741537975, + "learning_rate": 1.9051892303661834e-06, + "loss": 0.0049, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.005789702758193016, + "step": 2300, + "valid_targets_mean": 2142.9, + "valid_targets_min": 261 + }, + { + "epoch": 7.317460317460317, + "grad_norm": 0.16263356099794266, + "learning_rate": 1.8204778807900003e-06, + "loss": 0.0054, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.008312879130244255, + "step": 2305, + "valid_targets_mean": 2593.2, + "valid_targets_min": 265 + }, + { + "epoch": 7.333333333333333, + "grad_norm": 0.11500848498724002, + "learning_rate": 1.7376579694026896e-06, + "loss": 0.0053, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.00444476492702961, + "step": 2310, + "valid_targets_mean": 2889.4, + "valid_targets_min": 597 + }, + { + "epoch": 7.349206349206349, + "grad_norm": 0.11386148006656191, + "learning_rate": 1.6567327475595195e-06, + "loss": 0.0046, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.00467523280531168, + "step": 2315, + "valid_targets_mean": 2715.1, + "valid_targets_min": 255 + }, + { + "epoch": 7.365079365079365, + "grad_norm": 0.10564663344972974, + "learning_rate": 1.577705392233797e-06, + "loss": 0.0044, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0039878347888588905, + "step": 2320, + "valid_targets_mean": 2738.9, + "valid_targets_min": 479 + }, + { + "epoch": 7.380952380952381, + "grad_norm": 0.14596466349940437, + "learning_rate": 1.5005790058920943e-06, + "loss": 0.0046, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.00482321297749877, + "step": 2325, + "valid_targets_mean": 2278.7, + "valid_targets_min": 440 + }, + { + "epoch": 7.396825396825397, + "grad_norm": 0.15815995174795025, + "learning_rate": 1.4253566163725252e-06, + "loss": 0.0054, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.008869815617799759, + "step": 2330, + "valid_targets_mean": 2520.8, + "valid_targets_min": 471 + }, + { + "epoch": 7.412698412698413, + "grad_norm": 0.15219972127206582, + "learning_rate": 1.3520411767658059e-06, + "loss": 0.005, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004969125613570213, + "step": 2335, + "valid_targets_mean": 2238.5, + "valid_targets_min": 196 + }, + { + "epoch": 7.428571428571429, + "grad_norm": 0.12679396786089758, + "learning_rate": 1.2806355652993762e-06, + "loss": 0.0043, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.00437959237024188, + "step": 2340, + "valid_targets_mean": 2472.3, + "valid_targets_min": 429 + }, + { + "epoch": 7.444444444444445, + "grad_norm": 0.13239588431770893, + "learning_rate": 1.2111425852243785e-06, + "loss": 0.0046, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0049995798617601395, + "step": 2345, + "valid_targets_mean": 2110.2, + "valid_targets_min": 199 + }, + { + "epoch": 7.4603174603174605, + "grad_norm": 0.17194950905826895, + "learning_rate": 1.143564964705618e-06, + "loss": 0.0052, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.005149936769157648, + "step": 2350, + "valid_targets_mean": 2832.6, + "valid_targets_min": 203 + }, + { + "epoch": 7.476190476190476, + "grad_norm": 0.13042084506866755, + "learning_rate": 1.0779053567144427e-06, + "loss": 0.0047, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004457908682525158, + "step": 2355, + "valid_targets_mean": 2479.3, + "valid_targets_min": 205 + }, + { + "epoch": 7.492063492063492, + "grad_norm": 0.12420165067189844, + "learning_rate": 1.014166338924627e-06, + "loss": 0.0052, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004359714686870575, + "step": 2360, + "valid_targets_mean": 2606.1, + "valid_targets_min": 281 + }, + { + "epoch": 7.507936507936508, + "grad_norm": 0.12727412862064016, + "learning_rate": 9.523504136111306e-07, + "loss": 0.0045, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.005046484526246786, + "step": 2365, + "valid_targets_mean": 2563.5, + "valid_targets_min": 348 + }, + { + "epoch": 7.523809523809524, + "grad_norm": 0.16367637185707307, + "learning_rate": 8.92460007551904e-07, + "loss": 0.0057, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.008848870173096657, + "step": 2370, + "valid_targets_mean": 2358.9, + "valid_targets_min": 141 + }, + { + "epoch": 7.5396825396825395, + "grad_norm": 0.1377201736543532, + "learning_rate": 8.344974719326104e-07, + "loss": 0.0051, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.005841254256665707, + "step": 2375, + "valid_targets_mean": 2283.1, + "valid_targets_min": 713 + }, + { + "epoch": 7.555555555555555, + "grad_norm": 0.13900088142067465, + "learning_rate": 7.784650822542871e-07, + "loss": 0.0045, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.005029533989727497, + "step": 2380, + "valid_targets_mean": 2469.8, + "valid_targets_min": 203 + }, + { + "epoch": 7.571428571428571, + "grad_norm": 0.13895281918793312, + "learning_rate": 7.243650382440736e-07, + "loss": 0.0049, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004304093308746815, + "step": 2385, + "valid_targets_mean": 2505.7, + "valid_targets_min": 249 + }, + { + "epoch": 7.587301587301587, + "grad_norm": 0.1117210879178123, + "learning_rate": 6.721994637687967e-07, + "loss": 0.0046, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0043164994567632675, + "step": 2390, + "valid_targets_mean": 2568.0, + "valid_targets_min": 218 + }, + { + "epoch": 7.603174603174603, + "grad_norm": 0.16437369019465173, + "learning_rate": 6.219704067516374e-07, + "loss": 0.0054, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0057275728322565556, + "step": 2395, + "valid_targets_mean": 1991.8, + "valid_targets_min": 247 + }, + { + "epoch": 7.619047619047619, + "grad_norm": 0.11386782810781054, + "learning_rate": 5.736798390916898e-07, + "loss": 0.0055, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004491021391004324, + "step": 2400, + "valid_targets_mean": 2678.6, + "valid_targets_min": 194 + }, + { + "epoch": 7.634920634920634, + "grad_norm": 0.13886087588084992, + "learning_rate": 5.273296565865948e-07, + "loss": 0.005, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.005650711245834827, + "step": 2405, + "valid_targets_mean": 2258.8, + "valid_targets_min": 260 + }, + { + "epoch": 7.650793650793651, + "grad_norm": 0.10885128280665608, + "learning_rate": 4.829216788580726e-07, + "loss": 0.005, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004098647274076939, + "step": 2410, + "valid_targets_mean": 2962.7, + "valid_targets_min": 472 + }, + { + "epoch": 7.666666666666667, + "grad_norm": 0.10939793049463822, + "learning_rate": 4.404576492805179e-07, + "loss": 0.0057, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.00415010005235672, + "step": 2415, + "valid_targets_mean": 3144.0, + "valid_targets_min": 424 + }, + { + "epoch": 7.682539682539683, + "grad_norm": 0.10557362311620923, + "learning_rate": 3.999392349125386e-07, + "loss": 0.0044, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0036467863246798515, + "step": 2420, + "valid_targets_mean": 2919.8, + "valid_targets_min": 210 + }, + { + "epoch": 7.698412698412699, + "grad_norm": 0.11064503974604914, + "learning_rate": 3.613680264315189e-07, + "loss": 0.0044, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0040693264454603195, + "step": 2425, + "valid_targets_mean": 2565.4, + "valid_targets_min": 295 + }, + { + "epoch": 7.714285714285714, + "grad_norm": 0.1568796098379975, + "learning_rate": 3.247455380711806e-07, + "loss": 0.0045, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.005382425617426634, + "step": 2430, + "valid_targets_mean": 2266.0, + "valid_targets_min": 413 + }, + { + "epoch": 7.73015873015873, + "grad_norm": 0.11322868471768423, + "learning_rate": 2.900732075621082e-07, + "loss": 0.0053, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004472393076866865, + "step": 2435, + "valid_targets_mean": 2511.2, + "valid_targets_min": 245 + }, + { + "epoch": 7.746031746031746, + "grad_norm": 0.11150153238491746, + "learning_rate": 2.5735239607534434e-07, + "loss": 0.0045, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.003644701559096575, + "step": 2440, + "valid_targets_mean": 2734.3, + "valid_targets_min": 167 + }, + { + "epoch": 7.761904761904762, + "grad_norm": 0.13106156116851986, + "learning_rate": 2.2658438816892112e-07, + "loss": 0.0047, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004551380407065153, + "step": 2445, + "valid_targets_mean": 2632.0, + "valid_targets_min": 282 + }, + { + "epoch": 7.777777777777778, + "grad_norm": 0.09916014013948114, + "learning_rate": 1.9777039173746182e-07, + "loss": 0.0047, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004065266344696283, + "step": 2450, + "valid_targets_mean": 2905.0, + "valid_targets_min": 289 + }, + { + "epoch": 7.7936507936507935, + "grad_norm": 0.12565892852475183, + "learning_rate": 1.709115379647186e-07, + "loss": 0.0045, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004258735105395317, + "step": 2455, + "valid_targets_mean": 2910.7, + "valid_targets_min": 356 + }, + { + "epoch": 7.809523809523809, + "grad_norm": 0.1209405433734022, + "learning_rate": 1.460088812792082e-07, + "loss": 0.0048, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004565816838294268, + "step": 2460, + "valid_targets_mean": 2379.2, + "valid_targets_min": 265 + }, + { + "epoch": 7.825396825396825, + "grad_norm": 0.10695549919558085, + "learning_rate": 1.2306339931279499e-07, + "loss": 0.0051, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004238954745233059, + "step": 2465, + "valid_targets_mean": 2483.4, + "valid_targets_min": 234 + }, + { + "epoch": 7.841269841269841, + "grad_norm": 0.1439052511761759, + "learning_rate": 1.0207599286229941e-07, + "loss": 0.0048, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004376878961920738, + "step": 2470, + "valid_targets_mean": 2493.4, + "valid_targets_min": 195 + }, + { + "epoch": 7.857142857142857, + "grad_norm": 0.10749987603452196, + "learning_rate": 8.304748585417078e-08, + "loss": 0.0046, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.005094209220260382, + "step": 2475, + "valid_targets_mean": 2638.9, + "valid_targets_min": 223 + }, + { + "epoch": 7.8730158730158735, + "grad_norm": 0.1524185431933641, + "learning_rate": 6.597862531210197e-08, + "loss": 0.005, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.005541890859603882, + "step": 2480, + "valid_targets_mean": 2127.6, + "valid_targets_min": 236 + }, + { + "epoch": 7.888888888888889, + "grad_norm": 0.1413530465408386, + "learning_rate": 5.0870081327725194e-08, + "loss": 0.0045, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0046760221011936665, + "step": 2485, + "valid_targets_mean": 2399.0, + "valid_targets_min": 191 + }, + { + "epoch": 7.904761904761905, + "grad_norm": 0.1225197638005495, + "learning_rate": 3.7722447034305164e-08, + "loss": 0.0048, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0052412888035178185, + "step": 2490, + "valid_targets_mean": 2790.5, + "valid_targets_min": 194 + }, + { + "epoch": 7.920634920634921, + "grad_norm": 0.12110387602854578, + "learning_rate": 2.653623858344667e-08, + "loss": 0.0045, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.004263537935912609, + "step": 2495, + "valid_targets_mean": 2512.3, + "valid_targets_min": 270 + }, + { + "epoch": 7.936507936507937, + "grad_norm": 0.11858762290964402, + "learning_rate": 1.731189512482745e-08, + "loss": 0.0042, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.003940538503229618, + "step": 2500, + "valid_targets_mean": 2651.2, + "valid_targets_min": 335 + }, + { + "epoch": 7.9523809523809526, + "grad_norm": 0.10032930990230934, + "learning_rate": 1.0049778788967513e-08, + "loss": 0.0049, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0044016847386956215, + "step": 2505, + "valid_targets_mean": 2673.6, + "valid_targets_min": 220 + }, + { + "epoch": 7.968253968253968, + "grad_norm": 0.17880832570037253, + "learning_rate": 4.750174673018304e-09, + "loss": 0.0049, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.005777041893452406, + "step": 2510, + "valid_targets_mean": 2168.6, + "valid_targets_min": 380 + }, + { + "epoch": 7.984126984126984, + "grad_norm": 0.101799008142305, + "learning_rate": 1.4132908295549919e-09, + "loss": 0.0048, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.003868833649903536, + "step": 2515, + "valid_targets_mean": 2913.7, + "valid_targets_min": 247 + }, + { + "epoch": 8.0, + "grad_norm": 0.08617708171079032, + "learning_rate": 3.925825840522812e-11, + "loss": 0.0044, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.003110480960458517, + "step": 2520, + "valid_targets_mean": 2280.5, + "valid_targets_min": 275 + }, + { + "epoch": 8.0, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.003110480960458517, + "step": 2520, + "total_flos": 1155060078084096.0, + "train_loss": 0.14387717474665906, + "train_runtime": 35033.3193, + "train_samples_per_second": 2.295, + "train_steps_per_second": 0.072, + "valid_targets_mean": 2280.5, + "valid_targets_min": 275 + } + ], + "logging_steps": 5, + "max_steps": 2520, + "num_input_tokens_seen": 0, + "num_train_epochs": 8, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 1155060078084096.0, + "train_batch_size": 1, + "trial_name": null, + "trial_params": null +} diff --git a/training_args.bin b/training_args.bin new file mode 100644 index 0000000..685900f --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:faacfc9dc49c9b9c03e106bfe6543a7f09b6c0b589206d8e46c8f3400b909ef6 +size 8785 diff --git a/training_loss.png b/training_loss.png new file mode 100644 index 0000000..883e95e Binary files /dev/null and b/training_loss.png differ diff --git a/vocab.json b/vocab.json new file mode 100644 index 0000000..6c49fc6 --- /dev/null +++ b/vocab.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ca10d7e9fb3ed18575dd1e277a2579c16d108e32f27439684afa0e10b1440910 +size 2776833