commit 18ec8af48592074db427ebf1b38733569778caeb Author: ModelHub XC Date: Thu Aug 27 03:09:13 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: laion/Qwen3-8B_perturbed-docker-exp-taskmaster2-tasks_glm_4.7_traces_locetash_save-strategy_steps Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..a403a9d --- /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 + +merges.txt filter=lfs diff=lfs merge=lfs -text +model-00002-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +model-00001-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +model-00003-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text +vocab.json filter=lfs diff=lfs merge=lfs -text +training_args.bin filter=lfs diff=lfs merge=lfs -text +model-00004-of-00004.safetensors 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..cff8508 --- /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_perturbed-docker-exp-taskmaster2-tasks_glm_4.7_traces_locetash_save-strategy_steps + results: [] +--- + + + +# Qwen3-8B_perturbed-docker-exp-taskmaster2-tasks_glm_4.7_traces_locetash_save-strategy_steps + +This model is a fine-tuned version of [Qwen/Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B) on the DCAgent/perturbed-docker-exp-taskmaster2-tasks_glm_4.7_traces_locetash 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..654b8eb --- /dev/null +++ b/all_results.json @@ -0,0 +1,16 @@ +{ + "achieved_tflops_per_gpu": 0.001239108303511846, + "achieved_tflops_per_gpu_theoretical": 196.49846464374374, + "epoch": 8.0, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02538919635117054, + "mfu_percent": 0.0003971500972794378, + "mfu_percent_theoretical": 62.98027712940505, + "total_flos": 1518927301312512.0, + "train_loss": 0.13197601472825876, + "train_runtime": 38306.9648, + "train_samples_per_second": 1.867, + "train_steps_per_second": 0.058, + "valid_targets_mean": 3632.8, + "valid_targets_min": 732 +} \ 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..2b8fbc3 --- /dev/null +++ b/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ca0494b4fc1f8a7ca5197e5b3b98c9441c11ee47161d8d44691e27c6ab13ee17 +size 4902257696 diff --git a/model-00002-of-00004.safetensors b/model-00002-of-00004.safetensors new file mode 100644 index 0000000..c84d95d --- /dev/null +++ b/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:290df2ea1a45d93727d7a4c08531b2cb72f9d4d7f7914e6944fe816bd76a63d3 +size 4915960368 diff --git a/model-00003-of-00004.safetensors b/model-00003-of-00004.safetensors new file mode 100644 index 0000000..005f322 --- /dev/null +++ b/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2292abfbd3838cb25b58ac8c93cf5078689fef5108787a5df0d604b5f9327b12 +size 4983068496 diff --git a/model-00004-of-00004.safetensors b/model-00004-of-00004.safetensors new file mode 100644 index 0000000..e7b87d3 --- /dev/null +++ b/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:85f792f1a4c471db927690dd52433940dfb996c434dd37e705d74bf1f5b2d996 +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..ed0e1d8 --- /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/perturbed-docker-exp-taskmaster2-tasks_glm_4.7_traces_locetash", + "training_type": "SFT", + "training_parameters": "https://huggingface.co/laion/Qwen3-8B_perturbed-docker-exp-taskmaster2-tasks_glm_4.7_traces_locetash_save-strategy_steps/blob/main/config.json", + "wandb_link": "https://wandb.ai/dogml/dc-agent/runs/Qwen3-8B_perturbed-docker-exp-taskmaster2-tasks_glm_4.7_traces_locetash_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..654b8eb --- /dev/null +++ b/train_results.json @@ -0,0 +1,16 @@ +{ + "achieved_tflops_per_gpu": 0.001239108303511846, + "achieved_tflops_per_gpu_theoretical": 196.49846464374374, + "epoch": 8.0, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02538919635117054, + "mfu_percent": 0.0003971500972794378, + "mfu_percent_theoretical": 62.98027712940505, + "total_flos": 1518927301312512.0, + "train_loss": 0.13197601472825876, + "train_runtime": 38306.9648, + "train_samples_per_second": 1.867, + "train_steps_per_second": 0.058, + "valid_targets_mean": 3632.8, + "valid_targets_min": 732 +} \ No newline at end of file diff --git a/trainer_log.jsonl b/trainer_log.jsonl new file mode 100644 index 0000000..4e69267 --- /dev/null +++ b/trainer_log.jsonl @@ -0,0 +1,469 @@ +{"current_steps": 5, "total_steps": 2240, "loss": 0.7589, "lr": 3.3333333333333335e-05, "epoch": 0.017857142857142856, "percentage": 0.22, "elapsed_time": "0:05:25", "remaining_time": "1 day, 16:21:44"} +{"current_steps": 10, "total_steps": 2240, "loss": 0.5543, "lr": 7.500000000000001e-05, "epoch": 0.03571428571428571, "percentage": 0.45, "elapsed_time": "0:07:42", "remaining_time": "1 day, 4:39:36"} +{"current_steps": 15, "total_steps": 2240, "loss": 0.4812, "lr": 9.999980117587285e-05, "epoch": 0.05357142857142857, "percentage": 0.67, "elapsed_time": "0:09:42", "remaining_time": "23:59:36"} +{"current_steps": 20, "total_steps": 2240, "loss": 0.4195, "lr": 9.99975644226019e-05, "epoch": 0.07142857142857142, "percentage": 0.89, "elapsed_time": "0:11:26", "remaining_time": "21:09:31"} +{"current_steps": 25, "total_steps": 2240, "loss": 0.3984, "lr": 9.999284249745142e-05, "epoch": 0.08928571428571429, "percentage": 1.12, "elapsed_time": "0:13:20", "remaining_time": "19:41:25"} +{"current_steps": 30, "total_steps": 2240, "loss": 0.3894, "lr": 9.998563563512873e-05, "epoch": 0.10714285714285714, "percentage": 1.34, "elapsed_time": "0:14:52", "remaining_time": "18:15:37"} +{"current_steps": 35, "total_steps": 2240, "loss": 0.3957, "lr": 9.997594419385712e-05, "epoch": 0.125, "percentage": 1.56, "elapsed_time": "0:16:21", "remaining_time": "17:10:14"} +{"current_steps": 40, "total_steps": 2240, "loss": 0.373, "lr": 9.996376865535801e-05, "epoch": 0.14285714285714285, "percentage": 1.79, "elapsed_time": "0:17:53", "remaining_time": "16:23:43"} +{"current_steps": 45, "total_steps": 2240, "loss": 0.3651, "lr": 9.9949109624827e-05, "epoch": 0.16071428571428573, "percentage": 2.01, "elapsed_time": "0:19:12", "remaining_time": "15:36:56"} +{"current_steps": 50, "total_steps": 2240, "loss": 0.3637, "lr": 9.993196783090377e-05, "epoch": 0.17857142857142858, "percentage": 2.23, "elapsed_time": "0:20:36", "remaining_time": "15:02:27"} +{"current_steps": 55, "total_steps": 2240, "loss": 0.3527, "lr": 9.991234412563593e-05, "epoch": 0.19642857142857142, "percentage": 2.46, "elapsed_time": "0:21:59", "remaining_time": "14:33:48"} +{"current_steps": 60, "total_steps": 2240, "loss": 0.3611, "lr": 9.989023948443662e-05, "epoch": 0.21428571428571427, "percentage": 2.68, "elapsed_time": "0:23:24", "remaining_time": "14:10:17"} +{"current_steps": 65, "total_steps": 2240, "loss": 0.3648, "lr": 9.986565500603598e-05, "epoch": 0.23214285714285715, "percentage": 2.9, "elapsed_time": "0:24:46", "remaining_time": "13:48:52"} +{"current_steps": 70, "total_steps": 2240, "loss": 0.3618, "lr": 9.983859191242661e-05, "epoch": 0.25, "percentage": 3.12, "elapsed_time": "0:26:15", "remaining_time": "13:33:57"} +{"current_steps": 75, "total_steps": 2240, "loss": 0.3619, "lr": 9.980905154880288e-05, "epoch": 0.26785714285714285, "percentage": 3.35, "elapsed_time": "0:27:36", "remaining_time": "13:16:43"} +{"current_steps": 80, "total_steps": 2240, "loss": 0.3496, "lr": 9.97770353834939e-05, "epoch": 0.2857142857142857, "percentage": 3.57, "elapsed_time": "0:28:53", "remaining_time": "13:00:14"} +{"current_steps": 85, "total_steps": 2240, "loss": 0.3483, "lr": 9.974254500789065e-05, "epoch": 0.30357142857142855, "percentage": 3.79, "elapsed_time": "0:30:07", "remaining_time": "12:43:48"} +{"current_steps": 90, "total_steps": 2240, "loss": 0.358, "lr": 9.97055821363669e-05, "epoch": 0.32142857142857145, "percentage": 4.02, "elapsed_time": "0:31:29", "remaining_time": "12:32:19"} +{"current_steps": 95, "total_steps": 2240, "loss": 0.3357, "lr": 9.96661486061939e-05, "epoch": 0.3392857142857143, "percentage": 4.24, "elapsed_time": "0:32:41", "remaining_time": "12:18:16"} +{"current_steps": 100, "total_steps": 2240, "loss": 0.3519, "lr": 9.962424637744914e-05, "epoch": 0.35714285714285715, "percentage": 4.46, "elapsed_time": "0:33:54", "remaining_time": "12:05:37"} +{"current_steps": 105, "total_steps": 2240, "loss": 0.3408, "lr": 9.957987753291889e-05, "epoch": 0.375, "percentage": 4.69, "elapsed_time": "1:04:15", "remaining_time": "21:46:42"} +{"current_steps": 110, "total_steps": 2240, "loss": 0.3462, "lr": 9.953304427799469e-05, "epoch": 0.39285714285714285, "percentage": 4.91, "elapsed_time": "1:05:29", "remaining_time": "21:08:16"} +{"current_steps": 115, "total_steps": 2240, "loss": 0.3459, "lr": 9.948374894056368e-05, "epoch": 0.4107142857142857, "percentage": 5.13, "elapsed_time": "1:06:44", "remaining_time": "20:33:23"} +{"current_steps": 120, "total_steps": 2240, "loss": 0.3386, "lr": 9.943199397089296e-05, "epoch": 0.42857142857142855, "percentage": 5.36, "elapsed_time": "1:07:57", "remaining_time": "20:00:32"} +{"current_steps": 125, "total_steps": 2240, "loss": 0.3329, "lr": 9.937778194150771e-05, "epoch": 0.44642857142857145, "percentage": 5.58, "elapsed_time": "1:09:12", "remaining_time": "19:30:59"} +{"current_steps": 130, "total_steps": 2240, "loss": 0.3445, "lr": 9.932111554706345e-05, "epoch": 0.4642857142857143, "percentage": 5.8, "elapsed_time": "1:10:33", "remaining_time": "19:05:15"} +{"current_steps": 135, "total_steps": 2240, "loss": 0.34, "lr": 9.926199760421195e-05, "epoch": 0.48214285714285715, "percentage": 6.03, "elapsed_time": "1:11:48", "remaining_time": "18:39:33"} +{"current_steps": 140, "total_steps": 2240, "loss": 0.3272, "lr": 9.920043105146136e-05, "epoch": 0.5, "percentage": 6.25, "elapsed_time": "1:12:58", "remaining_time": "18:14:44"} +{"current_steps": 145, "total_steps": 2240, "loss": 0.3328, "lr": 9.913641894903006e-05, "epoch": 0.5178571428571429, "percentage": 6.47, "elapsed_time": "1:14:13", "remaining_time": "17:52:24"} +{"current_steps": 150, "total_steps": 2240, "loss": 0.3361, "lr": 9.906996447869454e-05, "epoch": 0.5357142857142857, "percentage": 6.7, "elapsed_time": "1:15:24", "remaining_time": "17:30:48"} +{"current_steps": 155, "total_steps": 2240, "loss": 0.3341, "lr": 9.900107094363138e-05, "epoch": 0.5535714285714286, "percentage": 6.92, "elapsed_time": "1:16:44", "remaining_time": "17:12:16"} +{"current_steps": 160, "total_steps": 2240, "loss": 0.3426, "lr": 9.892974176825286e-05, "epoch": 0.5714285714285714, "percentage": 7.14, "elapsed_time": "1:18:01", "remaining_time": "16:54:20"} +{"current_steps": 165, "total_steps": 2240, "loss": 0.3292, "lr": 9.885598049803693e-05, "epoch": 0.5892857142857143, "percentage": 7.37, "elapsed_time": "1:19:19", "remaining_time": "16:37:28"} +{"current_steps": 170, "total_steps": 2240, "loss": 0.3326, "lr": 9.877979079935086e-05, "epoch": 0.6071428571428571, "percentage": 7.59, "elapsed_time": "1:20:35", "remaining_time": "16:21:14"} +{"current_steps": 175, "total_steps": 2240, "loss": 0.3347, "lr": 9.870117645926906e-05, "epoch": 0.625, "percentage": 7.81, "elapsed_time": "1:21:54", "remaining_time": "16:06:28"} +{"current_steps": 180, "total_steps": 2240, "loss": 0.3253, "lr": 9.862014138538482e-05, "epoch": 0.6428571428571429, "percentage": 8.04, "elapsed_time": "1:23:02", "remaining_time": "15:50:18"} +{"current_steps": 185, "total_steps": 2240, "loss": 0.3359, "lr": 9.853668960561611e-05, "epoch": 0.6607142857142857, "percentage": 8.26, "elapsed_time": "1:24:19", "remaining_time": "15:36:38"} +{"current_steps": 190, "total_steps": 2240, "loss": 0.3299, "lr": 9.845082526800528e-05, "epoch": 0.6785714285714286, "percentage": 8.48, "elapsed_time": "1:25:38", "remaining_time": "15:24:01"} +{"current_steps": 195, "total_steps": 2240, "loss": 0.3285, "lr": 9.836255264051299e-05, "epoch": 0.6964285714285714, "percentage": 8.71, "elapsed_time": "1:26:53", "remaining_time": "15:11:16"} +{"current_steps": 200, "total_steps": 2240, "loss": 0.3294, "lr": 9.8271876110806e-05, "epoch": 0.7142857142857143, "percentage": 8.93, "elapsed_time": "1:28:09", "remaining_time": "14:59:17"} +{"current_steps": 205, "total_steps": 2240, "loss": 0.3327, "lr": 9.817880018603909e-05, "epoch": 0.7321428571428571, "percentage": 9.15, "elapsed_time": "2:04:07", "remaining_time": "20:32:06"} +{"current_steps": 210, "total_steps": 2240, "loss": 0.3261, "lr": 9.808332949263103e-05, "epoch": 0.75, "percentage": 9.38, "elapsed_time": "2:05:22", "remaining_time": "20:11:55"} +{"current_steps": 215, "total_steps": 2240, "loss": 0.3285, "lr": 9.798546877603468e-05, "epoch": 0.7678571428571429, "percentage": 9.6, "elapsed_time": "2:06:40", "remaining_time": "19:53:08"} +{"current_steps": 220, "total_steps": 2240, "loss": 0.3275, "lr": 9.788522290050094e-05, "epoch": 0.7857142857142857, "percentage": 9.82, "elapsed_time": "2:07:51", "remaining_time": "19:34:00"} +{"current_steps": 225, "total_steps": 2240, "loss": 0.3291, "lr": 9.778259684883719e-05, "epoch": 0.8035714285714286, "percentage": 10.04, "elapsed_time": "2:09:03", "remaining_time": "19:15:48"} +{"current_steps": 230, "total_steps": 2240, "loss": 0.3218, "lr": 9.767759572215945e-05, "epoch": 0.8214285714285714, "percentage": 10.27, "elapsed_time": "2:10:14", "remaining_time": "18:58:14"} +{"current_steps": 235, "total_steps": 2240, "loss": 0.3245, "lr": 9.75702247396389e-05, "epoch": 0.8392857142857143, "percentage": 10.49, "elapsed_time": "2:11:28", "remaining_time": "18:41:44"} +{"current_steps": 240, "total_steps": 2240, "loss": 0.3305, "lr": 9.746048923824245e-05, "epoch": 0.8571428571428571, "percentage": 10.71, "elapsed_time": "2:12:39", "remaining_time": "18:25:30"} +{"current_steps": 245, "total_steps": 2240, "loss": 0.331, "lr": 9.734839467246744e-05, "epoch": 0.875, "percentage": 10.94, "elapsed_time": "2:13:52", "remaining_time": "18:10:05"} +{"current_steps": 250, "total_steps": 2240, "loss": 0.3249, "lr": 9.723394661407053e-05, "epoch": 0.8928571428571429, "percentage": 11.16, "elapsed_time": "2:15:04", "remaining_time": "17:55:09"} +{"current_steps": 255, "total_steps": 2240, "loss": 0.3186, "lr": 9.711715075179076e-05, "epoch": 0.9107142857142857, "percentage": 11.38, "elapsed_time": "2:16:06", "remaining_time": "17:39:31"} +{"current_steps": 260, "total_steps": 2240, "loss": 0.3307, "lr": 9.699801289106676e-05, "epoch": 0.9285714285714286, "percentage": 11.61, "elapsed_time": "2:17:20", "remaining_time": "17:25:54"} +{"current_steps": 265, "total_steps": 2240, "loss": 0.3272, "lr": 9.687653895374823e-05, "epoch": 0.9464285714285714, "percentage": 11.83, "elapsed_time": "2:18:34", "remaining_time": "17:12:44"} +{"current_steps": 270, "total_steps": 2240, "loss": 0.3203, "lr": 9.675273497780155e-05, "epoch": 0.9642857142857143, "percentage": 12.05, "elapsed_time": "2:19:48", "remaining_time": "17:00:08"} +{"current_steps": 275, "total_steps": 2240, "loss": 0.3101, "lr": 9.662660711700967e-05, "epoch": 0.9821428571428571, "percentage": 12.28, "elapsed_time": "2:21:01", "remaining_time": "16:47:43"} +{"current_steps": 280, "total_steps": 2240, "loss": 0.3308, "lr": 9.649816164066623e-05, "epoch": 1.0, "percentage": 12.5, "elapsed_time": "2:22:15", "remaining_time": "16:35:49"} +{"current_steps": 285, "total_steps": 2240, "loss": 0.2804, "lr": 9.636740493326397e-05, "epoch": 1.0178571428571428, "percentage": 12.72, "elapsed_time": "2:23:29", "remaining_time": "16:24:19"} +{"current_steps": 290, "total_steps": 2240, "loss": 0.2641, "lr": 9.623434349417729e-05, "epoch": 1.0357142857142858, "percentage": 12.95, "elapsed_time": "2:24:38", "remaining_time": "16:12:36"} +{"current_steps": 295, "total_steps": 2240, "loss": 0.2751, "lr": 9.609898393733933e-05, "epoch": 1.0535714285714286, "percentage": 13.17, "elapsed_time": "2:25:49", "remaining_time": "16:01:29"} +{"current_steps": 300, "total_steps": 2240, "loss": 0.282, "lr": 9.59613329909131e-05, "epoch": 1.0714285714285714, "percentage": 13.39, "elapsed_time": "2:27:06", "remaining_time": "15:51:20"} +{"current_steps": 205, "total_steps": 2240, "loss": 0.3327, "lr": 9.817880018603909e-05, "epoch": 0.7321428571428571, "percentage": 9.15, "elapsed_time": "0:02:12", "remaining_time": "0:21:54"} +{"current_steps": 210, "total_steps": 2240, "loss": 0.3261, "lr": 9.808332949263103e-05, "epoch": 0.75, "percentage": 9.38, "elapsed_time": "0:03:34", "remaining_time": "0:34:29"} +{"current_steps": 215, "total_steps": 2240, "loss": 0.3284, "lr": 9.798546877603468e-05, "epoch": 0.7678571428571429, "percentage": 9.6, "elapsed_time": "0:04:54", "remaining_time": "0:46:10"} +{"current_steps": 220, "total_steps": 2240, "loss": 0.3292, "lr": 9.788522290050094e-05, "epoch": 0.7857142857142857, "percentage": 9.82, "elapsed_time": "0:06:07", "remaining_time": "0:56:11"} +{"current_steps": 225, "total_steps": 2240, "loss": 0.3316, "lr": 9.778259684883719e-05, "epoch": 0.8035714285714286, "percentage": 10.04, "elapsed_time": "0:07:19", "remaining_time": "1:05:39"} +{"current_steps": 230, "total_steps": 2240, "loss": 0.3217, "lr": 9.767759572215945e-05, "epoch": 0.8214285714285714, "percentage": 10.27, "elapsed_time": "0:08:32", "remaining_time": "1:14:35"} +{"current_steps": 235, "total_steps": 2240, "loss": 0.3252, "lr": 9.75702247396389e-05, "epoch": 0.8392857142857143, "percentage": 10.49, "elapsed_time": "0:09:47", "remaining_time": "1:23:31"} +{"current_steps": 240, "total_steps": 2240, "loss": 0.3277, "lr": 9.746048923824245e-05, "epoch": 0.8571428571428571, "percentage": 10.71, "elapsed_time": "0:10:58", "remaining_time": "1:31:24"} +{"current_steps": 245, "total_steps": 2240, "loss": 0.3307, "lr": 9.734839467246744e-05, "epoch": 0.875, "percentage": 10.94, "elapsed_time": "0:12:11", "remaining_time": "1:39:12"} +{"current_steps": 250, "total_steps": 2240, "loss": 0.325, "lr": 9.723394661407053e-05, "epoch": 0.8928571428571429, "percentage": 11.16, "elapsed_time": "0:13:23", "remaining_time": "1:46:36"} +{"current_steps": 255, "total_steps": 2240, "loss": 0.3188, "lr": 9.711715075179076e-05, "epoch": 0.9107142857142857, "percentage": 11.38, "elapsed_time": "0:14:25", "remaining_time": "1:52:15"} +{"current_steps": 260, "total_steps": 2240, "loss": 0.3312, "lr": 9.699801289106676e-05, "epoch": 0.9285714285714286, "percentage": 11.61, "elapsed_time": "0:15:36", "remaining_time": "1:58:52"} +{"current_steps": 265, "total_steps": 2240, "loss": 0.3275, "lr": 9.687653895374823e-05, "epoch": 0.9464285714285714, "percentage": 11.83, "elapsed_time": "0:16:49", "remaining_time": "2:05:24"} +{"current_steps": 270, "total_steps": 2240, "loss": 0.3205, "lr": 9.675273497780155e-05, "epoch": 0.9642857142857143, "percentage": 12.05, "elapsed_time": "0:18:04", "remaining_time": "2:11:51"} +{"current_steps": 275, "total_steps": 2240, "loss": 0.3103, "lr": 9.662660711700967e-05, "epoch": 0.9821428571428571, "percentage": 12.28, "elapsed_time": "0:19:16", "remaining_time": "2:17:42"} +{"current_steps": 280, "total_steps": 2240, "loss": 0.3312, "lr": 9.649816164066623e-05, "epoch": 1.0, "percentage": 12.5, "elapsed_time": "0:20:29", "remaining_time": "2:23:26"} +{"current_steps": 285, "total_steps": 2240, "loss": 0.281, "lr": 9.636740493326397e-05, "epoch": 1.0178571428571428, "percentage": 12.72, "elapsed_time": "0:21:43", "remaining_time": "2:28:58"} +{"current_steps": 290, "total_steps": 2240, "loss": 0.2645, "lr": 9.623434349417729e-05, "epoch": 1.0357142857142858, "percentage": 12.95, "elapsed_time": "0:22:51", "remaining_time": "2:33:44"} +{"current_steps": 295, "total_steps": 2240, "loss": 0.2761, "lr": 9.609898393733933e-05, "epoch": 1.0535714285714286, "percentage": 13.17, "elapsed_time": "0:24:03", "remaining_time": "2:38:34"} +{"current_steps": 300, "total_steps": 2240, "loss": 0.2816, "lr": 9.59613329909131e-05, "epoch": 1.0714285714285714, "percentage": 13.39, "elapsed_time": "0:25:19", "remaining_time": "2:43:46"} +{"current_steps": 305, "total_steps": 2240, "loss": 0.2817, "lr": 9.582139749695713e-05, "epoch": 1.0892857142857142, "percentage": 13.62, "elapsed_time": "0:39:00", "remaining_time": "4:07:25"} +{"current_steps": 310, "total_steps": 2240, "loss": 0.2723, "lr": 9.56791844110853e-05, "epoch": 1.1071428571428572, "percentage": 13.84, "elapsed_time": "0:40:12", "remaining_time": "4:10:22"} +{"current_steps": 315, "total_steps": 2240, "loss": 0.2858, "lr": 9.553470080212122e-05, "epoch": 1.125, "percentage": 14.06, "elapsed_time": "0:41:24", "remaining_time": "4:13:05"} +{"current_steps": 320, "total_steps": 2240, "loss": 0.2769, "lr": 9.538795385174672e-05, "epoch": 1.1428571428571428, "percentage": 14.29, "elapsed_time": "0:42:38", "remaining_time": "4:15:53"} +{"current_steps": 325, "total_steps": 2240, "loss": 0.2735, "lr": 9.523895085414501e-05, "epoch": 1.1607142857142858, "percentage": 14.51, "elapsed_time": "0:43:47", "remaining_time": "4:18:00"} +{"current_steps": 330, "total_steps": 2240, "loss": 0.2763, "lr": 9.508769921563809e-05, "epoch": 1.1785714285714286, "percentage": 14.73, "elapsed_time": "0:44:51", "remaining_time": "4:19:35"} +{"current_steps": 335, "total_steps": 2240, "loss": 0.2851, "lr": 9.493420645431852e-05, "epoch": 1.1964285714285714, "percentage": 14.96, "elapsed_time": "0:46:03", "remaining_time": "4:21:52"} +{"current_steps": 340, "total_steps": 2240, "loss": 0.2781, "lr": 9.477848019967583e-05, "epoch": 1.2142857142857142, "percentage": 15.18, "elapsed_time": "0:47:14", "remaining_time": "4:24:02"} +{"current_steps": 345, "total_steps": 2240, "loss": 0.2902, "lr": 9.462052819221726e-05, "epoch": 1.2321428571428572, "percentage": 15.4, "elapsed_time": "0:48:24", "remaining_time": "4:25:54"} +{"current_steps": 350, "total_steps": 2240, "loss": 0.2863, "lr": 9.4460358283083e-05, "epoch": 1.25, "percentage": 15.62, "elapsed_time": "0:49:37", "remaining_time": "4:28:00"} +{"current_steps": 355, "total_steps": 2240, "loss": 0.2751, "lr": 9.429797843365594e-05, "epoch": 1.2678571428571428, "percentage": 15.85, "elapsed_time": "0:50:52", "remaining_time": "4:30:09"} +{"current_steps": 360, "total_steps": 2240, "loss": 0.2796, "lr": 9.413339671516593e-05, "epoch": 1.2857142857142856, "percentage": 16.07, "elapsed_time": "0:52:04", "remaining_time": "4:31:57"} +{"current_steps": 365, "total_steps": 2240, "loss": 0.2801, "lr": 9.396662130828869e-05, "epoch": 1.3035714285714286, "percentage": 16.29, "elapsed_time": "0:53:14", "remaining_time": "4:33:31"} +{"current_steps": 370, "total_steps": 2240, "loss": 0.2772, "lr": 9.379766050273899e-05, "epoch": 1.3214285714285714, "percentage": 16.52, "elapsed_time": "0:54:21", "remaining_time": "4:34:41"} +{"current_steps": 375, "total_steps": 2240, "loss": 0.2768, "lr": 9.36265226968588e-05, "epoch": 1.3392857142857144, "percentage": 16.74, "elapsed_time": "0:55:32", "remaining_time": "4:36:11"} +{"current_steps": 380, "total_steps": 2240, "loss": 0.2781, "lr": 9.345321639719979e-05, "epoch": 1.3571428571428572, "percentage": 16.96, "elapsed_time": "0:56:42", "remaining_time": "4:37:35"} +{"current_steps": 385, "total_steps": 2240, "loss": 0.2829, "lr": 9.327775021810037e-05, "epoch": 1.375, "percentage": 17.19, "elapsed_time": "0:57:55", "remaining_time": "4:39:04"} +{"current_steps": 390, "total_steps": 2240, "loss": 0.2834, "lr": 9.31001328812577e-05, "epoch": 1.3928571428571428, "percentage": 17.41, "elapsed_time": "0:59:08", "remaining_time": "4:40:31"} +{"current_steps": 395, "total_steps": 2240, "loss": 0.2801, "lr": 9.292037321529404e-05, "epoch": 1.4107142857142856, "percentage": 17.63, "elapsed_time": "1:00:20", "remaining_time": "4:41:49"} +{"current_steps": 400, "total_steps": 2240, "loss": 0.2771, "lr": 9.273848015531795e-05, "epoch": 1.4285714285714286, "percentage": 17.86, "elapsed_time": "1:01:37", "remaining_time": "4:43:29"} +{"current_steps": 405, "total_steps": 2240, "loss": 0.2794, "lr": 9.255446274248023e-05, "epoch": 1.4464285714285714, "percentage": 18.08, "elapsed_time": "1:10:54", "remaining_time": "5:21:17"} +{"current_steps": 410, "total_steps": 2240, "loss": 0.2968, "lr": 9.236833012352442e-05, "epoch": 1.4642857142857144, "percentage": 18.3, "elapsed_time": "1:12:01", "remaining_time": "5:21:28"} +{"current_steps": 415, "total_steps": 2240, "loss": 0.2828, "lr": 9.218009155033218e-05, "epoch": 1.4821428571428572, "percentage": 18.53, "elapsed_time": "1:13:14", "remaining_time": "5:22:04"} +{"current_steps": 420, "total_steps": 2240, "loss": 0.2752, "lr": 9.198975637946347e-05, "epoch": 1.5, "percentage": 18.75, "elapsed_time": "1:14:30", "remaining_time": "5:22:53"} +{"current_steps": 425, "total_steps": 2240, "loss": 0.2835, "lr": 9.179733407169145e-05, "epoch": 1.5178571428571428, "percentage": 18.97, "elapsed_time": "1:15:43", "remaining_time": "5:23:24"} +{"current_steps": 430, "total_steps": 2240, "loss": 0.2794, "lr": 9.160283419153216e-05, "epoch": 1.5357142857142856, "percentage": 19.2, "elapsed_time": "1:16:55", "remaining_time": "5:23:48"} +{"current_steps": 435, "total_steps": 2240, "loss": 0.2756, "lr": 9.140626640676919e-05, "epoch": 1.5535714285714286, "percentage": 19.42, "elapsed_time": "1:18:03", "remaining_time": "5:23:54"} +{"current_steps": 440, "total_steps": 2240, "loss": 0.28, "lr": 9.12076404879731e-05, "epoch": 1.5714285714285714, "percentage": 19.64, "elapsed_time": "1:19:20", "remaining_time": "5:24:34"} +{"current_steps": 445, "total_steps": 2240, "loss": 0.2897, "lr": 9.10069663080158e-05, "epoch": 1.5892857142857144, "percentage": 19.87, "elapsed_time": "1:20:28", "remaining_time": "5:24:38"} +{"current_steps": 450, "total_steps": 2240, "loss": 0.2882, "lr": 9.080425384157971e-05, "epoch": 1.6071428571428572, "percentage": 20.09, "elapsed_time": "1:21:37", "remaining_time": "5:24:42"} +{"current_steps": 455, "total_steps": 2240, "loss": 0.2861, "lr": 9.059951316466208e-05, "epoch": 1.625, "percentage": 20.31, "elapsed_time": "1:22:48", "remaining_time": "5:24:52"} +{"current_steps": 460, "total_steps": 2240, "loss": 0.2794, "lr": 9.039275445407414e-05, "epoch": 1.6428571428571428, "percentage": 20.54, "elapsed_time": "1:24:04", "remaining_time": "5:25:18"} +{"current_steps": 465, "total_steps": 2240, "loss": 0.2741, "lr": 9.018398798693512e-05, "epoch": 1.6607142857142856, "percentage": 20.76, "elapsed_time": "1:25:16", "remaining_time": "5:25:30"} +{"current_steps": 470, "total_steps": 2240, "loss": 0.279, "lr": 8.997322414016158e-05, "epoch": 1.6785714285714286, "percentage": 20.98, "elapsed_time": "1:26:30", "remaining_time": "5:25:46"} +{"current_steps": 475, "total_steps": 2240, "loss": 0.2798, "lr": 8.976047338995155e-05, "epoch": 1.6964285714285714, "percentage": 21.21, "elapsed_time": "1:27:39", "remaining_time": "5:25:43"} +{"current_steps": 480, "total_steps": 2240, "loss": 0.281, "lr": 8.954574631126378e-05, "epoch": 1.7142857142857144, "percentage": 21.43, "elapsed_time": "1:28:51", "remaining_time": "5:25:48"} +{"current_steps": 485, "total_steps": 2240, "loss": 0.276, "lr": 8.932905357729214e-05, "epoch": 1.7321428571428572, "percentage": 21.65, "elapsed_time": "1:30:01", "remaining_time": "5:25:44"} +{"current_steps": 490, "total_steps": 2240, "loss": 0.2792, "lr": 8.911040595893505e-05, "epoch": 1.75, "percentage": 21.88, "elapsed_time": "1:31:11", "remaining_time": "5:25:41"} +{"current_steps": 495, "total_steps": 2240, "loss": 0.2788, "lr": 8.88898143242602e-05, "epoch": 1.7678571428571428, "percentage": 22.1, "elapsed_time": "1:32:23", "remaining_time": "5:25:43"} +{"current_steps": 500, "total_steps": 2240, "loss": 0.2791, "lr": 8.866728963796423e-05, "epoch": 1.7857142857142856, "percentage": 22.32, "elapsed_time": "1:33:37", "remaining_time": "5:25:49"} +{"current_steps": 505, "total_steps": 2240, "loss": 0.281, "lr": 8.844284296082776e-05, "epoch": 1.8035714285714286, "percentage": 22.54, "elapsed_time": "1:45:10", "remaining_time": "6:01:21"} +{"current_steps": 510, "total_steps": 2240, "loss": 0.2875, "lr": 8.821648544916567e-05, "epoch": 1.8214285714285714, "percentage": 22.77, "elapsed_time": "1:46:23", "remaining_time": "6:00:54"} +{"current_steps": 515, "total_steps": 2240, "loss": 0.2763, "lr": 8.79882283542725e-05, "epoch": 1.8392857142857144, "percentage": 22.99, "elapsed_time": "1:47:37", "remaining_time": "6:00:29"} +{"current_steps": 520, "total_steps": 2240, "loss": 0.2797, "lr": 8.775808302186314e-05, "epoch": 1.8571428571428572, "percentage": 23.21, "elapsed_time": "1:48:47", "remaining_time": "5:59:49"} +{"current_steps": 525, "total_steps": 2240, "loss": 0.2874, "lr": 8.752606089150903e-05, "epoch": 1.875, "percentage": 23.44, "elapsed_time": "1:50:01", "remaining_time": "5:59:24"} +{"current_steps": 530, "total_steps": 2240, "loss": 0.2774, "lr": 8.729217349606942e-05, "epoch": 1.8928571428571428, "percentage": 23.66, "elapsed_time": "1:51:15", "remaining_time": "5:58:57"} +{"current_steps": 535, "total_steps": 2240, "loss": 0.2891, "lr": 8.705643246111816e-05, "epoch": 1.9107142857142856, "percentage": 23.88, "elapsed_time": "1:52:28", "remaining_time": "5:58:26"} +{"current_steps": 540, "total_steps": 2240, "loss": 0.2796, "lr": 8.681884950436587e-05, "epoch": 1.9285714285714286, "percentage": 24.11, "elapsed_time": "1:53:43", "remaining_time": "5:58:01"} +{"current_steps": 545, "total_steps": 2240, "loss": 0.281, "lr": 8.657943643507746e-05, "epoch": 1.9464285714285714, "percentage": 24.33, "elapsed_time": "1:54:55", "remaining_time": "5:57:26"} +{"current_steps": 550, "total_steps": 2240, "loss": 0.2803, "lr": 8.633820515348517e-05, "epoch": 1.9642857142857144, "percentage": 24.55, "elapsed_time": "1:56:04", "remaining_time": "5:56:39"} +{"current_steps": 555, "total_steps": 2240, "loss": 0.2754, "lr": 8.609516765019698e-05, "epoch": 1.9821428571428572, "percentage": 24.78, "elapsed_time": "1:57:20", "remaining_time": "5:56:14"} +{"current_steps": 560, "total_steps": 2240, "loss": 0.2829, "lr": 8.58503360056008e-05, "epoch": 2.0, "percentage": 25.0, "elapsed_time": "1:58:31", "remaining_time": "5:55:34"} +{"current_steps": 565, "total_steps": 2240, "loss": 0.231, "lr": 8.560372238926372e-05, "epoch": 2.017857142857143, "percentage": 25.22, "elapsed_time": "1:59:43", "remaining_time": "5:54:57"} +{"current_steps": 570, "total_steps": 2240, "loss": 0.2185, "lr": 8.535533905932738e-05, "epoch": 2.0357142857142856, "percentage": 25.45, "elapsed_time": "2:00:52", "remaining_time": "5:54:08"} +{"current_steps": 575, "total_steps": 2240, "loss": 0.2212, "lr": 8.510519836189851e-05, "epoch": 2.0535714285714284, "percentage": 25.67, "elapsed_time": "2:02:04", "remaining_time": "5:53:29"} +{"current_steps": 580, "total_steps": 2240, "loss": 0.2128, "lr": 8.485331273043531e-05, "epoch": 2.0714285714285716, "percentage": 25.89, "elapsed_time": "2:03:17", "remaining_time": "5:52:52"} +{"current_steps": 585, "total_steps": 2240, "loss": 0.2194, "lr": 8.459969468512943e-05, "epoch": 2.0892857142857144, "percentage": 26.12, "elapsed_time": "2:04:31", "remaining_time": "5:52:18"} +{"current_steps": 590, "total_steps": 2240, "loss": 0.2288, "lr": 8.43443568322836e-05, "epoch": 2.107142857142857, "percentage": 26.34, "elapsed_time": "2:05:42", "remaining_time": "5:51:33"} +{"current_steps": 595, "total_steps": 2240, "loss": 0.2217, "lr": 8.40873118636851e-05, "epoch": 2.125, "percentage": 26.56, "elapsed_time": "2:06:51", "remaining_time": "5:50:42"} +{"current_steps": 600, "total_steps": 2240, "loss": 0.2247, "lr": 8.382857255597489e-05, "epoch": 2.142857142857143, "percentage": 26.79, "elapsed_time": "2:08:10", "remaining_time": "5:50:21"} +{"current_steps": 605, "total_steps": 2240, "loss": 0.23, "lr": 8.356815177001242e-05, "epoch": 2.1607142857142856, "percentage": 27.01, "elapsed_time": "2:22:05", "remaining_time": "6:23:59"} +{"current_steps": 610, "total_steps": 2240, "loss": 0.228, "lr": 8.330606245023657e-05, "epoch": 2.1785714285714284, "percentage": 27.23, "elapsed_time": "2:23:17", "remaining_time": "6:22:54"} +{"current_steps": 615, "total_steps": 2240, "loss": 0.2256, "lr": 8.304231762402204e-05, "epoch": 2.1964285714285716, "percentage": 27.46, "elapsed_time": "2:24:29", "remaining_time": "6:21:46"} +{"current_steps": 620, "total_steps": 2240, "loss": 0.2338, "lr": 8.277693040103192e-05, "epoch": 2.2142857142857144, "percentage": 27.68, "elapsed_time": "2:25:36", "remaining_time": "6:20:28"} +{"current_steps": 625, "total_steps": 2240, "loss": 0.2283, "lr": 8.250991397256609e-05, "epoch": 2.232142857142857, "percentage": 27.9, "elapsed_time": "2:26:49", "remaining_time": "6:19:23"} +{"current_steps": 630, "total_steps": 2240, "loss": 0.2272, "lr": 8.224128161090543e-05, "epoch": 2.25, "percentage": 28.12, "elapsed_time": "2:28:03", "remaining_time": "6:18:23"} +{"current_steps": 635, "total_steps": 2240, "loss": 0.2281, "lr": 8.197104666865218e-05, "epoch": 2.267857142857143, "percentage": 28.35, "elapsed_time": "2:29:15", "remaining_time": "6:17:15"} +{"current_steps": 640, "total_steps": 2240, "loss": 0.2252, "lr": 8.169922257806625e-05, "epoch": 2.2857142857142856, "percentage": 28.57, "elapsed_time": "2:30:25", "remaining_time": "6:16:03"} +{"current_steps": 645, "total_steps": 2240, "loss": 0.2292, "lr": 8.142582285039752e-05, "epoch": 2.3035714285714284, "percentage": 28.79, "elapsed_time": "2:31:41", "remaining_time": "6:15:06"} +{"current_steps": 650, "total_steps": 2240, "loss": 0.2281, "lr": 8.115086107521424e-05, "epoch": 2.3214285714285716, "percentage": 29.02, "elapsed_time": "2:32:46", "remaining_time": "6:13:42"} +{"current_steps": 655, "total_steps": 2240, "loss": 0.2336, "lr": 8.087435091972761e-05, "epoch": 2.3392857142857144, "percentage": 29.24, "elapsed_time": "2:33:59", "remaining_time": "6:12:37"} +{"current_steps": 660, "total_steps": 2240, "loss": 0.2306, "lr": 8.05963061281124e-05, "epoch": 2.357142857142857, "percentage": 29.46, "elapsed_time": "2:35:12", "remaining_time": "6:11:34"} +{"current_steps": 665, "total_steps": 2240, "loss": 0.233, "lr": 8.031674052082372e-05, "epoch": 2.375, "percentage": 29.69, "elapsed_time": "2:36:21", "remaining_time": "6:10:18"} +{"current_steps": 670, "total_steps": 2240, "loss": 0.2336, "lr": 8.003566799391024e-05, "epoch": 2.392857142857143, "percentage": 29.91, "elapsed_time": "2:37:34", "remaining_time": "6:09:13"} +{"current_steps": 675, "total_steps": 2240, "loss": 0.2376, "lr": 7.975310251832329e-05, "epoch": 2.4107142857142856, "percentage": 30.13, "elapsed_time": "2:38:43", "remaining_time": "6:07:59"} +{"current_steps": 680, "total_steps": 2240, "loss": 0.2287, "lr": 7.946905813922249e-05, "epoch": 2.4285714285714284, "percentage": 30.36, "elapsed_time": "2:39:52", "remaining_time": "6:06:47"} +{"current_steps": 685, "total_steps": 2240, "loss": 0.2297, "lr": 7.918354897527766e-05, "epoch": 2.4464285714285716, "percentage": 30.58, "elapsed_time": "2:41:02", "remaining_time": "6:05:33"} +{"current_steps": 690, "total_steps": 2240, "loss": 0.2351, "lr": 7.889658921796703e-05, "epoch": 2.4642857142857144, "percentage": 30.8, "elapsed_time": "2:42:17", "remaining_time": "6:04:34"} +{"current_steps": 695, "total_steps": 2240, "loss": 0.2288, "lr": 7.860819313087177e-05, "epoch": 2.482142857142857, "percentage": 31.03, "elapsed_time": "2:43:31", "remaining_time": "6:03:30"} +{"current_steps": 700, "total_steps": 2240, "loss": 0.2209, "lr": 7.831837504896707e-05, "epoch": 2.5, "percentage": 31.25, "elapsed_time": "2:44:39", "remaining_time": "6:02:15"} +{"current_steps": 705, "total_steps": 2240, "loss": 0.2292, "lr": 7.80271493779096e-05, "epoch": 2.517857142857143, "percentage": 31.47, "elapsed_time": "2:58:50", "remaining_time": "6:29:24"} +{"current_steps": 710, "total_steps": 2240, "loss": 0.2304, "lr": 7.773453059332145e-05, "epoch": 2.5357142857142856, "percentage": 31.7, "elapsed_time": "3:00:03", "remaining_time": "6:28:00"} +{"current_steps": 715, "total_steps": 2240, "loss": 0.2294, "lr": 7.744053324007063e-05, "epoch": 2.553571428571429, "percentage": 31.92, "elapsed_time": "3:01:13", "remaining_time": "6:26:31"} +{"current_steps": 720, "total_steps": 2240, "loss": 0.2347, "lr": 7.71451719315481e-05, "epoch": 2.571428571428571, "percentage": 32.14, "elapsed_time": "3:02:23", "remaining_time": "6:25:03"} +{"current_steps": 725, "total_steps": 2240, "loss": 0.2284, "lr": 7.684846134894133e-05, "epoch": 2.5892857142857144, "percentage": 32.37, "elapsed_time": "3:03:31", "remaining_time": "6:23:30"} +{"current_steps": 730, "total_steps": 2240, "loss": 0.2291, "lr": 7.655041624050467e-05, "epoch": 2.607142857142857, "percentage": 32.59, "elapsed_time": "3:04:36", "remaining_time": "6:21:52"} +{"current_steps": 735, "total_steps": 2240, "loss": 0.2368, "lr": 7.625105142082623e-05, "epoch": 2.625, "percentage": 32.81, "elapsed_time": "3:05:46", "remaining_time": "6:20:23"} +{"current_steps": 740, "total_steps": 2240, "loss": 0.2322, "lr": 7.595038177009144e-05, "epoch": 2.642857142857143, "percentage": 33.04, "elapsed_time": "3:07:01", "remaining_time": "6:19:05"} +{"current_steps": 745, "total_steps": 2240, "loss": 0.2334, "lr": 7.564842223334356e-05, "epoch": 2.6607142857142856, "percentage": 33.26, "elapsed_time": "3:08:15", "remaining_time": "6:17:47"} +{"current_steps": 750, "total_steps": 2240, "loss": 0.23, "lr": 7.534518781974065e-05, "epoch": 2.678571428571429, "percentage": 33.48, "elapsed_time": "3:09:25", "remaining_time": "6:16:19"} +{"current_steps": 755, "total_steps": 2240, "loss": 0.2305, "lr": 7.504069360180971e-05, "epoch": 2.696428571428571, "percentage": 33.71, "elapsed_time": "3:10:40", "remaining_time": "6:15:02"} +{"current_steps": 760, "total_steps": 2240, "loss": 0.2367, "lr": 7.473495471469733e-05, "epoch": 2.7142857142857144, "percentage": 33.93, "elapsed_time": "3:11:53", "remaining_time": "6:13:40"} +{"current_steps": 765, "total_steps": 2240, "loss": 0.2337, "lr": 7.442798635541749e-05, "epoch": 2.732142857142857, "percentage": 34.15, "elapsed_time": "3:13:07", "remaining_time": "6:12:21"} +{"current_steps": 770, "total_steps": 2240, "loss": 0.231, "lr": 7.411980378209611e-05, "epoch": 2.75, "percentage": 34.38, "elapsed_time": "3:14:18", "remaining_time": "6:10:56"} +{"current_steps": 775, "total_steps": 2240, "loss": 0.2367, "lr": 7.381042231321269e-05, "epoch": 2.767857142857143, "percentage": 34.6, "elapsed_time": "3:15:26", "remaining_time": "6:09:26"} +{"current_steps": 780, "total_steps": 2240, "loss": 0.2451, "lr": 7.349985732683886e-05, "epoch": 2.7857142857142856, "percentage": 34.82, "elapsed_time": "3:16:38", "remaining_time": "6:08:04"} +{"current_steps": 785, "total_steps": 2240, "loss": 0.2372, "lr": 7.318812425987395e-05, "epoch": 2.803571428571429, "percentage": 35.04, "elapsed_time": "3:17:56", "remaining_time": "6:06:52"} +{"current_steps": 790, "total_steps": 2240, "loss": 0.2275, "lr": 7.287523860727781e-05, "epoch": 2.821428571428571, "percentage": 35.27, "elapsed_time": "3:19:07", "remaining_time": "6:05:28"} +{"current_steps": 795, "total_steps": 2240, "loss": 0.2246, "lr": 7.256121592130055e-05, "epoch": 2.8392857142857144, "percentage": 35.49, "elapsed_time": "3:20:17", "remaining_time": "6:04:03"} +{"current_steps": 800, "total_steps": 2240, "loss": 0.2296, "lr": 7.224607181070941e-05, "epoch": 2.857142857142857, "percentage": 35.71, "elapsed_time": "3:21:26", "remaining_time": "6:02:36"} +{"current_steps": 805, "total_steps": 2240, "loss": 0.2344, "lr": 7.192982194001312e-05, "epoch": 2.875, "percentage": 35.94, "elapsed_time": "3:31:12", "remaining_time": "6:16:29"} +{"current_steps": 810, "total_steps": 2240, "loss": 0.2218, "lr": 7.161248202868309e-05, "epoch": 2.892857142857143, "percentage": 36.16, "elapsed_time": "3:32:20", "remaining_time": "6:14:52"} +{"current_steps": 815, "total_steps": 2240, "loss": 0.233, "lr": 7.129406785037214e-05, "epoch": 2.9107142857142856, "percentage": 36.38, "elapsed_time": "3:33:33", "remaining_time": "6:13:23"} +{"current_steps": 820, "total_steps": 2240, "loss": 0.2308, "lr": 7.09745952321305e-05, "epoch": 2.928571428571429, "percentage": 36.61, "elapsed_time": "3:34:45", "remaining_time": "6:11:54"} +{"current_steps": 825, "total_steps": 2240, "loss": 0.2345, "lr": 7.065408005361903e-05, "epoch": 2.946428571428571, "percentage": 36.83, "elapsed_time": "3:35:55", "remaining_time": "6:10:21"} +{"current_steps": 830, "total_steps": 2240, "loss": 0.2228, "lr": 7.033253824631991e-05, "epoch": 2.9642857142857144, "percentage": 37.05, "elapsed_time": "3:37:07", "remaining_time": "6:08:51"} +{"current_steps": 835, "total_steps": 2240, "loss": 0.2296, "lr": 7.000998579274487e-05, "epoch": 2.982142857142857, "percentage": 37.28, "elapsed_time": "3:38:18", "remaining_time": "6:07:19"} +{"current_steps": 840, "total_steps": 2240, "loss": 0.2328, "lr": 6.968643872564064e-05, "epoch": 3.0, "percentage": 37.5, "elapsed_time": "3:39:31", "remaining_time": "6:05:51"} +{"current_steps": 845, "total_steps": 2240, "loss": 0.1734, "lr": 6.936191312719203e-05, "epoch": 3.017857142857143, "percentage": 37.72, "elapsed_time": "3:40:45", "remaining_time": "6:04:27"} +{"current_steps": 850, "total_steps": 2240, "loss": 0.1718, "lr": 6.903642512822263e-05, "epoch": 3.0357142857142856, "percentage": 37.95, "elapsed_time": "3:41:53", "remaining_time": "6:02:52"} +{"current_steps": 855, "total_steps": 2240, "loss": 0.1735, "lr": 6.870999090739301e-05, "epoch": 3.0535714285714284, "percentage": 38.17, "elapsed_time": "3:43:08", "remaining_time": "6:01:27"} +{"current_steps": 860, "total_steps": 2240, "loss": 0.1745, "lr": 6.838262669039643e-05, "epoch": 3.0714285714285716, "percentage": 38.39, "elapsed_time": "3:44:21", "remaining_time": "6:00:00"} +{"current_steps": 865, "total_steps": 2240, "loss": 0.1695, "lr": 6.805434874915249e-05, "epoch": 3.0892857142857144, "percentage": 38.62, "elapsed_time": "3:45:30", "remaining_time": "5:58:28"} +{"current_steps": 870, "total_steps": 2240, "loss": 0.1687, "lr": 6.772517340099816e-05, "epoch": 3.107142857142857, "percentage": 38.84, "elapsed_time": "3:46:45", "remaining_time": "5:57:04"} +{"current_steps": 875, "total_steps": 2240, "loss": 0.1749, "lr": 6.739511700787683e-05, "epoch": 3.125, "percentage": 39.06, "elapsed_time": "3:47:55", "remaining_time": "5:55:33"} +{"current_steps": 880, "total_steps": 2240, "loss": 0.1789, "lr": 6.7064195975525e-05, "epoch": 3.142857142857143, "percentage": 39.29, "elapsed_time": "3:48:59", "remaining_time": "5:53:53"} +{"current_steps": 885, "total_steps": 2240, "loss": 0.1759, "lr": 6.67324267526568e-05, "epoch": 3.1607142857142856, "percentage": 39.51, "elapsed_time": "3:50:09", "remaining_time": "5:52:23"} +{"current_steps": 890, "total_steps": 2240, "loss": 0.1777, "lr": 6.639982583014637e-05, "epoch": 3.1785714285714284, "percentage": 39.73, "elapsed_time": "3:51:20", "remaining_time": "5:50:55"} +{"current_steps": 895, "total_steps": 2240, "loss": 0.1802, "lr": 6.606640974020823e-05, "epoch": 3.1964285714285716, "percentage": 39.96, "elapsed_time": "3:52:29", "remaining_time": "5:49:23"} +{"current_steps": 900, "total_steps": 2240, "loss": 0.1784, "lr": 6.573219505557548e-05, "epoch": 3.2142857142857144, "percentage": 40.18, "elapsed_time": "3:53:38", "remaining_time": "5:47:51"} +{"current_steps": 905, "total_steps": 2240, "loss": 0.1683, "lr": 6.539719838867604e-05, "epoch": 3.232142857142857, "percentage": 40.4, "elapsed_time": "3:57:54", "remaining_time": "5:50:57"} +{"current_steps": 910, "total_steps": 2240, "loss": 0.1777, "lr": 6.506143639080695e-05, "epoch": 3.25, "percentage": 40.62, "elapsed_time": "3:59:05", "remaining_time": "5:49:26"} +{"current_steps": 915, "total_steps": 2240, "loss": 0.1752, "lr": 6.472492575130671e-05, "epoch": 3.267857142857143, "percentage": 40.85, "elapsed_time": "4:00:13", "remaining_time": "5:47:51"} +{"current_steps": 920, "total_steps": 2240, "loss": 0.1792, "lr": 6.438768319672561e-05, "epoch": 3.2857142857142856, "percentage": 41.07, "elapsed_time": "4:01:22", "remaining_time": "5:46:19"} +{"current_steps": 925, "total_steps": 2240, "loss": 0.1772, "lr": 6.404972548999453e-05, "epoch": 3.3035714285714284, "percentage": 41.29, "elapsed_time": "4:02:33", "remaining_time": "5:44:49"} +{"current_steps": 930, "total_steps": 2240, "loss": 0.1697, "lr": 6.37110694295915e-05, "epoch": 3.3214285714285716, "percentage": 41.52, "elapsed_time": "4:03:42", "remaining_time": "5:43:17"} +{"current_steps": 935, "total_steps": 2240, "loss": 0.1764, "lr": 6.337173184870683e-05, "epoch": 3.3392857142857144, "percentage": 41.74, "elapsed_time": "4:04:50", "remaining_time": "5:41:44"} +{"current_steps": 940, "total_steps": 2240, "loss": 0.1781, "lr": 6.303172961440645e-05, "epoch": 3.357142857142857, "percentage": 41.96, "elapsed_time": "4:06:04", "remaining_time": "5:40:18"} +{"current_steps": 945, "total_steps": 2240, "loss": 0.1767, "lr": 6.269107962679344e-05, "epoch": 3.375, "percentage": 42.19, "elapsed_time": "4:07:16", "remaining_time": "5:38:50"} +{"current_steps": 950, "total_steps": 2240, "loss": 0.1801, "lr": 6.234979881816793e-05, "epoch": 3.392857142857143, "percentage": 42.41, "elapsed_time": "4:08:26", "remaining_time": "5:37:21"} +{"current_steps": 955, "total_steps": 2240, "loss": 0.1769, "lr": 6.200790415218568e-05, "epoch": 3.4107142857142856, "percentage": 42.63, "elapsed_time": "4:09:31", "remaining_time": "5:35:45"} +{"current_steps": 960, "total_steps": 2240, "loss": 0.1775, "lr": 6.166541262301468e-05, "epoch": 3.4285714285714284, "percentage": 42.86, "elapsed_time": "4:10:44", "remaining_time": "5:34:18"} +{"current_steps": 965, "total_steps": 2240, "loss": 0.1835, "lr": 6.13223412544905e-05, "epoch": 3.4464285714285716, "percentage": 43.08, "elapsed_time": "4:11:56", "remaining_time": "5:32:52"} +{"current_steps": 970, "total_steps": 2240, "loss": 0.175, "lr": 6.0978707099270214e-05, "epoch": 3.4642857142857144, "percentage": 43.3, "elapsed_time": "4:13:11", "remaining_time": "5:31:30"} +{"current_steps": 975, "total_steps": 2240, "loss": 0.1797, "lr": 6.0634527237984604e-05, "epoch": 3.482142857142857, "percentage": 43.53, "elapsed_time": "4:14:23", "remaining_time": "5:30:02"} +{"current_steps": 980, "total_steps": 2240, "loss": 0.1761, "lr": 6.028981877838925e-05, "epoch": 3.5, "percentage": 43.75, "elapsed_time": "4:15:35", "remaining_time": "5:28:37"} +{"current_steps": 985, "total_steps": 2240, "loss": 0.178, "lr": 5.994459885451422e-05, "epoch": 3.517857142857143, "percentage": 43.97, "elapsed_time": "4:16:45", "remaining_time": "5:27:08"} +{"current_steps": 990, "total_steps": 2240, "loss": 0.1773, "lr": 5.9598884625812315e-05, "epoch": 3.5357142857142856, "percentage": 44.2, "elapsed_time": "4:17:58", "remaining_time": "5:25:42"} +{"current_steps": 995, "total_steps": 2240, "loss": 0.1802, "lr": 5.925269327630615e-05, "epoch": 3.553571428571429, "percentage": 44.42, "elapsed_time": "4:19:10", "remaining_time": "5:24:18"} +{"current_steps": 1000, "total_steps": 2240, "loss": 0.1723, "lr": 5.890604201373411e-05, "epoch": 3.571428571428571, "percentage": 44.64, "elapsed_time": "4:20:20", "remaining_time": "5:22:49"} +{"current_steps": 1005, "total_steps": 2240, "loss": 0.1847, "lr": 5.855894806869493e-05, "epoch": 3.5892857142857144, "percentage": 44.87, "elapsed_time": "4:31:23", "remaining_time": "5:33:30"} +{"current_steps": 1010, "total_steps": 2240, "loss": 0.1768, "lr": 5.82114286937912e-05, "epoch": 3.607142857142857, "percentage": 45.09, "elapsed_time": "4:32:32", "remaining_time": "5:31:54"} +{"current_steps": 1015, "total_steps": 2240, "loss": 0.1736, "lr": 5.786350116277195e-05, "epoch": 3.625, "percentage": 45.31, "elapsed_time": "4:33:39", "remaining_time": "5:30:16"} +{"current_steps": 1020, "total_steps": 2240, "loss": 0.1785, "lr": 5.7515182769673915e-05, "epoch": 3.642857142857143, "percentage": 45.54, "elapsed_time": "4:34:49", "remaining_time": "5:28:42"} +{"current_steps": 1025, "total_steps": 2240, "loss": 0.1792, "lr": 5.716649082796198e-05, "epoch": 3.6607142857142856, "percentage": 45.76, "elapsed_time": "4:35:54", "remaining_time": "5:27:03"} +{"current_steps": 1030, "total_steps": 2240, "loss": 0.1803, "lr": 5.681744266966856e-05, "epoch": 3.678571428571429, "percentage": 45.98, "elapsed_time": "4:37:10", "remaining_time": "5:25:37"} +{"current_steps": 1035, "total_steps": 2240, "loss": 0.175, "lr": 5.6468055644532156e-05, "epoch": 3.696428571428571, "percentage": 46.21, "elapsed_time": "4:38:22", "remaining_time": "5:24:06"} +{"current_steps": 1040, "total_steps": 2240, "loss": 0.1758, "lr": 5.6118347119134916e-05, "epoch": 3.7142857142857144, "percentage": 46.43, "elapsed_time": "4:39:35", "remaining_time": "5:22:36"} +{"current_steps": 1045, "total_steps": 2240, "loss": 0.1849, "lr": 5.576833447603943e-05, "epoch": 3.732142857142857, "percentage": 46.65, "elapsed_time": "4:40:47", "remaining_time": "5:21:06"} +{"current_steps": 1050, "total_steps": 2240, "loss": 0.183, "lr": 5.541803511292474e-05, "epoch": 3.75, "percentage": 46.88, "elapsed_time": "4:41:59", "remaining_time": "5:19:35"} +{"current_steps": 1055, "total_steps": 2240, "loss": 0.1787, "lr": 5.506746644172154e-05, "epoch": 3.767857142857143, "percentage": 47.1, "elapsed_time": "4:43:09", "remaining_time": "5:18:02"} +{"current_steps": 1060, "total_steps": 2240, "loss": 0.1762, "lr": 5.471664588774671e-05, "epoch": 3.7857142857142856, "percentage": 47.32, "elapsed_time": "4:44:22", "remaining_time": "5:16:34"} +{"current_steps": 1065, "total_steps": 2240, "loss": 0.1814, "lr": 5.436559088883717e-05, "epoch": 3.803571428571429, "percentage": 47.54, "elapsed_time": "4:45:35", "remaining_time": "5:15:05"} +{"current_steps": 1070, "total_steps": 2240, "loss": 0.1811, "lr": 5.4014318894483175e-05, "epoch": 3.821428571428571, "percentage": 47.77, "elapsed_time": "4:46:45", "remaining_time": "5:13:33"} +{"current_steps": 1075, "total_steps": 2240, "loss": 0.1845, "lr": 5.3662847364960855e-05, "epoch": 3.8392857142857144, "percentage": 47.99, "elapsed_time": "4:47:55", "remaining_time": "5:12:01"} +{"current_steps": 1080, "total_steps": 2240, "loss": 0.1755, "lr": 5.331119377046446e-05, "epoch": 3.857142857142857, "percentage": 48.21, "elapsed_time": "4:49:05", "remaining_time": "5:10:30"} +{"current_steps": 1085, "total_steps": 2240, "loss": 0.1767, "lr": 5.295937559023794e-05, "epoch": 3.875, "percentage": 48.44, "elapsed_time": "4:50:18", "remaining_time": "5:09:02"} +{"current_steps": 1090, "total_steps": 2240, "loss": 0.1772, "lr": 5.260741031170605e-05, "epoch": 3.892857142857143, "percentage": 48.66, "elapsed_time": "4:51:30", "remaining_time": "5:07:33"} +{"current_steps": 1095, "total_steps": 2240, "loss": 0.1742, "lr": 5.225531542960528e-05, "epoch": 3.9107142857142856, "percentage": 48.88, "elapsed_time": "4:52:42", "remaining_time": "5:06:04"} +{"current_steps": 1100, "total_steps": 2240, "loss": 0.1799, "lr": 5.190310844511412e-05, "epoch": 3.928571428571429, "percentage": 49.11, "elapsed_time": "4:53:53", "remaining_time": "5:04:34"} +{"current_steps": 1105, "total_steps": 2240, "loss": 0.1785, "lr": 5.155080686498324e-05, "epoch": 3.946428571428571, "percentage": 49.33, "elapsed_time": "5:00:26", "remaining_time": "5:08:35"} +{"current_steps": 1110, "total_steps": 2240, "loss": 0.1807, "lr": 5.1198428200665227e-05, "epoch": 3.9642857142857144, "percentage": 49.55, "elapsed_time": "5:01:36", "remaining_time": "5:07:02"} +{"current_steps": 1115, "total_steps": 2240, "loss": 0.1704, "lr": 5.0845989967444255e-05, "epoch": 3.982142857142857, "percentage": 49.78, "elapsed_time": "5:02:45", "remaining_time": "5:05:28"} +{"current_steps": 1120, "total_steps": 2240, "loss": 0.1715, "lr": 5.0493509683565365e-05, "epoch": 4.0, "percentage": 50.0, "elapsed_time": "5:03:58", "remaining_time": "5:03:58"} +{"current_steps": 1125, "total_steps": 2240, "loss": 0.1216, "lr": 5.014100486936383e-05, "epoch": 4.017857142857143, "percentage": 50.22, "elapsed_time": "5:05:12", "remaining_time": "5:02:29"} +{"current_steps": 1130, "total_steps": 2240, "loss": 0.1248, "lr": 4.9788493046394136e-05, "epoch": 4.035714285714286, "percentage": 50.45, "elapsed_time": "5:06:23", "remaining_time": "5:00:58"} +{"current_steps": 1135, "total_steps": 2240, "loss": 0.1195, "lr": 4.9435991736559245e-05, "epoch": 4.053571428571429, "percentage": 50.67, "elapsed_time": "5:07:32", "remaining_time": "4:59:24"} +{"current_steps": 1140, "total_steps": 2240, "loss": 0.1214, "lr": 4.908351846123947e-05, "epoch": 4.071428571428571, "percentage": 50.89, "elapsed_time": "5:08:44", "remaining_time": "4:57:54"} +{"current_steps": 1145, "total_steps": 2240, "loss": 0.1217, "lr": 4.8731090740421685e-05, "epoch": 4.089285714285714, "percentage": 51.12, "elapsed_time": "5:09:56", "remaining_time": "4:56:24"} +{"current_steps": 1150, "total_steps": 2240, "loss": 0.1243, "lr": 4.837872609182846e-05, "epoch": 4.107142857142857, "percentage": 51.34, "elapsed_time": "5:11:09", "remaining_time": "4:54:55"} +{"current_steps": 1155, "total_steps": 2240, "loss": 0.1225, "lr": 4.802644203004723e-05, "epoch": 4.125, "percentage": 51.56, "elapsed_time": "5:12:21", "remaining_time": "4:53:26"} +{"current_steps": 1160, "total_steps": 2240, "loss": 0.1214, "lr": 4.767425606565987e-05, "epoch": 4.142857142857143, "percentage": 51.79, "elapsed_time": "5:13:32", "remaining_time": "4:51:54"} +{"current_steps": 1165, "total_steps": 2240, "loss": 0.1208, "lr": 4.7322185704372234e-05, "epoch": 4.160714285714286, "percentage": 52.01, "elapsed_time": "5:14:44", "remaining_time": "4:50:25"} +{"current_steps": 1170, "total_steps": 2240, "loss": 0.125, "lr": 4.697024844614396e-05, "epoch": 4.178571428571429, "percentage": 52.23, "elapsed_time": "5:15:56", "remaining_time": "4:48:56"} +{"current_steps": 1175, "total_steps": 2240, "loss": 0.1226, "lr": 4.661846178431873e-05, "epoch": 4.196428571428571, "percentage": 52.46, "elapsed_time": "5:17:06", "remaining_time": "4:47:25"} +{"current_steps": 1180, "total_steps": 2240, "loss": 0.1288, "lr": 4.626684320475473e-05, "epoch": 4.214285714285714, "percentage": 52.68, "elapsed_time": "5:18:15", "remaining_time": "4:45:53"} +{"current_steps": 1185, "total_steps": 2240, "loss": 0.1204, "lr": 4.591541018495542e-05, "epoch": 4.232142857142857, "percentage": 52.9, "elapsed_time": "5:19:26", "remaining_time": "4:44:23"} +{"current_steps": 1190, "total_steps": 2240, "loss": 0.1216, "lr": 4.556418019320087e-05, "epoch": 4.25, "percentage": 53.12, "elapsed_time": "5:20:38", "remaining_time": "4:42:55"} +{"current_steps": 1195, "total_steps": 2240, "loss": 0.1298, "lr": 4.521317068767949e-05, "epoch": 4.267857142857143, "percentage": 53.35, "elapsed_time": "5:21:51", "remaining_time": "4:41:27"} +{"current_steps": 1200, "total_steps": 2240, "loss": 0.1252, "lr": 4.4862399115620184e-05, "epoch": 4.285714285714286, "percentage": 53.57, "elapsed_time": "5:22:59", "remaining_time": "4:39:55"} +{"current_steps": 1205, "total_steps": 2240, "loss": 0.1246, "lr": 4.451188291242521e-05, "epoch": 4.303571428571429, "percentage": 53.79, "elapsed_time": "5:36:51", "remaining_time": "4:49:19"} +{"current_steps": 1210, "total_steps": 2240, "loss": 0.1301, "lr": 4.4161639500803515e-05, "epoch": 4.321428571428571, "percentage": 54.02, "elapsed_time": "5:38:00", "remaining_time": "4:47:43"} +{"current_steps": 1215, "total_steps": 2240, "loss": 0.1284, "lr": 4.381168628990464e-05, "epoch": 4.339285714285714, "percentage": 54.24, "elapsed_time": "5:39:07", "remaining_time": "4:46:05"} +{"current_steps": 1220, "total_steps": 2240, "loss": 0.1213, "lr": 4.346204067445348e-05, "epoch": 4.357142857142857, "percentage": 54.46, "elapsed_time": "5:40:19", "remaining_time": "4:44:31"} +{"current_steps": 1225, "total_steps": 2240, "loss": 0.1211, "lr": 4.311272003388568e-05, "epoch": 4.375, "percentage": 54.69, "elapsed_time": "5:41:30", "remaining_time": "4:42:57"} +{"current_steps": 1230, "total_steps": 2240, "loss": 0.1252, "lr": 4.2763741731483664e-05, "epoch": 4.392857142857143, "percentage": 54.91, "elapsed_time": "5:42:39", "remaining_time": "4:41:22"} +{"current_steps": 1235, "total_steps": 2240, "loss": 0.1265, "lr": 4.241512311351366e-05, "epoch": 4.410714285714286, "percentage": 55.13, "elapsed_time": "5:43:55", "remaining_time": "4:39:52"} +{"current_steps": 1240, "total_steps": 2240, "loss": 0.1304, "lr": 4.2066881508363523e-05, "epoch": 4.428571428571429, "percentage": 55.36, "elapsed_time": "5:45:09", "remaining_time": "4:38:21"} +{"current_steps": 1245, "total_steps": 2240, "loss": 0.1266, "lr": 4.1719034225681274e-05, "epoch": 4.446428571428571, "percentage": 55.58, "elapsed_time": "5:46:20", "remaining_time": "4:36:47"} +{"current_steps": 1250, "total_steps": 2240, "loss": 0.1339, "lr": 4.137159855551486e-05, "epoch": 4.464285714285714, "percentage": 55.8, "elapsed_time": "5:47:32", "remaining_time": "4:35:15"} +{"current_steps": 1255, "total_steps": 2240, "loss": 0.1251, "lr": 4.102459176745267e-05, "epoch": 4.482142857142857, "percentage": 56.03, "elapsed_time": "5:48:45", "remaining_time": "4:33:43"} +{"current_steps": 1260, "total_steps": 2240, "loss": 0.1307, "lr": 4.067803110976513e-05, "epoch": 4.5, "percentage": 56.25, "elapsed_time": "5:49:56", "remaining_time": "4:32:10"} +{"current_steps": 1265, "total_steps": 2240, "loss": 0.1244, "lr": 4.033193380854734e-05, "epoch": 4.517857142857143, "percentage": 56.47, "elapsed_time": "5:51:11", "remaining_time": "4:30:41"} +{"current_steps": 1270, "total_steps": 2240, "loss": 0.1264, "lr": 3.998631706686292e-05, "epoch": 4.535714285714286, "percentage": 56.7, "elapsed_time": "5:52:23", "remaining_time": "4:29:09"} +{"current_steps": 1275, "total_steps": 2240, "loss": 0.1211, "lr": 3.964119806388887e-05, "epoch": 4.553571428571429, "percentage": 56.92, "elapsed_time": "5:53:35", "remaining_time": "4:27:36"} +{"current_steps": 1280, "total_steps": 2240, "loss": 0.1255, "lr": 3.929659395406159e-05, "epoch": 4.571428571428571, "percentage": 57.14, "elapsed_time": "5:54:40", "remaining_time": "4:26:00"} +{"current_steps": 1285, "total_steps": 2240, "loss": 0.1275, "lr": 3.895252186622433e-05, "epoch": 4.589285714285714, "percentage": 57.37, "elapsed_time": "5:55:53", "remaining_time": "4:24:29"} +{"current_steps": 1290, "total_steps": 2240, "loss": 0.1196, "lr": 3.860899890277575e-05, "epoch": 4.607142857142857, "percentage": 57.59, "elapsed_time": "5:57:03", "remaining_time": "4:22:56"} +{"current_steps": 1295, "total_steps": 2240, "loss": 0.1245, "lr": 3.8266042138819746e-05, "epoch": 4.625, "percentage": 57.81, "elapsed_time": "5:58:11", "remaining_time": "4:21:22"} +{"current_steps": 1300, "total_steps": 2240, "loss": 0.1211, "lr": 3.7923668621316824e-05, "epoch": 4.642857142857143, "percentage": 58.04, "elapsed_time": "5:59:20", "remaining_time": "4:19:49"} +{"current_steps": 1305, "total_steps": 2240, "loss": 0.1254, "lr": 3.758189536823673e-05, "epoch": 4.660714285714286, "percentage": 58.26, "elapsed_time": "6:06:16", "remaining_time": "4:22:25"} +{"current_steps": 1310, "total_steps": 2240, "loss": 0.1232, "lr": 3.724073936771252e-05, "epoch": 4.678571428571429, "percentage": 58.48, "elapsed_time": "6:07:26", "remaining_time": "4:20:51"} +{"current_steps": 1315, "total_steps": 2240, "loss": 0.126, "lr": 3.6900217577196185e-05, "epoch": 4.696428571428571, "percentage": 58.71, "elapsed_time": "6:08:41", "remaining_time": "4:19:20"} +{"current_steps": 1320, "total_steps": 2240, "loss": 0.1245, "lr": 3.656034692261582e-05, "epoch": 4.714285714285714, "percentage": 58.93, "elapsed_time": "6:09:57", "remaining_time": "4:17:51"} +{"current_steps": 1325, "total_steps": 2240, "loss": 0.1244, "lr": 3.622114429753418e-05, "epoch": 4.732142857142857, "percentage": 59.15, "elapsed_time": "6:11:04", "remaining_time": "4:16:15"} +{"current_steps": 1330, "total_steps": 2240, "loss": 0.1241, "lr": 3.588262656230904e-05, "epoch": 4.75, "percentage": 59.38, "elapsed_time": "6:12:19", "remaining_time": "4:14:44"} +{"current_steps": 1335, "total_steps": 2240, "loss": 0.1277, "lr": 3.554481054325522e-05, "epoch": 4.767857142857143, "percentage": 59.6, "elapsed_time": "6:13:29", "remaining_time": "4:13:11"} +{"current_steps": 1340, "total_steps": 2240, "loss": 0.1278, "lr": 3.520771303180803e-05, "epoch": 4.785714285714286, "percentage": 59.82, "elapsed_time": "6:14:45", "remaining_time": "4:11:42"} +{"current_steps": 1345, "total_steps": 2240, "loss": 0.1196, "lr": 3.48713507836888e-05, "epoch": 4.803571428571429, "percentage": 60.04, "elapsed_time": "6:15:53", "remaining_time": "4:10:07"} +{"current_steps": 1350, "total_steps": 2240, "loss": 0.122, "lr": 3.4535740518071966e-05, "epoch": 4.821428571428571, "percentage": 60.27, "elapsed_time": "6:17:01", "remaining_time": "4:08:33"} +{"current_steps": 1355, "total_steps": 2240, "loss": 0.1229, "lr": 3.420089891675401e-05, "epoch": 4.839285714285714, "percentage": 60.49, "elapsed_time": "6:18:13", "remaining_time": "4:07:01"} +{"current_steps": 1360, "total_steps": 2240, "loss": 0.1282, "lr": 3.386684262332429e-05, "epoch": 4.857142857142857, "percentage": 60.71, "elapsed_time": "6:19:25", "remaining_time": "4:05:30"} +{"current_steps": 1365, "total_steps": 2240, "loss": 0.1269, "lr": 3.35335882423378e-05, "epoch": 4.875, "percentage": 60.94, "elapsed_time": "6:20:28", "remaining_time": "4:03:53"} +{"current_steps": 1370, "total_steps": 2240, "loss": 0.1214, "lr": 3.3201152338489726e-05, "epoch": 4.892857142857143, "percentage": 61.16, "elapsed_time": "6:21:37", "remaining_time": "4:02:20"} +{"current_steps": 1375, "total_steps": 2240, "loss": 0.1226, "lr": 3.2869551435792184e-05, "epoch": 4.910714285714286, "percentage": 61.38, "elapsed_time": "6:22:43", "remaining_time": "4:00:46"} +{"current_steps": 1380, "total_steps": 2240, "loss": 0.1232, "lr": 3.253880201675287e-05, "epoch": 4.928571428571429, "percentage": 61.61, "elapsed_time": "6:23:48", "remaining_time": "3:59:11"} +{"current_steps": 1385, "total_steps": 2240, "loss": 0.1251, "lr": 3.2208920521555676e-05, "epoch": 4.946428571428571, "percentage": 61.83, "elapsed_time": "6:24:58", "remaining_time": "3:57:39"} +{"current_steps": 1390, "total_steps": 2240, "loss": 0.1296, "lr": 3.187992334724363e-05, "epoch": 4.964285714285714, "percentage": 62.05, "elapsed_time": "6:26:10", "remaining_time": "3:56:09"} +{"current_steps": 1395, "total_steps": 2240, "loss": 0.1254, "lr": 3.1551826846903896e-05, "epoch": 4.982142857142857, "percentage": 62.28, "elapsed_time": "6:27:14", "remaining_time": "3:54:34"} +{"current_steps": 1400, "total_steps": 2240, "loss": 0.12, "lr": 3.122464732885476e-05, "epoch": 5.0, "percentage": 62.5, "elapsed_time": "6:28:21", "remaining_time": "3:53:01"} +{"current_steps": 1405, "total_steps": 2240, "loss": 0.0823, "lr": 3.0898401055835156e-05, "epoch": 5.017857142857143, "percentage": 62.72, "elapsed_time": "6:34:50", "remaining_time": "3:54:39"} +{"current_steps": 1410, "total_steps": 2240, "loss": 0.0837, "lr": 3.057310424419632e-05, "epoch": 5.035714285714286, "percentage": 62.95, "elapsed_time": "6:36:02", "remaining_time": "3:53:07"} +{"current_steps": 1415, "total_steps": 2240, "loss": 0.076, "lr": 3.0248773063095603e-05, "epoch": 5.053571428571429, "percentage": 63.17, "elapsed_time": "6:37:16", "remaining_time": "3:51:37"} +{"current_steps": 1420, "total_steps": 2240, "loss": 0.076, "lr": 2.9925423633692923e-05, "epoch": 5.071428571428571, "percentage": 63.39, "elapsed_time": "6:38:21", "remaining_time": "3:50:02"} +{"current_steps": 1425, "total_steps": 2240, "loss": 0.0762, "lr": 2.960307202834941e-05, "epoch": 5.089285714285714, "percentage": 63.62, "elapsed_time": "6:39:29", "remaining_time": "3:48:28"} +{"current_steps": 1430, "total_steps": 2240, "loss": 0.0775, "lr": 2.9281734269828408e-05, "epoch": 5.107142857142857, "percentage": 63.84, "elapsed_time": "6:40:39", "remaining_time": "3:46:56"} +{"current_steps": 1435, "total_steps": 2240, "loss": 0.0748, "lr": 2.896142633049922e-05, "epoch": 5.125, "percentage": 64.06, "elapsed_time": "6:41:48", "remaining_time": "3:45:24"} +{"current_steps": 1440, "total_steps": 2240, "loss": 0.0769, "lr": 2.8642164131543048e-05, "epoch": 5.142857142857143, "percentage": 64.29, "elapsed_time": "6:42:54", "remaining_time": "3:43:50"} +{"current_steps": 1445, "total_steps": 2240, "loss": 0.077, "lr": 2.8323963542161663e-05, "epoch": 5.160714285714286, "percentage": 64.51, "elapsed_time": "6:44:06", "remaining_time": "3:42:19"} +{"current_steps": 1450, "total_steps": 2240, "loss": 0.077, "lr": 2.800684037878867e-05, "epoch": 5.178571428571429, "percentage": 64.73, "elapsed_time": "6:45:19", "remaining_time": "3:40:50"} +{"current_steps": 1455, "total_steps": 2240, "loss": 0.0815, "lr": 2.7690810404303276e-05, "epoch": 5.196428571428571, "percentage": 64.96, "elapsed_time": "6:46:33", "remaining_time": "3:39:21"} +{"current_steps": 1460, "total_steps": 2240, "loss": 0.0767, "lr": 2.7375889327246744e-05, "epoch": 5.214285714285714, "percentage": 65.18, "elapsed_time": "6:47:49", "remaining_time": "3:37:52"} +{"current_steps": 1465, "total_steps": 2240, "loss": 0.0767, "lr": 2.7062092801041716e-05, "epoch": 5.232142857142857, "percentage": 65.4, "elapsed_time": "6:48:57", "remaining_time": "3:36:20"} +{"current_steps": 1470, "total_steps": 2240, "loss": 0.0794, "lr": 2.6749436423214e-05, "epoch": 5.25, "percentage": 65.62, "elapsed_time": "6:50:09", "remaining_time": "3:34:50"} +{"current_steps": 1475, "total_steps": 2240, "loss": 0.0781, "lr": 2.6437935734617393e-05, "epoch": 5.267857142857143, "percentage": 65.85, "elapsed_time": "6:51:18", "remaining_time": "3:33:19"} +{"current_steps": 1480, "total_steps": 2240, "loss": 0.0784, "lr": 2.612760621866113e-05, "epoch": 5.285714285714286, "percentage": 66.07, "elapsed_time": "6:52:30", "remaining_time": "3:31:49"} +{"current_steps": 1485, "total_steps": 2240, "loss": 0.0785, "lr": 2.581846330054034e-05, "epoch": 5.303571428571429, "percentage": 66.29, "elapsed_time": "6:53:40", "remaining_time": "3:30:19"} +{"current_steps": 1490, "total_steps": 2240, "loss": 0.0801, "lr": 2.551052234646929e-05, "epoch": 5.321428571428571, "percentage": 66.52, "elapsed_time": "6:54:54", "remaining_time": "3:28:50"} +{"current_steps": 1495, "total_steps": 2240, "loss": 0.076, "lr": 2.5203798662917555e-05, "epoch": 5.339285714285714, "percentage": 66.74, "elapsed_time": "6:56:11", "remaining_time": "3:27:23"} +{"current_steps": 1500, "total_steps": 2240, "loss": 0.0779, "lr": 2.489830749584924e-05, "epoch": 5.357142857142857, "percentage": 66.96, "elapsed_time": "6:57:24", "remaining_time": "3:25:55"} +{"current_steps": 1505, "total_steps": 2240, "loss": 0.0798, "lr": 2.4594064029965197e-05, "epoch": 5.375, "percentage": 67.19, "elapsed_time": "7:05:39", "remaining_time": "3:27:52"} +{"current_steps": 1510, "total_steps": 2240, "loss": 0.0787, "lr": 2.429108338794817e-05, "epoch": 5.392857142857143, "percentage": 67.41, "elapsed_time": "7:06:51", "remaining_time": "3:26:21"} +{"current_steps": 1515, "total_steps": 2240, "loss": 0.0791, "lr": 2.3989380629711194e-05, "epoch": 5.410714285714286, "percentage": 67.63, "elapsed_time": "7:08:04", "remaining_time": "3:24:51"} +{"current_steps": 1520, "total_steps": 2240, "loss": 0.0837, "lr": 2.3688970751649002e-05, "epoch": 5.428571428571429, "percentage": 67.86, "elapsed_time": "7:09:14", "remaining_time": "3:23:19"} +{"current_steps": 1525, "total_steps": 2240, "loss": 0.0764, "lr": 2.3389868685892573e-05, "epoch": 5.446428571428571, "percentage": 68.08, "elapsed_time": "7:10:24", "remaining_time": "3:21:47"} +{"current_steps": 1530, "total_steps": 2240, "loss": 0.0743, "lr": 2.309208929956694e-05, "epoch": 5.464285714285714, "percentage": 68.3, "elapsed_time": "7:11:35", "remaining_time": "3:20:16"} +{"current_steps": 1535, "total_steps": 2240, "loss": 0.0749, "lr": 2.2795647394052284e-05, "epoch": 5.482142857142857, "percentage": 68.53, "elapsed_time": "7:12:44", "remaining_time": "3:18:45"} +{"current_steps": 1540, "total_steps": 2240, "loss": 0.0761, "lr": 2.2500557704248083e-05, "epoch": 5.5, "percentage": 68.75, "elapsed_time": "7:13:58", "remaining_time": "3:17:15"} +{"current_steps": 1545, "total_steps": 2240, "loss": 0.0814, "lr": 2.220683489784081e-05, "epoch": 5.517857142857143, "percentage": 68.97, "elapsed_time": "7:15:12", "remaining_time": "3:15:46"} +{"current_steps": 1550, "total_steps": 2240, "loss": 0.0768, "lr": 2.1914493574574858e-05, "epoch": 5.535714285714286, "percentage": 69.2, "elapsed_time": "7:16:20", "remaining_time": "3:14:14"} +{"current_steps": 1555, "total_steps": 2240, "loss": 0.0806, "lr": 2.162354826552673e-05, "epoch": 5.553571428571429, "percentage": 69.42, "elapsed_time": "7:17:29", "remaining_time": "3:12:43"} +{"current_steps": 1560, "total_steps": 2240, "loss": 0.0799, "lr": 2.1334013432382894e-05, "epoch": 5.571428571428571, "percentage": 69.64, "elapsed_time": "7:18:33", "remaining_time": "3:11:09"} +{"current_steps": 1565, "total_steps": 2240, "loss": 0.0808, "lr": 2.1045903466720913e-05, "epoch": 5.589285714285714, "percentage": 69.87, "elapsed_time": "7:19:46", "remaining_time": "3:09:40"} +{"current_steps": 1570, "total_steps": 2240, "loss": 0.0782, "lr": 2.0759232689294044e-05, "epoch": 5.607142857142857, "percentage": 70.09, "elapsed_time": "7:20:57", "remaining_time": "3:08:10"} +{"current_steps": 1575, "total_steps": 2240, "loss": 0.0823, "lr": 2.0474015349319503e-05, "epoch": 5.625, "percentage": 70.31, "elapsed_time": "7:22:10", "remaining_time": "3:06:41"} +{"current_steps": 1580, "total_steps": 2240, "loss": 0.0764, "lr": 2.0190265623770143e-05, "epoch": 5.642857142857143, "percentage": 70.54, "elapsed_time": "7:23:19", "remaining_time": "3:05:11"} +{"current_steps": 1585, "total_steps": 2240, "loss": 0.0784, "lr": 1.990799761666975e-05, "epoch": 5.660714285714286, "percentage": 70.76, "elapsed_time": "7:24:26", "remaining_time": "3:03:39"} +{"current_steps": 1590, "total_steps": 2240, "loss": 0.0807, "lr": 1.962722535839202e-05, "epoch": 5.678571428571429, "percentage": 70.98, "elapsed_time": "7:25:36", "remaining_time": "3:02:09"} +{"current_steps": 1595, "total_steps": 2240, "loss": 0.0775, "lr": 1.9347962804963238e-05, "epoch": 5.696428571428571, "percentage": 71.21, "elapsed_time": "7:26:48", "remaining_time": "3:00:41"} +{"current_steps": 1600, "total_steps": 2240, "loss": 0.0798, "lr": 1.9070223837368412e-05, "epoch": 5.714285714285714, "percentage": 71.43, "elapsed_time": "7:28:00", "remaining_time": "2:59:12"} +{"current_steps": 1605, "total_steps": 2240, "loss": 0.0781, "lr": 1.8794022260861482e-05, "epoch": 5.732142857142857, "percentage": 71.65, "elapsed_time": "7:33:01", "remaining_time": "2:59:13"} +{"current_steps": 1610, "total_steps": 2240, "loss": 0.0783, "lr": 1.8519371804279046e-05, "epoch": 5.75, "percentage": 71.88, "elapsed_time": "7:34:12", "remaining_time": "2:57:44"} +{"current_steps": 1615, "total_steps": 2240, "loss": 0.0773, "lr": 1.8246286119357903e-05, "epoch": 5.767857142857143, "percentage": 72.1, "elapsed_time": "7:35:20", "remaining_time": "2:56:12"} +{"current_steps": 1620, "total_steps": 2240, "loss": 0.0766, "lr": 1.797477878005655e-05, "epoch": 5.785714285714286, "percentage": 72.32, "elapsed_time": "7:36:29", "remaining_time": "2:54:42"} +{"current_steps": 1625, "total_steps": 2240, "loss": 0.0773, "lr": 1.7704863281880496e-05, "epoch": 5.803571428571429, "percentage": 72.54, "elapsed_time": "7:37:39", "remaining_time": "2:53:12"} +{"current_steps": 1630, "total_steps": 2240, "loss": 0.0824, "lr": 1.743655304121136e-05, "epoch": 5.821428571428571, "percentage": 72.77, "elapsed_time": "7:38:48", "remaining_time": "2:51:42"} +{"current_steps": 1635, "total_steps": 2240, "loss": 0.0776, "lr": 1.7169861394640107e-05, "epoch": 5.839285714285714, "percentage": 72.99, "elapsed_time": "7:40:01", "remaining_time": "2:50:13"} +{"current_steps": 1640, "total_steps": 2240, "loss": 0.083, "lr": 1.6904801598304095e-05, "epoch": 5.857142857142857, "percentage": 73.21, "elapsed_time": "7:41:11", "remaining_time": "2:48:43"} +{"current_steps": 1645, "total_steps": 2240, "loss": 0.0795, "lr": 1.6641386827228105e-05, "epoch": 5.875, "percentage": 73.44, "elapsed_time": "7:42:20", "remaining_time": "2:47:13"} +{"current_steps": 1650, "total_steps": 2240, "loss": 0.0794, "lr": 1.637963017466961e-05, "epoch": 5.892857142857143, "percentage": 73.66, "elapsed_time": "7:43:32", "remaining_time": "2:45:45"} +{"current_steps": 1655, "total_steps": 2240, "loss": 0.0782, "lr": 1.61195446514678e-05, "epoch": 5.910714285714286, "percentage": 73.88, "elapsed_time": "7:44:48", "remaining_time": "2:44:17"} +{"current_steps": 1660, "total_steps": 2240, "loss": 0.0751, "lr": 1.586114318539697e-05, "epoch": 5.928571428571429, "percentage": 74.11, "elapsed_time": "7:46:01", "remaining_time": "2:42:49"} +{"current_steps": 1665, "total_steps": 2240, "loss": 0.0791, "lr": 1.560443862052393e-05, "epoch": 5.946428571428571, "percentage": 74.33, "elapsed_time": "7:47:14", "remaining_time": "2:41:21"} +{"current_steps": 1670, "total_steps": 2240, "loss": 0.0801, "lr": 1.534944371656955e-05, "epoch": 5.964285714285714, "percentage": 74.55, "elapsed_time": "7:48:24", "remaining_time": "2:39:52"} +{"current_steps": 1675, "total_steps": 2240, "loss": 0.071, "lr": 1.5096171148274546e-05, "epoch": 5.982142857142857, "percentage": 74.78, "elapsed_time": "7:49:31", "remaining_time": "2:38:22"} +{"current_steps": 1680, "total_steps": 2240, "loss": 0.0763, "lr": 1.4844633504769422e-05, "epoch": 6.0, "percentage": 75.0, "elapsed_time": "7:50:41", "remaining_time": "2:36:53"} +{"current_steps": 1685, "total_steps": 2240, "loss": 0.0494, "lr": 1.4594843288948773e-05, "epoch": 6.017857142857143, "percentage": 75.22, "elapsed_time": "7:51:50", "remaining_time": "2:35:24"} +{"current_steps": 1690, "total_steps": 2240, "loss": 0.0458, "lr": 1.4346812916849839e-05, "epoch": 6.035714285714286, "percentage": 75.45, "elapsed_time": "7:53:01", "remaining_time": "2:33:56"} +{"current_steps": 1695, "total_steps": 2240, "loss": 0.0447, "lr": 1.4100554717035241e-05, "epoch": 6.053571428571429, "percentage": 75.67, "elapsed_time": "7:54:12", "remaining_time": "2:32:28"} +{"current_steps": 1700, "total_steps": 2240, "loss": 0.0418, "lr": 1.385608092998032e-05, "epoch": 6.071428571428571, "percentage": 75.89, "elapsed_time": "7:55:21", "remaining_time": "2:30:59"} +{"current_steps": 1705, "total_steps": 2240, "loss": 0.0453, "lr": 1.3613403707464639e-05, "epoch": 6.089285714285714, "percentage": 76.12, "elapsed_time": "7:59:54", "remaining_time": "2:30:35"} +{"current_steps": 1710, "total_steps": 2240, "loss": 0.0428, "lr": 1.3372535111967949e-05, "epoch": 6.107142857142857, "percentage": 76.34, "elapsed_time": "8:01:07", "remaining_time": "2:29:07"} +{"current_steps": 1715, "total_steps": 2240, "loss": 0.047, "lr": 1.3133487116070643e-05, "epoch": 6.125, "percentage": 76.56, "elapsed_time": "8:02:19", "remaining_time": "2:27:38"} +{"current_steps": 1720, "total_steps": 2240, "loss": 0.0473, "lr": 1.289627160185869e-05, "epoch": 6.142857142857143, "percentage": 76.79, "elapsed_time": "8:03:33", "remaining_time": "2:26:11"} +{"current_steps": 1725, "total_steps": 2240, "loss": 0.0476, "lr": 1.2660900360332928e-05, "epoch": 6.160714285714286, "percentage": 77.01, "elapsed_time": "8:04:44", "remaining_time": "2:24:43"} +{"current_steps": 1730, "total_steps": 2240, "loss": 0.0445, "lr": 1.2427385090823073e-05, "epoch": 6.178571428571429, "percentage": 77.23, "elapsed_time": "8:05:52", "remaining_time": "2:23:14"} +{"current_steps": 1735, "total_steps": 2240, "loss": 0.0429, "lr": 1.2195737400406166e-05, "epoch": 6.196428571428571, "percentage": 77.46, "elapsed_time": "8:07:01", "remaining_time": "2:21:45"} +{"current_steps": 1740, "total_steps": 2240, "loss": 0.0461, "lr": 1.1965968803329585e-05, "epoch": 6.214285714285714, "percentage": 77.68, "elapsed_time": "8:08:08", "remaining_time": "2:20:16"} +{"current_steps": 1745, "total_steps": 2240, "loss": 0.0455, "lr": 1.1738090720438782e-05, "epoch": 6.232142857142857, "percentage": 77.9, "elapsed_time": "8:09:15", "remaining_time": "2:18:47"} +{"current_steps": 1750, "total_steps": 2240, "loss": 0.0455, "lr": 1.1512114478609598e-05, "epoch": 6.25, "percentage": 78.12, "elapsed_time": "8:10:25", "remaining_time": "2:17:19"} +{"current_steps": 1755, "total_steps": 2240, "loss": 0.0441, "lr": 1.1288051310185182e-05, "epoch": 6.267857142857143, "percentage": 78.35, "elapsed_time": "8:11:32", "remaining_time": "2:15:50"} +{"current_steps": 1760, "total_steps": 2240, "loss": 0.0445, "lr": 1.1065912352417768e-05, "epoch": 6.285714285714286, "percentage": 78.57, "elapsed_time": "8:12:40", "remaining_time": "2:14:22"} +{"current_steps": 1765, "total_steps": 2240, "loss": 0.0468, "lr": 1.0845708646915054e-05, "epoch": 6.303571428571429, "percentage": 78.79, "elapsed_time": "8:13:49", "remaining_time": "2:12:53"} +{"current_steps": 1770, "total_steps": 2240, "loss": 0.0464, "lr": 1.0627451139091321e-05, "epoch": 6.321428571428571, "percentage": 79.02, "elapsed_time": "8:15:02", "remaining_time": "2:11:27"} +{"current_steps": 1775, "total_steps": 2240, "loss": 0.0454, "lr": 1.0411150677623438e-05, "epoch": 6.339285714285714, "percentage": 79.24, "elapsed_time": "8:16:11", "remaining_time": "2:09:59"} +{"current_steps": 1780, "total_steps": 2240, "loss": 0.0445, "lr": 1.0196818013911625e-05, "epoch": 6.357142857142857, "percentage": 79.46, "elapsed_time": "8:17:22", "remaining_time": "2:08:32"} +{"current_steps": 1785, "total_steps": 2240, "loss": 0.0516, "lr": 9.98446380154499e-06, "epoch": 6.375, "percentage": 79.69, "elapsed_time": "8:18:35", "remaining_time": "2:07:05"} +{"current_steps": 1790, "total_steps": 2240, "loss": 0.0466, "lr": 9.774098595772058e-06, "epoch": 6.392857142857143, "percentage": 79.91, "elapsed_time": "8:19:44", "remaining_time": "2:05:37"} +{"current_steps": 1795, "total_steps": 2240, "loss": 0.0491, "lr": 9.56573285297605e-06, "epoch": 6.410714285714286, "percentage": 80.13, "elapsed_time": "8:20:49", "remaining_time": "2:04:09"} +{"current_steps": 1800, "total_steps": 2240, "loss": 0.0448, "lr": 9.359376930155157e-06, "epoch": 6.428571428571429, "percentage": 80.36, "elapsed_time": "8:22:00", "remaining_time": "2:02:42"} +{"current_steps": 1805, "total_steps": 2240, "loss": 0.0465, "lr": 9.15504108440774e-06, "epoch": 6.446428571428571, "percentage": 80.58, "elapsed_time": "8:27:03", "remaining_time": "2:02:11"} +{"current_steps": 1810, "total_steps": 2240, "loss": 0.0441, "lr": 8.95273547242253e-06, "epoch": 6.464285714285714, "percentage": 80.8, "elapsed_time": "8:28:15", "remaining_time": "2:00:44"} +{"current_steps": 1815, "total_steps": 2240, "loss": 0.0429, "lr": 8.752470149973684e-06, "epoch": 6.482142857142857, "percentage": 81.03, "elapsed_time": "8:29:31", "remaining_time": "1:59:18"} +{"current_steps": 1820, "total_steps": 2240, "loss": 0.0437, "lr": 8.55425507142108e-06, "epoch": 6.5, "percentage": 81.25, "elapsed_time": "8:30:41", "remaining_time": "1:57:51"} +{"current_steps": 1825, "total_steps": 2240, "loss": 0.0456, "lr": 8.358100089215426e-06, "epoch": 6.517857142857143, "percentage": 81.47, "elapsed_time": "8:31:54", "remaining_time": "1:56:24"} +{"current_steps": 1830, "total_steps": 2240, "loss": 0.046, "lr": 8.164014953408578e-06, "epoch": 6.535714285714286, "percentage": 81.7, "elapsed_time": "8:33:02", "remaining_time": "1:54:56"} +{"current_steps": 1835, "total_steps": 2240, "loss": 0.0464, "lr": 7.972009311168882e-06, "epoch": 6.553571428571429, "percentage": 81.92, "elapsed_time": "8:34:10", "remaining_time": "1:53:28"} +{"current_steps": 1840, "total_steps": 2240, "loss": 0.0426, "lr": 7.782092706301719e-06, "epoch": 6.571428571428571, "percentage": 82.14, "elapsed_time": "8:35:24", "remaining_time": "1:52:02"} +{"current_steps": 1845, "total_steps": 2240, "loss": 0.0451, "lr": 7.5942745787750065e-06, "epoch": 6.589285714285714, "percentage": 82.37, "elapsed_time": "8:36:38", "remaining_time": "1:50:36"} +{"current_steps": 1850, "total_steps": 2240, "loss": 0.0436, "lr": 7.4085642642501005e-06, "epoch": 6.607142857142857, "percentage": 82.59, "elapsed_time": "8:37:51", "remaining_time": "1:49:10"} +{"current_steps": 1855, "total_steps": 2240, "loss": 0.0444, "lr": 7.224970993617686e-06, "epoch": 6.625, "percentage": 82.81, "elapsed_time": "8:39:06", "remaining_time": "1:47:44"} +{"current_steps": 1860, "total_steps": 2240, "loss": 0.0462, "lr": 7.04350389253895e-06, "epoch": 6.642857142857143, "percentage": 83.04, "elapsed_time": "8:40:22", "remaining_time": "1:46:18"} +{"current_steps": 1865, "total_steps": 2240, "loss": 0.0447, "lr": 6.864171980991985e-06, "epoch": 6.660714285714286, "percentage": 83.26, "elapsed_time": "8:41:34", "remaining_time": "1:44:52"} +{"current_steps": 1870, "total_steps": 2240, "loss": 0.0469, "lr": 6.686984172823491e-06, "epoch": 6.678571428571429, "percentage": 83.48, "elapsed_time": "8:42:41", "remaining_time": "1:43:25"} +{"current_steps": 1875, "total_steps": 2240, "loss": 0.0426, "lr": 6.5119492753056565e-06, "epoch": 6.696428571428571, "percentage": 83.71, "elapsed_time": "8:43:50", "remaining_time": "1:41:58"} +{"current_steps": 1880, "total_steps": 2240, "loss": 0.0419, "lr": 6.339075988698367e-06, "epoch": 6.714285714285714, "percentage": 83.93, "elapsed_time": "8:45:04", "remaining_time": "1:40:32"} +{"current_steps": 1885, "total_steps": 2240, "loss": 0.0457, "lr": 6.168372905816821e-06, "epoch": 6.732142857142857, "percentage": 84.15, "elapsed_time": "8:46:11", "remaining_time": "1:39:05"} +{"current_steps": 1890, "total_steps": 2240, "loss": 0.0421, "lr": 5.9998485116043614e-06, "epoch": 6.75, "percentage": 84.38, "elapsed_time": "8:47:13", "remaining_time": "1:37:38"} +{"current_steps": 1895, "total_steps": 2240, "loss": 0.0456, "lr": 5.833511182710716e-06, "epoch": 6.767857142857143, "percentage": 84.6, "elapsed_time": "8:48:25", "remaining_time": "1:36:12"} +{"current_steps": 1900, "total_steps": 2240, "loss": 0.0443, "lr": 5.6693691870756905e-06, "epoch": 6.785714285714286, "percentage": 84.82, "elapsed_time": "8:49:39", "remaining_time": "1:34:46"} +{"current_steps": 1905, "total_steps": 2240, "loss": 0.045, "lr": 5.507430683518161e-06, "epoch": 6.803571428571429, "percentage": 85.04, "elapsed_time": "8:57:36", "remaining_time": "1:34:32"} +{"current_steps": 1910, "total_steps": 2240, "loss": 0.0429, "lr": 5.3477037213304995e-06, "epoch": 6.821428571428571, "percentage": 85.27, "elapsed_time": "8:58:51", "remaining_time": "1:33:06"} +{"current_steps": 1915, "total_steps": 2240, "loss": 0.0453, "lr": 5.19019623987857e-06, "epoch": 6.839285714285714, "percentage": 85.49, "elapsed_time": "9:00:00", "remaining_time": "1:31:38"} +{"current_steps": 1920, "total_steps": 2240, "loss": 0.0432, "lr": 5.034916068206996e-06, "epoch": 6.857142857142857, "percentage": 85.71, "elapsed_time": "9:01:12", "remaining_time": "1:30:12"} +{"current_steps": 1925, "total_steps": 2240, "loss": 0.0438, "lr": 4.881870924650062e-06, "epoch": 6.875, "percentage": 85.94, "elapsed_time": "9:02:20", "remaining_time": "1:28:44"} +{"current_steps": 1930, "total_steps": 2240, "loss": 0.0422, "lr": 4.731068416448081e-06, "epoch": 6.892857142857143, "percentage": 86.16, "elapsed_time": "9:03:33", "remaining_time": "1:27:18"} +{"current_steps": 1935, "total_steps": 2240, "loss": 0.044, "lr": 4.5825160393692445e-06, "epoch": 6.910714285714286, "percentage": 86.38, "elapsed_time": "9:04:39", "remaining_time": "1:25:50"} +{"current_steps": 1940, "total_steps": 2240, "loss": 0.0443, "lr": 4.43622117733703e-06, "epoch": 6.928571428571429, "percentage": 86.61, "elapsed_time": "9:05:49", "remaining_time": "1:24:24"} +{"current_steps": 1945, "total_steps": 2240, "loss": 0.0439, "lr": 4.292191102063192e-06, "epoch": 6.946428571428571, "percentage": 86.83, "elapsed_time": "9:07:01", "remaining_time": "1:22:58"} +{"current_steps": 1950, "total_steps": 2240, "loss": 0.0441, "lr": 4.150432972686352e-06, "epoch": 6.964285714285714, "percentage": 87.05, "elapsed_time": "9:08:08", "remaining_time": "1:21:31"} +{"current_steps": 1955, "total_steps": 2240, "loss": 0.0432, "lr": 4.010953835416037e-06, "epoch": 6.982142857142857, "percentage": 87.28, "elapsed_time": "9:09:10", "remaining_time": "1:20:03"} +{"current_steps": 1960, "total_steps": 2240, "loss": 0.0431, "lr": 3.87376062318257e-06, "epoch": 7.0, "percentage": 87.5, "elapsed_time": "9:10:23", "remaining_time": "1:18:37"} +{"current_steps": 1965, "total_steps": 2240, "loss": 0.0303, "lr": 3.7388601552924062e-06, "epoch": 7.017857142857143, "percentage": 87.72, "elapsed_time": "9:11:30", "remaining_time": "1:17:10"} +{"current_steps": 1970, "total_steps": 2240, "loss": 0.0293, "lr": 3.606259137089141e-06, "epoch": 7.035714285714286, "percentage": 87.95, "elapsed_time": "9:12:41", "remaining_time": "1:15:44"} +{"current_steps": 1975, "total_steps": 2240, "loss": 0.0285, "lr": 3.4759641596202762e-06, "epoch": 7.053571428571429, "percentage": 88.17, "elapsed_time": "9:13:55", "remaining_time": "1:14:19"} +{"current_steps": 1980, "total_steps": 2240, "loss": 0.0278, "lr": 3.347981699309588e-06, "epoch": 7.071428571428571, "percentage": 88.39, "elapsed_time": "9:14:59", "remaining_time": "1:12:52"} +{"current_steps": 1985, "total_steps": 2240, "loss": 0.028, "lr": 3.2223181176351426e-06, "epoch": 7.089285714285714, "percentage": 88.62, "elapsed_time": "9:16:13", "remaining_time": "1:11:27"} +{"current_steps": 1990, "total_steps": 2240, "loss": 0.0288, "lr": 3.098979660813217e-06, "epoch": 7.107142857142857, "percentage": 88.84, "elapsed_time": "9:17:27", "remaining_time": "1:10:01"} +{"current_steps": 1995, "total_steps": 2240, "loss": 0.0283, "lr": 2.977972459487738e-06, "epoch": 7.125, "percentage": 89.06, "elapsed_time": "9:18:41", "remaining_time": "1:08:36"} +{"current_steps": 2000, "total_steps": 2240, "loss": 0.0303, "lr": 2.8593025284255614e-06, "epoch": 7.142857142857143, "percentage": 89.29, "elapsed_time": "9:19:52", "remaining_time": "1:07:11"} +{"current_steps": 2005, "total_steps": 2240, "loss": 0.0287, "lr": 2.7429757662175314e-06, "epoch": 7.160714285714286, "percentage": 89.51, "elapsed_time": "9:25:25", "remaining_time": "1:06:16"} +{"current_steps": 2010, "total_steps": 2240, "loss": 0.0275, "lr": 2.6289979549852795e-06, "epoch": 7.178571428571429, "percentage": 89.73, "elapsed_time": "9:26:40", "remaining_time": "1:04:50"} +{"current_steps": 2015, "total_steps": 2240, "loss": 0.0268, "lr": 2.5173747600937993e-06, "epoch": 7.196428571428571, "percentage": 89.96, "elapsed_time": "9:27:50", "remaining_time": "1:03:24"} +{"current_steps": 2020, "total_steps": 2240, "loss": 0.027, "lr": 2.408111729869844e-06, "epoch": 7.214285714285714, "percentage": 90.18, "elapsed_time": "9:28:59", "remaining_time": "1:01:58"} +{"current_steps": 2025, "total_steps": 2240, "loss": 0.0294, "lr": 2.3012142953261928e-06, "epoch": 7.232142857142857, "percentage": 90.4, "elapsed_time": "9:30:09", "remaining_time": "1:00:32"} +{"current_steps": 2030, "total_steps": 2240, "loss": 0.0273, "lr": 2.19668776989162e-06, "epoch": 7.25, "percentage": 90.62, "elapsed_time": "9:31:18", "remaining_time": "0:59:06"} +{"current_steps": 2035, "total_steps": 2240, "loss": 0.0274, "lr": 2.0945373491468466e-06, "epoch": 7.267857142857143, "percentage": 90.85, "elapsed_time": "9:32:27", "remaining_time": "0:57:40"} +{"current_steps": 2040, "total_steps": 2240, "loss": 0.0283, "lr": 1.9947681105662806e-06, "epoch": 7.285714285714286, "percentage": 91.07, "elapsed_time": "9:33:40", "remaining_time": "0:56:14"} +{"current_steps": 2045, "total_steps": 2240, "loss": 0.0295, "lr": 1.8973850132655956e-06, "epoch": 7.303571428571429, "percentage": 91.29, "elapsed_time": "9:34:49", "remaining_time": "0:54:48"} +{"current_steps": 2050, "total_steps": 2240, "loss": 0.027, "lr": 1.8023928977552839e-06, "epoch": 7.321428571428571, "percentage": 91.52, "elapsed_time": "9:35:59", "remaining_time": "0:53:23"} +{"current_steps": 2055, "total_steps": 2240, "loss": 0.0261, "lr": 1.7097964857000327e-06, "epoch": 7.339285714285714, "percentage": 91.74, "elapsed_time": "9:37:10", "remaining_time": "0:51:57"} +{"current_steps": 2060, "total_steps": 2240, "loss": 0.0267, "lr": 1.619600379684022e-06, "epoch": 7.357142857142857, "percentage": 91.96, "elapsed_time": "9:38:21", "remaining_time": "0:50:32"} +{"current_steps": 2065, "total_steps": 2240, "loss": 0.0275, "lr": 1.5318090629821757e-06, "epoch": 7.375, "percentage": 92.19, "elapsed_time": "9:39:28", "remaining_time": "0:49:06"} +{"current_steps": 2070, "total_steps": 2240, "loss": 0.0283, "lr": 1.4464268993372831e-06, "epoch": 7.392857142857143, "percentage": 92.41, "elapsed_time": "9:40:43", "remaining_time": "0:47:41"} +{"current_steps": 2075, "total_steps": 2240, "loss": 0.0277, "lr": 1.3634581327431229e-06, "epoch": 7.410714285714286, "percentage": 92.63, "elapsed_time": "9:41:56", "remaining_time": "0:46:16"} +{"current_steps": 2080, "total_steps": 2240, "loss": 0.0282, "lr": 1.282906887233487e-06, "epoch": 7.428571428571429, "percentage": 92.86, "elapsed_time": "9:43:01", "remaining_time": "0:44:50"} +{"current_steps": 2085, "total_steps": 2240, "loss": 0.0282, "lr": 1.2047771666772122e-06, "epoch": 7.446428571428571, "percentage": 93.08, "elapsed_time": "9:44:15", "remaining_time": "0:43:26"} +{"current_steps": 2090, "total_steps": 2240, "loss": 0.0285, "lr": 1.1290728545791773e-06, "epoch": 7.464285714285714, "percentage": 93.3, "elapsed_time": "9:45:29", "remaining_time": "0:42:01"} +{"current_steps": 2095, "total_steps": 2240, "loss": 0.0281, "lr": 1.0557977138872132e-06, "epoch": 7.482142857142857, "percentage": 93.53, "elapsed_time": "9:46:38", "remaining_time": "0:40:36"} +{"current_steps": 2100, "total_steps": 2240, "loss": 0.0292, "lr": 9.849553868051364e-07, "epoch": 7.5, "percentage": 93.75, "elapsed_time": "9:47:50", "remaining_time": "0:39:11"} +{"current_steps": 2105, "total_steps": 2240, "loss": 0.0264, "lr": 9.165493946116432e-07, "epoch": 7.517857142857143, "percentage": 93.97, "elapsed_time": "9:55:42", "remaining_time": "0:38:12"} +{"current_steps": 2110, "total_steps": 2240, "loss": 0.0273, "lr": 8.505831374853279e-07, "epoch": 7.535714285714286, "percentage": 94.2, "elapsed_time": "9:56:56", "remaining_time": "0:36:46"} +{"current_steps": 2115, "total_steps": 2240, "loss": 0.0284, "lr": 7.870598943356622e-07, "epoch": 7.553571428571429, "percentage": 94.42, "elapsed_time": "9:58:09", "remaining_time": "0:35:21"} +{"current_steps": 2120, "total_steps": 2240, "loss": 0.0269, "lr": 7.259828226399978e-07, "epoch": 7.571428571428571, "percentage": 94.64, "elapsed_time": "9:59:23", "remaining_time": "0:33:55"} +{"current_steps": 2125, "total_steps": 2240, "loss": 0.0272, "lr": 6.673549582866367e-07, "epoch": 7.589285714285714, "percentage": 94.87, "elapsed_time": "10:00:28", "remaining_time": "0:32:29"} +{"current_steps": 2130, "total_steps": 2240, "loss": 0.0265, "lr": 6.111792154239404e-07, "epoch": 7.607142857142857, "percentage": 95.09, "elapsed_time": "10:01:39", "remaining_time": "0:31:04"} +{"current_steps": 2135, "total_steps": 2240, "loss": 0.03, "lr": 5.574583863154403e-07, "epoch": 7.625, "percentage": 95.31, "elapsed_time": "10:02:53", "remaining_time": "0:29:39"} +{"current_steps": 2140, "total_steps": 2240, "loss": 0.0271, "lr": 5.061951412010879e-07, "epoch": 7.642857142857143, "percentage": 95.54, "elapsed_time": "10:04:07", "remaining_time": "0:28:13"} +{"current_steps": 2145, "total_steps": 2240, "loss": 0.0281, "lr": 4.573920281645161e-07, "epoch": 7.660714285714286, "percentage": 95.76, "elapsed_time": "10:05:16", "remaining_time": "0:26:48"} +{"current_steps": 2150, "total_steps": 2240, "loss": 0.0261, "lr": 4.1105147300636303e-07, "epoch": 7.678571428571429, "percentage": 95.98, "elapsed_time": "10:06:27", "remaining_time": "0:25:23"} +{"current_steps": 2155, "total_steps": 2240, "loss": 0.0275, "lr": 3.67175779123724e-07, "epoch": 7.696428571428571, "percentage": 96.21, "elapsed_time": "10:07:36", "remaining_time": "0:23:57"} +{"current_steps": 2160, "total_steps": 2240, "loss": 0.0273, "lr": 3.2576712739565416e-07, "epoch": 7.714285714285714, "percentage": 96.43, "elapsed_time": "10:08:47", "remaining_time": "0:22:32"} +{"current_steps": 2165, "total_steps": 2240, "loss": 0.0278, "lr": 2.8682757607474407e-07, "epoch": 7.732142857142857, "percentage": 96.65, "elapsed_time": "10:09:56", "remaining_time": "0:21:07"} +{"current_steps": 2170, "total_steps": 2240, "loss": 0.0285, "lr": 2.5035906068482937e-07, "epoch": 7.75, "percentage": 96.88, "elapsed_time": "10:11:06", "remaining_time": "0:19:42"} +{"current_steps": 2175, "total_steps": 2240, "loss": 0.0312, "lr": 2.1636339392479553e-07, "epoch": 7.767857142857143, "percentage": 97.1, "elapsed_time": "10:12:13", "remaining_time": "0:18:17"} +{"current_steps": 2180, "total_steps": 2240, "loss": 0.0273, "lr": 1.848422655784554e-07, "epoch": 7.785714285714286, "percentage": 97.32, "elapsed_time": "10:13:23", "remaining_time": "0:16:52"} +{"current_steps": 2185, "total_steps": 2240, "loss": 0.0266, "lr": 1.557972424305665e-07, "epoch": 7.803571428571429, "percentage": 97.54, "elapsed_time": "10:14:32", "remaining_time": "0:15:28"} +{"current_steps": 2190, "total_steps": 2240, "loss": 0.0274, "lr": 1.2922976818894317e-07, "epoch": 7.821428571428571, "percentage": 97.77, "elapsed_time": "10:15:43", "remaining_time": "0:14:03"} +{"current_steps": 2195, "total_steps": 2240, "loss": 0.0275, "lr": 1.0514116341271418e-07, "epoch": 7.839285714285714, "percentage": 97.99, "elapsed_time": "10:16:51", "remaining_time": "0:12:38"} +{"current_steps": 2200, "total_steps": 2240, "loss": 0.0286, "lr": 8.353262544666951e-08, "epoch": 7.857142857142857, "percentage": 98.21, "elapsed_time": "10:18:02", "remaining_time": "0:11:14"} +{"current_steps": 2205, "total_steps": 2240, "loss": 0.0269, "lr": 6.440522836174135e-08, "epoch": 7.875, "percentage": 98.44, "elapsed_time": "10:22:46", "remaining_time": "0:09:53"} +{"current_steps": 2210, "total_steps": 2240, "loss": 0.0263, "lr": 4.775992290163567e-08, "epoch": 7.892857142857143, "percentage": 98.66, "elapsed_time": "10:23:59", "remaining_time": "0:08:28"} +{"current_steps": 2215, "total_steps": 2240, "loss": 0.0251, "lr": 3.359753643555341e-08, "epoch": 7.910714285714286, "percentage": 98.88, "elapsed_time": "10:25:13", "remaining_time": "0:07:03"} +{"current_steps": 2220, "total_steps": 2240, "loss": 0.0277, "lr": 2.191877291707889e-08, "epoch": 7.928571428571429, "percentage": 99.11, "elapsed_time": "10:26:27", "remaining_time": "0:05:38"} +{"current_steps": 2225, "total_steps": 2240, "loss": 0.0302, "lr": 1.2724212849180017e-08, "epoch": 7.946428571428571, "percentage": 99.33, "elapsed_time": "10:27:40", "remaining_time": "0:04:13"} +{"current_steps": 2230, "total_steps": 2240, "loss": 0.0269, "lr": 6.0143132553591716e-09, "epoch": 7.964285714285714, "percentage": 99.55, "elapsed_time": "10:28:52", "remaining_time": "0:02:49"} +{"current_steps": 2235, "total_steps": 2240, "loss": 0.0279, "lr": 1.7894076569435757e-09, "epoch": 7.982142857142857, "percentage": 99.78, "elapsed_time": "10:30:04", "remaining_time": "0:01:24"} +{"current_steps": 2240, "total_steps": 2240, "loss": 0.0257, "lr": 4.970605649301519e-11, "epoch": 8.0, "percentage": 100.0, "elapsed_time": "10:31:14", "remaining_time": "0:00:00"} +{"current_steps": 2240, "total_steps": 2240, "epoch": 8.0, "percentage": 100.0, "elapsed_time": "10:38:24", "remaining_time": "0:00:00"} diff --git a/trainer_state.json b/trainer_state.json new file mode 100644 index 0000000..7f2306a --- /dev/null +++ b/trainer_state.json @@ -0,0 +1,4975 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 8.0, + "eval_steps": 500, + "global_step": 2240, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.017857142857142856, + "grad_norm": 2.2143791888772313, + "learning_rate": 3.3333333333333335e-05, + "loss": 0.7589, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.6283497214317322, + "step": 5, + "valid_targets_mean": 3923.2, + "valid_targets_min": 1502 + }, + { + "epoch": 0.03571428571428571, + "grad_norm": 1.045606578232033, + "learning_rate": 7.500000000000001e-05, + "loss": 0.5543, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.5275813937187195, + "step": 10, + "valid_targets_mean": 3801.8, + "valid_targets_min": 1809 + }, + { + "epoch": 0.05357142857142857, + "grad_norm": 0.6804034955361882, + "learning_rate": 9.999980117587285e-05, + "loss": 0.4812, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4718513488769531, + "step": 15, + "valid_targets_mean": 3839.1, + "valid_targets_min": 701 + }, + { + "epoch": 0.07142857142857142, + "grad_norm": 0.5376873859780553, + "learning_rate": 9.99975644226019e-05, + "loss": 0.4195, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.4006612002849579, + "step": 20, + "valid_targets_mean": 4341.0, + "valid_targets_min": 1577 + }, + { + "epoch": 0.08928571428571429, + "grad_norm": 0.5080170486946795, + "learning_rate": 9.999284249745142e-05, + "loss": 0.3984, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3883684277534485, + "step": 25, + "valid_targets_mean": 4493.2, + "valid_targets_min": 628 + }, + { + "epoch": 0.10714285714285714, + "grad_norm": 0.4796878065639005, + "learning_rate": 9.998563563512873e-05, + "loss": 0.3894, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.39119023084640503, + "step": 30, + "valid_targets_mean": 4029.5, + "valid_targets_min": 1126 + }, + { + "epoch": 0.125, + "grad_norm": 0.4584485789433166, + "learning_rate": 9.997594419385712e-05, + "loss": 0.3957, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.39080631732940674, + "step": 35, + "valid_targets_mean": 4221.4, + "valid_targets_min": 1256 + }, + { + "epoch": 0.14285714285714285, + "grad_norm": 0.49765643549056565, + "learning_rate": 9.996376865535801e-05, + "loss": 0.373, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.38425832986831665, + "step": 40, + "valid_targets_mean": 4279.8, + "valid_targets_min": 1760 + }, + { + "epoch": 0.16071428571428573, + "grad_norm": 0.4197472888117808, + "learning_rate": 9.9949109624827e-05, + "loss": 0.3651, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.34883612394332886, + "step": 45, + "valid_targets_mean": 4182.0, + "valid_targets_min": 1065 + }, + { + "epoch": 0.17857142857142858, + "grad_norm": 0.4456537714766154, + "learning_rate": 9.993196783090377e-05, + "loss": 0.3637, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.35759520530700684, + "step": 50, + "valid_targets_mean": 4269.8, + "valid_targets_min": 744 + }, + { + "epoch": 0.19642857142857142, + "grad_norm": 0.42387027707803576, + "learning_rate": 9.991234412563593e-05, + "loss": 0.3527, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3638553023338318, + "step": 55, + "valid_targets_mean": 4110.2, + "valid_targets_min": 1697 + }, + { + "epoch": 0.21428571428571427, + "grad_norm": 0.4331419804202472, + "learning_rate": 9.989023948443662e-05, + "loss": 0.3611, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.34972161054611206, + "step": 60, + "valid_targets_mean": 4123.8, + "valid_targets_min": 1327 + }, + { + "epoch": 0.23214285714285715, + "grad_norm": 0.44017062541139595, + "learning_rate": 9.986565500603598e-05, + "loss": 0.3648, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.348086953163147, + "step": 65, + "valid_targets_mean": 4128.8, + "valid_targets_min": 553 + }, + { + "epoch": 0.25, + "grad_norm": 0.44356231076156627, + "learning_rate": 9.983859191242661e-05, + "loss": 0.3618, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3596280813217163, + "step": 70, + "valid_targets_mean": 4374.8, + "valid_targets_min": 780 + }, + { + "epoch": 0.26785714285714285, + "grad_norm": 0.3746280363703801, + "learning_rate": 9.980905154880288e-05, + "loss": 0.3619, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3581845462322235, + "step": 75, + "valid_targets_mean": 4323.8, + "valid_targets_min": 788 + }, + { + "epoch": 0.2857142857142857, + "grad_norm": 0.3924110629496763, + "learning_rate": 9.97770353834939e-05, + "loss": 0.3496, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.36474257707595825, + "step": 80, + "valid_targets_mean": 4063.3, + "valid_targets_min": 903 + }, + { + "epoch": 0.30357142857142855, + "grad_norm": 0.38505934203448594, + "learning_rate": 9.974254500789065e-05, + "loss": 0.3483, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3599857985973358, + "step": 85, + "valid_targets_mean": 4027.1, + "valid_targets_min": 946 + }, + { + "epoch": 0.32142857142857145, + "grad_norm": 0.398414567706717, + "learning_rate": 9.97055821363669e-05, + "loss": 0.358, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.38219308853149414, + "step": 90, + "valid_targets_mean": 3981.4, + "valid_targets_min": 949 + }, + { + "epoch": 0.3392857142857143, + "grad_norm": 0.38616087711889674, + "learning_rate": 9.96661486061939e-05, + "loss": 0.3357, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.32120615243911743, + "step": 95, + "valid_targets_mean": 4083.1, + "valid_targets_min": 716 + }, + { + "epoch": 0.35714285714285715, + "grad_norm": 0.4052083186171738, + "learning_rate": 9.962424637744914e-05, + "loss": 0.3519, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.34588366746902466, + "step": 100, + "valid_targets_mean": 4165.5, + "valid_targets_min": 929 + }, + { + "epoch": 0.375, + "grad_norm": 0.4113305815656917, + "learning_rate": 9.957987753291889e-05, + "loss": 0.3408, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3552583158016205, + "step": 105, + "valid_targets_mean": 3359.1, + "valid_targets_min": 878 + }, + { + "epoch": 0.39285714285714285, + "grad_norm": 0.42055378406206817, + "learning_rate": 9.953304427799469e-05, + "loss": 0.3462, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3395262360572815, + "step": 110, + "valid_targets_mean": 3937.2, + "valid_targets_min": 1641 + }, + { + "epoch": 0.4107142857142857, + "grad_norm": 0.4198064379300064, + "learning_rate": 9.948374894056368e-05, + "loss": 0.3459, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.35403895378112793, + "step": 115, + "valid_targets_mean": 4178.4, + "valid_targets_min": 1161 + }, + { + "epoch": 0.42857142857142855, + "grad_norm": 0.39462209473570964, + "learning_rate": 9.943199397089296e-05, + "loss": 0.3386, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.34414926171302795, + "step": 120, + "valid_targets_mean": 3716.6, + "valid_targets_min": 1338 + }, + { + "epoch": 0.44642857142857145, + "grad_norm": 0.4161448502322806, + "learning_rate": 9.937778194150771e-05, + "loss": 0.3329, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.36363863945007324, + "step": 125, + "valid_targets_mean": 3618.8, + "valid_targets_min": 860 + }, + { + "epoch": 0.4642857142857143, + "grad_norm": 0.38347601281668314, + "learning_rate": 9.932111554706345e-05, + "loss": 0.3445, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3507039546966553, + "step": 130, + "valid_targets_mean": 4651.9, + "valid_targets_min": 1361 + }, + { + "epoch": 0.48214285714285715, + "grad_norm": 0.33758621227358415, + "learning_rate": 9.926199760421195e-05, + "loss": 0.34, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3302854597568512, + "step": 135, + "valid_targets_mean": 3754.7, + "valid_targets_min": 1124 + }, + { + "epoch": 0.5, + "grad_norm": 0.3725493130576605, + "learning_rate": 9.920043105146136e-05, + "loss": 0.3272, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.31262439489364624, + "step": 140, + "valid_targets_mean": 3260.9, + "valid_targets_min": 1009 + }, + { + "epoch": 0.5178571428571429, + "grad_norm": 0.3669697608298975, + "learning_rate": 9.913641894903006e-05, + "loss": 0.3328, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3365815281867981, + "step": 145, + "valid_targets_mean": 3906.8, + "valid_targets_min": 630 + }, + { + "epoch": 0.5357142857142857, + "grad_norm": 0.3219368757340431, + "learning_rate": 9.906996447869454e-05, + "loss": 0.3361, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3143583834171295, + "step": 150, + "valid_targets_mean": 3700.3, + "valid_targets_min": 1427 + }, + { + "epoch": 0.5535714285714286, + "grad_norm": 0.3343532940339292, + "learning_rate": 9.900107094363138e-05, + "loss": 0.3341, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.33497369289398193, + "step": 155, + "valid_targets_mean": 4139.4, + "valid_targets_min": 824 + }, + { + "epoch": 0.5714285714285714, + "grad_norm": 0.3883775859048715, + "learning_rate": 9.892974176825286e-05, + "loss": 0.3426, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3533402681350708, + "step": 160, + "valid_targets_mean": 4143.4, + "valid_targets_min": 1687 + }, + { + "epoch": 0.5892857142857143, + "grad_norm": 0.3565535130225987, + "learning_rate": 9.885598049803693e-05, + "loss": 0.3292, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.32782241702079773, + "step": 165, + "valid_targets_mean": 3839.8, + "valid_targets_min": 773 + }, + { + "epoch": 0.6071428571428571, + "grad_norm": 0.3388734365099713, + "learning_rate": 9.877979079935086e-05, + "loss": 0.3326, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.32009679079055786, + "step": 170, + "valid_targets_mean": 3931.8, + "valid_targets_min": 694 + }, + { + "epoch": 0.625, + "grad_norm": 0.41178246030035925, + "learning_rate": 9.870117645926906e-05, + "loss": 0.3347, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3536105751991272, + "step": 175, + "valid_targets_mean": 3873.2, + "valid_targets_min": 649 + }, + { + "epoch": 0.6428571428571429, + "grad_norm": 0.34137864366933385, + "learning_rate": 9.862014138538482e-05, + "loss": 0.3253, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.31523963809013367, + "step": 180, + "valid_targets_mean": 3763.6, + "valid_targets_min": 1378 + }, + { + "epoch": 0.6607142857142857, + "grad_norm": 0.32330618571687525, + "learning_rate": 9.853668960561611e-05, + "loss": 0.3359, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3308245539665222, + "step": 185, + "valid_targets_mean": 4124.9, + "valid_targets_min": 1088 + }, + { + "epoch": 0.6785714285714286, + "grad_norm": 0.3180286168936679, + "learning_rate": 9.845082526800528e-05, + "loss": 0.3299, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.32368218898773193, + "step": 190, + "valid_targets_mean": 4221.8, + "valid_targets_min": 1047 + }, + { + "epoch": 0.6964285714285714, + "grad_norm": 0.3235670725571024, + "learning_rate": 9.836255264051299e-05, + "loss": 0.3285, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.30624061822891235, + "step": 195, + "valid_targets_mean": 4307.4, + "valid_targets_min": 1189 + }, + { + "epoch": 0.7142857142857143, + "grad_norm": 0.338079101621634, + "learning_rate": 9.8271876110806e-05, + "loss": 0.3294, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3417225480079651, + "step": 200, + "valid_targets_mean": 4331.5, + "valid_targets_min": 1325 + }, + { + "epoch": 0.7321428571428571, + "grad_norm": 0.35727140262149987, + "learning_rate": 9.817880018603909e-05, + "loss": 0.3327, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3296249508857727, + "step": 205, + "valid_targets_mean": 3664.5, + "valid_targets_min": 896 + }, + { + "epoch": 0.75, + "grad_norm": 0.34175342948530363, + "learning_rate": 9.808332949263103e-05, + "loss": 0.3261, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.34093087911605835, + "step": 210, + "valid_targets_mean": 4452.3, + "valid_targets_min": 1486 + }, + { + "epoch": 0.7678571428571429, + "grad_norm": 0.3205276032358539, + "learning_rate": 9.798546877603468e-05, + "loss": 0.3284, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3253118395805359, + "step": 215, + "valid_targets_mean": 3928.4, + "valid_targets_min": 778 + }, + { + "epoch": 0.7857142857142857, + "grad_norm": 4.449910131544663, + "learning_rate": 9.788522290050094e-05, + "loss": 0.3292, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3441806435585022, + "step": 220, + "valid_targets_mean": 4150.8, + "valid_targets_min": 1466 + }, + { + "epoch": 0.8035714285714286, + "grad_norm": 0.3520805191490191, + "learning_rate": 9.778259684883719e-05, + "loss": 0.3316, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3302420973777771, + "step": 225, + "valid_targets_mean": 4138.9, + "valid_targets_min": 920 + }, + { + "epoch": 0.8214285714285714, + "grad_norm": 0.31373847032156543, + "learning_rate": 9.767759572215945e-05, + "loss": 0.3217, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.32329776883125305, + "step": 230, + "valid_targets_mean": 4336.2, + "valid_targets_min": 682 + }, + { + "epoch": 0.8392857142857143, + "grad_norm": 0.3656870097046931, + "learning_rate": 9.75702247396389e-05, + "loss": 0.3252, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.33107879757881165, + "step": 235, + "valid_targets_mean": 4153.0, + "valid_targets_min": 1143 + }, + { + "epoch": 0.8571428571428571, + "grad_norm": 0.35532493332253884, + "learning_rate": 9.746048923824245e-05, + "loss": 0.3277, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3462904095649719, + "step": 240, + "valid_targets_mean": 4539.9, + "valid_targets_min": 779 + }, + { + "epoch": 0.875, + "grad_norm": 0.3545366134729719, + "learning_rate": 9.734839467246744e-05, + "loss": 0.3307, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3441724181175232, + "step": 245, + "valid_targets_mean": 3941.3, + "valid_targets_min": 928 + }, + { + "epoch": 0.8928571428571429, + "grad_norm": 0.40848602320115285, + "learning_rate": 9.723394661407053e-05, + "loss": 0.325, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.34841489791870117, + "step": 250, + "valid_targets_mean": 3455.7, + "valid_targets_min": 1159 + }, + { + "epoch": 0.9107142857142857, + "grad_norm": 0.37130021145062575, + "learning_rate": 9.711715075179076e-05, + "loss": 0.3188, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3177517354488373, + "step": 255, + "valid_targets_mean": 3474.8, + "valid_targets_min": 1354 + }, + { + "epoch": 0.9285714285714286, + "grad_norm": 0.3259400830503473, + "learning_rate": 9.699801289106676e-05, + "loss": 0.3312, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3229742646217346, + "step": 260, + "valid_targets_mean": 4038.4, + "valid_targets_min": 820 + }, + { + "epoch": 0.9464285714285714, + "grad_norm": 0.3127541801830022, + "learning_rate": 9.687653895374823e-05, + "loss": 0.3275, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3187171220779419, + "step": 265, + "valid_targets_mean": 4271.7, + "valid_targets_min": 1884 + }, + { + "epoch": 0.9642857142857143, + "grad_norm": 0.3004092250599939, + "learning_rate": 9.675273497780155e-05, + "loss": 0.3205, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3269023299217224, + "step": 270, + "valid_targets_mean": 4342.6, + "valid_targets_min": 902 + }, + { + "epoch": 0.9821428571428571, + "grad_norm": 0.3090665254851116, + "learning_rate": 9.662660711700967e-05, + "loss": 0.3103, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.30684539675712585, + "step": 275, + "valid_targets_mean": 4178.2, + "valid_targets_min": 1458 + }, + { + "epoch": 1.0, + "grad_norm": 0.3200346730173292, + "learning_rate": 9.649816164066623e-05, + "loss": 0.3312, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.33374136686325073, + "step": 280, + "valid_targets_mean": 3672.4, + "valid_targets_min": 601 + }, + { + "epoch": 1.0178571428571428, + "grad_norm": 0.34517629620426826, + "learning_rate": 9.636740493326397e-05, + "loss": 0.281, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2977467179298401, + "step": 285, + "valid_targets_mean": 4474.2, + "valid_targets_min": 795 + }, + { + "epoch": 1.0357142857142858, + "grad_norm": 0.29292127724446076, + "learning_rate": 9.623434349417729e-05, + "loss": 0.2645, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.26123446226119995, + "step": 290, + "valid_targets_mean": 4160.8, + "valid_targets_min": 960 + }, + { + "epoch": 1.0535714285714286, + "grad_norm": 0.33800624212708835, + "learning_rate": 9.609898393733933e-05, + "loss": 0.2761, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.28165048360824585, + "step": 295, + "valid_targets_mean": 4013.4, + "valid_targets_min": 1234 + }, + { + "epoch": 1.0714285714285714, + "grad_norm": 0.3023727407552516, + "learning_rate": 9.59613329909131e-05, + "loss": 0.2816, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2674732208251953, + "step": 300, + "valid_targets_mean": 4057.6, + "valid_targets_min": 729 + }, + { + "epoch": 1.0892857142857142, + "grad_norm": 0.30002665570628373, + "learning_rate": 9.582139749695713e-05, + "loss": 0.2817, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.27096495032310486, + "step": 305, + "valid_targets_mean": 3638.5, + "valid_targets_min": 1370 + }, + { + "epoch": 1.1071428571428572, + "grad_norm": 0.324477356487815, + "learning_rate": 9.56791844110853e-05, + "loss": 0.2723, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.28439098596572876, + "step": 310, + "valid_targets_mean": 3994.7, + "valid_targets_min": 1633 + }, + { + "epoch": 1.125, + "grad_norm": 0.3253539045198033, + "learning_rate": 9.553470080212122e-05, + "loss": 0.2858, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2925952672958374, + "step": 315, + "valid_targets_mean": 4254.1, + "valid_targets_min": 867 + }, + { + "epoch": 1.1428571428571428, + "grad_norm": 0.32347990356976075, + "learning_rate": 9.538795385174672e-05, + "loss": 0.2769, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.27409157156944275, + "step": 320, + "valid_targets_mean": 3832.7, + "valid_targets_min": 780 + }, + { + "epoch": 1.1607142857142858, + "grad_norm": 0.3104593428416724, + "learning_rate": 9.523895085414501e-05, + "loss": 0.2735, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.25909847021102905, + "step": 325, + "valid_targets_mean": 3732.6, + "valid_targets_min": 1300 + }, + { + "epoch": 1.1785714285714286, + "grad_norm": 0.32715583481802546, + "learning_rate": 9.508769921563809e-05, + "loss": 0.2763, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2759383022785187, + "step": 330, + "valid_targets_mean": 3947.2, + "valid_targets_min": 1674 + }, + { + "epoch": 1.1964285714285714, + "grad_norm": 0.3451547957041938, + "learning_rate": 9.493420645431852e-05, + "loss": 0.2851, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2902929186820984, + "step": 335, + "valid_targets_mean": 4012.7, + "valid_targets_min": 1389 + }, + { + "epoch": 1.2142857142857142, + "grad_norm": 0.35446757439165194, + "learning_rate": 9.477848019967583e-05, + "loss": 0.2781, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2935929596424103, + "step": 340, + "valid_targets_mean": 3453.9, + "valid_targets_min": 926 + }, + { + "epoch": 1.2321428571428572, + "grad_norm": 0.3278767929484546, + "learning_rate": 9.462052819221726e-05, + "loss": 0.2902, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.29519742727279663, + "step": 345, + "valid_targets_mean": 4377.2, + "valid_targets_min": 748 + }, + { + "epoch": 1.25, + "grad_norm": 0.3141469785066057, + "learning_rate": 9.4460358283083e-05, + "loss": 0.2863, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3076077401638031, + "step": 350, + "valid_targets_mean": 3883.5, + "valid_targets_min": 1577 + }, + { + "epoch": 1.2678571428571428, + "grad_norm": 0.29540838389589275, + "learning_rate": 9.429797843365594e-05, + "loss": 0.2751, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.28724199533462524, + "step": 355, + "valid_targets_mean": 4575.3, + "valid_targets_min": 732 + }, + { + "epoch": 1.2857142857142856, + "grad_norm": 0.30255705801440214, + "learning_rate": 9.413339671516593e-05, + "loss": 0.2796, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.282297283411026, + "step": 360, + "valid_targets_mean": 4083.5, + "valid_targets_min": 1527 + }, + { + "epoch": 1.3035714285714286, + "grad_norm": 0.31946287205388435, + "learning_rate": 9.396662130828869e-05, + "loss": 0.2801, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.27121782302856445, + "step": 365, + "valid_targets_mean": 3983.3, + "valid_targets_min": 942 + }, + { + "epoch": 1.3214285714285714, + "grad_norm": 0.3161112548673974, + "learning_rate": 9.379766050273899e-05, + "loss": 0.2772, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2704480290412903, + "step": 370, + "valid_targets_mean": 3990.7, + "valid_targets_min": 1301 + }, + { + "epoch": 1.3392857142857144, + "grad_norm": 0.33878327261925667, + "learning_rate": 9.36265226968588e-05, + "loss": 0.2768, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.27773725986480713, + "step": 375, + "valid_targets_mean": 3821.6, + "valid_targets_min": 1636 + }, + { + "epoch": 1.3571428571428572, + "grad_norm": 0.32149726251644833, + "learning_rate": 9.345321639719979e-05, + "loss": 0.2781, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3027724027633667, + "step": 380, + "valid_targets_mean": 3755.8, + "valid_targets_min": 1588 + }, + { + "epoch": 1.375, + "grad_norm": 0.26983559471000856, + "learning_rate": 9.327775021810037e-05, + "loss": 0.2829, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.28533998131752014, + "step": 385, + "valid_targets_mean": 4883.8, + "valid_targets_min": 1621 + }, + { + "epoch": 1.3928571428571428, + "grad_norm": 0.3034738822017227, + "learning_rate": 9.31001328812577e-05, + "loss": 0.2834, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2803567945957184, + "step": 390, + "valid_targets_mean": 3837.8, + "valid_targets_min": 773 + }, + { + "epoch": 1.4107142857142856, + "grad_norm": 0.2900489145425836, + "learning_rate": 9.292037321529404e-05, + "loss": 0.2801, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2895487844944, + "step": 395, + "valid_targets_mean": 4024.4, + "valid_targets_min": 1322 + }, + { + "epoch": 1.4285714285714286, + "grad_norm": 0.30801221100789417, + "learning_rate": 9.273848015531795e-05, + "loss": 0.2771, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.279008686542511, + "step": 400, + "valid_targets_mean": 3967.7, + "valid_targets_min": 691 + }, + { + "epoch": 1.4464285714285714, + "grad_norm": 0.27770932626085026, + "learning_rate": 9.255446274248023e-05, + "loss": 0.2794, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.26959356665611267, + "step": 405, + "valid_targets_mean": 4549.9, + "valid_targets_min": 1048 + }, + { + "epoch": 1.4642857142857144, + "grad_norm": 0.2937936622420831, + "learning_rate": 9.236833012352442e-05, + "loss": 0.2968, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.29291456937789917, + "step": 410, + "valid_targets_mean": 3857.6, + "valid_targets_min": 1329 + }, + { + "epoch": 1.4821428571428572, + "grad_norm": 0.30086879589171994, + "learning_rate": 9.218009155033218e-05, + "loss": 0.2828, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2887283265590668, + "step": 415, + "valid_targets_mean": 4369.2, + "valid_targets_min": 1676 + }, + { + "epoch": 1.5, + "grad_norm": 0.30993326987536407, + "learning_rate": 9.198975637946347e-05, + "loss": 0.2752, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.286073237657547, + "step": 420, + "valid_targets_mean": 4086.6, + "valid_targets_min": 1349 + }, + { + "epoch": 1.5178571428571428, + "grad_norm": 0.31096503950204796, + "learning_rate": 9.179733407169145e-05, + "loss": 0.2835, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3040330111980438, + "step": 425, + "valid_targets_mean": 4739.1, + "valid_targets_min": 928 + }, + { + "epoch": 1.5357142857142856, + "grad_norm": 0.27613993704313583, + "learning_rate": 9.160283419153216e-05, + "loss": 0.2794, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2678983211517334, + "step": 430, + "valid_targets_mean": 3657.5, + "valid_targets_min": 1615 + }, + { + "epoch": 1.5535714285714286, + "grad_norm": 0.3027506465678719, + "learning_rate": 9.140626640676919e-05, + "loss": 0.2756, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2832491397857666, + "step": 435, + "valid_targets_mean": 4023.2, + "valid_targets_min": 1020 + }, + { + "epoch": 1.5714285714285714, + "grad_norm": 0.2746711947459874, + "learning_rate": 9.12076404879731e-05, + "loss": 0.28, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.28107601404190063, + "step": 440, + "valid_targets_mean": 4040.1, + "valid_targets_min": 1161 + }, + { + "epoch": 1.5892857142857144, + "grad_norm": 0.300269057710733, + "learning_rate": 9.10069663080158e-05, + "loss": 0.2897, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2961501479148865, + "step": 445, + "valid_targets_mean": 4180.1, + "valid_targets_min": 1188 + }, + { + "epoch": 1.6071428571428572, + "grad_norm": 0.2927151696397954, + "learning_rate": 9.080425384157971e-05, + "loss": 0.2882, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2890561819076538, + "step": 450, + "valid_targets_mean": 3919.0, + "valid_targets_min": 982 + }, + { + "epoch": 1.625, + "grad_norm": 0.270367262690994, + "learning_rate": 9.059951316466208e-05, + "loss": 0.2861, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2827759385108948, + "step": 455, + "valid_targets_mean": 4359.0, + "valid_targets_min": 1809 + }, + { + "epoch": 1.6428571428571428, + "grad_norm": 0.2961526538388308, + "learning_rate": 9.039275445407414e-05, + "loss": 0.2794, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2863369882106781, + "step": 460, + "valid_targets_mean": 4010.0, + "valid_targets_min": 1013 + }, + { + "epoch": 1.6607142857142856, + "grad_norm": 0.2827589674719424, + "learning_rate": 9.018398798693512e-05, + "loss": 0.2741, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.27826663851737976, + "step": 465, + "valid_targets_mean": 4303.2, + "valid_targets_min": 885 + }, + { + "epoch": 1.6785714285714286, + "grad_norm": 0.28604333715699365, + "learning_rate": 8.997322414016158e-05, + "loss": 0.279, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.29337555170059204, + "step": 470, + "valid_targets_mean": 4326.4, + "valid_targets_min": 1355 + }, + { + "epoch": 1.6964285714285714, + "grad_norm": 0.26669613943802206, + "learning_rate": 8.976047338995155e-05, + "loss": 0.2798, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2845120429992676, + "step": 475, + "valid_targets_mean": 4743.3, + "valid_targets_min": 1432 + }, + { + "epoch": 1.7142857142857144, + "grad_norm": 0.3075103842217294, + "learning_rate": 8.954574631126378e-05, + "loss": 0.281, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2861628234386444, + "step": 480, + "valid_targets_mean": 3601.3, + "valid_targets_min": 649 + }, + { + "epoch": 1.7321428571428572, + "grad_norm": 0.2537728906482412, + "learning_rate": 8.932905357729214e-05, + "loss": 0.276, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.281796395778656, + "step": 485, + "valid_targets_mean": 4734.3, + "valid_targets_min": 1318 + }, + { + "epoch": 1.75, + "grad_norm": 0.2665464345466697, + "learning_rate": 8.911040595893505e-05, + "loss": 0.2792, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2716560959815979, + "step": 490, + "valid_targets_mean": 3803.8, + "valid_targets_min": 1412 + }, + { + "epoch": 1.7678571428571428, + "grad_norm": 0.2471094014680912, + "learning_rate": 8.88898143242602e-05, + "loss": 0.2788, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2866831123828888, + "step": 495, + "valid_targets_mean": 5028.0, + "valid_targets_min": 868 + }, + { + "epoch": 1.7857142857142856, + "grad_norm": 0.2843669256068658, + "learning_rate": 8.866728963796423e-05, + "loss": 0.2791, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.29680773615837097, + "step": 500, + "valid_targets_mean": 4357.9, + "valid_targets_min": 711 + }, + { + "epoch": 1.8035714285714286, + "grad_norm": 0.28768990388704035, + "learning_rate": 8.844284296082776e-05, + "loss": 0.281, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.28727036714553833, + "step": 505, + "valid_targets_mean": 3947.9, + "valid_targets_min": 721 + }, + { + "epoch": 1.8214285714285714, + "grad_norm": 0.2616621943996813, + "learning_rate": 8.821648544916567e-05, + "loss": 0.2875, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.27664506435394287, + "step": 510, + "valid_targets_mean": 4528.7, + "valid_targets_min": 1949 + }, + { + "epoch": 1.8392857142857144, + "grad_norm": 0.2958030209996585, + "learning_rate": 8.79882283542725e-05, + "loss": 0.2763, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2893030643463135, + "step": 515, + "valid_targets_mean": 4298.3, + "valid_targets_min": 1494 + }, + { + "epoch": 1.8571428571428572, + "grad_norm": 0.28209834949958607, + "learning_rate": 8.775808302186314e-05, + "loss": 0.2797, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2799920439720154, + "step": 520, + "valid_targets_mean": 4215.8, + "valid_targets_min": 824 + }, + { + "epoch": 1.875, + "grad_norm": 0.2918990373232483, + "learning_rate": 8.752606089150903e-05, + "loss": 0.2874, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2988679111003876, + "step": 525, + "valid_targets_mean": 4380.1, + "valid_targets_min": 1494 + }, + { + "epoch": 1.8928571428571428, + "grad_norm": 0.29809691037806973, + "learning_rate": 8.729217349606942e-05, + "loss": 0.2774, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2863641679286957, + "step": 530, + "valid_targets_mean": 4880.6, + "valid_targets_min": 1076 + }, + { + "epoch": 1.9107142857142856, + "grad_norm": 0.28085345956845326, + "learning_rate": 8.705643246111816e-05, + "loss": 0.2891, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2826314866542816, + "step": 535, + "valid_targets_mean": 3880.8, + "valid_targets_min": 1226 + }, + { + "epoch": 1.9285714285714286, + "grad_norm": 0.2963891733516634, + "learning_rate": 8.681884950436587e-05, + "loss": 0.2796, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.28725090622901917, + "step": 540, + "valid_targets_mean": 3308.3, + "valid_targets_min": 617 + }, + { + "epoch": 1.9464285714285714, + "grad_norm": 0.27494989411880194, + "learning_rate": 8.657943643507746e-05, + "loss": 0.281, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.26711052656173706, + "step": 545, + "valid_targets_mean": 3776.5, + "valid_targets_min": 706 + }, + { + "epoch": 1.9642857142857144, + "grad_norm": 0.2560289775051106, + "learning_rate": 8.633820515348517e-05, + "loss": 0.2803, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.27995413541793823, + "step": 550, + "valid_targets_mean": 4268.9, + "valid_targets_min": 2286 + }, + { + "epoch": 1.9821428571428572, + "grad_norm": 0.27129891738048895, + "learning_rate": 8.609516765019698e-05, + "loss": 0.2754, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.28502243757247925, + "step": 555, + "valid_targets_mean": 3829.7, + "valid_targets_min": 943 + }, + { + "epoch": 2.0, + "grad_norm": 0.26428844546413777, + "learning_rate": 8.58503360056008e-05, + "loss": 0.2829, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.26649177074432373, + "step": 560, + "valid_targets_mean": 4592.5, + "valid_targets_min": 1634 + }, + { + "epoch": 2.017857142857143, + "grad_norm": 0.342076187133186, + "learning_rate": 8.560372238926372e-05, + "loss": 0.231, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23271307349205017, + "step": 565, + "valid_targets_mean": 4750.8, + "valid_targets_min": 1142 + }, + { + "epoch": 2.0357142857142856, + "grad_norm": 0.2591601056664035, + "learning_rate": 8.535533905932738e-05, + "loss": 0.2185, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20991137623786926, + "step": 570, + "valid_targets_mean": 3943.1, + "valid_targets_min": 756 + }, + { + "epoch": 2.0535714285714284, + "grad_norm": 0.2719176227918457, + "learning_rate": 8.510519836189851e-05, + "loss": 0.2212, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2201782464981079, + "step": 575, + "valid_targets_mean": 4093.7, + "valid_targets_min": 1364 + }, + { + "epoch": 2.0714285714285716, + "grad_norm": 0.2766993918032138, + "learning_rate": 8.485331273043531e-05, + "loss": 0.2128, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20157991349697113, + "step": 580, + "valid_targets_mean": 3884.0, + "valid_targets_min": 976 + }, + { + "epoch": 2.0892857142857144, + "grad_norm": 0.28289976797188854, + "learning_rate": 8.459969468512943e-05, + "loss": 0.2194, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2258216291666031, + "step": 585, + "valid_targets_mean": 3976.0, + "valid_targets_min": 1254 + }, + { + "epoch": 2.107142857142857, + "grad_norm": 0.2786573673112787, + "learning_rate": 8.43443568322836e-05, + "loss": 0.2288, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24203696846961975, + "step": 590, + "valid_targets_mean": 4164.1, + "valid_targets_min": 1495 + }, + { + "epoch": 2.125, + "grad_norm": 0.2630375685065281, + "learning_rate": 8.40873118636851e-05, + "loss": 0.2217, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21527183055877686, + "step": 595, + "valid_targets_mean": 4820.2, + "valid_targets_min": 923 + }, + { + "epoch": 2.142857142857143, + "grad_norm": 0.27411051143798987, + "learning_rate": 8.382857255597489e-05, + "loss": 0.2247, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21411019563674927, + "step": 600, + "valid_targets_mean": 3970.6, + "valid_targets_min": 546 + }, + { + "epoch": 2.1607142857142856, + "grad_norm": 0.2818013334082892, + "learning_rate": 8.356815177001242e-05, + "loss": 0.23, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2396702766418457, + "step": 605, + "valid_targets_mean": 4244.2, + "valid_targets_min": 701 + }, + { + "epoch": 2.1785714285714284, + "grad_norm": 0.26681441971155045, + "learning_rate": 8.330606245023657e-05, + "loss": 0.228, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22952930629253387, + "step": 610, + "valid_targets_mean": 4677.6, + "valid_targets_min": 1153 + }, + { + "epoch": 2.1964285714285716, + "grad_norm": 0.3204304446425023, + "learning_rate": 8.304231762402204e-05, + "loss": 0.2256, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22708728909492493, + "step": 615, + "valid_targets_mean": 3806.8, + "valid_targets_min": 942 + }, + { + "epoch": 2.2142857142857144, + "grad_norm": 0.2947138506224873, + "learning_rate": 8.277693040103192e-05, + "loss": 0.2338, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23883011937141418, + "step": 620, + "valid_targets_mean": 4111.0, + "valid_targets_min": 526 + }, + { + "epoch": 2.232142857142857, + "grad_norm": 0.27599504080019766, + "learning_rate": 8.250991397256609e-05, + "loss": 0.2283, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2176445722579956, + "step": 625, + "valid_targets_mean": 4158.2, + "valid_targets_min": 910 + }, + { + "epoch": 2.25, + "grad_norm": 0.29677815247089595, + "learning_rate": 8.224128161090543e-05, + "loss": 0.2272, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23432499170303345, + "step": 630, + "valid_targets_mean": 4380.6, + "valid_targets_min": 649 + }, + { + "epoch": 2.267857142857143, + "grad_norm": 0.26701601857548446, + "learning_rate": 8.197104666865218e-05, + "loss": 0.2281, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20332011580467224, + "step": 635, + "valid_targets_mean": 4002.0, + "valid_targets_min": 735 + }, + { + "epoch": 2.2857142857142856, + "grad_norm": 0.2794854143750737, + "learning_rate": 8.169922257806625e-05, + "loss": 0.2252, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22544199228286743, + "step": 640, + "valid_targets_mean": 4170.1, + "valid_targets_min": 1241 + }, + { + "epoch": 2.3035714285714284, + "grad_norm": 0.2781616811634506, + "learning_rate": 8.142582285039752e-05, + "loss": 0.2292, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23094919323921204, + "step": 645, + "valid_targets_mean": 4609.7, + "valid_targets_min": 1634 + }, + { + "epoch": 2.3214285714285716, + "grad_norm": 0.2936106664520855, + "learning_rate": 8.115086107521424e-05, + "loss": 0.2281, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24029132723808289, + "step": 650, + "valid_targets_mean": 3905.7, + "valid_targets_min": 1457 + }, + { + "epoch": 2.3392857142857144, + "grad_norm": 0.27959402587966176, + "learning_rate": 8.087435091972761e-05, + "loss": 0.2336, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23994141817092896, + "step": 655, + "valid_targets_mean": 4219.6, + "valid_targets_min": 759 + }, + { + "epoch": 2.357142857142857, + "grad_norm": 0.30007047440429424, + "learning_rate": 8.05963061281124e-05, + "loss": 0.2306, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2326786071062088, + "step": 660, + "valid_targets_mean": 4520.8, + "valid_targets_min": 926 + }, + { + "epoch": 2.375, + "grad_norm": 0.2769829773261415, + "learning_rate": 8.031674052082372e-05, + "loss": 0.233, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21899914741516113, + "step": 665, + "valid_targets_mean": 3920.0, + "valid_targets_min": 1306 + }, + { + "epoch": 2.392857142857143, + "grad_norm": 0.3318007501596546, + "learning_rate": 8.003566799391024e-05, + "loss": 0.2336, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23069044947624207, + "step": 670, + "valid_targets_mean": 3423.6, + "valid_targets_min": 1174 + }, + { + "epoch": 2.4107142857142856, + "grad_norm": 0.3147244965672268, + "learning_rate": 7.975310251832329e-05, + "loss": 0.2376, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2449096441268921, + "step": 675, + "valid_targets_mean": 3822.1, + "valid_targets_min": 944 + }, + { + "epoch": 2.4285714285714284, + "grad_norm": 0.2845619114038091, + "learning_rate": 7.946905813922249e-05, + "loss": 0.2287, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22111892700195312, + "step": 680, + "valid_targets_mean": 4054.1, + "valid_targets_min": 1188 + }, + { + "epoch": 2.4464285714285716, + "grad_norm": 0.28150460983655995, + "learning_rate": 7.918354897527766e-05, + "loss": 0.2297, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2350059449672699, + "step": 685, + "valid_targets_mean": 4716.6, + "valid_targets_min": 1771 + }, + { + "epoch": 2.4642857142857144, + "grad_norm": 0.2838096630512811, + "learning_rate": 7.889658921796703e-05, + "loss": 0.2351, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23345017433166504, + "step": 690, + "valid_targets_mean": 3914.8, + "valid_targets_min": 1404 + }, + { + "epoch": 2.482142857142857, + "grad_norm": 0.2647263836909652, + "learning_rate": 7.860819313087177e-05, + "loss": 0.2288, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23521623015403748, + "step": 695, + "valid_targets_mean": 4487.4, + "valid_targets_min": 1086 + }, + { + "epoch": 2.5, + "grad_norm": 0.2733656061562082, + "learning_rate": 7.831837504896707e-05, + "loss": 0.2209, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.205388605594635, + "step": 700, + "valid_targets_mean": 3501.2, + "valid_targets_min": 928 + }, + { + "epoch": 2.517857142857143, + "grad_norm": 0.28752162203154863, + "learning_rate": 7.80271493779096e-05, + "loss": 0.2292, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23121970891952515, + "step": 705, + "valid_targets_mean": 3692.7, + "valid_targets_min": 860 + }, + { + "epoch": 2.5357142857142856, + "grad_norm": 0.28429461410539203, + "learning_rate": 7.773453059332145e-05, + "loss": 0.2304, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2292163223028183, + "step": 710, + "valid_targets_mean": 4528.9, + "valid_targets_min": 1568 + }, + { + "epoch": 2.553571428571429, + "grad_norm": 0.27743745553251714, + "learning_rate": 7.744053324007063e-05, + "loss": 0.2294, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22714003920555115, + "step": 715, + "valid_targets_mean": 3678.2, + "valid_targets_min": 975 + }, + { + "epoch": 2.571428571428571, + "grad_norm": 0.28509001670278994, + "learning_rate": 7.71451719315481e-05, + "loss": 0.2347, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23330581188201904, + "step": 720, + "valid_targets_mean": 4192.9, + "valid_targets_min": 635 + }, + { + "epoch": 2.5892857142857144, + "grad_norm": 0.2817519019435897, + "learning_rate": 7.684846134894133e-05, + "loss": 0.2284, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23951683938503265, + "step": 725, + "valid_targets_mean": 4106.0, + "valid_targets_min": 719 + }, + { + "epoch": 2.607142857142857, + "grad_norm": 0.27345571093528337, + "learning_rate": 7.655041624050467e-05, + "loss": 0.2291, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2236991673707962, + "step": 730, + "valid_targets_mean": 4008.7, + "valid_targets_min": 1626 + }, + { + "epoch": 2.625, + "grad_norm": 0.27244674495714866, + "learning_rate": 7.625105142082623e-05, + "loss": 0.2368, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2419646680355072, + "step": 735, + "valid_targets_mean": 4142.3, + "valid_targets_min": 1032 + }, + { + "epoch": 2.642857142857143, + "grad_norm": 0.25647714180223846, + "learning_rate": 7.595038177009144e-05, + "loss": 0.2322, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22575123608112335, + "step": 740, + "valid_targets_mean": 4246.4, + "valid_targets_min": 1231 + }, + { + "epoch": 2.6607142857142856, + "grad_norm": 0.2763437468946581, + "learning_rate": 7.564842223334356e-05, + "loss": 0.2334, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23697665333747864, + "step": 745, + "valid_targets_mean": 3998.6, + "valid_targets_min": 1060 + }, + { + "epoch": 2.678571428571429, + "grad_norm": 0.2863963365342055, + "learning_rate": 7.534518781974065e-05, + "loss": 0.23, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2318994104862213, + "step": 750, + "valid_targets_mean": 4046.4, + "valid_targets_min": 752 + }, + { + "epoch": 2.696428571428571, + "grad_norm": 0.2756630470120407, + "learning_rate": 7.504069360180971e-05, + "loss": 0.2305, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23213598132133484, + "step": 755, + "valid_targets_mean": 4041.9, + "valid_targets_min": 1580 + }, + { + "epoch": 2.7142857142857144, + "grad_norm": 0.29430289997131637, + "learning_rate": 7.473495471469733e-05, + "loss": 0.2367, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.239883691072464, + "step": 760, + "valid_targets_mean": 3572.8, + "valid_targets_min": 748 + }, + { + "epoch": 2.732142857142857, + "grad_norm": 0.28964763447455116, + "learning_rate": 7.442798635541749e-05, + "loss": 0.2337, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24638831615447998, + "step": 765, + "valid_targets_mean": 3717.0, + "valid_targets_min": 647 + }, + { + "epoch": 2.75, + "grad_norm": 0.28398665189885464, + "learning_rate": 7.411980378209611e-05, + "loss": 0.231, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24520479142665863, + "step": 770, + "valid_targets_mean": 3927.2, + "valid_targets_min": 1020 + }, + { + "epoch": 2.767857142857143, + "grad_norm": 0.2763335857962852, + "learning_rate": 7.381042231321269e-05, + "loss": 0.2367, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23126724362373352, + "step": 775, + "valid_targets_mean": 4046.0, + "valid_targets_min": 1438 + }, + { + "epoch": 2.7857142857142856, + "grad_norm": 0.27362571597529123, + "learning_rate": 7.349985732683886e-05, + "loss": 0.2451, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.25000321865081787, + "step": 780, + "valid_targets_mean": 4404.0, + "valid_targets_min": 955 + }, + { + "epoch": 2.803571428571429, + "grad_norm": 0.29235887525618376, + "learning_rate": 7.318812425987395e-05, + "loss": 0.2372, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2301982045173645, + "step": 785, + "valid_targets_mean": 3738.8, + "valid_targets_min": 1271 + }, + { + "epoch": 2.821428571428571, + "grad_norm": 0.27655410904571226, + "learning_rate": 7.287523860727781e-05, + "loss": 0.2275, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2272815853357315, + "step": 790, + "valid_targets_mean": 3962.1, + "valid_targets_min": 622 + }, + { + "epoch": 2.8392857142857144, + "grad_norm": 0.24846363705556654, + "learning_rate": 7.256121592130055e-05, + "loss": 0.2246, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21925005316734314, + "step": 795, + "valid_targets_mean": 4467.1, + "valid_targets_min": 1329 + }, + { + "epoch": 2.857142857142857, + "grad_norm": 0.25248835602837744, + "learning_rate": 7.224607181070941e-05, + "loss": 0.2296, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2129673957824707, + "step": 800, + "valid_targets_mean": 4262.2, + "valid_targets_min": 1676 + }, + { + "epoch": 2.875, + "grad_norm": 0.26551307723687684, + "learning_rate": 7.192982194001312e-05, + "loss": 0.2344, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23995287716388702, + "step": 805, + "valid_targets_mean": 4005.6, + "valid_targets_min": 1306 + }, + { + "epoch": 2.892857142857143, + "grad_norm": 0.2836671386904135, + "learning_rate": 7.161248202868309e-05, + "loss": 0.2218, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24358542263507843, + "step": 810, + "valid_targets_mean": 3929.9, + "valid_targets_min": 723 + }, + { + "epoch": 2.9107142857142856, + "grad_norm": 0.28698631435601263, + "learning_rate": 7.129406785037214e-05, + "loss": 0.233, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22982844710350037, + "step": 815, + "valid_targets_mean": 3644.5, + "valid_targets_min": 785 + }, + { + "epoch": 2.928571428571429, + "grad_norm": 0.2826927287490477, + "learning_rate": 7.09745952321305e-05, + "loss": 0.2308, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24791894853115082, + "step": 820, + "valid_targets_mean": 4062.5, + "valid_targets_min": 748 + }, + { + "epoch": 2.946428571428571, + "grad_norm": 0.27529644267658454, + "learning_rate": 7.065408005361903e-05, + "loss": 0.2345, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24385912716388702, + "step": 825, + "valid_targets_mean": 4057.3, + "valid_targets_min": 1355 + }, + { + "epoch": 2.9642857142857144, + "grad_norm": 0.24499889291823151, + "learning_rate": 7.033253824631991e-05, + "loss": 0.2228, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22231167554855347, + "step": 830, + "valid_targets_mean": 4576.9, + "valid_targets_min": 903 + }, + { + "epoch": 2.982142857142857, + "grad_norm": 0.24422903308054847, + "learning_rate": 7.000998579274487e-05, + "loss": 0.2296, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2222803235054016, + "step": 835, + "valid_targets_mean": 4633.5, + "valid_targets_min": 1042 + }, + { + "epoch": 3.0, + "grad_norm": 0.3013045867283355, + "learning_rate": 6.968643872564064e-05, + "loss": 0.2328, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2250707447528839, + "step": 840, + "valid_targets_mean": 3316.1, + "valid_targets_min": 715 + }, + { + "epoch": 3.017857142857143, + "grad_norm": 0.36888355514819277, + "learning_rate": 6.936191312719203e-05, + "loss": 0.1734, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18953615427017212, + "step": 845, + "valid_targets_mean": 4303.2, + "valid_targets_min": 915 + }, + { + "epoch": 3.0357142857142856, + "grad_norm": 0.2743608827751915, + "learning_rate": 6.903642512822263e-05, + "loss": 0.1718, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16002383828163147, + "step": 850, + "valid_targets_mean": 4635.4, + "valid_targets_min": 1385 + }, + { + "epoch": 3.0535714285714284, + "grad_norm": 0.2949178115736851, + "learning_rate": 6.870999090739301e-05, + "loss": 0.1735, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17243778705596924, + "step": 855, + "valid_targets_mean": 4107.0, + "valid_targets_min": 1631 + }, + { + "epoch": 3.0714285714285716, + "grad_norm": 0.29558901205386284, + "learning_rate": 6.838262669039643e-05, + "loss": 0.1745, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1815764605998993, + "step": 860, + "valid_targets_mean": 3878.2, + "valid_targets_min": 1559 + }, + { + "epoch": 3.0892857142857144, + "grad_norm": 0.28472658734981565, + "learning_rate": 6.805434874915249e-05, + "loss": 0.1695, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15954339504241943, + "step": 865, + "valid_targets_mean": 3988.9, + "valid_targets_min": 1336 + }, + { + "epoch": 3.107142857142857, + "grad_norm": 0.28455293881829763, + "learning_rate": 6.772517340099816e-05, + "loss": 0.1687, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1778276264667511, + "step": 870, + "valid_targets_mean": 4087.2, + "valid_targets_min": 1615 + }, + { + "epoch": 3.125, + "grad_norm": 0.29906849485130366, + "learning_rate": 6.739511700787683e-05, + "loss": 0.1749, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18001751601696014, + "step": 875, + "valid_targets_mean": 4335.7, + "valid_targets_min": 530 + }, + { + "epoch": 3.142857142857143, + "grad_norm": 0.29891495714854055, + "learning_rate": 6.7064195975525e-05, + "loss": 0.1789, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1715604066848755, + "step": 880, + "valid_targets_mean": 3988.3, + "valid_targets_min": 732 + }, + { + "epoch": 3.1607142857142856, + "grad_norm": 0.3015940788820379, + "learning_rate": 6.67324267526568e-05, + "loss": 0.1759, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16995128989219666, + "step": 885, + "valid_targets_mean": 3274.9, + "valid_targets_min": 698 + }, + { + "epoch": 3.1785714285714284, + "grad_norm": 0.2887035705764202, + "learning_rate": 6.639982583014637e-05, + "loss": 0.1777, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1806672215461731, + "step": 890, + "valid_targets_mean": 4508.6, + "valid_targets_min": 735 + }, + { + "epoch": 3.1964285714285716, + "grad_norm": 0.3059279725982452, + "learning_rate": 6.606640974020823e-05, + "loss": 0.1802, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17828229069709778, + "step": 895, + "valid_targets_mean": 4020.5, + "valid_targets_min": 824 + }, + { + "epoch": 3.2142857142857144, + "grad_norm": 0.28577223351690517, + "learning_rate": 6.573219505557548e-05, + "loss": 0.1784, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18053409457206726, + "step": 900, + "valid_targets_mean": 4198.5, + "valid_targets_min": 836 + }, + { + "epoch": 3.232142857142857, + "grad_norm": 0.30099103878237704, + "learning_rate": 6.539719838867604e-05, + "loss": 0.1683, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16517090797424316, + "step": 905, + "valid_targets_mean": 4438.6, + "valid_targets_min": 715 + }, + { + "epoch": 3.25, + "grad_norm": 0.30147019285920784, + "learning_rate": 6.506143639080695e-05, + "loss": 0.1777, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1623014211654663, + "step": 910, + "valid_targets_mean": 3888.4, + "valid_targets_min": 539 + }, + { + "epoch": 3.267857142857143, + "grad_norm": 0.30187178272438475, + "learning_rate": 6.472492575130671e-05, + "loss": 0.1752, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18048198521137238, + "step": 915, + "valid_targets_mean": 4569.2, + "valid_targets_min": 1788 + }, + { + "epoch": 3.2857142857142856, + "grad_norm": 0.2949425020713681, + "learning_rate": 6.438768319672561e-05, + "loss": 0.1792, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17755785584449768, + "step": 920, + "valid_targets_mean": 3876.5, + "valid_targets_min": 773 + }, + { + "epoch": 3.3035714285714284, + "grad_norm": 0.2813486601344545, + "learning_rate": 6.404972548999453e-05, + "loss": 0.1772, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19014155864715576, + "step": 925, + "valid_targets_mean": 3937.8, + "valid_targets_min": 779 + }, + { + "epoch": 3.3214285714285716, + "grad_norm": 0.29136913304596107, + "learning_rate": 6.37110694295915e-05, + "loss": 0.1697, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17238101363182068, + "step": 930, + "valid_targets_mean": 3864.3, + "valid_targets_min": 1686 + }, + { + "epoch": 3.3392857142857144, + "grad_norm": 0.30664582906264054, + "learning_rate": 6.337173184870683e-05, + "loss": 0.1764, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16993115842342377, + "step": 935, + "valid_targets_mean": 3501.4, + "valid_targets_min": 832 + }, + { + "epoch": 3.357142857142857, + "grad_norm": 0.26315110459279134, + "learning_rate": 6.303172961440645e-05, + "loss": 0.1781, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17361228168010712, + "step": 940, + "valid_targets_mean": 4797.4, + "valid_targets_min": 1259 + }, + { + "epoch": 3.375, + "grad_norm": 0.282137175833168, + "learning_rate": 6.269107962679344e-05, + "loss": 0.1767, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1820008009672165, + "step": 945, + "valid_targets_mean": 4040.9, + "valid_targets_min": 1506 + }, + { + "epoch": 3.392857142857143, + "grad_norm": 0.263903533187103, + "learning_rate": 6.234979881816793e-05, + "loss": 0.1801, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1839650720357895, + "step": 950, + "valid_targets_mean": 4767.8, + "valid_targets_min": 1463 + }, + { + "epoch": 3.4107142857142856, + "grad_norm": 0.29927560095320194, + "learning_rate": 6.200790415218568e-05, + "loss": 0.1769, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17318886518478394, + "step": 955, + "valid_targets_mean": 3689.7, + "valid_targets_min": 1254 + }, + { + "epoch": 3.4285714285714284, + "grad_norm": 0.27627340639348663, + "learning_rate": 6.166541262301468e-05, + "loss": 0.1775, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18372446298599243, + "step": 960, + "valid_targets_mean": 4918.7, + "valid_targets_min": 1791 + }, + { + "epoch": 3.4464285714285716, + "grad_norm": 0.30119506003284013, + "learning_rate": 6.13223412544905e-05, + "loss": 0.1835, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1705668568611145, + "step": 965, + "valid_targets_mean": 3409.4, + "valid_targets_min": 979 + }, + { + "epoch": 3.4642857142857144, + "grad_norm": 0.2932496586638858, + "learning_rate": 6.0978707099270214e-05, + "loss": 0.175, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18003857135772705, + "step": 970, + "valid_targets_mean": 4257.7, + "valid_targets_min": 942 + }, + { + "epoch": 3.482142857142857, + "grad_norm": 0.2782506995796531, + "learning_rate": 6.0634527237984604e-05, + "loss": 0.1797, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18446490168571472, + "step": 975, + "valid_targets_mean": 4364.7, + "valid_targets_min": 971 + }, + { + "epoch": 3.5, + "grad_norm": 0.3099050593712785, + "learning_rate": 6.028981877838925e-05, + "loss": 0.1761, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1693999171257019, + "step": 980, + "valid_targets_mean": 3708.5, + "valid_targets_min": 840 + }, + { + "epoch": 3.517857142857143, + "grad_norm": 0.28926992677835955, + "learning_rate": 5.994459885451422e-05, + "loss": 0.178, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16710270941257477, + "step": 985, + "valid_targets_mean": 3700.2, + "valid_targets_min": 1355 + }, + { + "epoch": 3.5357142857142856, + "grad_norm": 0.2836517919187999, + "learning_rate": 5.9598884625812315e-05, + "loss": 0.1773, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1771649420261383, + "step": 990, + "valid_targets_mean": 3893.0, + "valid_targets_min": 1533 + }, + { + "epoch": 3.553571428571429, + "grad_norm": 0.2898241109281251, + "learning_rate": 5.925269327630615e-05, + "loss": 0.1802, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18730324506759644, + "step": 995, + "valid_targets_mean": 4069.4, + "valid_targets_min": 1540 + }, + { + "epoch": 3.571428571428571, + "grad_norm": 0.30985368243041084, + "learning_rate": 5.890604201373411e-05, + "loss": 0.1723, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1658320128917694, + "step": 1000, + "valid_targets_mean": 3816.3, + "valid_targets_min": 654 + }, + { + "epoch": 3.5892857142857144, + "grad_norm": 0.28229940773328654, + "learning_rate": 5.855894806869493e-05, + "loss": 0.1847, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18986308574676514, + "step": 1005, + "valid_targets_mean": 4387.7, + "valid_targets_min": 752 + }, + { + "epoch": 3.607142857142857, + "grad_norm": 0.30482724301755476, + "learning_rate": 5.82114286937912e-05, + "loss": 0.1768, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16868126392364502, + "step": 1010, + "valid_targets_mean": 3607.3, + "valid_targets_min": 873 + }, + { + "epoch": 3.625, + "grad_norm": 0.2841782198251889, + "learning_rate": 5.786350116277195e-05, + "loss": 0.1736, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1747998744249344, + "step": 1015, + "valid_targets_mean": 4285.5, + "valid_targets_min": 1672 + }, + { + "epoch": 3.642857142857143, + "grad_norm": 0.27288382845384174, + "learning_rate": 5.7515182769673915e-05, + "loss": 0.1785, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16694341599941254, + "step": 1020, + "valid_targets_mean": 4297.9, + "valid_targets_min": 626 + }, + { + "epoch": 3.6607142857142856, + "grad_norm": 0.31694801804057104, + "learning_rate": 5.716649082796198e-05, + "loss": 0.1792, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17280970513820648, + "step": 1025, + "valid_targets_mean": 3639.3, + "valid_targets_min": 740 + }, + { + "epoch": 3.678571428571429, + "grad_norm": 0.30803221958484994, + "learning_rate": 5.681744266966856e-05, + "loss": 0.1803, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18715696036815643, + "step": 1030, + "valid_targets_mean": 3359.3, + "valid_targets_min": 1427 + }, + { + "epoch": 3.696428571428571, + "grad_norm": 0.28087694955377235, + "learning_rate": 5.6468055644532156e-05, + "loss": 0.175, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17530888319015503, + "step": 1035, + "valid_targets_mean": 3581.6, + "valid_targets_min": 1188 + }, + { + "epoch": 3.7142857142857144, + "grad_norm": 0.2827245499761631, + "learning_rate": 5.6118347119134916e-05, + "loss": 0.1758, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17095929384231567, + "step": 1040, + "valid_targets_mean": 4282.7, + "valid_targets_min": 779 + }, + { + "epoch": 3.732142857142857, + "grad_norm": 0.2864589954586194, + "learning_rate": 5.576833447603943e-05, + "loss": 0.1849, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18526536226272583, + "step": 1045, + "valid_targets_mean": 4631.1, + "valid_targets_min": 932 + }, + { + "epoch": 3.75, + "grad_norm": 0.29839060591616573, + "learning_rate": 5.541803511292474e-05, + "loss": 0.183, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1914333999156952, + "step": 1050, + "valid_targets_mean": 3823.6, + "valid_targets_min": 723 + }, + { + "epoch": 3.767857142857143, + "grad_norm": 0.27512542215177754, + "learning_rate": 5.506746644172154e-05, + "loss": 0.1787, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16921429336071014, + "step": 1055, + "valid_targets_mean": 3671.9, + "valid_targets_min": 1528 + }, + { + "epoch": 3.7857142857142856, + "grad_norm": 0.287025523735746, + "learning_rate": 5.471664588774671e-05, + "loss": 0.1762, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19496572017669678, + "step": 1060, + "valid_targets_mean": 4692.2, + "valid_targets_min": 1688 + }, + { + "epoch": 3.803571428571429, + "grad_norm": 0.2784694847243461, + "learning_rate": 5.436559088883717e-05, + "loss": 0.1814, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18207921087741852, + "step": 1065, + "valid_targets_mean": 3979.0, + "valid_targets_min": 1435 + }, + { + "epoch": 3.821428571428571, + "grad_norm": 0.3110022048416664, + "learning_rate": 5.4014318894483175e-05, + "loss": 0.1811, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18802762031555176, + "step": 1070, + "valid_targets_mean": 3546.8, + "valid_targets_min": 731 + }, + { + "epoch": 3.8392857142857144, + "grad_norm": 0.30785860600083625, + "learning_rate": 5.3662847364960855e-05, + "loss": 0.1845, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1960841566324234, + "step": 1075, + "valid_targets_mean": 3787.0, + "valid_targets_min": 755 + }, + { + "epoch": 3.857142857142857, + "grad_norm": 0.2573481530109878, + "learning_rate": 5.331119377046446e-05, + "loss": 0.1755, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17424535751342773, + "step": 1080, + "valid_targets_mean": 4731.3, + "valid_targets_min": 878 + }, + { + "epoch": 3.875, + "grad_norm": 0.28220342262631487, + "learning_rate": 5.295937559023794e-05, + "loss": 0.1767, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17703908681869507, + "step": 1085, + "valid_targets_mean": 4261.6, + "valid_targets_min": 1349 + }, + { + "epoch": 3.892857142857143, + "grad_norm": 0.2940665061569505, + "learning_rate": 5.260741031170605e-05, + "loss": 0.1772, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16015812754631042, + "step": 1090, + "valid_targets_mean": 4217.4, + "valid_targets_min": 649 + }, + { + "epoch": 3.9107142857142856, + "grad_norm": 0.2892615443879139, + "learning_rate": 5.225531542960528e-05, + "loss": 0.1742, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18002673983573914, + "step": 1095, + "valid_targets_mean": 4396.1, + "valid_targets_min": 635 + }, + { + "epoch": 3.928571428571429, + "grad_norm": 0.2844523156160923, + "learning_rate": 5.190310844511412e-05, + "loss": 0.1799, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17435865104198456, + "step": 1100, + "valid_targets_mean": 4064.3, + "valid_targets_min": 1587 + }, + { + "epoch": 3.946428571428571, + "grad_norm": 0.2794612377756448, + "learning_rate": 5.155080686498324e-05, + "loss": 0.1785, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17053505778312683, + "step": 1105, + "valid_targets_mean": 4957.3, + "valid_targets_min": 1522 + }, + { + "epoch": 3.9642857142857144, + "grad_norm": 0.31960522165351335, + "learning_rate": 5.1198428200665227e-05, + "loss": 0.1807, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19623860716819763, + "step": 1110, + "valid_targets_mean": 3665.9, + "valid_targets_min": 647 + }, + { + "epoch": 3.982142857142857, + "grad_norm": 0.26489906105265987, + "learning_rate": 5.0845989967444255e-05, + "loss": 0.1704, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1709238737821579, + "step": 1115, + "valid_targets_mean": 4770.1, + "valid_targets_min": 1088 + }, + { + "epoch": 4.0, + "grad_norm": 0.28991999424732645, + "learning_rate": 5.0493509683565365e-05, + "loss": 0.1715, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13934820890426636, + "step": 1120, + "valid_targets_mean": 3832.1, + "valid_targets_min": 1444 + }, + { + "epoch": 4.017857142857143, + "grad_norm": 0.451534621770474, + "learning_rate": 5.014100486936383e-05, + "loss": 0.1216, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11967265605926514, + "step": 1125, + "valid_targets_mean": 4510.9, + "valid_targets_min": 668 + }, + { + "epoch": 4.035714285714286, + "grad_norm": 0.31869542031728276, + "learning_rate": 4.9788493046394136e-05, + "loss": 0.1248, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12776055932044983, + "step": 1130, + "valid_targets_mean": 3481.6, + "valid_targets_min": 1427 + }, + { + "epoch": 4.053571428571429, + "grad_norm": 0.3275684299181593, + "learning_rate": 4.9435991736559245e-05, + "loss": 0.1195, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13159064948558807, + "step": 1135, + "valid_targets_mean": 4423.1, + "valid_targets_min": 1458 + }, + { + "epoch": 4.071428571428571, + "grad_norm": 0.27607322783336474, + "learning_rate": 4.908351846123947e-05, + "loss": 0.1214, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11679871380329132, + "step": 1140, + "valid_targets_mean": 4784.1, + "valid_targets_min": 1057 + }, + { + "epoch": 4.089285714285714, + "grad_norm": 0.312437064743053, + "learning_rate": 4.8731090740421685e-05, + "loss": 0.1217, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12207663059234619, + "step": 1145, + "valid_targets_mean": 4172.3, + "valid_targets_min": 1435 + }, + { + "epoch": 4.107142857142857, + "grad_norm": 0.33140749499535, + "learning_rate": 4.837872609182846e-05, + "loss": 0.1243, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12098836153745651, + "step": 1150, + "valid_targets_mean": 3562.7, + "valid_targets_min": 700 + }, + { + "epoch": 4.125, + "grad_norm": 0.31548516365293605, + "learning_rate": 4.802644203004723e-05, + "loss": 0.1225, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11879569292068481, + "step": 1155, + "valid_targets_mean": 3779.4, + "valid_targets_min": 1094 + }, + { + "epoch": 4.142857142857143, + "grad_norm": 0.3168397120550745, + "learning_rate": 4.767425606565987e-05, + "loss": 0.1214, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1203106939792633, + "step": 1160, + "valid_targets_mean": 3891.2, + "valid_targets_min": 685 + }, + { + "epoch": 4.160714285714286, + "grad_norm": 0.3350388879507561, + "learning_rate": 4.7322185704372234e-05, + "loss": 0.1208, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12125179171562195, + "step": 1165, + "valid_targets_mean": 3572.4, + "valid_targets_min": 738 + }, + { + "epoch": 4.178571428571429, + "grad_norm": 0.3116230575392623, + "learning_rate": 4.697024844614396e-05, + "loss": 0.125, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1259845346212387, + "step": 1170, + "valid_targets_mean": 4102.0, + "valid_targets_min": 1573 + }, + { + "epoch": 4.196428571428571, + "grad_norm": 0.32083995600086124, + "learning_rate": 4.661846178431873e-05, + "loss": 0.1226, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11833621561527252, + "step": 1175, + "valid_targets_mean": 3753.1, + "valid_targets_min": 1395 + }, + { + "epoch": 4.214285714285714, + "grad_norm": 0.31654278436640576, + "learning_rate": 4.626684320475473e-05, + "loss": 0.1288, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1299327313899994, + "step": 1180, + "valid_targets_mean": 4084.4, + "valid_targets_min": 1568 + }, + { + "epoch": 4.232142857142857, + "grad_norm": 0.3128244320348968, + "learning_rate": 4.591541018495542e-05, + "loss": 0.1204, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10938730090856552, + "step": 1185, + "valid_targets_mean": 3991.2, + "valid_targets_min": 1256 + }, + { + "epoch": 4.25, + "grad_norm": 0.3065740834373241, + "learning_rate": 4.556418019320087e-05, + "loss": 0.1216, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11140576004981995, + "step": 1190, + "valid_targets_mean": 4395.6, + "valid_targets_min": 1477 + }, + { + "epoch": 4.267857142857143, + "grad_norm": 0.31978949222293923, + "learning_rate": 4.521317068767949e-05, + "loss": 0.1298, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13997787237167358, + "step": 1195, + "valid_targets_mean": 3860.8, + "valid_targets_min": 836 + }, + { + "epoch": 4.285714285714286, + "grad_norm": 0.35349677784290057, + "learning_rate": 4.4862399115620184e-05, + "loss": 0.1252, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13185806572437286, + "step": 1200, + "valid_targets_mean": 3925.2, + "valid_targets_min": 799 + }, + { + "epoch": 4.303571428571429, + "grad_norm": 0.3244248562590004, + "learning_rate": 4.451188291242521e-05, + "loss": 0.1246, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12347427755594254, + "step": 1205, + "valid_targets_mean": 3967.4, + "valid_targets_min": 646 + }, + { + "epoch": 4.321428571428571, + "grad_norm": 0.30723219910694577, + "learning_rate": 4.4161639500803515e-05, + "loss": 0.1301, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12852035462856293, + "step": 1210, + "valid_targets_mean": 4419.7, + "valid_targets_min": 1143 + }, + { + "epoch": 4.339285714285714, + "grad_norm": 0.3363607312732738, + "learning_rate": 4.381168628990464e-05, + "loss": 0.1284, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1323574185371399, + "step": 1215, + "valid_targets_mean": 3381.8, + "valid_targets_min": 630 + }, + { + "epoch": 4.357142857142857, + "grad_norm": 0.3098634754883557, + "learning_rate": 4.346204067445348e-05, + "loss": 0.1213, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11363941431045532, + "step": 1220, + "valid_targets_mean": 4274.2, + "valid_targets_min": 1371 + }, + { + "epoch": 4.375, + "grad_norm": 0.32154880540342107, + "learning_rate": 4.311272003388568e-05, + "loss": 0.1211, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1257430911064148, + "step": 1225, + "valid_targets_mean": 4325.1, + "valid_targets_min": 657 + }, + { + "epoch": 4.392857142857143, + "grad_norm": 0.28637371109478005, + "learning_rate": 4.2763741731483664e-05, + "loss": 0.1252, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12955023348331451, + "step": 1230, + "valid_targets_mean": 5250.8, + "valid_targets_min": 1438 + }, + { + "epoch": 4.410714285714286, + "grad_norm": 0.31422460467122065, + "learning_rate": 4.241512311351366e-05, + "loss": 0.1265, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11697547137737274, + "step": 1235, + "valid_targets_mean": 4267.8, + "valid_targets_min": 1499 + }, + { + "epoch": 4.428571428571429, + "grad_norm": 0.3056796048265474, + "learning_rate": 4.2066881508363523e-05, + "loss": 0.1304, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13574860990047455, + "step": 1240, + "valid_targets_mean": 4882.5, + "valid_targets_min": 2100 + }, + { + "epoch": 4.446428571428571, + "grad_norm": 0.3216849805396823, + "learning_rate": 4.1719034225681274e-05, + "loss": 0.1266, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1248731017112732, + "step": 1245, + "valid_targets_mean": 3816.4, + "valid_targets_min": 528 + }, + { + "epoch": 4.464285714285714, + "grad_norm": 0.3143370105415563, + "learning_rate": 4.137159855551486e-05, + "loss": 0.1339, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11978290975093842, + "step": 1250, + "valid_targets_mean": 4493.4, + "valid_targets_min": 874 + }, + { + "epoch": 4.482142857142857, + "grad_norm": 0.33373032285127163, + "learning_rate": 4.102459176745267e-05, + "loss": 0.1251, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1251649409532547, + "step": 1255, + "valid_targets_mean": 3819.5, + "valid_targets_min": 837 + }, + { + "epoch": 4.5, + "grad_norm": 0.30593502005900564, + "learning_rate": 4.067803110976513e-05, + "loss": 0.1307, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1278524100780487, + "step": 1260, + "valid_targets_mean": 4171.4, + "valid_targets_min": 1564 + }, + { + "epoch": 4.517857142857143, + "grad_norm": 0.3222173010025613, + "learning_rate": 4.033193380854734e-05, + "loss": 0.1244, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11454290896654129, + "step": 1265, + "valid_targets_mean": 3779.9, + "valid_targets_min": 742 + }, + { + "epoch": 4.535714285714286, + "grad_norm": 0.3303574927974592, + "learning_rate": 3.998631706686292e-05, + "loss": 0.1264, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1289159059524536, + "step": 1270, + "valid_targets_mean": 3778.0, + "valid_targets_min": 755 + }, + { + "epoch": 4.553571428571429, + "grad_norm": 0.31890491075495686, + "learning_rate": 3.964119806388887e-05, + "loss": 0.1211, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12130078673362732, + "step": 1275, + "valid_targets_mean": 3904.1, + "valid_targets_min": 1516 + }, + { + "epoch": 4.571428571428571, + "grad_norm": 0.32075685342886695, + "learning_rate": 3.929659395406159e-05, + "loss": 0.1255, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12681984901428223, + "step": 1280, + "valid_targets_mean": 4103.4, + "valid_targets_min": 1497 + }, + { + "epoch": 4.589285714285714, + "grad_norm": 0.30516634302964996, + "learning_rate": 3.895252186622433e-05, + "loss": 0.1275, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13202255964279175, + "step": 1285, + "valid_targets_mean": 4647.9, + "valid_targets_min": 1808 + }, + { + "epoch": 4.607142857142857, + "grad_norm": 0.31526161090705346, + "learning_rate": 3.860899890277575e-05, + "loss": 0.1196, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11984704434871674, + "step": 1290, + "valid_targets_mean": 4053.4, + "valid_targets_min": 1332 + }, + { + "epoch": 4.625, + "grad_norm": 0.2971140343455729, + "learning_rate": 3.8266042138819746e-05, + "loss": 0.1245, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11468343436717987, + "step": 1295, + "valid_targets_mean": 3809.1, + "valid_targets_min": 1391 + }, + { + "epoch": 4.642857142857143, + "grad_norm": 0.29336236692116224, + "learning_rate": 3.7923668621316824e-05, + "loss": 0.1211, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12110649049282074, + "step": 1300, + "valid_targets_mean": 4419.2, + "valid_targets_min": 1444 + }, + { + "epoch": 4.660714285714286, + "grad_norm": 0.32346533785021364, + "learning_rate": 3.758189536823673e-05, + "loss": 0.1254, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13088198006153107, + "step": 1305, + "valid_targets_mean": 3517.6, + "valid_targets_min": 626 + }, + { + "epoch": 4.678571428571429, + "grad_norm": 0.31441159333543034, + "learning_rate": 3.724073936771252e-05, + "loss": 0.1232, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12440180033445358, + "step": 1310, + "valid_targets_mean": 3832.8, + "valid_targets_min": 1451 + }, + { + "epoch": 4.696428571428571, + "grad_norm": 0.3186480697373718, + "learning_rate": 3.6900217577196185e-05, + "loss": 0.126, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13353362679481506, + "step": 1315, + "valid_targets_mean": 4094.6, + "valid_targets_min": 735 + }, + { + "epoch": 4.714285714285714, + "grad_norm": 0.3020743749620538, + "learning_rate": 3.656034692261582e-05, + "loss": 0.1245, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13044799864292145, + "step": 1320, + "valid_targets_mean": 4323.6, + "valid_targets_min": 1301 + }, + { + "epoch": 4.732142857142857, + "grad_norm": 0.3421631818815327, + "learning_rate": 3.622114429753418e-05, + "loss": 0.1244, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13268694281578064, + "step": 1325, + "valid_targets_mean": 3505.2, + "valid_targets_min": 622 + }, + { + "epoch": 4.75, + "grad_norm": 0.3423456249386629, + "learning_rate": 3.588262656230904e-05, + "loss": 0.1241, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13105139136314392, + "step": 1330, + "valid_targets_mean": 3470.0, + "valid_targets_min": 530 + }, + { + "epoch": 4.767857142857143, + "grad_norm": 0.30620685170143885, + "learning_rate": 3.554481054325522e-05, + "loss": 0.1277, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1297651082277298, + "step": 1335, + "valid_targets_mean": 3925.0, + "valid_targets_min": 1728 + }, + { + "epoch": 4.785714285714286, + "grad_norm": 0.2868301774018994, + "learning_rate": 3.520771303180803e-05, + "loss": 0.1278, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12289422750473022, + "step": 1340, + "valid_targets_mean": 4697.8, + "valid_targets_min": 1159 + }, + { + "epoch": 4.803571428571429, + "grad_norm": 0.2923485857312998, + "learning_rate": 3.48713507836888e-05, + "loss": 0.1196, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11775079369544983, + "step": 1345, + "valid_targets_mean": 4619.3, + "valid_targets_min": 1605 + }, + { + "epoch": 4.821428571428571, + "grad_norm": 0.30833598321387995, + "learning_rate": 3.4535740518071966e-05, + "loss": 0.122, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12246265262365341, + "step": 1350, + "valid_targets_mean": 4220.8, + "valid_targets_min": 1254 + }, + { + "epoch": 4.839285714285714, + "grad_norm": 0.29392845652129446, + "learning_rate": 3.420089891675401e-05, + "loss": 0.1229, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10792181640863419, + "step": 1355, + "valid_targets_mean": 4164.7, + "valid_targets_min": 695 + }, + { + "epoch": 4.857142857142857, + "grad_norm": 0.2976928081052809, + "learning_rate": 3.386684262332429e-05, + "loss": 0.1282, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1328098177909851, + "step": 1360, + "valid_targets_mean": 4361.8, + "valid_targets_min": 1532 + }, + { + "epoch": 4.875, + "grad_norm": 0.32963911556753467, + "learning_rate": 3.35335882423378e-05, + "loss": 0.1269, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1411687731742859, + "step": 1365, + "valid_targets_mean": 3795.0, + "valid_targets_min": 868 + }, + { + "epoch": 4.892857142857143, + "grad_norm": 0.3061787070012623, + "learning_rate": 3.3201152338489726e-05, + "loss": 0.1214, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13152283430099487, + "step": 1370, + "valid_targets_mean": 4392.7, + "valid_targets_min": 1593 + }, + { + "epoch": 4.910714285714286, + "grad_norm": 0.30129774933052, + "learning_rate": 3.2869551435792184e-05, + "loss": 0.1226, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11523544788360596, + "step": 1375, + "valid_targets_mean": 3903.0, + "valid_targets_min": 915 + }, + { + "epoch": 4.928571428571429, + "grad_norm": 0.30784140914840247, + "learning_rate": 3.253880201675287e-05, + "loss": 0.1232, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11999917775392532, + "step": 1380, + "valid_targets_mean": 4264.9, + "valid_targets_min": 623 + }, + { + "epoch": 4.946428571428571, + "grad_norm": 0.31174757627134553, + "learning_rate": 3.2208920521555676e-05, + "loss": 0.1251, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12148493528366089, + "step": 1385, + "valid_targets_mean": 4068.2, + "valid_targets_min": 1405 + }, + { + "epoch": 4.964285714285714, + "grad_norm": 0.33685982146125953, + "learning_rate": 3.187992334724363e-05, + "loss": 0.1296, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13812394440174103, + "step": 1390, + "valid_targets_mean": 3466.9, + "valid_targets_min": 617 + }, + { + "epoch": 4.982142857142857, + "grad_norm": 0.31973777275111004, + "learning_rate": 3.1551826846903896e-05, + "loss": 0.1254, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.12065494060516357, + "step": 1395, + "valid_targets_mean": 3906.6, + "valid_targets_min": 733 + }, + { + "epoch": 5.0, + "grad_norm": 0.2994819596511423, + "learning_rate": 3.122464732885476e-05, + "loss": 0.12, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10494282841682434, + "step": 1400, + "valid_targets_mean": 4060.3, + "valid_targets_min": 744 + }, + { + "epoch": 5.017857142857143, + "grad_norm": 0.6895531207869822, + "learning_rate": 3.0898401055835156e-05, + "loss": 0.0823, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08173909783363342, + "step": 1405, + "valid_targets_mean": 3932.3, + "valid_targets_min": 938 + }, + { + "epoch": 5.035714285714286, + "grad_norm": 0.308257835133526, + "learning_rate": 3.057310424419632e-05, + "loss": 0.0837, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07695461809635162, + "step": 1410, + "valid_targets_mean": 4120.6, + "valid_targets_min": 1708 + }, + { + "epoch": 5.053571428571429, + "grad_norm": 0.3648171824246545, + "learning_rate": 3.0248773063095603e-05, + "loss": 0.076, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08023419976234436, + "step": 1415, + "valid_targets_mean": 3667.2, + "valid_targets_min": 691 + }, + { + "epoch": 5.071428571428571, + "grad_norm": 0.3433617630578246, + "learning_rate": 2.9925423633692923e-05, + "loss": 0.076, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08550803363323212, + "step": 1420, + "valid_targets_mean": 3896.2, + "valid_targets_min": 868 + }, + { + "epoch": 5.089285714285714, + "grad_norm": 0.3221600725668014, + "learning_rate": 2.960307202834941e-05, + "loss": 0.0762, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07856301963329315, + "step": 1425, + "valid_targets_mean": 4333.6, + "valid_targets_min": 786 + }, + { + "epoch": 5.107142857142857, + "grad_norm": 0.3198964123687821, + "learning_rate": 2.9281734269828408e-05, + "loss": 0.0775, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0831381231546402, + "step": 1430, + "valid_targets_mean": 4443.3, + "valid_targets_min": 1331 + }, + { + "epoch": 5.125, + "grad_norm": 0.3273563067974785, + "learning_rate": 2.896142633049922e-05, + "loss": 0.0748, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07411238551139832, + "step": 1435, + "valid_targets_mean": 3886.0, + "valid_targets_min": 738 + }, + { + "epoch": 5.142857142857143, + "grad_norm": 0.3044094111140198, + "learning_rate": 2.8642164131543048e-05, + "loss": 0.0769, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0747397392988205, + "step": 1440, + "valid_targets_mean": 4391.4, + "valid_targets_min": 1094 + }, + { + "epoch": 5.160714285714286, + "grad_norm": 0.35020721440807756, + "learning_rate": 2.8323963542161663e-05, + "loss": 0.077, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.09041236340999603, + "step": 1445, + "valid_targets_mean": 3875.8, + "valid_targets_min": 1609 + }, + { + "epoch": 5.178571428571429, + "grad_norm": 0.3360841433627317, + "learning_rate": 2.800684037878867e-05, + "loss": 0.077, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07672949880361557, + "step": 1450, + "valid_targets_mean": 3548.8, + "valid_targets_min": 719 + }, + { + "epoch": 5.196428571428571, + "grad_norm": 0.335243405340086, + "learning_rate": 2.7690810404303276e-05, + "loss": 0.0815, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08165518939495087, + "step": 1455, + "valid_targets_mean": 3935.0, + "valid_targets_min": 1304 + }, + { + "epoch": 5.214285714285714, + "grad_norm": 0.3228656216287706, + "learning_rate": 2.7375889327246744e-05, + "loss": 0.0767, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07716818153858185, + "step": 1460, + "valid_targets_mean": 4342.4, + "valid_targets_min": 1373 + }, + { + "epoch": 5.232142857142857, + "grad_norm": 0.30602600993967805, + "learning_rate": 2.7062092801041716e-05, + "loss": 0.0767, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07576284557580948, + "step": 1465, + "valid_targets_mean": 4187.2, + "valid_targets_min": 715 + }, + { + "epoch": 5.25, + "grad_norm": 0.37919790138615983, + "learning_rate": 2.6749436423214e-05, + "loss": 0.0794, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0794055387377739, + "step": 1470, + "valid_targets_mean": 4306.7, + "valid_targets_min": 851 + }, + { + "epoch": 5.267857142857143, + "grad_norm": 0.31112794556315865, + "learning_rate": 2.6437935734617393e-05, + "loss": 0.0781, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07946112751960754, + "step": 1475, + "valid_targets_mean": 4588.7, + "valid_targets_min": 1451 + }, + { + "epoch": 5.285714285714286, + "grad_norm": 0.31463210971547867, + "learning_rate": 2.612760621866113e-05, + "loss": 0.0784, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07934070378541946, + "step": 1480, + "valid_targets_mean": 4180.7, + "valid_targets_min": 779 + }, + { + "epoch": 5.303571428571429, + "grad_norm": 0.320225067301147, + "learning_rate": 2.581846330054034e-05, + "loss": 0.0785, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07398343086242676, + "step": 1485, + "valid_targets_mean": 4172.8, + "valid_targets_min": 1403 + }, + { + "epoch": 5.321428571428571, + "grad_norm": 0.31651267928872295, + "learning_rate": 2.551052234646929e-05, + "loss": 0.0801, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08124566078186035, + "step": 1490, + "valid_targets_mean": 4220.7, + "valid_targets_min": 1476 + }, + { + "epoch": 5.339285714285714, + "grad_norm": 0.318255472621264, + "learning_rate": 2.5203798662917555e-05, + "loss": 0.076, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07714743167161942, + "step": 1495, + "valid_targets_mean": 4190.4, + "valid_targets_min": 1142 + }, + { + "epoch": 5.357142857142857, + "grad_norm": 0.33975709493337775, + "learning_rate": 2.489830749584924e-05, + "loss": 0.0779, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0697803795337677, + "step": 1500, + "valid_targets_mean": 3611.3, + "valid_targets_min": 1618 + }, + { + "epoch": 5.375, + "grad_norm": 0.32820382646406937, + "learning_rate": 2.4594064029965197e-05, + "loss": 0.0798, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07638822495937347, + "step": 1505, + "valid_targets_mean": 4132.7, + "valid_targets_min": 1155 + }, + { + "epoch": 5.392857142857143, + "grad_norm": 0.3540109637161371, + "learning_rate": 2.429108338794817e-05, + "loss": 0.0787, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08276736736297607, + "step": 1510, + "valid_targets_mean": 3988.2, + "valid_targets_min": 943 + }, + { + "epoch": 5.410714285714286, + "grad_norm": 0.3578200985747836, + "learning_rate": 2.3989380629711194e-05, + "loss": 0.0791, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08381788432598114, + "step": 1515, + "valid_targets_mean": 3404.6, + "valid_targets_min": 672 + }, + { + "epoch": 5.428571428571429, + "grad_norm": 0.3218477229817908, + "learning_rate": 2.3688970751649002e-05, + "loss": 0.0837, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08048512041568756, + "step": 1520, + "valid_targets_mean": 3888.3, + "valid_targets_min": 1188 + }, + { + "epoch": 5.446428571428571, + "grad_norm": 0.3300434508814151, + "learning_rate": 2.3389868685892573e-05, + "loss": 0.0764, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07927368581295013, + "step": 1525, + "valid_targets_mean": 3909.9, + "valid_targets_min": 1486 + }, + { + "epoch": 5.464285714285714, + "grad_norm": 0.30726628897860647, + "learning_rate": 2.309208929956694e-05, + "loss": 0.0743, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07424400746822357, + "step": 1530, + "valid_targets_mean": 4278.7, + "valid_targets_min": 735 + }, + { + "epoch": 5.482142857142857, + "grad_norm": 0.2865995807445488, + "learning_rate": 2.2795647394052284e-05, + "loss": 0.0749, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06469792127609253, + "step": 1535, + "valid_targets_mean": 4355.5, + "valid_targets_min": 1455 + }, + { + "epoch": 5.5, + "grad_norm": 0.3264845505865478, + "learning_rate": 2.2500557704248083e-05, + "loss": 0.0761, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07780517637729645, + "step": 1540, + "valid_targets_mean": 4292.3, + "valid_targets_min": 711 + }, + { + "epoch": 5.517857142857143, + "grad_norm": 0.3154717281910029, + "learning_rate": 2.220683489784081e-05, + "loss": 0.0814, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08107089251279831, + "step": 1545, + "valid_targets_mean": 4626.5, + "valid_targets_min": 1527 + }, + { + "epoch": 5.535714285714286, + "grad_norm": 0.29810732769867293, + "learning_rate": 2.1914493574574858e-05, + "loss": 0.0768, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07099950313568115, + "step": 1550, + "valid_targets_mean": 4301.5, + "valid_targets_min": 1124 + }, + { + "epoch": 5.553571428571429, + "grad_norm": 0.33309410301089987, + "learning_rate": 2.162354826552673e-05, + "loss": 0.0806, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07673013210296631, + "step": 1555, + "valid_targets_mean": 3712.4, + "valid_targets_min": 530 + }, + { + "epoch": 5.571428571428571, + "grad_norm": 0.3488466182358611, + "learning_rate": 2.1334013432382894e-05, + "loss": 0.0799, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07404391467571259, + "step": 1560, + "valid_targets_mean": 3314.2, + "valid_targets_min": 764 + }, + { + "epoch": 5.589285714285714, + "grad_norm": 0.30167910299742445, + "learning_rate": 2.1045903466720913e-05, + "loss": 0.0808, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07387515902519226, + "step": 1565, + "valid_targets_mean": 4527.9, + "valid_targets_min": 896 + }, + { + "epoch": 5.607142857142857, + "grad_norm": 0.331159472333701, + "learning_rate": 2.0759232689294044e-05, + "loss": 0.0782, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07977255433797836, + "step": 1570, + "valid_targets_mean": 3897.1, + "valid_targets_min": 903 + }, + { + "epoch": 5.625, + "grad_norm": 0.35940782160274726, + "learning_rate": 2.0474015349319503e-05, + "loss": 0.0823, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.09296411275863647, + "step": 1575, + "valid_targets_mean": 3816.1, + "valid_targets_min": 664 + }, + { + "epoch": 5.642857142857143, + "grad_norm": 0.30894802512426395, + "learning_rate": 2.0190265623770143e-05, + "loss": 0.0764, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0791928768157959, + "step": 1580, + "valid_targets_mean": 4218.3, + "valid_targets_min": 732 + }, + { + "epoch": 5.660714285714286, + "grad_norm": 0.32164922417655917, + "learning_rate": 1.990799761666975e-05, + "loss": 0.0784, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08298400044441223, + "step": 1585, + "valid_targets_mean": 4553.2, + "valid_targets_min": 1560 + }, + { + "epoch": 5.678571428571429, + "grad_norm": 0.32348854042942393, + "learning_rate": 1.962722535839202e-05, + "loss": 0.0807, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08424737304449081, + "step": 1590, + "valid_targets_mean": 4280.0, + "valid_targets_min": 580 + }, + { + "epoch": 5.696428571428571, + "grad_norm": 0.3223358842716645, + "learning_rate": 1.9347962804963238e-05, + "loss": 0.0775, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08354779332876205, + "step": 1595, + "valid_targets_mean": 4233.2, + "valid_targets_min": 1661 + }, + { + "epoch": 5.714285714285714, + "grad_norm": 0.34322995935945416, + "learning_rate": 1.9070223837368412e-05, + "loss": 0.0798, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08479490876197815, + "step": 1600, + "valid_targets_mean": 4020.3, + "valid_targets_min": 836 + }, + { + "epoch": 5.732142857142857, + "grad_norm": 0.3234315363924524, + "learning_rate": 1.8794022260861482e-05, + "loss": 0.0781, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07609676569700241, + "step": 1605, + "valid_targets_mean": 3915.0, + "valid_targets_min": 1600 + }, + { + "epoch": 5.75, + "grad_norm": 0.31836053518331936, + "learning_rate": 1.8519371804279046e-05, + "loss": 0.0783, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08317436277866364, + "step": 1610, + "valid_targets_mean": 4347.5, + "valid_targets_min": 1306 + }, + { + "epoch": 5.767857142857143, + "grad_norm": 0.3460059635579521, + "learning_rate": 1.8246286119357903e-05, + "loss": 0.0773, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07543785870075226, + "step": 1615, + "valid_targets_mean": 3351.2, + "valid_targets_min": 1573 + }, + { + "epoch": 5.785714285714286, + "grad_norm": 0.3101390452055134, + "learning_rate": 1.797477878005655e-05, + "loss": 0.0766, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0691436380147934, + "step": 1620, + "valid_targets_mean": 3907.1, + "valid_targets_min": 1587 + }, + { + "epoch": 5.803571428571429, + "grad_norm": 0.3244743631959807, + "learning_rate": 1.7704863281880496e-05, + "loss": 0.0773, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08170685172080994, + "step": 1625, + "valid_targets_mean": 3930.0, + "valid_targets_min": 1770 + }, + { + "epoch": 5.821428571428571, + "grad_norm": 0.3338666069308065, + "learning_rate": 1.743655304121136e-05, + "loss": 0.0824, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08442583680152893, + "step": 1630, + "valid_targets_mean": 4034.2, + "valid_targets_min": 1809 + }, + { + "epoch": 5.839285714285714, + "grad_norm": 0.32492420368431646, + "learning_rate": 1.7169861394640107e-05, + "loss": 0.0776, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07858459651470184, + "step": 1635, + "valid_targets_mean": 3990.6, + "valid_targets_min": 1578 + }, + { + "epoch": 5.857142857142857, + "grad_norm": 0.3596885837994889, + "learning_rate": 1.6904801598304095e-05, + "loss": 0.083, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0926271304488182, + "step": 1640, + "valid_targets_mean": 3676.3, + "valid_targets_min": 729 + }, + { + "epoch": 5.875, + "grad_norm": 0.33296059014478036, + "learning_rate": 1.6641386827228105e-05, + "loss": 0.0795, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08574201166629791, + "step": 1645, + "valid_targets_mean": 4530.5, + "valid_targets_min": 1656 + }, + { + "epoch": 5.892857142857143, + "grad_norm": 0.32781758731573796, + "learning_rate": 1.637963017466961e-05, + "loss": 0.0794, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08394479751586914, + "step": 1650, + "valid_targets_mean": 3995.4, + "valid_targets_min": 1319 + }, + { + "epoch": 5.910714285714286, + "grad_norm": 0.30127371707688244, + "learning_rate": 1.61195446514678e-05, + "loss": 0.0782, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07440881431102753, + "step": 1655, + "valid_targets_mean": 4835.7, + "valid_targets_min": 733 + }, + { + "epoch": 5.928571428571429, + "grad_norm": 0.3075362685774865, + "learning_rate": 1.586114318539697e-05, + "loss": 0.0751, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07055199146270752, + "step": 1660, + "valid_targets_mean": 3962.7, + "valid_targets_min": 1329 + }, + { + "epoch": 5.946428571428571, + "grad_norm": 0.31217462168206894, + "learning_rate": 1.560443862052393e-05, + "loss": 0.0791, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.08014695346355438, + "step": 1665, + "valid_targets_mean": 4384.1, + "valid_targets_min": 910 + }, + { + "epoch": 5.964285714285714, + "grad_norm": 0.3581031397083817, + "learning_rate": 1.534944371656955e-05, + "loss": 0.0801, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0818948745727539, + "step": 1670, + "valid_targets_mean": 3801.5, + "valid_targets_min": 703 + }, + { + "epoch": 5.982142857142857, + "grad_norm": 0.2824198991546861, + "learning_rate": 1.5096171148274546e-05, + "loss": 0.071, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07383203506469727, + "step": 1675, + "valid_targets_mean": 5191.9, + "valid_targets_min": 1418 + }, + { + "epoch": 6.0, + "grad_norm": 0.29215381073719615, + "learning_rate": 1.4844633504769422e-05, + "loss": 0.0763, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06313540041446686, + "step": 1680, + "valid_targets_mean": 4091.1, + "valid_targets_min": 824 + }, + { + "epoch": 6.017857142857143, + "grad_norm": 0.33128677829620434, + "learning_rate": 1.4594843288948773e-05, + "loss": 0.0494, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04688066989183426, + "step": 1685, + "valid_targets_mean": 3909.4, + "valid_targets_min": 1727 + }, + { + "epoch": 6.035714285714286, + "grad_norm": 0.32277792926022025, + "learning_rate": 1.4346812916849839e-05, + "loss": 0.0458, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04678421840071678, + "step": 1690, + "valid_targets_mean": 4666.6, + "valid_targets_min": 1778 + }, + { + "epoch": 6.053571428571429, + "grad_norm": 0.3024827157078677, + "learning_rate": 1.4100554717035241e-05, + "loss": 0.0447, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04976955056190491, + "step": 1695, + "valid_targets_mean": 4191.4, + "valid_targets_min": 1465 + }, + { + "epoch": 6.071428571428571, + "grad_norm": 0.3242059330228126, + "learning_rate": 1.385608092998032e-05, + "loss": 0.0418, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.046975165605545044, + "step": 1700, + "valid_targets_mean": 3965.9, + "valid_targets_min": 1398 + }, + { + "epoch": 6.089285714285714, + "grad_norm": 0.2962955558533571, + "learning_rate": 1.3613403707464639e-05, + "loss": 0.0453, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.046070873737335205, + "step": 1705, + "valid_targets_mean": 3866.3, + "valid_targets_min": 626 + }, + { + "epoch": 6.107142857142857, + "grad_norm": 0.25811918191317496, + "learning_rate": 1.3372535111967949e-05, + "loss": 0.0428, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03580906242132187, + "step": 1710, + "valid_targets_mean": 4262.9, + "valid_targets_min": 843 + }, + { + "epoch": 6.125, + "grad_norm": 0.31320492455858356, + "learning_rate": 1.3133487116070643e-05, + "loss": 0.047, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04627719148993492, + "step": 1715, + "valid_targets_mean": 3930.0, + "valid_targets_min": 1212 + }, + { + "epoch": 6.142857142857143, + "grad_norm": 0.27285108861959784, + "learning_rate": 1.289627160185869e-05, + "loss": 0.0473, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.042734332382678986, + "step": 1720, + "valid_targets_mean": 4337.3, + "valid_targets_min": 896 + }, + { + "epoch": 6.160714285714286, + "grad_norm": 0.28411027727199856, + "learning_rate": 1.2660900360332928e-05, + "loss": 0.0476, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04408092796802521, + "step": 1725, + "valid_targets_mean": 4725.9, + "valid_targets_min": 1901 + }, + { + "epoch": 6.178571428571429, + "grad_norm": 0.295791635289143, + "learning_rate": 1.2427385090823073e-05, + "loss": 0.0445, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04541953653097153, + "step": 1730, + "valid_targets_mean": 4368.9, + "valid_targets_min": 649 + }, + { + "epoch": 6.196428571428571, + "grad_norm": 0.3166340845840027, + "learning_rate": 1.2195737400406166e-05, + "loss": 0.0429, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04301552474498749, + "step": 1735, + "valid_targets_mean": 3893.9, + "valid_targets_min": 716 + }, + { + "epoch": 6.214285714285714, + "grad_norm": 0.30779377142376796, + "learning_rate": 1.1965968803329585e-05, + "loss": 0.0461, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04589950293302536, + "step": 1740, + "valid_targets_mean": 3924.6, + "valid_targets_min": 873 + }, + { + "epoch": 6.232142857142857, + "grad_norm": 0.2910638794587935, + "learning_rate": 1.1738090720438782e-05, + "loss": 0.0455, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04226720705628395, + "step": 1745, + "valid_targets_mean": 4091.4, + "valid_targets_min": 1524 + }, + { + "epoch": 6.25, + "grad_norm": 0.3242031786451, + "learning_rate": 1.1512114478609598e-05, + "loss": 0.0455, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04951293766498566, + "step": 1750, + "valid_targets_mean": 3823.4, + "valid_targets_min": 826 + }, + { + "epoch": 6.267857142857143, + "grad_norm": 0.3057689916945106, + "learning_rate": 1.1288051310185182e-05, + "loss": 0.0441, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04831494018435478, + "step": 1755, + "valid_targets_mean": 4375.2, + "valid_targets_min": 1437 + }, + { + "epoch": 6.285714285714286, + "grad_norm": 0.2850917461167024, + "learning_rate": 1.1065912352417768e-05, + "loss": 0.0445, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.044211000204086304, + "step": 1760, + "valid_targets_mean": 4469.8, + "valid_targets_min": 1577 + }, + { + "epoch": 6.303571428571429, + "grad_norm": 0.32143495868772404, + "learning_rate": 1.0845708646915054e-05, + "loss": 0.0468, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04553980380296707, + "step": 1765, + "valid_targets_mean": 4249.5, + "valid_targets_min": 1653 + }, + { + "epoch": 6.321428571428571, + "grad_norm": 0.30244920578865875, + "learning_rate": 1.0627451139091321e-05, + "loss": 0.0464, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04243556410074234, + "step": 1770, + "valid_targets_mean": 3491.2, + "valid_targets_min": 717 + }, + { + "epoch": 6.339285714285714, + "grad_norm": 0.3328415537025107, + "learning_rate": 1.0411150677623438e-05, + "loss": 0.0454, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.048559363931417465, + "step": 1775, + "valid_targets_mean": 3556.2, + "valid_targets_min": 559 + }, + { + "epoch": 6.357142857142857, + "grad_norm": 0.3119612788548375, + "learning_rate": 1.0196818013911625e-05, + "loss": 0.0445, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04680598899722099, + "step": 1780, + "valid_targets_mean": 3996.8, + "valid_targets_min": 1626 + }, + { + "epoch": 6.375, + "grad_norm": 0.3387795897243579, + "learning_rate": 9.98446380154499e-06, + "loss": 0.0516, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.05588133633136749, + "step": 1785, + "valid_targets_mean": 3472.1, + "valid_targets_min": 658 + }, + { + "epoch": 6.392857142857143, + "grad_norm": 0.2981059943839166, + "learning_rate": 9.774098595772058e-06, + "loss": 0.0466, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0470689982175827, + "step": 1790, + "valid_targets_mean": 4551.4, + "valid_targets_min": 1779 + }, + { + "epoch": 6.410714285714286, + "grad_norm": 0.37706594196408666, + "learning_rate": 9.56573285297605e-06, + "loss": 0.0491, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.058256857097148895, + "step": 1795, + "valid_targets_mean": 3202.8, + "valid_targets_min": 1325 + }, + { + "epoch": 6.428571428571429, + "grad_norm": 0.3076226099755853, + "learning_rate": 9.359376930155157e-06, + "loss": 0.0448, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0463181808590889, + "step": 1800, + "valid_targets_mean": 4044.7, + "valid_targets_min": 1364 + }, + { + "epoch": 6.446428571428571, + "grad_norm": 0.31811673914159155, + "learning_rate": 9.15504108440774e-06, + "loss": 0.0465, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04420042037963867, + "step": 1805, + "valid_targets_mean": 3828.1, + "valid_targets_min": 635 + }, + { + "epoch": 6.464285714285714, + "grad_norm": 0.33670025997047826, + "learning_rate": 8.95273547242253e-06, + "loss": 0.0441, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.055958278477191925, + "step": 1810, + "valid_targets_mean": 3902.7, + "valid_targets_min": 1533 + }, + { + "epoch": 6.482142857142857, + "grad_norm": 0.3163194889679004, + "learning_rate": 8.752470149973684e-06, + "loss": 0.0429, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04517853260040283, + "step": 1815, + "valid_targets_mean": 3710.7, + "valid_targets_min": 1403 + }, + { + "epoch": 6.5, + "grad_norm": 0.2801185603695465, + "learning_rate": 8.55425507142108e-06, + "loss": 0.0437, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03667178004980087, + "step": 1820, + "valid_targets_mean": 4392.8, + "valid_targets_min": 1543 + }, + { + "epoch": 6.517857142857143, + "grad_norm": 0.29716054911941264, + "learning_rate": 8.358100089215426e-06, + "loss": 0.0456, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04201003164052963, + "step": 1825, + "valid_targets_mean": 4288.2, + "valid_targets_min": 1890 + }, + { + "epoch": 6.535714285714286, + "grad_norm": 0.3033236749012966, + "learning_rate": 8.164014953408578e-06, + "loss": 0.046, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.046882886439561844, + "step": 1830, + "valid_targets_mean": 4121.1, + "valid_targets_min": 1047 + }, + { + "epoch": 6.553571428571429, + "grad_norm": 0.31238757240923204, + "learning_rate": 7.972009311168882e-06, + "loss": 0.0464, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04627194255590439, + "step": 1835, + "valid_targets_mean": 3875.4, + "valid_targets_min": 1122 + }, + { + "epoch": 6.571428571428571, + "grad_norm": 0.3345411558246883, + "learning_rate": 7.782092706301719e-06, + "loss": 0.0426, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.047288376837968826, + "step": 1840, + "valid_targets_mean": 3608.0, + "valid_targets_min": 998 + }, + { + "epoch": 6.589285714285714, + "grad_norm": 0.27854846628966456, + "learning_rate": 7.5942745787750065e-06, + "loss": 0.0451, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03810355067253113, + "step": 1845, + "valid_targets_mean": 4442.4, + "valid_targets_min": 1489 + }, + { + "epoch": 6.607142857142857, + "grad_norm": 0.2771888164657521, + "learning_rate": 7.4085642642501005e-06, + "loss": 0.0436, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.041286811232566833, + "step": 1850, + "valid_targets_mean": 4297.5, + "valid_targets_min": 1639 + }, + { + "epoch": 6.625, + "grad_norm": 0.31833418128680374, + "learning_rate": 7.224970993617686e-06, + "loss": 0.0444, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04912199452519417, + "step": 1855, + "valid_targets_mean": 3792.6, + "valid_targets_min": 1223 + }, + { + "epoch": 6.642857142857143, + "grad_norm": 0.28500398844467983, + "learning_rate": 7.04350389253895e-06, + "loss": 0.0462, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04187515377998352, + "step": 1860, + "valid_targets_mean": 4411.6, + "valid_targets_min": 982 + }, + { + "epoch": 6.660714285714286, + "grad_norm": 0.3046838227536038, + "learning_rate": 6.864171980991985e-06, + "loss": 0.0447, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04445522651076317, + "step": 1865, + "valid_targets_mean": 3998.7, + "valid_targets_min": 768 + }, + { + "epoch": 6.678571428571429, + "grad_norm": 0.34177412988985256, + "learning_rate": 6.686984172823491e-06, + "loss": 0.0469, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.043909791857004166, + "step": 1870, + "valid_targets_mean": 3358.7, + "valid_targets_min": 1170 + }, + { + "epoch": 6.696428571428571, + "grad_norm": 0.3081403779008394, + "learning_rate": 6.5119492753056565e-06, + "loss": 0.0426, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04129282385110855, + "step": 1875, + "valid_targets_mean": 3783.2, + "valid_targets_min": 755 + }, + { + "epoch": 6.714285714285714, + "grad_norm": 0.29344883551798356, + "learning_rate": 6.339075988698367e-06, + "loss": 0.0419, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03774677962064743, + "step": 1880, + "valid_targets_mean": 3536.6, + "valid_targets_min": 767 + }, + { + "epoch": 6.732142857142857, + "grad_norm": 0.3122233995162009, + "learning_rate": 6.168372905816821e-06, + "loss": 0.0457, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.051206715404987335, + "step": 1885, + "valid_targets_mean": 4062.1, + "valid_targets_min": 1181 + }, + { + "epoch": 6.75, + "grad_norm": 0.3024455602704436, + "learning_rate": 5.9998485116043614e-06, + "loss": 0.0421, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04325060173869133, + "step": 1890, + "valid_targets_mean": 3738.1, + "valid_targets_min": 740 + }, + { + "epoch": 6.767857142857143, + "grad_norm": 0.29202107330476984, + "learning_rate": 5.833511182710716e-06, + "loss": 0.0456, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04033234715461731, + "step": 1895, + "valid_targets_mean": 4103.8, + "valid_targets_min": 706 + }, + { + "epoch": 6.785714285714286, + "grad_norm": 0.3153219707379342, + "learning_rate": 5.6693691870756905e-06, + "loss": 0.0443, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04557320848107338, + "step": 1900, + "valid_targets_mean": 4071.7, + "valid_targets_min": 1457 + }, + { + "epoch": 6.803571428571429, + "grad_norm": 0.3239445898532858, + "learning_rate": 5.507430683518161e-06, + "loss": 0.045, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.048414990305900574, + "step": 1905, + "valid_targets_mean": 3420.9, + "valid_targets_min": 677 + }, + { + "epoch": 6.821428571428571, + "grad_norm": 0.2729590900182919, + "learning_rate": 5.3477037213304995e-06, + "loss": 0.0429, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.041574254631996155, + "step": 1910, + "valid_targets_mean": 4623.2, + "valid_targets_min": 1355 + }, + { + "epoch": 6.839285714285714, + "grad_norm": 0.3166006926315605, + "learning_rate": 5.19019623987857e-06, + "loss": 0.0453, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04982426017522812, + "step": 1915, + "valid_targets_mean": 3700.3, + "valid_targets_min": 1403 + }, + { + "epoch": 6.857142857142857, + "grad_norm": 0.2941772873722246, + "learning_rate": 5.034916068206996e-06, + "loss": 0.0432, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.039322469383478165, + "step": 1920, + "valid_targets_mean": 4301.0, + "valid_targets_min": 1016 + }, + { + "epoch": 6.875, + "grad_norm": 0.31952800546673077, + "learning_rate": 4.881870924650062e-06, + "loss": 0.0438, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.044378332793712616, + "step": 1925, + "valid_targets_mean": 3555.5, + "valid_targets_min": 1126 + }, + { + "epoch": 6.892857142857143, + "grad_norm": 0.28833642877919735, + "learning_rate": 4.731068416448081e-06, + "loss": 0.0422, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04003613442182541, + "step": 1930, + "valid_targets_mean": 4343.4, + "valid_targets_min": 732 + }, + { + "epoch": 6.910714285714286, + "grad_norm": 0.3752187637346029, + "learning_rate": 4.5825160393692445e-06, + "loss": 0.044, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04771614447236061, + "step": 1935, + "valid_targets_mean": 3208.1, + "valid_targets_min": 664 + }, + { + "epoch": 6.928571428571429, + "grad_norm": 0.29643123397187077, + "learning_rate": 4.43622117733703e-06, + "loss": 0.0443, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.040253691375255585, + "step": 1940, + "valid_targets_mean": 3875.8, + "valid_targets_min": 1463 + }, + { + "epoch": 6.946428571428571, + "grad_norm": 0.2981666393648047, + "learning_rate": 4.292191102063192e-06, + "loss": 0.0439, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04336239770054817, + "step": 1945, + "valid_targets_mean": 4253.1, + "valid_targets_min": 1050 + }, + { + "epoch": 6.964285714285714, + "grad_norm": 0.29540049175948035, + "learning_rate": 4.150432972686352e-06, + "loss": 0.0441, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04280786216259003, + "step": 1950, + "valid_targets_mean": 3587.3, + "valid_targets_min": 759 + }, + { + "epoch": 6.982142857142857, + "grad_norm": 0.2670918467783886, + "learning_rate": 4.010953835416037e-06, + "loss": 0.0432, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03771406039595604, + "step": 1955, + "valid_targets_mean": 4613.3, + "valid_targets_min": 718 + }, + { + "epoch": 7.0, + "grad_norm": 0.2476789548915118, + "learning_rate": 3.87376062318257e-06, + "loss": 0.0431, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03568948060274124, + "step": 1960, + "valid_targets_mean": 4422.9, + "valid_targets_min": 1210 + }, + { + "epoch": 7.017857142857143, + "grad_norm": 0.2402449280017575, + "learning_rate": 3.7388601552924062e-06, + "loss": 0.0303, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03295475244522095, + "step": 1965, + "valid_targets_mean": 4295.0, + "valid_targets_min": 1618 + }, + { + "epoch": 7.035714285714286, + "grad_norm": 0.28020921647160824, + "learning_rate": 3.606259137089141e-06, + "loss": 0.0293, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.029505660757422447, + "step": 1970, + "valid_targets_mean": 4480.2, + "valid_targets_min": 1188 + }, + { + "epoch": 7.053571428571429, + "grad_norm": 0.3028387876476253, + "learning_rate": 3.4759641596202762e-06, + "loss": 0.0285, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03014300763607025, + "step": 1975, + "valid_targets_mean": 4058.9, + "valid_targets_min": 1302 + }, + { + "epoch": 7.071428571428571, + "grad_norm": 0.2472396501411902, + "learning_rate": 3.347981699309588e-06, + "loss": 0.0278, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02828243002295494, + "step": 1980, + "valid_targets_mean": 4124.2, + "valid_targets_min": 688 + }, + { + "epoch": 7.089285714285714, + "grad_norm": 0.2703004284444484, + "learning_rate": 3.2223181176351426e-06, + "loss": 0.028, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03063022345304489, + "step": 1985, + "valid_targets_mean": 3820.6, + "valid_targets_min": 1259 + }, + { + "epoch": 7.107142857142857, + "grad_norm": 0.26838235499836266, + "learning_rate": 3.098979660813217e-06, + "loss": 0.0288, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0317617729306221, + "step": 1990, + "valid_targets_mean": 4288.8, + "valid_targets_min": 1471 + }, + { + "epoch": 7.125, + "grad_norm": 0.2377212567699478, + "learning_rate": 2.977972459487738e-06, + "loss": 0.0283, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02683611959218979, + "step": 1995, + "valid_targets_mean": 4518.8, + "valid_targets_min": 1252 + }, + { + "epoch": 7.142857142857143, + "grad_norm": 0.25550182965351664, + "learning_rate": 2.8593025284255614e-06, + "loss": 0.0303, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.030834609642624855, + "step": 2000, + "valid_targets_mean": 4345.4, + "valid_targets_min": 1042 + }, + { + "epoch": 7.160714285714286, + "grad_norm": 0.25811691155876393, + "learning_rate": 2.7429757662175314e-06, + "loss": 0.0287, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.029558485373854637, + "step": 2005, + "valid_targets_mean": 3989.8, + "valid_targets_min": 1573 + }, + { + "epoch": 7.178571428571429, + "grad_norm": 0.23620614147659477, + "learning_rate": 2.6289979549852795e-06, + "loss": 0.0275, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.026751425117254257, + "step": 2010, + "valid_targets_mean": 4712.8, + "valid_targets_min": 1678 + }, + { + "epoch": 7.196428571428571, + "grad_norm": 0.26336723811506235, + "learning_rate": 2.5173747600937993e-06, + "loss": 0.0268, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.028679480776190758, + "step": 2015, + "valid_targets_mean": 3954.4, + "valid_targets_min": 1756 + }, + { + "epoch": 7.214285714285714, + "grad_norm": 0.26252053084850574, + "learning_rate": 2.408111729869844e-06, + "loss": 0.027, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03053813800215721, + "step": 2020, + "valid_targets_mean": 4182.7, + "valid_targets_min": 1391 + }, + { + "epoch": 7.232142857142857, + "grad_norm": 0.26291895983257224, + "learning_rate": 2.3012142953261928e-06, + "loss": 0.0294, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02912769839167595, + "step": 2025, + "valid_targets_mean": 3967.3, + "valid_targets_min": 1153 + }, + { + "epoch": 7.25, + "grad_norm": 0.2707817756634944, + "learning_rate": 2.19668776989162e-06, + "loss": 0.0273, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02793579176068306, + "step": 2030, + "valid_targets_mean": 3973.5, + "valid_targets_min": 1439 + }, + { + "epoch": 7.267857142857143, + "grad_norm": 0.27404498709534575, + "learning_rate": 2.0945373491468466e-06, + "loss": 0.0274, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.030848775058984756, + "step": 2035, + "valid_targets_mean": 3605.4, + "valid_targets_min": 1181 + }, + { + "epoch": 7.285714285714286, + "grad_norm": 0.25644289885954996, + "learning_rate": 1.9947681105662806e-06, + "loss": 0.0283, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.028455138206481934, + "step": 2040, + "valid_targets_mean": 4225.5, + "valid_targets_min": 1019 + }, + { + "epoch": 7.303571428571429, + "grad_norm": 0.24788849511384553, + "learning_rate": 1.8973850132655956e-06, + "loss": 0.0295, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.026823539286851883, + "step": 2045, + "valid_targets_mean": 4545.6, + "valid_targets_min": 654 + }, + { + "epoch": 7.321428571428571, + "grad_norm": 0.23023188044419493, + "learning_rate": 1.8023928977552839e-06, + "loss": 0.027, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.024156294763088226, + "step": 2050, + "valid_targets_mean": 4390.5, + "valid_targets_min": 971 + }, + { + "epoch": 7.339285714285714, + "grad_norm": 0.24487460197524116, + "learning_rate": 1.7097964857000327e-06, + "loss": 0.0261, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.025889787822961807, + "step": 2055, + "valid_targets_mean": 3515.4, + "valid_targets_min": 701 + }, + { + "epoch": 7.357142857142857, + "grad_norm": 0.23295292400779435, + "learning_rate": 1.619600379684022e-06, + "loss": 0.0267, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.026460006833076477, + "step": 2060, + "valid_targets_mean": 4376.2, + "valid_targets_min": 1405 + }, + { + "epoch": 7.375, + "grad_norm": 0.25816462763652187, + "learning_rate": 1.5318090629821757e-06, + "loss": 0.0275, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02663073129951954, + "step": 2065, + "valid_targets_mean": 3985.7, + "valid_targets_min": 785 + }, + { + "epoch": 7.392857142857143, + "grad_norm": 0.26253550806655407, + "learning_rate": 1.4464268993372831e-06, + "loss": 0.0283, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.029157299548387527, + "step": 2070, + "valid_targets_mean": 4350.7, + "valid_targets_min": 979 + }, + { + "epoch": 7.410714285714286, + "grad_norm": 0.2401398112867429, + "learning_rate": 1.3634581327431229e-06, + "loss": 0.0277, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.026041582226753235, + "step": 2075, + "valid_targets_mean": 4322.2, + "valid_targets_min": 539 + }, + { + "epoch": 7.428571428571429, + "grad_norm": 0.23176291437483168, + "learning_rate": 1.282906887233487e-06, + "loss": 0.0282, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.025533093139529228, + "step": 2080, + "valid_targets_mean": 4151.3, + "valid_targets_min": 1338 + }, + { + "epoch": 7.446428571428571, + "grad_norm": 0.2615132783259965, + "learning_rate": 1.2047771666772122e-06, + "loss": 0.0282, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.031607020646333694, + "step": 2085, + "valid_targets_mean": 4072.1, + "valid_targets_min": 868 + }, + { + "epoch": 7.464285714285714, + "grad_norm": 0.2832672289481666, + "learning_rate": 1.1290728545791773e-06, + "loss": 0.0285, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.031227465718984604, + "step": 2090, + "valid_targets_mean": 3998.8, + "valid_targets_min": 1453 + }, + { + "epoch": 7.482142857142857, + "grad_norm": 0.26075429365735775, + "learning_rate": 1.0557977138872132e-06, + "loss": 0.0281, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0277554951608181, + "step": 2095, + "valid_targets_mean": 4246.7, + "valid_targets_min": 794 + }, + { + "epoch": 7.5, + "grad_norm": 0.24886649010465955, + "learning_rate": 9.849553868051364e-07, + "loss": 0.0292, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.029004719108343124, + "step": 2100, + "valid_targets_mean": 4587.9, + "valid_targets_min": 1423 + }, + { + "epoch": 7.517857142857143, + "grad_norm": 0.24636360160608198, + "learning_rate": 9.165493946116432e-07, + "loss": 0.0264, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02765900269150734, + "step": 2105, + "valid_targets_mean": 4292.6, + "valid_targets_min": 647 + }, + { + "epoch": 7.535714285714286, + "grad_norm": 0.238511909526691, + "learning_rate": 8.505831374853279e-07, + "loss": 0.0273, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.026020430028438568, + "step": 2110, + "valid_targets_mean": 4522.2, + "valid_targets_min": 1432 + }, + { + "epoch": 7.553571428571429, + "grad_norm": 0.2679364234988592, + "learning_rate": 7.870598943356622e-07, + "loss": 0.0284, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.029078487306833267, + "step": 2115, + "valid_targets_mean": 3563.8, + "valid_targets_min": 622 + }, + { + "epoch": 7.571428571428571, + "grad_norm": 0.2273914222608692, + "learning_rate": 7.259828226399978e-07, + "loss": 0.0269, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.023035066202282906, + "step": 2120, + "valid_targets_mean": 4171.8, + "valid_targets_min": 748 + }, + { + "epoch": 7.589285714285714, + "grad_norm": 0.2790259964939605, + "learning_rate": 6.673549582866367e-07, + "loss": 0.0272, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03183634206652641, + "step": 2125, + "valid_targets_mean": 3474.6, + "valid_targets_min": 744 + }, + { + "epoch": 7.607142857142857, + "grad_norm": 0.28110522530065396, + "learning_rate": 6.111792154239404e-07, + "loss": 0.0265, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.026433158665895462, + "step": 2130, + "valid_targets_mean": 4318.2, + "valid_targets_min": 1460 + }, + { + "epoch": 7.625, + "grad_norm": 0.2918742405825761, + "learning_rate": 5.574583863154403e-07, + "loss": 0.03, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.029343387112021446, + "step": 2135, + "valid_targets_mean": 4260.1, + "valid_targets_min": 526 + }, + { + "epoch": 7.642857142857143, + "grad_norm": 0.23961087496613925, + "learning_rate": 5.061951412010879e-07, + "loss": 0.0271, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02575666829943657, + "step": 2140, + "valid_targets_mean": 4550.4, + "valid_targets_min": 1612 + }, + { + "epoch": 7.660714285714286, + "grad_norm": 0.29059265527695227, + "learning_rate": 4.573920281645161e-07, + "loss": 0.0281, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.030294449999928474, + "step": 2145, + "valid_targets_mean": 3374.5, + "valid_targets_min": 748 + }, + { + "epoch": 7.678571428571429, + "grad_norm": 0.2314680079446443, + "learning_rate": 4.1105147300636303e-07, + "loss": 0.0261, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.023414522409439087, + "step": 2150, + "valid_targets_mean": 4249.2, + "valid_targets_min": 1124 + }, + { + "epoch": 7.696428571428571, + "grad_norm": 0.27967946639968444, + "learning_rate": 3.67175779123724e-07, + "loss": 0.0275, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.027103278785943985, + "step": 2155, + "valid_targets_mean": 3515.4, + "valid_targets_min": 1403 + }, + { + "epoch": 7.714285714285714, + "grad_norm": 0.25318541674090805, + "learning_rate": 3.2576712739565416e-07, + "loss": 0.0273, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.027268027886748314, + "step": 2160, + "valid_targets_mean": 4233.0, + "valid_targets_min": 1513 + }, + { + "epoch": 7.732142857142857, + "grad_norm": 0.2457910859704936, + "learning_rate": 2.8682757607474407e-07, + "loss": 0.0278, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.024341989308595657, + "step": 2165, + "valid_targets_mean": 3975.7, + "valid_targets_min": 1438 + }, + { + "epoch": 7.75, + "grad_norm": 0.2428621631422345, + "learning_rate": 2.5035906068482937e-07, + "loss": 0.0285, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.023659780621528625, + "step": 2170, + "valid_targets_mean": 3809.1, + "valid_targets_min": 910 + }, + { + "epoch": 7.767857142857143, + "grad_norm": 0.29480998073755216, + "learning_rate": 2.1636339392479553e-07, + "loss": 0.0312, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03240353614091873, + "step": 2175, + "valid_targets_mean": 3849.1, + "valid_targets_min": 711 + }, + { + "epoch": 7.785714285714286, + "grad_norm": 0.2868068952090629, + "learning_rate": 1.848422655784554e-07, + "loss": 0.0273, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.028937779366970062, + "step": 2180, + "valid_targets_mean": 3705.9, + "valid_targets_min": 601 + }, + { + "epoch": 7.803571428571429, + "grad_norm": 0.22954205773896127, + "learning_rate": 1.557972424305665e-07, + "loss": 0.0266, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.023847095668315887, + "step": 2185, + "valid_targets_mean": 4292.6, + "valid_targets_min": 1489 + }, + { + "epoch": 7.821428571428571, + "grad_norm": 0.2862732642090174, + "learning_rate": 1.2922976818894317e-07, + "loss": 0.0274, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.030526213347911835, + "step": 2190, + "valid_targets_mean": 4125.9, + "valid_targets_min": 1562 + }, + { + "epoch": 7.839285714285714, + "grad_norm": 0.23121469380521936, + "learning_rate": 1.0514116341271418e-07, + "loss": 0.0275, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02458646520972252, + "step": 2195, + "valid_targets_mean": 4182.2, + "valid_targets_min": 672 + }, + { + "epoch": 7.857142857142857, + "grad_norm": 0.2822696277632301, + "learning_rate": 8.353262544666951e-08, + "loss": 0.0286, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.032052166759967804, + "step": 2200, + "valid_targets_mean": 3759.6, + "valid_targets_min": 1468 + }, + { + "epoch": 7.875, + "grad_norm": 0.2544669964532589, + "learning_rate": 6.440522836174135e-08, + "loss": 0.0269, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.026916902512311935, + "step": 2205, + "valid_targets_mean": 3926.3, + "valid_targets_min": 1710 + }, + { + "epoch": 7.892857142857143, + "grad_norm": 0.22879593211883528, + "learning_rate": 4.775992290163567e-08, + "loss": 0.0263, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.023991543799638748, + "step": 2210, + "valid_targets_mean": 4248.2, + "valid_targets_min": 824 + }, + { + "epoch": 7.910714285714286, + "grad_norm": 0.20783417458487638, + "learning_rate": 3.359753643555341e-08, + "loss": 0.0251, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02183828502893448, + "step": 2215, + "valid_targets_mean": 4822.1, + "valid_targets_min": 1783 + }, + { + "epoch": 7.928571428571429, + "grad_norm": 0.23350831188752086, + "learning_rate": 2.191877291707889e-08, + "loss": 0.0277, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.024739433079957962, + "step": 2220, + "valid_targets_mean": 4504.9, + "valid_targets_min": 717 + }, + { + "epoch": 7.946428571428571, + "grad_norm": 0.2855603160247391, + "learning_rate": 1.2724212849180017e-08, + "loss": 0.0302, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0332905612885952, + "step": 2225, + "valid_targets_mean": 3765.7, + "valid_targets_min": 1212 + }, + { + "epoch": 7.964285714285714, + "grad_norm": 0.25970453601157745, + "learning_rate": 6.0143132553591716e-09, + "loss": 0.0269, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02955716848373413, + "step": 2230, + "valid_targets_mean": 3971.6, + "valid_targets_min": 1354 + }, + { + "epoch": 7.982142857142857, + "grad_norm": 0.26079981089905174, + "learning_rate": 1.7894076569435757e-09, + "loss": 0.0279, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03021353855729103, + "step": 2235, + "valid_targets_mean": 4439.9, + "valid_targets_min": 732 + }, + { + "epoch": 8.0, + "grad_norm": 0.231373875163851, + "learning_rate": 4.970605649301519e-11, + "loss": 0.0257, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02538919635117054, + "step": 2240, + "valid_targets_mean": 3632.8, + "valid_targets_min": 732 + }, + { + "epoch": 8.0, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02538919635117054, + "step": 2240, + "total_flos": 1518927301312512.0, + "train_loss": 0.13197601472825876, + "train_runtime": 38306.9648, + "train_samples_per_second": 1.867, + "train_steps_per_second": 0.058, + "valid_targets_mean": 3632.8, + "valid_targets_min": 732 + } + ], + "logging_steps": 5, + "max_steps": 2240, + "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": 1518927301312512.0, + "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..cc7778a --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7a01b75a5fb7adfd7a2e98bffd02b1a47258eac70fda90d732a60dbeb169ba57 +size 8913 diff --git a/training_loss.png b/training_loss.png new file mode 100644 index 0000000..43c1213 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