commit ccc206d4be228e4f1c5956629626e795178f1ff1 Author: ModelHub XC Date: Thu Aug 20 16:21:13 2026 +0800 初始化项目,由ModelHub XC社区提供模型 Model: laion/Qwen3-8B_exp-swd-swesmith-wo-docker_glm_4.7_traces_locetash_save-strategy_steps Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..9feea38 --- /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-00004-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +training_args.bin filter=lfs diff=lfs merge=lfs -text +model-00003-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +vocab.json filter=lfs diff=lfs merge=lfs -text +model-00001-of-00004.safetensors filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text +model-00002-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..d4bbd88 --- /dev/null +++ b/README.md @@ -0,0 +1,60 @@ +--- +library_name: transformers +license: apache-2.0 +base_model: Qwen/Qwen3-8B +tags: +- llama-factory +- full +- generated_from_trainer +model-index: +- name: Qwen3-8B_exp-swd-swesmith-wo-docker_glm_4.7_traces_locetash_save-strategy_steps + results: [] +--- + + + +# Qwen3-8B_exp-swd-swesmith-wo-docker_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/exp-swd-swesmith-wo-docker_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..bc05a6c --- /dev/null +++ b/all_results.json @@ -0,0 +1,16 @@ +{ + "achieved_tflops_per_gpu": 0.0020004562843468124, + "achieved_tflops_per_gpu_theoretical": 189.7226555180974, + "epoch": 8.0, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.012045512907207012, + "mfu_percent": 0.0006411718860085937, + "mfu_percent_theoretical": 60.80854343528763, + "total_flos": 2887866713112576.0, + "train_loss": 0.1099344994546549, + "train_runtime": 45112.6253, + "train_samples_per_second": 1.778, + "train_steps_per_second": 0.056, + "valid_targets_mean": 4840.4, + "valid_targets_min": 726 +} \ 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..1d85e43 --- /dev/null +++ b/model-00001-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:857bfb50d3549ce7920227c4a62037a86558f82393b94fe7a56e84a5478aaed8 +size 4902257696 diff --git a/model-00002-of-00004.safetensors b/model-00002-of-00004.safetensors new file mode 100644 index 0000000..14d1b90 --- /dev/null +++ b/model-00002-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:de34de5bad264afe70b6e3412b1bb760cacd14f2e042cc8b2c3d48bc5d98e55b +size 4915960368 diff --git a/model-00003-of-00004.safetensors b/model-00003-of-00004.safetensors new file mode 100644 index 0000000..4e656b6 --- /dev/null +++ b/model-00003-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e1ac833128d148de5ad665ac657537ae733db97035a2317b3215a771d4848add +size 4983068496 diff --git a/model-00004-of-00004.safetensors b/model-00004-of-00004.safetensors new file mode 100644 index 0000000..7cec239 --- /dev/null +++ b/model-00004-of-00004.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9b0b498ef094333d3aa77d14599bdf14bb903ddf1abcd47a9fe96b3ae1058ab3 +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..953f47e --- /dev/null +++ b/run_summary.json @@ -0,0 +1,12 @@ +{ + "agent_name": null, + "training_start": null, + "training_end": null, + "created_by": "DCAgent", + "base_model_name": "Qwen/Qwen3-8B", + "dataset_name": "DCAgent/exp-swd-swesmith-wo-docker_glm_4.7_traces_locetash", + "training_type": "SFT", + "training_parameters": "https://huggingface.co/laion/Qwen3-8B_exp-swd-swesmith-wo-docker_glm_4.7_traces_locetash_save-strategy_steps/blob/main/config.json", + "wandb_link": "https://wandb.ai/dogml/dc-agent/runs/Qwen3-8B_exp-swd-swesmith-wo-docker_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..bc05a6c --- /dev/null +++ b/train_results.json @@ -0,0 +1,16 @@ +{ + "achieved_tflops_per_gpu": 0.0020004562843468124, + "achieved_tflops_per_gpu_theoretical": 189.7226555180974, + "epoch": 8.0, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.012045512907207012, + "mfu_percent": 0.0006411718860085937, + "mfu_percent_theoretical": 60.80854343528763, + "total_flos": 2887866713112576.0, + "train_loss": 0.1099344994546549, + "train_runtime": 45112.6253, + "train_samples_per_second": 1.778, + "train_steps_per_second": 0.056, + "valid_targets_mean": 4840.4, + "valid_targets_min": 726 +} \ No newline at end of file diff --git a/trainer_log.jsonl b/trainer_log.jsonl new file mode 100644 index 0000000..0b64ec0 --- /dev/null +++ b/trainer_log.jsonl @@ -0,0 +1,503 @@ +{"current_steps": 5, "total_steps": 2512, "loss": 0.6402, "lr": 3.0769230769230774e-05, "epoch": 0.01592356687898089, "percentage": 0.2, "elapsed_time": "0:01:59", "remaining_time": "16:38:23"} +{"current_steps": 10, "total_steps": 2512, "loss": 0.4346, "lr": 6.923076923076924e-05, "epoch": 0.03184713375796178, "percentage": 0.4, "elapsed_time": "0:03:14", "remaining_time": "13:31:34"} +{"current_steps": 15, "total_steps": 2512, "loss": 0.3816, "lr": 9.999996048998592e-05, "epoch": 0.04777070063694268, "percentage": 0.6, "elapsed_time": "0:04:27", "remaining_time": "12:21:34"} +{"current_steps": 20, "total_steps": 2512, "loss": 0.3323, "lr": 9.999857764604897e-05, "epoch": 0.06369426751592357, "percentage": 0.8, "elapsed_time": "0:05:38", "remaining_time": "11:42:13"} +{"current_steps": 25, "total_steps": 2512, "loss": 0.3195, "lr": 9.999521936384848e-05, "epoch": 0.07961783439490445, "percentage": 1.0, "elapsed_time": "0:06:50", "remaining_time": "11:21:22"} +{"current_steps": 30, "total_steps": 2512, "loss": 0.3016, "lr": 9.998988577606981e-05, "epoch": 0.09554140127388536, "percentage": 1.19, "elapsed_time": "0:08:00", "remaining_time": "11:02:26"} +{"current_steps": 35, "total_steps": 2512, "loss": 0.2895, "lr": 9.998257709344246e-05, "epoch": 0.11146496815286625, "percentage": 1.39, "elapsed_time": "0:09:10", "remaining_time": "10:49:35"} +{"current_steps": 40, "total_steps": 2512, "loss": 0.2816, "lr": 9.99732936047316e-05, "epoch": 0.12738853503184713, "percentage": 1.59, "elapsed_time": "0:10:20", "remaining_time": "10:39:35"} +{"current_steps": 45, "total_steps": 2512, "loss": 0.2908, "lr": 9.99620356767269e-05, "epoch": 0.14331210191082802, "percentage": 1.79, "elapsed_time": "0:11:27", "remaining_time": "10:28:32"} +{"current_steps": 50, "total_steps": 2512, "loss": 0.2846, "lr": 9.994880375422783e-05, "epoch": 0.1592356687898089, "percentage": 1.99, "elapsed_time": "0:12:34", "remaining_time": "10:19:28"} +{"current_steps": 55, "total_steps": 2512, "loss": 0.2811, "lr": 9.993359836002618e-05, "epoch": 0.1751592356687898, "percentage": 2.19, "elapsed_time": "0:13:46", "remaining_time": "10:15:05"} +{"current_steps": 60, "total_steps": 2512, "loss": 0.277, "lr": 9.99164200948854e-05, "epoch": 0.1910828025477707, "percentage": 2.39, "elapsed_time": "0:14:56", "remaining_time": "10:10:34"} +{"current_steps": 65, "total_steps": 2512, "loss": 0.2818, "lr": 9.989726963751682e-05, "epoch": 0.2070063694267516, "percentage": 2.59, "elapsed_time": "0:16:06", "remaining_time": "10:06:29"} +{"current_steps": 70, "total_steps": 2512, "loss": 0.2709, "lr": 9.987614774455291e-05, "epoch": 0.2229299363057325, "percentage": 2.79, "elapsed_time": "0:17:18", "remaining_time": "10:03:33"} +{"current_steps": 75, "total_steps": 2512, "loss": 0.2791, "lr": 9.985305525051732e-05, "epoch": 0.23885350318471338, "percentage": 2.99, "elapsed_time": "0:18:30", "remaining_time": "10:01:16"} +{"current_steps": 80, "total_steps": 2512, "loss": 0.2675, "lr": 9.98279930677919e-05, "epoch": 0.25477707006369427, "percentage": 3.18, "elapsed_time": "0:19:38", "remaining_time": "9:57:04"} +{"current_steps": 85, "total_steps": 2512, "loss": 0.2702, "lr": 9.980096218658075e-05, "epoch": 0.27070063694267515, "percentage": 3.38, "elapsed_time": "0:20:50", "remaining_time": "9:55:07"} +{"current_steps": 90, "total_steps": 2512, "loss": 0.2748, "lr": 9.977196367487099e-05, "epoch": 0.28662420382165604, "percentage": 3.58, "elapsed_time": "0:21:59", "remaining_time": "9:51:37"} +{"current_steps": 95, "total_steps": 2512, "loss": 0.2663, "lr": 9.974099867839057e-05, "epoch": 0.30254777070063693, "percentage": 3.78, "elapsed_time": "0:23:10", "remaining_time": "9:49:45"} +{"current_steps": 100, "total_steps": 2512, "loss": 0.2707, "lr": 9.970806842056311e-05, "epoch": 0.3184713375796178, "percentage": 3.98, "elapsed_time": "0:24:20", "remaining_time": "9:47:02"} +{"current_steps": 105, "total_steps": 2512, "loss": 0.2688, "lr": 9.967317420245943e-05, "epoch": 0.3343949044585987, "percentage": 4.18, "elapsed_time": "0:44:46", "remaining_time": "17:06:31"} +{"current_steps": 110, "total_steps": 2512, "loss": 0.265, "lr": 9.963631740274622e-05, "epoch": 0.3503184713375796, "percentage": 4.38, "elapsed_time": "0:45:57", "remaining_time": "16:43:32"} +{"current_steps": 115, "total_steps": 2512, "loss": 0.2675, "lr": 9.959749947763158e-05, "epoch": 0.3662420382165605, "percentage": 4.58, "elapsed_time": "0:47:06", "remaining_time": "16:21:58"} +{"current_steps": 120, "total_steps": 2512, "loss": 0.2564, "lr": 9.955672196080739e-05, "epoch": 0.3821656050955414, "percentage": 4.78, "elapsed_time": "0:48:16", "remaining_time": "16:02:21"} +{"current_steps": 125, "total_steps": 2512, "loss": 0.2675, "lr": 9.951398646338883e-05, "epoch": 0.3980891719745223, "percentage": 4.98, "elapsed_time": "0:49:28", "remaining_time": "15:44:53"} +{"current_steps": 130, "total_steps": 2512, "loss": 0.248, "lr": 9.946929467385069e-05, "epoch": 0.4140127388535032, "percentage": 5.18, "elapsed_time": "0:50:36", "remaining_time": "15:27:17"} +{"current_steps": 135, "total_steps": 2512, "loss": 0.2656, "lr": 9.94226483579606e-05, "epoch": 0.4299363057324841, "percentage": 5.37, "elapsed_time": "0:51:45", "remaining_time": "15:11:24"} +{"current_steps": 140, "total_steps": 2512, "loss": 0.2458, "lr": 9.937404935870938e-05, "epoch": 0.445859872611465, "percentage": 5.57, "elapsed_time": "0:52:52", "remaining_time": "14:55:59"} +{"current_steps": 145, "total_steps": 2512, "loss": 0.2523, "lr": 9.932349959623803e-05, "epoch": 0.46178343949044587, "percentage": 5.77, "elapsed_time": "0:54:03", "remaining_time": "14:42:27"} +{"current_steps": 150, "total_steps": 2512, "loss": 0.2608, "lr": 9.927100106776212e-05, "epoch": 0.47770700636942676, "percentage": 5.97, "elapsed_time": "0:55:10", "remaining_time": "14:28:53"} +{"current_steps": 155, "total_steps": 2512, "loss": 0.265, "lr": 9.92165558474927e-05, "epoch": 0.49363057324840764, "percentage": 6.17, "elapsed_time": "0:56:19", "remaining_time": "14:16:34"} +{"current_steps": 160, "total_steps": 2512, "loss": 0.2487, "lr": 9.916016608655435e-05, "epoch": 0.5095541401273885, "percentage": 6.37, "elapsed_time": "0:57:25", "remaining_time": "14:04:04"} +{"current_steps": 165, "total_steps": 2512, "loss": 0.2566, "lr": 9.91018340129003e-05, "epoch": 0.5254777070063694, "percentage": 6.57, "elapsed_time": "0:58:34", "remaining_time": "13:53:16"} +{"current_steps": 170, "total_steps": 2512, "loss": 0.2502, "lr": 9.904156193122432e-05, "epoch": 0.5414012738853503, "percentage": 6.77, "elapsed_time": "0:59:43", "remaining_time": "13:42:52"} +{"current_steps": 175, "total_steps": 2512, "loss": 0.2647, "lr": 9.897935222286967e-05, "epoch": 0.5573248407643312, "percentage": 6.97, "elapsed_time": "1:00:51", "remaining_time": "13:32:46"} +{"current_steps": 180, "total_steps": 2512, "loss": 0.2599, "lr": 9.891520734573505e-05, "epoch": 0.5732484076433121, "percentage": 7.17, "elapsed_time": "1:02:03", "remaining_time": "13:23:55"} +{"current_steps": 185, "total_steps": 2512, "loss": 0.2604, "lr": 9.884912983417743e-05, "epoch": 0.589171974522293, "percentage": 7.36, "elapsed_time": "1:03:12", "remaining_time": "13:15:00"} +{"current_steps": 190, "total_steps": 2512, "loss": 0.2449, "lr": 9.878112229891199e-05, "epoch": 0.6050955414012739, "percentage": 7.56, "elapsed_time": "1:04:20", "remaining_time": "13:06:15"} +{"current_steps": 195, "total_steps": 2512, "loss": 0.249, "lr": 9.87111874269089e-05, "epoch": 0.6210191082802548, "percentage": 7.76, "elapsed_time": "1:05:28", "remaining_time": "12:58:03"} +{"current_steps": 200, "total_steps": 2512, "loss": 0.2574, "lr": 9.86393279812872e-05, "epoch": 0.6369426751592356, "percentage": 7.96, "elapsed_time": "1:06:37", "remaining_time": "12:50:08"} +{"current_steps": 205, "total_steps": 2512, "loss": 0.2595, "lr": 9.856554680120565e-05, "epoch": 0.6528662420382165, "percentage": 8.16, "elapsed_time": "1:19:52", "remaining_time": "14:58:57"} +{"current_steps": 210, "total_steps": 2512, "loss": 0.2508, "lr": 9.848984680175049e-05, "epoch": 0.6687898089171974, "percentage": 8.36, "elapsed_time": "1:21:05", "remaining_time": "14:48:54"} +{"current_steps": 215, "total_steps": 2512, "loss": 0.2553, "lr": 9.841223097382029e-05, "epoch": 0.6847133757961783, "percentage": 8.56, "elapsed_time": "1:22:17", "remaining_time": "14:39:08"} +{"current_steps": 220, "total_steps": 2512, "loss": 0.2476, "lr": 9.833270238400783e-05, "epoch": 0.7006369426751592, "percentage": 8.76, "elapsed_time": "1:23:28", "remaining_time": "14:29:43"} +{"current_steps": 225, "total_steps": 2512, "loss": 0.2518, "lr": 9.825126417447889e-05, "epoch": 0.7165605095541401, "percentage": 8.96, "elapsed_time": "1:24:37", "remaining_time": "14:20:11"} +{"current_steps": 230, "total_steps": 2512, "loss": 0.2501, "lr": 9.816791956284809e-05, "epoch": 0.732484076433121, "percentage": 9.16, "elapsed_time": "1:25:48", "remaining_time": "14:11:26"} +{"current_steps": 235, "total_steps": 2512, "loss": 0.2544, "lr": 9.808267184205183e-05, "epoch": 0.7484076433121019, "percentage": 9.36, "elapsed_time": "1:26:55", "remaining_time": "14:02:10"} +{"current_steps": 240, "total_steps": 2512, "loss": 0.2522, "lr": 9.799552438021807e-05, "epoch": 0.7643312101910829, "percentage": 9.55, "elapsed_time": "1:28:07", "remaining_time": "13:54:13"} +{"current_steps": 245, "total_steps": 2512, "loss": 0.2466, "lr": 9.79064806205334e-05, "epoch": 0.7802547770700637, "percentage": 9.75, "elapsed_time": "1:29:19", "remaining_time": "13:46:35"} +{"current_steps": 250, "total_steps": 2512, "loss": 0.2536, "lr": 9.781554408110691e-05, "epoch": 0.7961783439490446, "percentage": 9.95, "elapsed_time": "1:30:25", "remaining_time": "13:38:06"} +{"current_steps": 255, "total_steps": 2512, "loss": 0.2454, "lr": 9.772271835483118e-05, "epoch": 0.8121019108280255, "percentage": 10.15, "elapsed_time": "1:31:34", "remaining_time": "13:30:32"} +{"current_steps": 260, "total_steps": 2512, "loss": 0.2449, "lr": 9.762800710924038e-05, "epoch": 0.8280254777070064, "percentage": 10.35, "elapsed_time": "1:32:42", "remaining_time": "13:22:56"} +{"current_steps": 265, "total_steps": 2512, "loss": 0.2498, "lr": 9.753141408636532e-05, "epoch": 0.8439490445859873, "percentage": 10.55, "elapsed_time": "1:33:53", "remaining_time": "13:16:03"} +{"current_steps": 270, "total_steps": 2512, "loss": 0.2609, "lr": 9.743294310258564e-05, "epoch": 0.8598726114649682, "percentage": 10.75, "elapsed_time": "1:35:03", "remaining_time": "13:09:16"} +{"current_steps": 275, "total_steps": 2512, "loss": 0.2395, "lr": 9.733259804847901e-05, "epoch": 0.8757961783439491, "percentage": 10.95, "elapsed_time": "1:36:16", "remaining_time": "13:03:05"} +{"current_steps": 280, "total_steps": 2512, "loss": 0.2567, "lr": 9.723038288866738e-05, "epoch": 0.89171974522293, "percentage": 11.15, "elapsed_time": "1:37:26", "remaining_time": "12:56:47"} +{"current_steps": 285, "total_steps": 2512, "loss": 0.2417, "lr": 9.712630166166039e-05, "epoch": 0.9076433121019108, "percentage": 11.35, "elapsed_time": "1:38:36", "remaining_time": "12:50:31"} +{"current_steps": 290, "total_steps": 2512, "loss": 0.2544, "lr": 9.70203584796958e-05, "epoch": 0.9235668789808917, "percentage": 11.54, "elapsed_time": "1:39:48", "remaining_time": "12:44:45"} +{"current_steps": 295, "total_steps": 2512, "loss": 0.2438, "lr": 9.6912557528577e-05, "epoch": 0.9394904458598726, "percentage": 11.74, "elapsed_time": "1:40:56", "remaining_time": "12:38:37"} +{"current_steps": 300, "total_steps": 2512, "loss": 0.2487, "lr": 9.680290306750761e-05, "epoch": 0.9554140127388535, "percentage": 11.94, "elapsed_time": "1:42:03", "remaining_time": "12:32:30"} +{"current_steps": 305, "total_steps": 2512, "loss": 0.242, "lr": 9.669139942892323e-05, "epoch": 0.9713375796178344, "percentage": 12.14, "elapsed_time": "1:52:40", "remaining_time": "13:35:17"} +{"current_steps": 310, "total_steps": 2512, "loss": 0.2385, "lr": 9.657805101832029e-05, "epoch": 0.9872611464968153, "percentage": 12.34, "elapsed_time": "1:53:50", "remaining_time": "13:28:39"} +{"current_steps": 315, "total_steps": 2512, "loss": 0.2349, "lr": 9.646286231408194e-05, "epoch": 1.0031847133757963, "percentage": 12.54, "elapsed_time": "1:55:01", "remaining_time": "13:22:18"} +{"current_steps": 320, "total_steps": 2512, "loss": 0.2006, "lr": 9.63458378673011e-05, "epoch": 1.019108280254777, "percentage": 12.74, "elapsed_time": "1:56:09", "remaining_time": "13:15:42"} +{"current_steps": 325, "total_steps": 2512, "loss": 0.2014, "lr": 9.62269823016007e-05, "epoch": 1.035031847133758, "percentage": 12.94, "elapsed_time": "1:57:17", "remaining_time": "13:09:14"} +{"current_steps": 330, "total_steps": 2512, "loss": 0.2018, "lr": 9.610630031295098e-05, "epoch": 1.0509554140127388, "percentage": 13.14, "elapsed_time": "1:58:26", "remaining_time": "13:03:06"} +{"current_steps": 335, "total_steps": 2512, "loss": 0.2066, "lr": 9.598379666948394e-05, "epoch": 1.0668789808917198, "percentage": 13.34, "elapsed_time": "1:59:34", "remaining_time": "12:57:01"} +{"current_steps": 340, "total_steps": 2512, "loss": 0.1968, "lr": 9.585947621130496e-05, "epoch": 1.0828025477707006, "percentage": 13.54, "elapsed_time": "2:00:45", "remaining_time": "12:51:24"} +{"current_steps": 345, "total_steps": 2512, "loss": 0.1962, "lr": 9.573334385030157e-05, "epoch": 1.0987261146496816, "percentage": 13.73, "elapsed_time": "2:01:50", "remaining_time": "12:45:21"} +{"current_steps": 350, "total_steps": 2512, "loss": 0.2086, "lr": 9.56054045699494e-05, "epoch": 1.1146496815286624, "percentage": 13.93, "elapsed_time": "2:03:01", "remaining_time": "12:39:57"} +{"current_steps": 355, "total_steps": 2512, "loss": 0.2018, "lr": 9.547566342511521e-05, "epoch": 1.1305732484076434, "percentage": 14.13, "elapsed_time": "2:04:11", "remaining_time": "12:34:33"} +{"current_steps": 360, "total_steps": 2512, "loss": 0.195, "lr": 9.534412554185725e-05, "epoch": 1.1464968152866242, "percentage": 14.33, "elapsed_time": "2:05:20", "remaining_time": "12:29:14"} +{"current_steps": 365, "total_steps": 2512, "loss": 0.2032, "lr": 9.521079611722277e-05, "epoch": 1.1624203821656052, "percentage": 14.53, "elapsed_time": "2:06:28", "remaining_time": "12:23:58"} +{"current_steps": 370, "total_steps": 2512, "loss": 0.2185, "lr": 9.507568041904251e-05, "epoch": 1.178343949044586, "percentage": 14.73, "elapsed_time": "2:07:36", "remaining_time": "12:18:46"} +{"current_steps": 375, "total_steps": 2512, "loss": 0.2057, "lr": 9.493878378572276e-05, "epoch": 1.194267515923567, "percentage": 14.93, "elapsed_time": "2:08:46", "remaining_time": "12:13:52"} +{"current_steps": 380, "total_steps": 2512, "loss": 0.2051, "lr": 9.480011162603434e-05, "epoch": 1.2101910828025477, "percentage": 15.13, "elapsed_time": "2:09:58", "remaining_time": "12:09:11"} +{"current_steps": 385, "total_steps": 2512, "loss": 0.2024, "lr": 9.465966941889892e-05, "epoch": 1.2261146496815287, "percentage": 15.33, "elapsed_time": "2:11:06", "remaining_time": "12:04:17"} +{"current_steps": 390, "total_steps": 2512, "loss": 0.2004, "lr": 9.451746271317252e-05, "epoch": 1.2420382165605095, "percentage": 15.53, "elapsed_time": "2:12:19", "remaining_time": "11:59:56"} +{"current_steps": 395, "total_steps": 2512, "loss": 0.216, "lr": 9.437349712742634e-05, "epoch": 1.2579617834394905, "percentage": 15.72, "elapsed_time": "2:13:30", "remaining_time": "11:55:30"} +{"current_steps": 400, "total_steps": 2512, "loss": 0.2096, "lr": 9.422777834972474e-05, "epoch": 1.2738853503184713, "percentage": 15.92, "elapsed_time": "2:14:41", "remaining_time": "11:51:08"} +{"current_steps": 405, "total_steps": 2512, "loss": 0.2018, "lr": 9.408031213740045e-05, "epoch": 1.2898089171974523, "percentage": 16.12, "elapsed_time": "2:22:27", "remaining_time": "12:21:09"} +{"current_steps": 410, "total_steps": 2512, "loss": 0.2012, "lr": 9.393110431682721e-05, "epoch": 1.305732484076433, "percentage": 16.32, "elapsed_time": "2:23:38", "remaining_time": "12:16:23"} +{"current_steps": 415, "total_steps": 2512, "loss": 0.2078, "lr": 9.378016078318947e-05, "epoch": 1.321656050955414, "percentage": 16.52, "elapsed_time": "2:24:45", "remaining_time": "12:11:25"} +{"current_steps": 420, "total_steps": 2512, "loss": 0.2043, "lr": 9.362748750024954e-05, "epoch": 1.3375796178343948, "percentage": 16.72, "elapsed_time": "2:25:54", "remaining_time": "12:06:44"} +{"current_steps": 425, "total_steps": 2512, "loss": 0.2016, "lr": 9.347309050011187e-05, "epoch": 1.3535031847133758, "percentage": 16.92, "elapsed_time": "2:26:59", "remaining_time": "12:01:50"} +{"current_steps": 430, "total_steps": 2512, "loss": 0.2087, "lr": 9.331697588298487e-05, "epoch": 1.3694267515923566, "percentage": 17.12, "elapsed_time": "2:28:07", "remaining_time": "11:57:12"} +{"current_steps": 435, "total_steps": 2512, "loss": 0.2112, "lr": 9.315914981693974e-05, "epoch": 1.3853503184713376, "percentage": 17.32, "elapsed_time": "2:29:18", "remaining_time": "11:52:52"} +{"current_steps": 440, "total_steps": 2512, "loss": 0.2035, "lr": 9.29996185376669e-05, "epoch": 1.4012738853503186, "percentage": 17.52, "elapsed_time": "2:30:28", "remaining_time": "11:48:37"} +{"current_steps": 445, "total_steps": 2512, "loss": 0.1951, "lr": 9.283838834822948e-05, "epoch": 1.4171974522292994, "percentage": 17.71, "elapsed_time": "2:31:37", "remaining_time": "11:44:15"} +{"current_steps": 450, "total_steps": 2512, "loss": 0.2064, "lr": 9.267546561881442e-05, "epoch": 1.4331210191082802, "percentage": 17.91, "elapsed_time": "2:32:47", "remaining_time": "11:40:07"} +{"current_steps": 455, "total_steps": 2512, "loss": 0.2103, "lr": 9.251085678648072e-05, "epoch": 1.4490445859872612, "percentage": 18.11, "elapsed_time": "2:33:59", "remaining_time": "11:36:12"} +{"current_steps": 460, "total_steps": 2512, "loss": 0.2054, "lr": 9.234456835490509e-05, "epoch": 1.4649681528662422, "percentage": 18.31, "elapsed_time": "2:35:09", "remaining_time": "11:32:06"} +{"current_steps": 465, "total_steps": 2512, "loss": 0.2003, "lr": 9.217660689412504e-05, "epoch": 1.480891719745223, "percentage": 18.51, "elapsed_time": "2:36:20", "remaining_time": "11:28:15"} +{"current_steps": 470, "total_steps": 2512, "loss": 0.197, "lr": 9.200697904027928e-05, "epoch": 1.4968152866242037, "percentage": 18.71, "elapsed_time": "2:37:29", "remaining_time": "11:24:15"} +{"current_steps": 475, "total_steps": 2512, "loss": 0.2077, "lr": 9.183569149534554e-05, "epoch": 1.5127388535031847, "percentage": 18.91, "elapsed_time": "2:38:38", "remaining_time": "11:20:18"} +{"current_steps": 480, "total_steps": 2512, "loss": 0.2038, "lr": 9.166275102687573e-05, "epoch": 1.5286624203821657, "percentage": 19.11, "elapsed_time": "2:39:47", "remaining_time": "11:16:28"} +{"current_steps": 485, "total_steps": 2512, "loss": 0.2009, "lr": 9.148816446772858e-05, "epoch": 1.5445859872611465, "percentage": 19.31, "elapsed_time": "2:40:57", "remaining_time": "11:12:43"} +{"current_steps": 490, "total_steps": 2512, "loss": 0.2, "lr": 9.131193871579975e-05, "epoch": 1.5605095541401273, "percentage": 19.51, "elapsed_time": "2:42:05", "remaining_time": "11:08:53"} +{"current_steps": 495, "total_steps": 2512, "loss": 0.2044, "lr": 9.113408073374913e-05, "epoch": 1.5764331210191083, "percentage": 19.71, "elapsed_time": "2:43:14", "remaining_time": "11:05:09"} +{"current_steps": 500, "total_steps": 2512, "loss": 0.2096, "lr": 9.09545975487259e-05, "epoch": 1.5923566878980893, "percentage": 19.9, "elapsed_time": "2:44:20", "remaining_time": "11:01:19"} +{"current_steps": 505, "total_steps": 2512, "loss": 0.2083, "lr": 9.07734962520908e-05, "epoch": 1.60828025477707, "percentage": 20.1, "elapsed_time": "2:58:01", "remaining_time": "11:47:30"} +{"current_steps": 510, "total_steps": 2512, "loss": 0.2083, "lr": 9.059078399913604e-05, "epoch": 1.6242038216560508, "percentage": 20.3, "elapsed_time": "2:59:10", "remaining_time": "11:43:20"} +{"current_steps": 515, "total_steps": 2512, "loss": 0.206, "lr": 9.040646800880244e-05, "epoch": 1.6401273885350318, "percentage": 20.5, "elapsed_time": "3:00:19", "remaining_time": "11:39:13"} +{"current_steps": 520, "total_steps": 2512, "loss": 0.21, "lr": 9.022055556339439e-05, "epoch": 1.6560509554140128, "percentage": 20.7, "elapsed_time": "3:01:27", "remaining_time": "11:35:07"} +{"current_steps": 525, "total_steps": 2512, "loss": 0.203, "lr": 9.0033054008292e-05, "epoch": 1.6719745222929936, "percentage": 20.9, "elapsed_time": "3:02:35", "remaining_time": "11:31:04"} +{"current_steps": 530, "total_steps": 2512, "loss": 0.1995, "lr": 8.984397075166095e-05, "epoch": 1.6878980891719744, "percentage": 21.1, "elapsed_time": "3:03:43", "remaining_time": "11:27:02"} +{"current_steps": 535, "total_steps": 2512, "loss": 0.2081, "lr": 8.965331326415976e-05, "epoch": 1.7038216560509554, "percentage": 21.3, "elapsed_time": "3:04:54", "remaining_time": "11:23:17"} +{"current_steps": 540, "total_steps": 2512, "loss": 0.2066, "lr": 8.946108907864461e-05, "epoch": 1.7197452229299364, "percentage": 21.5, "elapsed_time": "3:06:05", "remaining_time": "11:19:34"} +{"current_steps": 545, "total_steps": 2512, "loss": 0.2067, "lr": 8.926730578987182e-05, "epoch": 1.7356687898089171, "percentage": 21.7, "elapsed_time": "3:07:15", "remaining_time": "11:15:50"} +{"current_steps": 550, "total_steps": 2512, "loss": 0.2085, "lr": 8.907197105419761e-05, "epoch": 1.7515923566878981, "percentage": 21.89, "elapsed_time": "3:08:23", "remaining_time": "11:12:01"} +{"current_steps": 555, "total_steps": 2512, "loss": 0.2035, "lr": 8.887509258927578e-05, "epoch": 1.767515923566879, "percentage": 22.09, "elapsed_time": "3:09:29", "remaining_time": "11:08:10"} +{"current_steps": 560, "total_steps": 2512, "loss": 0.206, "lr": 8.867667817375266e-05, "epoch": 1.78343949044586, "percentage": 22.29, "elapsed_time": "3:10:35", "remaining_time": "11:04:21"} +{"current_steps": 565, "total_steps": 2512, "loss": 0.2062, "lr": 8.847673564695982e-05, "epoch": 1.799363057324841, "percentage": 22.49, "elapsed_time": "3:11:44", "remaining_time": "11:00:45"} +{"current_steps": 570, "total_steps": 2512, "loss": 0.2041, "lr": 8.827527290860436e-05, "epoch": 1.8152866242038217, "percentage": 22.69, "elapsed_time": "3:12:53", "remaining_time": "10:57:12"} +{"current_steps": 575, "total_steps": 2512, "loss": 0.203, "lr": 8.807229791845673e-05, "epoch": 1.8312101910828025, "percentage": 22.89, "elapsed_time": "3:13:58", "remaining_time": "10:53:26"} +{"current_steps": 580, "total_steps": 2512, "loss": 0.206, "lr": 8.78678186960363e-05, "epoch": 1.8471337579617835, "percentage": 23.09, "elapsed_time": "3:15:07", "remaining_time": "10:49:56"} +{"current_steps": 585, "total_steps": 2512, "loss": 0.2036, "lr": 8.766184332029452e-05, "epoch": 1.8630573248407645, "percentage": 23.29, "elapsed_time": "3:16:15", "remaining_time": "10:46:27"} +{"current_steps": 590, "total_steps": 2512, "loss": 0.1995, "lr": 8.745437992929566e-05, "epoch": 1.8789808917197452, "percentage": 23.49, "elapsed_time": "3:17:22", "remaining_time": "10:42:57"} +{"current_steps": 595, "total_steps": 2512, "loss": 0.2095, "lr": 8.72454367198953e-05, "epoch": 1.894904458598726, "percentage": 23.69, "elapsed_time": "3:18:33", "remaining_time": "10:39:42"} +{"current_steps": 600, "total_steps": 2512, "loss": 0.1997, "lr": 8.703502194741648e-05, "epoch": 1.910828025477707, "percentage": 23.89, "elapsed_time": "3:19:44", "remaining_time": "10:36:29"} +{"current_steps": 605, "total_steps": 2512, "loss": 0.212, "lr": 8.68231439253236e-05, "epoch": 1.926751592356688, "percentage": 24.08, "elapsed_time": "3:25:34", "remaining_time": "10:47:58"} +{"current_steps": 610, "total_steps": 2512, "loss": 0.2072, "lr": 8.660981102489378e-05, "epoch": 1.9426751592356688, "percentage": 24.28, "elapsed_time": "3:26:42", "remaining_time": "10:44:31"} +{"current_steps": 615, "total_steps": 2512, "loss": 0.2029, "lr": 8.639503167488634e-05, "epoch": 1.9585987261146496, "percentage": 24.48, "elapsed_time": "3:27:55", "remaining_time": "10:41:20"} +{"current_steps": 620, "total_steps": 2512, "loss": 0.2124, "lr": 8.617881436120956e-05, "epoch": 1.9745222929936306, "percentage": 24.68, "elapsed_time": "3:29:02", "remaining_time": "10:37:54"} +{"current_steps": 625, "total_steps": 2512, "loss": 0.2025, "lr": 8.596116762658554e-05, "epoch": 1.9904458598726116, "percentage": 24.88, "elapsed_time": "3:30:10", "remaining_time": "10:34:33"} +{"current_steps": 630, "total_steps": 2512, "loss": 0.1858, "lr": 8.57421000702127e-05, "epoch": 2.0063694267515926, "percentage": 25.08, "elapsed_time": "3:31:20", "remaining_time": "10:31:20"} +{"current_steps": 635, "total_steps": 2512, "loss": 0.1483, "lr": 8.552162034742592e-05, "epoch": 2.022292993630573, "percentage": 25.28, "elapsed_time": "3:32:29", "remaining_time": "10:28:05"} +{"current_steps": 640, "total_steps": 2512, "loss": 0.1526, "lr": 8.529973716935455e-05, "epoch": 2.038216560509554, "percentage": 25.48, "elapsed_time": "3:33:36", "remaining_time": "10:24:47"} +{"current_steps": 645, "total_steps": 2512, "loss": 0.1482, "lr": 8.507645930257846e-05, "epoch": 2.054140127388535, "percentage": 25.68, "elapsed_time": "3:34:46", "remaining_time": "10:21:39"} +{"current_steps": 650, "total_steps": 2512, "loss": 0.1498, "lr": 8.485179556878138e-05, "epoch": 2.070063694267516, "percentage": 25.88, "elapsed_time": "3:35:55", "remaining_time": "10:18:33"} +{"current_steps": 655, "total_steps": 2512, "loss": 0.1506, "lr": 8.462575484440254e-05, "epoch": 2.0859872611464967, "percentage": 26.07, "elapsed_time": "3:37:06", "remaining_time": "10:15:31"} +{"current_steps": 660, "total_steps": 2512, "loss": 0.1502, "lr": 8.439834606028594e-05, "epoch": 2.1019108280254777, "percentage": 26.27, "elapsed_time": "3:38:13", "remaining_time": "10:12:20"} +{"current_steps": 665, "total_steps": 2512, "loss": 0.1485, "lr": 8.416957820132742e-05, "epoch": 2.1178343949044587, "percentage": 26.47, "elapsed_time": "3:39:23", "remaining_time": "10:09:22"} +{"current_steps": 670, "total_steps": 2512, "loss": 0.1528, "lr": 8.393946030611978e-05, "epoch": 2.1337579617834397, "percentage": 26.67, "elapsed_time": "3:40:34", "remaining_time": "10:06:26"} +{"current_steps": 675, "total_steps": 2512, "loss": 0.1527, "lr": 8.370800146659557e-05, "epoch": 2.1496815286624202, "percentage": 26.87, "elapsed_time": "3:41:42", "remaining_time": "10:03:22"} +{"current_steps": 680, "total_steps": 2512, "loss": 0.151, "lr": 8.347521082766783e-05, "epoch": 2.1656050955414012, "percentage": 27.07, "elapsed_time": "3:42:54", "remaining_time": "10:00:32"} +{"current_steps": 685, "total_steps": 2512, "loss": 0.1591, "lr": 8.324109758686896e-05, "epoch": 2.1815286624203822, "percentage": 27.27, "elapsed_time": "3:44:03", "remaining_time": "9:57:36"} +{"current_steps": 690, "total_steps": 2512, "loss": 0.1583, "lr": 8.300567099398718e-05, "epoch": 2.1974522292993632, "percentage": 27.47, "elapsed_time": "3:45:13", "remaining_time": "9:54:43"} +{"current_steps": 695, "total_steps": 2512, "loss": 0.1565, "lr": 8.276894035070107e-05, "epoch": 2.213375796178344, "percentage": 27.67, "elapsed_time": "3:46:23", "remaining_time": "9:51:53"} +{"current_steps": 700, "total_steps": 2512, "loss": 0.1539, "lr": 8.25309150102121e-05, "epoch": 2.229299363057325, "percentage": 27.87, "elapsed_time": "3:47:32", "remaining_time": "9:48:59"} +{"current_steps": 705, "total_steps": 2512, "loss": 0.1483, "lr": 8.229160437687515e-05, "epoch": 2.245222929936306, "percentage": 28.07, "elapsed_time": "3:58:43", "remaining_time": "10:11:52"} +{"current_steps": 710, "total_steps": 2512, "loss": 0.1571, "lr": 8.205101790582681e-05, "epoch": 2.261146496815287, "percentage": 28.26, "elapsed_time": "3:59:49", "remaining_time": "10:08:42"} +{"current_steps": 715, "total_steps": 2512, "loss": 0.1497, "lr": 8.180916510261187e-05, "epoch": 2.2770700636942673, "percentage": 28.46, "elapsed_time": "4:00:58", "remaining_time": "10:05:37"} +{"current_steps": 720, "total_steps": 2512, "loss": 0.1461, "lr": 8.156605552280784e-05, "epoch": 2.2929936305732483, "percentage": 28.66, "elapsed_time": "4:02:10", "remaining_time": "10:02:45"} +{"current_steps": 725, "total_steps": 2512, "loss": 0.1502, "lr": 8.132169877164724e-05, "epoch": 2.3089171974522293, "percentage": 28.86, "elapsed_time": "4:03:19", "remaining_time": "9:59:45"} +{"current_steps": 730, "total_steps": 2512, "loss": 0.1508, "lr": 8.107610450363825e-05, "epoch": 2.3248407643312103, "percentage": 29.06, "elapsed_time": "4:04:30", "remaining_time": "9:56:52"} +{"current_steps": 735, "total_steps": 2512, "loss": 0.1509, "lr": 8.082928242218322e-05, "epoch": 2.340764331210191, "percentage": 29.26, "elapsed_time": "4:05:41", "remaining_time": "9:53:59"} +{"current_steps": 740, "total_steps": 2512, "loss": 0.153, "lr": 8.05812422791952e-05, "epoch": 2.356687898089172, "percentage": 29.46, "elapsed_time": "4:06:50", "remaining_time": "9:51:03"} +{"current_steps": 745, "total_steps": 2512, "loss": 0.1537, "lr": 8.033199387471277e-05, "epoch": 2.372611464968153, "percentage": 29.66, "elapsed_time": "4:07:58", "remaining_time": "9:48:10"} +{"current_steps": 750, "total_steps": 2512, "loss": 0.1554, "lr": 8.00815470565128e-05, "epoch": 2.388535031847134, "percentage": 29.86, "elapsed_time": "4:09:08", "remaining_time": "9:45:19"} +{"current_steps": 755, "total_steps": 2512, "loss": 0.1559, "lr": 7.98299117197213e-05, "epoch": 2.404458598726115, "percentage": 30.06, "elapsed_time": "4:10:19", "remaining_time": "9:42:31"} +{"current_steps": 760, "total_steps": 2512, "loss": 0.1522, "lr": 7.957709780642258e-05, "epoch": 2.4203821656050954, "percentage": 30.25, "elapsed_time": "4:11:29", "remaining_time": "9:39:44"} +{"current_steps": 765, "total_steps": 2512, "loss": 0.1604, "lr": 7.932311530526631e-05, "epoch": 2.4363057324840764, "percentage": 30.45, "elapsed_time": "4:12:36", "remaining_time": "9:36:51"} +{"current_steps": 770, "total_steps": 2512, "loss": 0.1572, "lr": 7.9067974251073e-05, "epoch": 2.4522292993630574, "percentage": 30.65, "elapsed_time": "4:13:45", "remaining_time": "9:34:04"} +{"current_steps": 775, "total_steps": 2512, "loss": 0.1554, "lr": 7.881168472443743e-05, "epoch": 2.468152866242038, "percentage": 30.85, "elapsed_time": "4:14:52", "remaining_time": "9:31:15"} +{"current_steps": 780, "total_steps": 2512, "loss": 0.1565, "lr": 7.855425685133039e-05, "epoch": 2.484076433121019, "percentage": 31.05, "elapsed_time": "4:15:59", "remaining_time": "9:28:25"} +{"current_steps": 785, "total_steps": 2512, "loss": 0.1571, "lr": 7.829570080269864e-05, "epoch": 2.5, "percentage": 31.25, "elapsed_time": "4:17:05", "remaining_time": "9:25:37"} +{"current_steps": 790, "total_steps": 2512, "loss": 0.152, "lr": 7.8036026794063e-05, "epoch": 2.515923566878981, "percentage": 31.45, "elapsed_time": "4:18:14", "remaining_time": "9:22:54"} +{"current_steps": 795, "total_steps": 2512, "loss": 0.153, "lr": 7.777524508511481e-05, "epoch": 2.531847133757962, "percentage": 31.65, "elapsed_time": "4:19:22", "remaining_time": "9:20:10"} +{"current_steps": 800, "total_steps": 2512, "loss": 0.154, "lr": 7.75133659793105e-05, "epoch": 2.5477707006369426, "percentage": 31.85, "elapsed_time": "4:20:31", "remaining_time": "9:17:30"} +{"current_steps": 805, "total_steps": 2512, "loss": 0.1521, "lr": 7.725039982346448e-05, "epoch": 2.5636942675159236, "percentage": 32.05, "elapsed_time": "4:33:15", "remaining_time": "9:39:26"} +{"current_steps": 810, "total_steps": 2512, "loss": 0.1551, "lr": 7.698635700734048e-05, "epoch": 2.5796178343949046, "percentage": 32.25, "elapsed_time": "4:34:21", "remaining_time": "9:36:30"} +{"current_steps": 815, "total_steps": 2512, "loss": 0.1549, "lr": 7.672124796324087e-05, "epoch": 2.595541401273885, "percentage": 32.44, "elapsed_time": "4:35:28", "remaining_time": "9:33:36"} +{"current_steps": 820, "total_steps": 2512, "loss": 0.1522, "lr": 7.645508316559466e-05, "epoch": 2.611464968152866, "percentage": 32.64, "elapsed_time": "4:36:38", "remaining_time": "9:30:49"} +{"current_steps": 825, "total_steps": 2512, "loss": 0.1556, "lr": 7.61878731305435e-05, "epoch": 2.627388535031847, "percentage": 32.84, "elapsed_time": "4:37:50", "remaining_time": "9:28:07"} +{"current_steps": 830, "total_steps": 2512, "loss": 0.1544, "lr": 7.591962841552627e-05, "epoch": 2.643312101910828, "percentage": 33.04, "elapsed_time": "4:38:56", "remaining_time": "9:25:16"} +{"current_steps": 835, "total_steps": 2512, "loss": 0.1619, "lr": 7.56503596188619e-05, "epoch": 2.659235668789809, "percentage": 33.24, "elapsed_time": "4:40:05", "remaining_time": "9:22:30"} +{"current_steps": 840, "total_steps": 2512, "loss": 0.1542, "lr": 7.53800773793308e-05, "epoch": 2.6751592356687897, "percentage": 33.44, "elapsed_time": "4:41:15", "remaining_time": "9:19:49"} +{"current_steps": 845, "total_steps": 2512, "loss": 0.1566, "lr": 7.510879237575423e-05, "epoch": 2.6910828025477707, "percentage": 33.64, "elapsed_time": "4:42:22", "remaining_time": "9:17:03"} +{"current_steps": 850, "total_steps": 2512, "loss": 0.1595, "lr": 7.483651532657266e-05, "epoch": 2.7070063694267517, "percentage": 33.84, "elapsed_time": "4:43:29", "remaining_time": "9:14:18"} +{"current_steps": 855, "total_steps": 2512, "loss": 0.1574, "lr": 7.456325698942214e-05, "epoch": 2.722929936305732, "percentage": 34.04, "elapsed_time": "4:44:36", "remaining_time": "9:11:34"} +{"current_steps": 860, "total_steps": 2512, "loss": 0.1511, "lr": 7.428902816070931e-05, "epoch": 2.738853503184713, "percentage": 34.24, "elapsed_time": "4:45:42", "remaining_time": "9:08:49"} +{"current_steps": 865, "total_steps": 2512, "loss": 0.1599, "lr": 7.401383967518478e-05, "epoch": 2.754777070063694, "percentage": 34.43, "elapsed_time": "4:46:49", "remaining_time": "9:06:07"} +{"current_steps": 870, "total_steps": 2512, "loss": 0.1581, "lr": 7.373770240551514e-05, "epoch": 2.770700636942675, "percentage": 34.63, "elapsed_time": "4:47:54", "remaining_time": "9:03:22"} +{"current_steps": 875, "total_steps": 2512, "loss": 0.1553, "lr": 7.346062726185332e-05, "epoch": 2.786624203821656, "percentage": 34.83, "elapsed_time": "4:49:06", "remaining_time": "9:00:53"} +{"current_steps": 880, "total_steps": 2512, "loss": 0.1591, "lr": 7.318262519140755e-05, "epoch": 2.802547770700637, "percentage": 35.03, "elapsed_time": "4:50:14", "remaining_time": "8:58:16"} +{"current_steps": 885, "total_steps": 2512, "loss": 0.1558, "lr": 7.290370717800882e-05, "epoch": 2.8184713375796178, "percentage": 35.23, "elapsed_time": "4:51:22", "remaining_time": "8:55:40"} +{"current_steps": 890, "total_steps": 2512, "loss": 0.1585, "lr": 7.262388424167698e-05, "epoch": 2.8343949044585988, "percentage": 35.43, "elapsed_time": "4:52:34", "remaining_time": "8:53:12"} +{"current_steps": 895, "total_steps": 2512, "loss": 0.1537, "lr": 7.23431674381852e-05, "epoch": 2.8503184713375798, "percentage": 35.63, "elapsed_time": "4:53:47", "remaining_time": "8:50:47"} +{"current_steps": 900, "total_steps": 2512, "loss": 0.1541, "lr": 7.20615678586233e-05, "epoch": 2.8662420382165603, "percentage": 35.83, "elapsed_time": "4:54:53", "remaining_time": "8:48:10"} +{"current_steps": 905, "total_steps": 2512, "loss": 0.1504, "lr": 7.177909662895949e-05, "epoch": 2.8821656050955413, "percentage": 36.03, "elapsed_time": "5:01:28", "remaining_time": "8:55:19"} +{"current_steps": 910, "total_steps": 2512, "loss": 0.1597, "lr": 7.149576490960073e-05, "epoch": 2.8980891719745223, "percentage": 36.23, "elapsed_time": "5:02:36", "remaining_time": "8:52:44"} +{"current_steps": 915, "total_steps": 2512, "loss": 0.1535, "lr": 7.121158389495186e-05, "epoch": 2.9140127388535033, "percentage": 36.43, "elapsed_time": "5:03:42", "remaining_time": "8:50:04"} +{"current_steps": 920, "total_steps": 2512, "loss": 0.1508, "lr": 7.092656481297332e-05, "epoch": 2.9299363057324843, "percentage": 36.62, "elapsed_time": "5:04:51", "remaining_time": "8:47:32"} +{"current_steps": 925, "total_steps": 2512, "loss": 0.1497, "lr": 7.064071892473744e-05, "epoch": 2.945859872611465, "percentage": 36.82, "elapsed_time": "5:05:58", "remaining_time": "8:44:56"} +{"current_steps": 930, "total_steps": 2512, "loss": 0.1493, "lr": 7.035405752398358e-05, "epoch": 2.961783439490446, "percentage": 37.02, "elapsed_time": "5:07:04", "remaining_time": "8:42:20"} +{"current_steps": 935, "total_steps": 2512, "loss": 0.1614, "lr": 7.006659193667195e-05, "epoch": 2.977707006369427, "percentage": 37.22, "elapsed_time": "5:08:11", "remaining_time": "8:39:48"} +{"current_steps": 940, "total_steps": 2512, "loss": 0.1559, "lr": 6.977833352053607e-05, "epoch": 2.9936305732484074, "percentage": 37.42, "elapsed_time": "5:09:21", "remaining_time": "8:37:20"} +{"current_steps": 945, "total_steps": 2512, "loss": 0.1212, "lr": 6.948929366463396e-05, "epoch": 3.0095541401273884, "percentage": 37.62, "elapsed_time": "5:10:33", "remaining_time": "8:34:57"} +{"current_steps": 950, "total_steps": 2512, "loss": 0.1055, "lr": 6.919948378889837e-05, "epoch": 3.0254777070063694, "percentage": 37.82, "elapsed_time": "5:11:40", "remaining_time": "8:32:27"} +{"current_steps": 955, "total_steps": 2512, "loss": 0.1, "lr": 6.890891534368538e-05, "epoch": 3.0414012738853504, "percentage": 38.02, "elapsed_time": "5:12:48", "remaining_time": "8:29:59"} +{"current_steps": 960, "total_steps": 2512, "loss": 0.1045, "lr": 6.861759980932204e-05, "epoch": 3.0573248407643314, "percentage": 38.22, "elapsed_time": "5:13:55", "remaining_time": "8:27:30"} +{"current_steps": 965, "total_steps": 2512, "loss": 0.1044, "lr": 6.832554869565284e-05, "epoch": 3.073248407643312, "percentage": 38.42, "elapsed_time": "5:15:02", "remaining_time": "8:25:03"} +{"current_steps": 970, "total_steps": 2512, "loss": 0.1064, "lr": 6.803277354158494e-05, "epoch": 3.089171974522293, "percentage": 38.61, "elapsed_time": "5:16:12", "remaining_time": "8:22:41"} +{"current_steps": 975, "total_steps": 2512, "loss": 0.1024, "lr": 6.773928591463223e-05, "epoch": 3.105095541401274, "percentage": 38.81, "elapsed_time": "5:17:25", "remaining_time": "8:20:23"} +{"current_steps": 980, "total_steps": 2512, "loss": 0.1019, "lr": 6.744509741045835e-05, "epoch": 3.121019108280255, "percentage": 39.01, "elapsed_time": "5:18:32", "remaining_time": "8:17:57"} +{"current_steps": 985, "total_steps": 2512, "loss": 0.1048, "lr": 6.715021965241848e-05, "epoch": 3.1369426751592355, "percentage": 39.21, "elapsed_time": "5:19:42", "remaining_time": "8:15:37"} +{"current_steps": 990, "total_steps": 2512, "loss": 0.1025, "lr": 6.68546642911002e-05, "epoch": 3.1528662420382165, "percentage": 39.41, "elapsed_time": "5:20:49", "remaining_time": "8:13:13"} +{"current_steps": 995, "total_steps": 2512, "loss": 0.1016, "lr": 6.655844300386308e-05, "epoch": 3.1687898089171975, "percentage": 39.61, "elapsed_time": "5:22:01", "remaining_time": "8:10:57"} +{"current_steps": 1000, "total_steps": 2512, "loss": 0.1029, "lr": 6.626156749437736e-05, "epoch": 3.1847133757961785, "percentage": 39.81, "elapsed_time": "5:23:08", "remaining_time": "8:08:34"} +{"current_steps": 1005, "total_steps": 2512, "loss": 0.1014, "lr": 6.596404949216153e-05, "epoch": 3.200636942675159, "percentage": 40.01, "elapsed_time": "5:26:20", "remaining_time": "8:09:20"} +{"current_steps": 1010, "total_steps": 2512, "loss": 0.102, "lr": 6.56659007521189e-05, "epoch": 3.21656050955414, "percentage": 40.21, "elapsed_time": "5:27:26", "remaining_time": "8:06:57"} +{"current_steps": 1015, "total_steps": 2512, "loss": 0.1069, "lr": 6.536713305407314e-05, "epoch": 3.232484076433121, "percentage": 40.41, "elapsed_time": "5:28:36", "remaining_time": "8:04:38"} +{"current_steps": 1020, "total_steps": 2512, "loss": 0.1043, "lr": 6.506775820230295e-05, "epoch": 3.248407643312102, "percentage": 40.61, "elapsed_time": "5:29:45", "remaining_time": "8:02:20"} +{"current_steps": 1025, "total_steps": 2512, "loss": 0.1017, "lr": 6.476778802507549e-05, "epoch": 3.2643312101910826, "percentage": 40.8, "elapsed_time": "5:30:52", "remaining_time": "8:00:00"} +{"current_steps": 1030, "total_steps": 2512, "loss": 0.1084, "lr": 6.446723437417928e-05, "epoch": 3.2802547770700636, "percentage": 41.0, "elapsed_time": "5:32:00", "remaining_time": "7:57:42"} +{"current_steps": 1035, "total_steps": 2512, "loss": 0.107, "lr": 6.416610912445574e-05, "epoch": 3.2961783439490446, "percentage": 41.2, "elapsed_time": "5:33:11", "remaining_time": "7:55:29"} +{"current_steps": 1040, "total_steps": 2512, "loss": 0.11, "lr": 6.386442417333011e-05, "epoch": 3.3121019108280256, "percentage": 41.4, "elapsed_time": "5:34:21", "remaining_time": "7:53:14"} +{"current_steps": 1045, "total_steps": 2512, "loss": 0.1057, "lr": 6.356219144034143e-05, "epoch": 3.328025477707006, "percentage": 41.6, "elapsed_time": "5:35:27", "remaining_time": "7:50:55"} +{"current_steps": 1050, "total_steps": 2512, "loss": 0.1044, "lr": 6.325942286667149e-05, "epoch": 3.343949044585987, "percentage": 41.8, "elapsed_time": "5:36:36", "remaining_time": "7:48:41"} +{"current_steps": 1055, "total_steps": 2512, "loss": 0.1031, "lr": 6.295613041467306e-05, "epoch": 3.359872611464968, "percentage": 42.0, "elapsed_time": "5:37:45", "remaining_time": "7:46:27"} +{"current_steps": 1060, "total_steps": 2512, "loss": 0.1028, "lr": 6.265232606739734e-05, "epoch": 3.375796178343949, "percentage": 42.2, "elapsed_time": "5:38:54", "remaining_time": "7:44:14"} +{"current_steps": 1065, "total_steps": 2512, "loss": 0.1087, "lr": 6.234802182812043e-05, "epoch": 3.3917197452229297, "percentage": 42.4, "elapsed_time": "5:40:03", "remaining_time": "7:42:01"} +{"current_steps": 1070, "total_steps": 2512, "loss": 0.1042, "lr": 6.20432297198691e-05, "epoch": 3.4076433121019107, "percentage": 42.6, "elapsed_time": "5:41:08", "remaining_time": "7:39:44"} +{"current_steps": 1075, "total_steps": 2512, "loss": 0.1094, "lr": 6.17379617849458e-05, "epoch": 3.4235668789808917, "percentage": 42.79, "elapsed_time": "5:42:16", "remaining_time": "7:37:32"} +{"current_steps": 1080, "total_steps": 2512, "loss": 0.1062, "lr": 6.14322300844528e-05, "epoch": 3.4394904458598727, "percentage": 42.99, "elapsed_time": "5:43:24", "remaining_time": "7:35:20"} +{"current_steps": 1085, "total_steps": 2512, "loss": 0.1065, "lr": 6.112604669781572e-05, "epoch": 3.4554140127388537, "percentage": 43.19, "elapsed_time": "5:44:34", "remaining_time": "7:33:11"} +{"current_steps": 1090, "total_steps": 2512, "loss": 0.104, "lr": 6.0819423722306253e-05, "epoch": 3.4713375796178343, "percentage": 43.39, "elapsed_time": "5:45:43", "remaining_time": "7:31:01"} +{"current_steps": 1095, "total_steps": 2512, "loss": 0.1097, "lr": 6.0512373272564184e-05, "epoch": 3.4872611464968153, "percentage": 43.59, "elapsed_time": "5:46:49", "remaining_time": "7:28:48"} +{"current_steps": 1100, "total_steps": 2512, "loss": 0.1041, "lr": 6.020490748011877e-05, "epoch": 3.5031847133757963, "percentage": 43.79, "elapsed_time": "5:47:59", "remaining_time": "7:26:41"} +{"current_steps": 1105, "total_steps": 2512, "loss": 0.1059, "lr": 5.9897038492909394e-05, "epoch": 3.519108280254777, "percentage": 43.99, "elapsed_time": "5:55:31", "remaining_time": "7:32:41"} +{"current_steps": 1110, "total_steps": 2512, "loss": 0.106, "lr": 5.958877847480564e-05, "epoch": 3.535031847133758, "percentage": 44.19, "elapsed_time": "5:56:39", "remaining_time": "7:30:28"} +{"current_steps": 1115, "total_steps": 2512, "loss": 0.1085, "lr": 5.928013960512668e-05, "epoch": 3.550955414012739, "percentage": 44.39, "elapsed_time": "5:57:44", "remaining_time": "7:28:13"} +{"current_steps": 1120, "total_steps": 2512, "loss": 0.1076, "lr": 5.897113407816006e-05, "epoch": 3.56687898089172, "percentage": 44.59, "elapsed_time": "5:58:54", "remaining_time": "7:26:03"} +{"current_steps": 1125, "total_steps": 2512, "loss": 0.1024, "lr": 5.86617741026799e-05, "epoch": 3.582802547770701, "percentage": 44.79, "elapsed_time": "6:00:02", "remaining_time": "7:23:53"} +{"current_steps": 1130, "total_steps": 2512, "loss": 0.1125, "lr": 5.8352071901464566e-05, "epoch": 3.5987261146496814, "percentage": 44.98, "elapsed_time": "6:01:15", "remaining_time": "7:21:48"} +{"current_steps": 1135, "total_steps": 2512, "loss": 0.1082, "lr": 5.804203971081371e-05, "epoch": 3.6146496815286624, "percentage": 45.18, "elapsed_time": "6:02:23", "remaining_time": "7:19:39"} +{"current_steps": 1140, "total_steps": 2512, "loss": 0.1056, "lr": 5.773168978006486e-05, "epoch": 3.6305732484076434, "percentage": 45.38, "elapsed_time": "6:03:30", "remaining_time": "7:17:29"} +{"current_steps": 1145, "total_steps": 2512, "loss": 0.1023, "lr": 5.742103437110937e-05, "epoch": 3.646496815286624, "percentage": 45.58, "elapsed_time": "6:04:37", "remaining_time": "7:15:19"} +{"current_steps": 1150, "total_steps": 2512, "loss": 0.1071, "lr": 5.711008575790806e-05, "epoch": 3.662420382165605, "percentage": 45.78, "elapsed_time": "6:05:43", "remaining_time": "7:13:08"} +{"current_steps": 1155, "total_steps": 2512, "loss": 0.1041, "lr": 5.679885622600616e-05, "epoch": 3.678343949044586, "percentage": 45.98, "elapsed_time": "6:06:48", "remaining_time": "7:10:57"} +{"current_steps": 1160, "total_steps": 2512, "loss": 0.1106, "lr": 5.6487358072048e-05, "epoch": 3.694267515923567, "percentage": 46.18, "elapsed_time": "6:07:54", "remaining_time": "7:08:48"} +{"current_steps": 1165, "total_steps": 2512, "loss": 0.1062, "lr": 5.617560360329114e-05, "epoch": 3.710191082802548, "percentage": 46.38, "elapsed_time": "6:09:01", "remaining_time": "7:06:41"} +{"current_steps": 1170, "total_steps": 2512, "loss": 0.1089, "lr": 5.58636051371201e-05, "epoch": 3.7261146496815285, "percentage": 46.58, "elapsed_time": "6:10:11", "remaining_time": "7:04:36"} +{"current_steps": 1175, "total_steps": 2512, "loss": 0.1041, "lr": 5.5551375000559716e-05, "epoch": 3.7420382165605095, "percentage": 46.78, "elapsed_time": "6:11:18", "remaining_time": "7:02:29"} +{"current_steps": 1180, "total_steps": 2512, "loss": 0.1114, "lr": 5.523892552978808e-05, "epoch": 3.7579617834394905, "percentage": 46.97, "elapsed_time": "6:12:22", "remaining_time": "7:00:20"} +{"current_steps": 1185, "total_steps": 2512, "loss": 0.1061, "lr": 5.492626906964917e-05, "epoch": 3.7738853503184715, "percentage": 47.17, "elapsed_time": "6:13:33", "remaining_time": "6:58:19"} +{"current_steps": 1190, "total_steps": 2512, "loss": 0.1046, "lr": 5.4613417973165106e-05, "epoch": 3.789808917197452, "percentage": 47.37, "elapsed_time": "6:14:41", "remaining_time": "6:56:15"} +{"current_steps": 1195, "total_steps": 2512, "loss": 0.1067, "lr": 5.430038460104803e-05, "epoch": 3.805732484076433, "percentage": 47.57, "elapsed_time": "6:15:49", "remaining_time": "6:54:12"} +{"current_steps": 1200, "total_steps": 2512, "loss": 0.1074, "lr": 5.39871813212118e-05, "epoch": 3.821656050955414, "percentage": 47.77, "elapsed_time": "6:16:55", "remaining_time": "6:52:05"} +{"current_steps": 1205, "total_steps": 2512, "loss": 0.1038, "lr": 5.36738205082833e-05, "epoch": 3.837579617834395, "percentage": 47.97, "elapsed_time": "6:22:02", "remaining_time": "6:54:22"} +{"current_steps": 1210, "total_steps": 2512, "loss": 0.1106, "lr": 5.3360314543113544e-05, "epoch": 3.853503184713376, "percentage": 48.17, "elapsed_time": "6:23:11", "remaining_time": "6:52:19"} +{"current_steps": 1215, "total_steps": 2512, "loss": 0.1057, "lr": 5.304667581228846e-05, "epoch": 3.8694267515923566, "percentage": 48.37, "elapsed_time": "6:24:17", "remaining_time": "6:50:13"} +{"current_steps": 1220, "total_steps": 2512, "loss": 0.1048, "lr": 5.2732916707639576e-05, "epoch": 3.8853503184713376, "percentage": 48.57, "elapsed_time": "6:25:26", "remaining_time": "6:48:11"} +{"current_steps": 1225, "total_steps": 2512, "loss": 0.1057, "lr": 5.2419049625754346e-05, "epoch": 3.9012738853503186, "percentage": 48.77, "elapsed_time": "6:26:33", "remaining_time": "6:46:07"} +{"current_steps": 1230, "total_steps": 2512, "loss": 0.1098, "lr": 5.210508696748642e-05, "epoch": 3.917197452229299, "percentage": 48.96, "elapsed_time": "6:27:45", "remaining_time": "6:44:09"} +{"current_steps": 1235, "total_steps": 2512, "loss": 0.1034, "lr": 5.17910411374656e-05, "epoch": 3.93312101910828, "percentage": 49.16, "elapsed_time": "6:28:53", "remaining_time": "6:42:07"} +{"current_steps": 1240, "total_steps": 2512, "loss": 0.1089, "lr": 5.1476924543607884e-05, "epoch": 3.949044585987261, "percentage": 49.36, "elapsed_time": "6:30:02", "remaining_time": "6:40:05"} +{"current_steps": 1245, "total_steps": 2512, "loss": 0.1115, "lr": 5.116274959662508e-05, "epoch": 3.964968152866242, "percentage": 49.56, "elapsed_time": "6:31:11", "remaining_time": "6:38:06"} +{"current_steps": 1250, "total_steps": 2512, "loss": 0.1033, "lr": 5.0848528709534525e-05, "epoch": 3.980891719745223, "percentage": 49.76, "elapsed_time": "6:32:20", "remaining_time": "6:36:06"} +{"current_steps": 1255, "total_steps": 2512, "loss": 0.1102, "lr": 5.053427429716867e-05, "epoch": 3.9968152866242037, "percentage": 49.96, "elapsed_time": "6:33:32", "remaining_time": "6:34:10"} +{"current_steps": 1260, "total_steps": 2512, "loss": 0.0717, "lr": 5.021999877568452e-05, "epoch": 4.012738853503185, "percentage": 50.16, "elapsed_time": "6:34:41", "remaining_time": "6:32:11"} +{"current_steps": 1265, "total_steps": 2512, "loss": 0.0669, "lr": 4.990571456207311e-05, "epoch": 4.028662420382165, "percentage": 50.36, "elapsed_time": "6:35:45", "remaining_time": "6:30:07"} +{"current_steps": 1270, "total_steps": 2512, "loss": 0.0653, "lr": 4.959143407366891e-05, "epoch": 4.044585987261146, "percentage": 50.56, "elapsed_time": "6:36:53", "remaining_time": "6:28:08"} +{"current_steps": 1275, "total_steps": 2512, "loss": 0.0672, "lr": 4.927716972765918e-05, "epoch": 4.060509554140127, "percentage": 50.76, "elapsed_time": "6:38:00", "remaining_time": "6:26:08"} +{"current_steps": 1280, "total_steps": 2512, "loss": 0.0643, "lr": 4.8962933940593444e-05, "epoch": 4.076433121019108, "percentage": 50.96, "elapsed_time": "6:39:07", "remaining_time": "6:24:09"} +{"current_steps": 1285, "total_steps": 2512, "loss": 0.0676, "lr": 4.864873912789282e-05, "epoch": 4.092356687898089, "percentage": 51.15, "elapsed_time": "6:40:14", "remaining_time": "6:22:10"} +{"current_steps": 1290, "total_steps": 2512, "loss": 0.0642, "lr": 4.833459770335958e-05, "epoch": 4.10828025477707, "percentage": 51.35, "elapsed_time": "6:41:20", "remaining_time": "6:20:11"} +{"current_steps": 1295, "total_steps": 2512, "loss": 0.0634, "lr": 4.802052207868654e-05, "epoch": 4.124203821656051, "percentage": 51.55, "elapsed_time": "6:42:29", "remaining_time": "6:18:15"} +{"current_steps": 1300, "total_steps": 2512, "loss": 0.0651, "lr": 4.7706524662966874e-05, "epoch": 4.140127388535032, "percentage": 51.75, "elapsed_time": "6:43:36", "remaining_time": "6:16:17"} +{"current_steps": 1305, "total_steps": 2512, "loss": 0.0657, "lr": 4.739261786220368e-05, "epoch": 4.156050955414012, "percentage": 51.95, "elapsed_time": "6:47:59", "remaining_time": "6:17:21"} +{"current_steps": 1310, "total_steps": 2512, "loss": 0.0647, "lr": 4.707881407881988e-05, "epoch": 4.171974522292993, "percentage": 52.15, "elapsed_time": "6:49:05", "remaining_time": "6:15:21"} +{"current_steps": 1315, "total_steps": 2512, "loss": 0.0665, "lr": 4.67651257111682e-05, "epoch": 4.187898089171974, "percentage": 52.35, "elapsed_time": "6:50:11", "remaining_time": "6:13:23"} +{"current_steps": 1320, "total_steps": 2512, "loss": 0.0682, "lr": 4.6451565153041274e-05, "epoch": 4.203821656050955, "percentage": 52.55, "elapsed_time": "6:51:20", "remaining_time": "6:11:26"} +{"current_steps": 1325, "total_steps": 2512, "loss": 0.0687, "lr": 4.613814479318203e-05, "epoch": 4.219745222929936, "percentage": 52.75, "elapsed_time": "6:52:27", "remaining_time": "6:09:30"} +{"current_steps": 1330, "total_steps": 2512, "loss": 0.0687, "lr": 4.582487701479417e-05, "epoch": 4.235668789808917, "percentage": 52.95, "elapsed_time": "6:53:35", "remaining_time": "6:07:33"} +{"current_steps": 1335, "total_steps": 2512, "loss": 0.0684, "lr": 4.551177419505288e-05, "epoch": 4.251592356687898, "percentage": 53.14, "elapsed_time": "6:54:41", "remaining_time": "6:05:36"} +{"current_steps": 1340, "total_steps": 2512, "loss": 0.0674, "lr": 4.5198848704615914e-05, "epoch": 4.267515923566879, "percentage": 53.34, "elapsed_time": "6:55:49", "remaining_time": "6:03:41"} +{"current_steps": 1345, "total_steps": 2512, "loss": 0.0646, "lr": 4.4886112907134715e-05, "epoch": 4.2834394904458595, "percentage": 53.54, "elapsed_time": "6:56:59", "remaining_time": "6:01:48"} +{"current_steps": 1350, "total_steps": 2512, "loss": 0.0685, "lr": 4.457357915876599e-05, "epoch": 4.2993630573248405, "percentage": 53.74, "elapsed_time": "6:58:06", "remaining_time": "5:59:53"} +{"current_steps": 1355, "total_steps": 2512, "loss": 0.0664, "lr": 4.4261259807683534e-05, "epoch": 4.3152866242038215, "percentage": 53.94, "elapsed_time": "6:59:18", "remaining_time": "5:58:01"} +{"current_steps": 1360, "total_steps": 2512, "loss": 0.069, "lr": 4.394916719359028e-05, "epoch": 4.3312101910828025, "percentage": 54.14, "elapsed_time": "7:00:26", "remaining_time": "5:56:07"} +{"current_steps": 1365, "total_steps": 2512, "loss": 0.065, "lr": 4.363731364723084e-05, "epoch": 4.3471337579617835, "percentage": 54.34, "elapsed_time": "7:01:36", "remaining_time": "5:54:16"} +{"current_steps": 1370, "total_steps": 2512, "loss": 0.0676, "lr": 4.332571148990429e-05, "epoch": 4.3630573248407645, "percentage": 54.54, "elapsed_time": "7:02:46", "remaining_time": "5:52:24"} +{"current_steps": 1375, "total_steps": 2512, "loss": 0.0665, "lr": 4.3014373032977336e-05, "epoch": 4.3789808917197455, "percentage": 54.74, "elapsed_time": "7:03:58", "remaining_time": "5:50:35"} +{"current_steps": 1380, "total_steps": 2512, "loss": 0.0672, "lr": 4.270331057739793e-05, "epoch": 4.3949044585987265, "percentage": 54.94, "elapsed_time": "7:05:05", "remaining_time": "5:48:41"} +{"current_steps": 1385, "total_steps": 2512, "loss": 0.0659, "lr": 4.239253641320921e-05, "epoch": 4.4108280254777075, "percentage": 55.14, "elapsed_time": "7:06:14", "remaining_time": "5:46:50"} +{"current_steps": 1390, "total_steps": 2512, "loss": 0.0674, "lr": 4.208206281906399e-05, "epoch": 4.426751592356688, "percentage": 55.33, "elapsed_time": "7:07:22", "remaining_time": "5:44:58"} +{"current_steps": 1395, "total_steps": 2512, "loss": 0.0659, "lr": 4.177190206173958e-05, "epoch": 4.442675159235669, "percentage": 55.53, "elapsed_time": "7:08:31", "remaining_time": "5:43:07"} +{"current_steps": 1400, "total_steps": 2512, "loss": 0.0681, "lr": 4.146206639565312e-05, "epoch": 4.45859872611465, "percentage": 55.73, "elapsed_time": "7:09:40", "remaining_time": "5:41:16"} +{"current_steps": 1405, "total_steps": 2512, "loss": 0.0664, "lr": 4.1152568062377474e-05, "epoch": 4.474522292993631, "percentage": 55.93, "elapsed_time": "7:20:30", "remaining_time": "5:47:04"} +{"current_steps": 1410, "total_steps": 2512, "loss": 0.0656, "lr": 4.084341929015749e-05, "epoch": 4.490445859872612, "percentage": 56.13, "elapsed_time": "7:21:40", "remaining_time": "5:45:11"} +{"current_steps": 1415, "total_steps": 2512, "loss": 0.0692, "lr": 4.05346322934269e-05, "epoch": 4.506369426751593, "percentage": 56.33, "elapsed_time": "7:22:46", "remaining_time": "5:43:15"} +{"current_steps": 1420, "total_steps": 2512, "loss": 0.0643, "lr": 4.022621927232577e-05, "epoch": 4.522292993630574, "percentage": 56.53, "elapsed_time": "7:23:54", "remaining_time": "5:41:22"} +{"current_steps": 1425, "total_steps": 2512, "loss": 0.0666, "lr": 3.991819241221835e-05, "epoch": 4.538216560509554, "percentage": 56.73, "elapsed_time": "7:25:03", "remaining_time": "5:39:29"} +{"current_steps": 1430, "total_steps": 2512, "loss": 0.0695, "lr": 3.961056388321177e-05, "epoch": 4.554140127388535, "percentage": 56.93, "elapsed_time": "7:26:14", "remaining_time": "5:37:38"} +{"current_steps": 1435, "total_steps": 2512, "loss": 0.0671, "lr": 3.930334583967514e-05, "epoch": 4.570063694267516, "percentage": 57.13, "elapsed_time": "7:27:24", "remaining_time": "5:35:47"} +{"current_steps": 1440, "total_steps": 2512, "loss": 0.0674, "lr": 3.8996550419759315e-05, "epoch": 4.585987261146497, "percentage": 57.32, "elapsed_time": "7:28:33", "remaining_time": "5:33:55"} +{"current_steps": 1445, "total_steps": 2512, "loss": 0.0669, "lr": 3.869018974491735e-05, "epoch": 4.601910828025478, "percentage": 57.52, "elapsed_time": "7:29:43", "remaining_time": "5:32:04"} +{"current_steps": 1450, "total_steps": 2512, "loss": 0.0668, "lr": 3.838427591942557e-05, "epoch": 4.617834394904459, "percentage": 57.72, "elapsed_time": "7:30:48", "remaining_time": "5:30:10"} +{"current_steps": 1455, "total_steps": 2512, "loss": 0.0704, "lr": 3.807882102990532e-05, "epoch": 4.63375796178344, "percentage": 57.92, "elapsed_time": "7:31:54", "remaining_time": "5:28:17"} +{"current_steps": 1460, "total_steps": 2512, "loss": 0.0672, "lr": 3.7773837144845455e-05, "epoch": 4.649681528662421, "percentage": 58.12, "elapsed_time": "7:33:02", "remaining_time": "5:26:26"} +{"current_steps": 1465, "total_steps": 2512, "loss": 0.0677, "lr": 3.746933631412548e-05, "epoch": 4.665605095541402, "percentage": 58.32, "elapsed_time": "7:34:14", "remaining_time": "5:24:37"} +{"current_steps": 1470, "total_steps": 2512, "loss": 0.0645, "lr": 3.716533056853948e-05, "epoch": 4.681528662420382, "percentage": 58.52, "elapsed_time": "7:35:21", "remaining_time": "5:22:46"} +{"current_steps": 1475, "total_steps": 2512, "loss": 0.068, "lr": 3.686183191932078e-05, "epoch": 4.697452229299363, "percentage": 58.72, "elapsed_time": "7:36:29", "remaining_time": "5:20:56"} +{"current_steps": 1480, "total_steps": 2512, "loss": 0.0685, "lr": 3.6558852357667416e-05, "epoch": 4.713375796178344, "percentage": 58.92, "elapsed_time": "7:37:36", "remaining_time": "5:19:05"} +{"current_steps": 1485, "total_steps": 2512, "loss": 0.067, "lr": 3.625640385426826e-05, "epoch": 4.729299363057325, "percentage": 59.12, "elapsed_time": "7:38:46", "remaining_time": "5:17:16"} +{"current_steps": 1490, "total_steps": 2512, "loss": 0.0679, "lr": 3.595449835883018e-05, "epoch": 4.745222929936306, "percentage": 59.32, "elapsed_time": "7:39:51", "remaining_time": "5:15:24"} +{"current_steps": 1495, "total_steps": 2512, "loss": 0.0664, "lr": 3.5653147799605866e-05, "epoch": 4.761146496815287, "percentage": 59.51, "elapsed_time": "7:41:00", "remaining_time": "5:13:36"} +{"current_steps": 1500, "total_steps": 2512, "loss": 0.0661, "lr": 3.535236408292253e-05, "epoch": 4.777070063694268, "percentage": 59.71, "elapsed_time": "7:42:10", "remaining_time": "5:11:49"} +{"current_steps": 1505, "total_steps": 2512, "loss": 0.0666, "lr": 3.505215909271149e-05, "epoch": 4.792993630573249, "percentage": 59.91, "elapsed_time": "7:52:07", "remaining_time": "5:15:54"} +{"current_steps": 1510, "total_steps": 2512, "loss": 0.0667, "lr": 3.4752544690038647e-05, "epoch": 4.80891719745223, "percentage": 60.11, "elapsed_time": "7:53:14", "remaining_time": "5:14:01"} +{"current_steps": 1515, "total_steps": 2512, "loss": 0.0653, "lr": 3.4453532712635855e-05, "epoch": 4.82484076433121, "percentage": 60.31, "elapsed_time": "7:54:21", "remaining_time": "5:12:10"} +{"current_steps": 1520, "total_steps": 2512, "loss": 0.0675, "lr": 3.415513497443322e-05, "epoch": 4.840764331210191, "percentage": 60.51, "elapsed_time": "7:55:30", "remaining_time": "5:10:19"} +{"current_steps": 1525, "total_steps": 2512, "loss": 0.067, "lr": 3.385736326509232e-05, "epoch": 4.856687898089172, "percentage": 60.71, "elapsed_time": "7:56:40", "remaining_time": "5:08:30"} +{"current_steps": 1530, "total_steps": 2512, "loss": 0.0659, "lr": 3.3560229349540394e-05, "epoch": 4.872611464968153, "percentage": 60.91, "elapsed_time": "7:57:49", "remaining_time": "5:06:40"} +{"current_steps": 1535, "total_steps": 2512, "loss": 0.0661, "lr": 3.3263744967505534e-05, "epoch": 4.888535031847134, "percentage": 61.11, "elapsed_time": "7:58:56", "remaining_time": "5:04:50"} +{"current_steps": 1540, "total_steps": 2512, "loss": 0.0667, "lr": 3.296792183305283e-05, "epoch": 4.904458598726115, "percentage": 61.31, "elapsed_time": "8:00:00", "remaining_time": "5:02:58"} +{"current_steps": 1545, "total_steps": 2512, "loss": 0.0664, "lr": 3.2672771634121543e-05, "epoch": 4.920382165605096, "percentage": 61.5, "elapsed_time": "8:01:08", "remaining_time": "5:01:08"} +{"current_steps": 1550, "total_steps": 2512, "loss": 0.0667, "lr": 3.237830603206333e-05, "epoch": 4.936305732484076, "percentage": 61.7, "elapsed_time": "8:02:15", "remaining_time": "4:59:18"} +{"current_steps": 1555, "total_steps": 2512, "loss": 0.0667, "lr": 3.208453666118151e-05, "epoch": 4.952229299363057, "percentage": 61.9, "elapsed_time": "8:03:24", "remaining_time": "4:57:30"} +{"current_steps": 1560, "total_steps": 2512, "loss": 0.0646, "lr": 3.1791475128271383e-05, "epoch": 4.968152866242038, "percentage": 62.1, "elapsed_time": "8:04:31", "remaining_time": "4:55:40"} +{"current_steps": 1565, "total_steps": 2512, "loss": 0.0658, "lr": 3.149913301216164e-05, "epoch": 4.984076433121019, "percentage": 62.3, "elapsed_time": "8:05:39", "remaining_time": "4:53:52"} +{"current_steps": 1570, "total_steps": 2512, "loss": 0.0617, "lr": 3.1207521863256914e-05, "epoch": 5.0, "percentage": 62.5, "elapsed_time": "8:06:47", "remaining_time": "4:52:04"} +{"current_steps": 1575, "total_steps": 2512, "loss": 0.0402, "lr": 3.0916653203081367e-05, "epoch": 5.015923566878981, "percentage": 62.7, "elapsed_time": "8:07:54", "remaining_time": "4:50:16"} +{"current_steps": 1580, "total_steps": 2512, "loss": 0.0397, "lr": 3.0626538523823565e-05, "epoch": 5.031847133757962, "percentage": 62.9, "elapsed_time": "8:09:00", "remaining_time": "4:48:27"} +{"current_steps": 1585, "total_steps": 2512, "loss": 0.039, "lr": 3.033718928788233e-05, "epoch": 5.047770700636943, "percentage": 63.1, "elapsed_time": "8:10:05", "remaining_time": "4:46:38"} +{"current_steps": 1590, "total_steps": 2512, "loss": 0.0395, "lr": 3.004861692741391e-05, "epoch": 5.063694267515924, "percentage": 63.3, "elapsed_time": "8:11:14", "remaining_time": "4:44:51"} +{"current_steps": 1595, "total_steps": 2512, "loss": 0.0388, "lr": 2.976083284388031e-05, "epoch": 5.079617834394904, "percentage": 63.5, "elapsed_time": "8:12:21", "remaining_time": "4:43:04"} +{"current_steps": 1600, "total_steps": 2512, "loss": 0.0397, "lr": 2.947384840759878e-05, "epoch": 5.095541401273885, "percentage": 63.69, "elapsed_time": "8:13:31", "remaining_time": "4:41:18"} +{"current_steps": 1605, "total_steps": 2512, "loss": 0.0391, "lr": 2.918767495729261e-05, "epoch": 5.111464968152866, "percentage": 63.89, "elapsed_time": "8:27:48", "remaining_time": "4:46:58"} +{"current_steps": 1610, "total_steps": 2512, "loss": 0.0384, "lr": 2.8902323799643116e-05, "epoch": 5.127388535031847, "percentage": 64.09, "elapsed_time": "8:28:58", "remaining_time": "4:45:09"} +{"current_steps": 1615, "total_steps": 2512, "loss": 0.0389, "lr": 2.8617806208842924e-05, "epoch": 5.143312101910828, "percentage": 64.29, "elapsed_time": "8:30:09", "remaining_time": "4:43:21"} +{"current_steps": 1620, "total_steps": 2512, "loss": 0.0387, "lr": 2.83341334261505e-05, "epoch": 5.159235668789809, "percentage": 64.49, "elapsed_time": "8:31:15", "remaining_time": "4:41:30"} +{"current_steps": 1625, "total_steps": 2512, "loss": 0.0388, "lr": 2.8051316659446104e-05, "epoch": 5.17515923566879, "percentage": 64.69, "elapsed_time": "8:32:20", "remaining_time": "4:39:39"} +{"current_steps": 1630, "total_steps": 2512, "loss": 0.0381, "lr": 2.7769367082788812e-05, "epoch": 5.191082802547771, "percentage": 64.89, "elapsed_time": "8:33:27", "remaining_time": "4:37:49"} +{"current_steps": 1635, "total_steps": 2512, "loss": 0.0394, "lr": 2.74882958359752e-05, "epoch": 5.207006369426751, "percentage": 65.09, "elapsed_time": "8:34:34", "remaining_time": "4:36:00"} +{"current_steps": 1640, "total_steps": 2512, "loss": 0.0389, "lr": 2.7208114024099053e-05, "epoch": 5.222929936305732, "percentage": 65.29, "elapsed_time": "8:35:40", "remaining_time": "4:34:11"} +{"current_steps": 1645, "total_steps": 2512, "loss": 0.0407, "lr": 2.6928832717112757e-05, "epoch": 5.238853503184713, "percentage": 65.49, "elapsed_time": "8:36:50", "remaining_time": "4:32:24"} +{"current_steps": 1650, "total_steps": 2512, "loss": 0.0397, "lr": 2.6650462949389765e-05, "epoch": 5.254777070063694, "percentage": 65.68, "elapsed_time": "8:37:56", "remaining_time": "4:30:35"} +{"current_steps": 1655, "total_steps": 2512, "loss": 0.0399, "lr": 2.6373015719288806e-05, "epoch": 5.270700636942675, "percentage": 65.88, "elapsed_time": "8:39:05", "remaining_time": "4:28:47"} +{"current_steps": 1660, "total_steps": 2512, "loss": 0.0397, "lr": 2.609650198871917e-05, "epoch": 5.286624203821656, "percentage": 66.08, "elapsed_time": "8:40:13", "remaining_time": "4:27:00"} +{"current_steps": 1665, "total_steps": 2512, "loss": 0.0391, "lr": 2.5820932682707732e-05, "epoch": 5.302547770700637, "percentage": 66.28, "elapsed_time": "8:41:22", "remaining_time": "4:25:13"} +{"current_steps": 1670, "total_steps": 2512, "loss": 0.0395, "lr": 2.5546318688967254e-05, "epoch": 5.318471337579618, "percentage": 66.48, "elapsed_time": "8:42:28", "remaining_time": "4:23:25"} +{"current_steps": 1675, "total_steps": 2512, "loss": 0.0396, "lr": 2.5272670857466165e-05, "epoch": 5.334394904458598, "percentage": 66.68, "elapsed_time": "8:43:36", "remaining_time": "4:21:38"} +{"current_steps": 1680, "total_steps": 2512, "loss": 0.0403, "lr": 2.500000000000001e-05, "epoch": 5.350318471337579, "percentage": 66.88, "elapsed_time": "8:44:46", "remaining_time": "4:19:53"} +{"current_steps": 1685, "total_steps": 2512, "loss": 0.0401, "lr": 2.4728316889764132e-05, "epoch": 5.36624203821656, "percentage": 67.08, "elapsed_time": "8:45:54", "remaining_time": "4:18:06"} +{"current_steps": 1690, "total_steps": 2512, "loss": 0.0385, "lr": 2.4457632260928142e-05, "epoch": 5.382165605095541, "percentage": 67.28, "elapsed_time": "8:47:01", "remaining_time": "4:16:20"} +{"current_steps": 1695, "total_steps": 2512, "loss": 0.0397, "lr": 2.418795680821171e-05, "epoch": 5.398089171974522, "percentage": 67.48, "elapsed_time": "8:48:11", "remaining_time": "4:14:35"} +{"current_steps": 1700, "total_steps": 2512, "loss": 0.0388, "lr": 2.3919301186462122e-05, "epoch": 5.414012738853503, "percentage": 67.68, "elapsed_time": "8:49:18", "remaining_time": "4:12:49"} +{"current_steps": 1705, "total_steps": 2512, "loss": 0.0387, "lr": 2.3651676010233205e-05, "epoch": 5.429936305732484, "percentage": 67.87, "elapsed_time": "8:59:28", "remaining_time": "4:15:20"} +{"current_steps": 1710, "total_steps": 2512, "loss": 0.0392, "lr": 2.338509185336605e-05, "epoch": 5.445859872611465, "percentage": 68.07, "elapsed_time": "9:00:34", "remaining_time": "4:13:31"} +{"current_steps": 1715, "total_steps": 2512, "loss": 0.0382, "lr": 2.3119559248571128e-05, "epoch": 5.461783439490446, "percentage": 68.27, "elapsed_time": "9:01:41", "remaining_time": "4:11:44"} +{"current_steps": 1720, "total_steps": 2512, "loss": 0.0377, "lr": 2.2855088687012267e-05, "epoch": 5.477707006369426, "percentage": 68.47, "elapsed_time": "9:02:52", "remaining_time": "4:09:58"} +{"current_steps": 1725, "total_steps": 2512, "loss": 0.0382, "lr": 2.2591690617892024e-05, "epoch": 5.493630573248407, "percentage": 68.67, "elapsed_time": "9:04:01", "remaining_time": "4:08:12"} +{"current_steps": 1730, "total_steps": 2512, "loss": 0.038, "lr": 2.2329375448038963e-05, "epoch": 5.509554140127388, "percentage": 68.87, "elapsed_time": "9:05:10", "remaining_time": "4:06:26"} +{"current_steps": 1735, "total_steps": 2512, "loss": 0.0382, "lr": 2.206815354149636e-05, "epoch": 5.525477707006369, "percentage": 69.07, "elapsed_time": "9:06:19", "remaining_time": "4:04:39"} +{"current_steps": 1740, "total_steps": 2512, "loss": 0.0383, "lr": 2.1808035219112815e-05, "epoch": 5.54140127388535, "percentage": 69.27, "elapsed_time": "9:07:25", "remaining_time": "4:02:52"} +{"current_steps": 1745, "total_steps": 2512, "loss": 0.0377, "lr": 2.15490307581344e-05, "epoch": 5.557324840764331, "percentage": 69.47, "elapsed_time": "9:08:33", "remaining_time": "4:01:07"} +{"current_steps": 1750, "total_steps": 2512, "loss": 0.0385, "lr": 2.1291150391798702e-05, "epoch": 5.573248407643312, "percentage": 69.67, "elapsed_time": "9:09:41", "remaining_time": "3:59:21"} +{"current_steps": 1755, "total_steps": 2512, "loss": 0.0386, "lr": 2.1034404308930406e-05, "epoch": 5.5891719745222925, "percentage": 69.86, "elapsed_time": "9:10:48", "remaining_time": "3:57:35"} +{"current_steps": 1760, "total_steps": 2512, "loss": 0.0381, "lr": 2.0778802653538825e-05, "epoch": 5.6050955414012735, "percentage": 70.06, "elapsed_time": "9:11:54", "remaining_time": "3:55:48"} +{"current_steps": 1765, "total_steps": 2512, "loss": 0.0393, "lr": 2.0524355524417017e-05, "epoch": 5.6210191082802545, "percentage": 70.26, "elapsed_time": "9:13:01", "remaining_time": "3:54:03"} +{"current_steps": 1770, "total_steps": 2512, "loss": 0.0381, "lr": 2.0271072974742893e-05, "epoch": 5.6369426751592355, "percentage": 70.46, "elapsed_time": "9:14:08", "remaining_time": "3:52:18"} +{"current_steps": 1775, "total_steps": 2512, "loss": 0.0388, "lr": 2.00189650116819e-05, "epoch": 5.6528662420382165, "percentage": 70.66, "elapsed_time": "9:15:19", "remaining_time": "3:50:34"} +{"current_steps": 1780, "total_steps": 2512, "loss": 0.0384, "lr": 1.976804159599176e-05, "epoch": 5.6687898089171975, "percentage": 70.86, "elapsed_time": "9:16:26", "remaining_time": "3:48:49"} +{"current_steps": 1785, "total_steps": 2512, "loss": 0.0384, "lr": 1.9518312641628794e-05, "epoch": 5.6847133757961785, "percentage": 71.06, "elapsed_time": "9:17:37", "remaining_time": "3:47:06"} +{"current_steps": 1790, "total_steps": 2512, "loss": 0.0382, "lr": 1.9269788015356338e-05, "epoch": 5.7006369426751595, "percentage": 71.26, "elapsed_time": "9:18:48", "remaining_time": "3:45:23"} +{"current_steps": 1795, "total_steps": 2512, "loss": 0.0393, "lr": 1.902247753635486e-05, "epoch": 5.7165605095541405, "percentage": 71.46, "elapsed_time": "9:20:00", "remaining_time": "3:43:41"} +{"current_steps": 1800, "total_steps": 2512, "loss": 0.0392, "lr": 1.8776390975833968e-05, "epoch": 5.732484076433121, "percentage": 71.66, "elapsed_time": "9:21:10", "remaining_time": "3:41:58"} +{"current_steps": 1805, "total_steps": 2512, "loss": 0.0393, "lr": 1.8531538056646414e-05, "epoch": 5.748407643312102, "percentage": 71.86, "elapsed_time": "9:28:28", "remaining_time": "3:42:39"} +{"current_steps": 1810, "total_steps": 2512, "loss": 0.0381, "lr": 1.8287928452903873e-05, "epoch": 5.764331210191083, "percentage": 72.05, "elapsed_time": "9:29:33", "remaining_time": "3:40:54"} +{"current_steps": 1815, "total_steps": 2512, "loss": 0.0376, "lr": 1.804557178959485e-05, "epoch": 5.780254777070064, "percentage": 72.25, "elapsed_time": "9:30:39", "remaining_time": "3:39:08"} +{"current_steps": 1820, "total_steps": 2512, "loss": 0.0377, "lr": 1.7804477642204222e-05, "epoch": 5.796178343949045, "percentage": 72.45, "elapsed_time": "9:31:46", "remaining_time": "3:37:24"} +{"current_steps": 1825, "total_steps": 2512, "loss": 0.0388, "lr": 1.7564655536335074e-05, "epoch": 5.812101910828026, "percentage": 72.65, "elapsed_time": "9:32:52", "remaining_time": "3:35:39"} +{"current_steps": 1830, "total_steps": 2512, "loss": 0.0389, "lr": 1.7326114947332207e-05, "epoch": 5.828025477707007, "percentage": 72.85, "elapsed_time": "9:34:02", "remaining_time": "3:33:55"} +{"current_steps": 1835, "total_steps": 2512, "loss": 0.0389, "lr": 1.7088865299907898e-05, "epoch": 5.843949044585988, "percentage": 73.05, "elapsed_time": "9:35:11", "remaining_time": "3:32:12"} +{"current_steps": 1840, "total_steps": 2512, "loss": 0.0401, "lr": 1.68529159677694e-05, "epoch": 5.859872611464969, "percentage": 73.25, "elapsed_time": "9:36:22", "remaining_time": "3:30:30"} +{"current_steps": 1845, "total_steps": 2512, "loss": 0.038, "lr": 1.6618276273248716e-05, "epoch": 5.875796178343949, "percentage": 73.45, "elapsed_time": "9:37:30", "remaining_time": "3:28:46"} +{"current_steps": 1850, "total_steps": 2512, "loss": 0.0379, "lr": 1.6384955486934156e-05, "epoch": 5.89171974522293, "percentage": 73.65, "elapsed_time": "9:38:35", "remaining_time": "3:27:02"} +{"current_steps": 1855, "total_steps": 2512, "loss": 0.0385, "lr": 1.6152962827304163e-05, "epoch": 5.907643312101911, "percentage": 73.85, "elapsed_time": "9:39:45", "remaining_time": "3:25:20"} +{"current_steps": 1860, "total_steps": 2512, "loss": 0.0377, "lr": 1.5922307460363007e-05, "epoch": 5.923566878980892, "percentage": 74.04, "elapsed_time": "9:40:55", "remaining_time": "3:23:38"} +{"current_steps": 1865, "total_steps": 2512, "loss": 0.0392, "lr": 1.5692998499278656e-05, "epoch": 5.939490445859873, "percentage": 74.24, "elapsed_time": "9:42:03", "remaining_time": "3:21:55"} +{"current_steps": 1870, "total_steps": 2512, "loss": 0.0377, "lr": 1.5465045004022798e-05, "epoch": 5.955414012738854, "percentage": 74.44, "elapsed_time": "9:43:10", "remaining_time": "3:20:12"} +{"current_steps": 1875, "total_steps": 2512, "loss": 0.0394, "lr": 1.523845598101274e-05, "epoch": 5.971337579617835, "percentage": 74.64, "elapsed_time": "9:44:19", "remaining_time": "3:18:30"} +{"current_steps": 1880, "total_steps": 2512, "loss": 0.0377, "lr": 1.5013240382755712e-05, "epoch": 5.987261146496815, "percentage": 74.84, "elapsed_time": "9:45:28", "remaining_time": "3:16:49"} +{"current_steps": 1885, "total_steps": 2512, "loss": 0.034, "lr": 1.4789407107495012e-05, "epoch": 6.003184713375796, "percentage": 75.04, "elapsed_time": "9:46:32", "remaining_time": "3:15:05"} +{"current_steps": 1890, "total_steps": 2512, "loss": 0.0236, "lr": 1.456696499885859e-05, "epoch": 6.019108280254777, "percentage": 75.24, "elapsed_time": "9:47:39", "remaining_time": "3:13:23"} +{"current_steps": 1895, "total_steps": 2512, "loss": 0.0225, "lr": 1.4345922845509473e-05, "epoch": 6.035031847133758, "percentage": 75.44, "elapsed_time": "9:48:49", "remaining_time": "3:11:42"} +{"current_steps": 1900, "total_steps": 2512, "loss": 0.0229, "lr": 1.412628938079869e-05, "epoch": 6.050955414012739, "percentage": 75.64, "elapsed_time": "9:49:56", "remaining_time": "3:10:01"} +{"current_steps": 1905, "total_steps": 2512, "loss": 0.0229, "lr": 1.3908073282420064e-05, "epoch": 6.06687898089172, "percentage": 75.84, "elapsed_time": "9:55:03", "remaining_time": "3:09:36"} +{"current_steps": 1910, "total_steps": 2512, "loss": 0.023, "lr": 1.3691283172067493e-05, "epoch": 6.082802547770701, "percentage": 76.04, "elapsed_time": "9:56:09", "remaining_time": "3:07:53"} +{"current_steps": 1915, "total_steps": 2512, "loss": 0.0224, "lr": 1.3475927615094198e-05, "epoch": 6.098726114649682, "percentage": 76.23, "elapsed_time": "9:57:16", "remaining_time": "3:06:11"} +{"current_steps": 1920, "total_steps": 2512, "loss": 0.0231, "lr": 1.3262015120174397e-05, "epoch": 6.114649681528663, "percentage": 76.43, "elapsed_time": "9:58:23", "remaining_time": "3:04:30"} +{"current_steps": 1925, "total_steps": 2512, "loss": 0.0219, "lr": 1.3049554138967051e-05, "epoch": 6.130573248407643, "percentage": 76.63, "elapsed_time": "9:59:30", "remaining_time": "3:02:48"} +{"current_steps": 1930, "total_steps": 2512, "loss": 0.0228, "lr": 1.2838553065781994e-05, "epoch": 6.146496815286624, "percentage": 76.83, "elapsed_time": "10:00:40", "remaining_time": "3:01:08"} +{"current_steps": 1935, "total_steps": 2512, "loss": 0.0231, "lr": 1.262902023724824e-05, "epoch": 6.162420382165605, "percentage": 77.03, "elapsed_time": "10:01:53", "remaining_time": "2:59:28"} +{"current_steps": 1940, "total_steps": 2512, "loss": 0.0227, "lr": 1.2420963931984647e-05, "epoch": 6.178343949044586, "percentage": 77.23, "elapsed_time": "10:03:02", "remaining_time": "2:57:48"} +{"current_steps": 1945, "total_steps": 2512, "loss": 0.0222, "lr": 1.221439237027277e-05, "epoch": 6.194267515923567, "percentage": 77.43, "elapsed_time": "10:04:08", "remaining_time": "2:56:07"} +{"current_steps": 1950, "total_steps": 2512, "loss": 0.0234, "lr": 1.2009313713732146e-05, "epoch": 6.210191082802548, "percentage": 77.63, "elapsed_time": "10:05:19", "remaining_time": "2:54:27"} +{"current_steps": 1955, "total_steps": 2512, "loss": 0.0223, "lr": 1.1805736064997747e-05, "epoch": 6.226114649681529, "percentage": 77.83, "elapsed_time": "10:06:27", "remaining_time": "2:52:47"} +{"current_steps": 1960, "total_steps": 2512, "loss": 0.0225, "lr": 1.160366746739996e-05, "epoch": 6.24203821656051, "percentage": 78.03, "elapsed_time": "10:07:35", "remaining_time": "2:51:07"} +{"current_steps": 1965, "total_steps": 2512, "loss": 0.0218, "lr": 1.1403115904646645e-05, "epoch": 6.25796178343949, "percentage": 78.22, "elapsed_time": "10:08:45", "remaining_time": "2:49:27"} +{"current_steps": 1970, "total_steps": 2512, "loss": 0.0231, "lr": 1.1204089300507847e-05, "epoch": 6.273885350318471, "percentage": 78.42, "elapsed_time": "10:09:51", "remaining_time": "2:47:47"} +{"current_steps": 1975, "total_steps": 2512, "loss": 0.0225, "lr": 1.1006595518502666e-05, "epoch": 6.289808917197452, "percentage": 78.62, "elapsed_time": "10:10:59", "remaining_time": "2:46:07"} +{"current_steps": 1980, "total_steps": 2512, "loss": 0.0234, "lr": 1.0810642361588519e-05, "epoch": 6.305732484076433, "percentage": 78.82, "elapsed_time": "10:12:06", "remaining_time": "2:44:27"} +{"current_steps": 1985, "total_steps": 2512, "loss": 0.0225, "lr": 1.0616237571852949e-05, "epoch": 6.321656050955414, "percentage": 79.02, "elapsed_time": "10:13:15", "remaining_time": "2:42:48"} +{"current_steps": 1990, "total_steps": 2512, "loss": 0.022, "lr": 1.0423388830207647e-05, "epoch": 6.337579617834395, "percentage": 79.22, "elapsed_time": "10:14:21", "remaining_time": "2:41:09"} +{"current_steps": 1995, "total_steps": 2512, "loss": 0.0224, "lr": 1.0232103756085032e-05, "epoch": 6.353503184713376, "percentage": 79.42, "elapsed_time": "10:15:32", "remaining_time": "2:39:31"} +{"current_steps": 2000, "total_steps": 2512, "loss": 0.0217, "lr": 1.0042389907137184e-05, "epoch": 6.369426751592357, "percentage": 79.62, "elapsed_time": "10:16:43", "remaining_time": "2:37:52"} +{"current_steps": 2005, "total_steps": 2512, "loss": 0.0225, "lr": 9.854254778937272e-06, "epoch": 6.385350318471337, "percentage": 79.82, "elapsed_time": "10:21:47", "remaining_time": "2:37:13"} +{"current_steps": 2010, "total_steps": 2512, "loss": 0.022, "lr": 9.667705804683347e-06, "epoch": 6.401273885350318, "percentage": 80.02, "elapsed_time": "10:22:55", "remaining_time": "2:35:34"} +{"current_steps": 2015, "total_steps": 2512, "loss": 0.0224, "lr": 9.48275035490474e-06, "epoch": 6.417197452229299, "percentage": 80.21, "elapsed_time": "10:24:04", "remaining_time": "2:33:55"} +{"current_steps": 2020, "total_steps": 2512, "loss": 0.0226, "lr": 9.299395737170757e-06, "epoch": 6.43312101910828, "percentage": 80.41, "elapsed_time": "10:25:12", "remaining_time": "2:32:16"} +{"current_steps": 2025, "total_steps": 2512, "loss": 0.0222, "lr": 9.117649195802053e-06, "epoch": 6.449044585987261, "percentage": 80.61, "elapsed_time": "10:26:19", "remaining_time": "2:30:37"} +{"current_steps": 2030, "total_steps": 2512, "loss": 0.0226, "lr": 8.93751791158432e-06, "epoch": 6.464968152866242, "percentage": 80.81, "elapsed_time": "10:27:25", "remaining_time": "2:28:58"} +{"current_steps": 2035, "total_steps": 2512, "loss": 0.0228, "lr": 8.759009001484664e-06, "epoch": 6.480891719745223, "percentage": 81.01, "elapsed_time": "10:28:34", "remaining_time": "2:27:20"} +{"current_steps": 2040, "total_steps": 2512, "loss": 0.0217, "lr": 8.582129518370325e-06, "epoch": 6.496815286624204, "percentage": 81.21, "elapsed_time": "10:29:43", "remaining_time": "2:25:42"} +{"current_steps": 2045, "total_steps": 2512, "loss": 0.0219, "lr": 8.4068864507301e-06, "epoch": 6.512738853503185, "percentage": 81.41, "elapsed_time": "10:30:51", "remaining_time": "2:24:03"} +{"current_steps": 2050, "total_steps": 2512, "loss": 0.0226, "lr": 8.233286722398176e-06, "epoch": 6.528662420382165, "percentage": 81.61, "elapsed_time": "10:31:59", "remaining_time": "2:22:25"} +{"current_steps": 2055, "total_steps": 2512, "loss": 0.0226, "lr": 8.061337192280567e-06, "epoch": 6.544585987261146, "percentage": 81.81, "elapsed_time": "10:33:11", "remaining_time": "2:20:48"} +{"current_steps": 2060, "total_steps": 2512, "loss": 0.0219, "lr": 7.891044654084185e-06, "epoch": 6.560509554140127, "percentage": 82.01, "elapsed_time": "10:34:19", "remaining_time": "2:19:11"} +{"current_steps": 2065, "total_steps": 2512, "loss": 0.0221, "lr": 7.72241583604832e-06, "epoch": 6.576433121019108, "percentage": 82.21, "elapsed_time": "10:35:31", "remaining_time": "2:17:34"} +{"current_steps": 2070, "total_steps": 2512, "loss": 0.0227, "lr": 7.555457400678917e-06, "epoch": 6.592356687898089, "percentage": 82.4, "elapsed_time": "10:36:38", "remaining_time": "2:15:56"} +{"current_steps": 2075, "total_steps": 2512, "loss": 0.0221, "lr": 7.390175944485239e-06, "epoch": 6.60828025477707, "percentage": 82.6, "elapsed_time": "10:37:48", "remaining_time": "2:14:19"} +{"current_steps": 2080, "total_steps": 2512, "loss": 0.0215, "lr": 7.22657799771933e-06, "epoch": 6.624203821656051, "percentage": 82.8, "elapsed_time": "10:38:57", "remaining_time": "2:12:42"} +{"current_steps": 2085, "total_steps": 2512, "loss": 0.0222, "lr": 7.06467002411792e-06, "epoch": 6.640127388535031, "percentage": 83.0, "elapsed_time": "10:40:04", "remaining_time": "2:11:05"} +{"current_steps": 2090, "total_steps": 2512, "loss": 0.0217, "lr": 6.904458420647131e-06, "epoch": 6.656050955414012, "percentage": 83.2, "elapsed_time": "10:41:08", "remaining_time": "2:09:27"} +{"current_steps": 2095, "total_steps": 2512, "loss": 0.022, "lr": 6.745949517249639e-06, "epoch": 6.671974522292993, "percentage": 83.4, "elapsed_time": "10:42:16", "remaining_time": "2:07:50"} +{"current_steps": 2100, "total_steps": 2512, "loss": 0.0219, "lr": 6.589149576594683e-06, "epoch": 6.687898089171974, "percentage": 83.6, "elapsed_time": "10:43:23", "remaining_time": "2:06:13"} +{"current_steps": 2105, "total_steps": 2512, "loss": 0.0223, "lr": 6.43406479383053e-06, "epoch": 6.703821656050955, "percentage": 83.8, "elapsed_time": "10:48:27", "remaining_time": "2:05:22"} +{"current_steps": 2110, "total_steps": 2512, "loss": 0.0226, "lr": 6.280701296339775e-06, "epoch": 6.719745222929936, "percentage": 84.0, "elapsed_time": "10:49:35", "remaining_time": "2:03:45"} +{"current_steps": 2115, "total_steps": 2512, "loss": 0.0216, "lr": 6.129065143497215e-06, "epoch": 6.735668789808917, "percentage": 84.2, "elapsed_time": "10:50:41", "remaining_time": "2:02:08"} +{"current_steps": 2120, "total_steps": 2512, "loss": 0.0225, "lr": 5.979162326430438e-06, "epoch": 6.751592356687898, "percentage": 84.39, "elapsed_time": "10:51:48", "remaining_time": "2:00:31"} +{"current_steps": 2125, "total_steps": 2512, "loss": 0.0219, "lr": 5.830998767783135e-06, "epoch": 6.767515923566879, "percentage": 84.59, "elapsed_time": "10:52:54", "remaining_time": "1:58:54"} +{"current_steps": 2130, "total_steps": 2512, "loss": 0.0218, "lr": 5.6845803214811145e-06, "epoch": 6.7834394904458595, "percentage": 84.79, "elapsed_time": "10:54:00", "remaining_time": "1:57:17"} +{"current_steps": 2135, "total_steps": 2512, "loss": 0.0213, "lr": 5.539912772500944e-06, "epoch": 6.7993630573248405, "percentage": 84.99, "elapsed_time": "10:55:11", "remaining_time": "1:55:41"} +{"current_steps": 2140, "total_steps": 2512, "loss": 0.0215, "lr": 5.397001836641469e-06, "epoch": 6.8152866242038215, "percentage": 85.19, "elapsed_time": "10:56:21", "remaining_time": "1:54:05"} +{"current_steps": 2145, "total_steps": 2512, "loss": 0.0212, "lr": 5.255853160297947e-06, "epoch": 6.8312101910828025, "percentage": 85.39, "elapsed_time": "10:57:29", "remaining_time": "1:52:29"} +{"current_steps": 2150, "total_steps": 2512, "loss": 0.0215, "lr": 5.116472320238908e-06, "epoch": 6.8471337579617835, "percentage": 85.59, "elapsed_time": "10:58:38", "remaining_time": "1:50:53"} +{"current_steps": 2155, "total_steps": 2512, "loss": 0.0215, "lr": 4.978864823385937e-06, "epoch": 6.8630573248407645, "percentage": 85.79, "elapsed_time": "10:59:46", "remaining_time": "1:49:17"} +{"current_steps": 2160, "total_steps": 2512, "loss": 0.0207, "lr": 4.843036106595961e-06, "epoch": 6.8789808917197455, "percentage": 85.99, "elapsed_time": "11:00:53", "remaining_time": "1:47:42"} +{"current_steps": 2165, "total_steps": 2512, "loss": 0.0222, "lr": 4.708991536446556e-06, "epoch": 6.8949044585987265, "percentage": 86.19, "elapsed_time": "11:01:59", "remaining_time": "1:46:06"} +{"current_steps": 2170, "total_steps": 2512, "loss": 0.0215, "lr": 4.576736409023813e-06, "epoch": 6.9108280254777075, "percentage": 86.39, "elapsed_time": "11:03:07", "remaining_time": "1:44:30"} +{"current_steps": 2175, "total_steps": 2512, "loss": 0.0223, "lr": 4.446275949713186e-06, "epoch": 6.926751592356688, "percentage": 86.58, "elapsed_time": "11:04:16", "remaining_time": "1:42:55"} +{"current_steps": 2180, "total_steps": 2512, "loss": 0.0211, "lr": 4.317615312992962e-06, "epoch": 6.942675159235669, "percentage": 86.78, "elapsed_time": "11:05:28", "remaining_time": "1:41:20"} +{"current_steps": 2185, "total_steps": 2512, "loss": 0.0219, "lr": 4.190759582230641e-06, "epoch": 6.95859872611465, "percentage": 86.98, "elapsed_time": "11:06:38", "remaining_time": "1:39:46"} +{"current_steps": 2190, "total_steps": 2512, "loss": 0.0211, "lr": 4.065713769482082e-06, "epoch": 6.974522292993631, "percentage": 87.18, "elapsed_time": "11:07:44", "remaining_time": "1:38:10"} +{"current_steps": 2195, "total_steps": 2512, "loss": 0.0218, "lr": 3.942482815293513e-06, "epoch": 6.990445859872612, "percentage": 87.38, "elapsed_time": "11:08:50", "remaining_time": "1:36:35"} +{"current_steps": 2200, "total_steps": 2512, "loss": 0.0185, "lr": 3.8210715885062585e-06, "epoch": 7.006369426751593, "percentage": 87.58, "elapsed_time": "11:09:59", "remaining_time": "1:35:00"} +{"current_steps": 2205, "total_steps": 2512, "loss": 0.015, "lr": 3.7014848860644656e-06, "epoch": 7.022292993630574, "percentage": 87.78, "elapsed_time": "11:15:08", "remaining_time": "1:33:59"} +{"current_steps": 2210, "total_steps": 2512, "loss": 0.0147, "lr": 3.5837274328254735e-06, "epoch": 7.038216560509555, "percentage": 87.98, "elapsed_time": "11:16:19", "remaining_time": "1:32:25"} +{"current_steps": 2215, "total_steps": 2512, "loss": 0.0152, "lr": 3.467803881373233e-06, "epoch": 7.054140127388535, "percentage": 88.18, "elapsed_time": "11:17:26", "remaining_time": "1:30:50"} +{"current_steps": 2220, "total_steps": 2512, "loss": 0.0149, "lr": 3.3537188118343998e-06, "epoch": 7.070063694267516, "percentage": 88.38, "elapsed_time": "11:18:35", "remaining_time": "1:29:15"} +{"current_steps": 2225, "total_steps": 2512, "loss": 0.015, "lr": 3.2414767316974514e-06, "epoch": 7.085987261146497, "percentage": 88.57, "elapsed_time": "11:19:42", "remaining_time": "1:27:40"} +{"current_steps": 2230, "total_steps": 2512, "loss": 0.0145, "lr": 3.131082075634523e-06, "epoch": 7.101910828025478, "percentage": 88.77, "elapsed_time": "11:20:52", "remaining_time": "1:26:06"} +{"current_steps": 2235, "total_steps": 2512, "loss": 0.0152, "lr": 3.022539205326258e-06, "epoch": 7.117834394904459, "percentage": 88.97, "elapsed_time": "11:22:01", "remaining_time": "1:24:31"} +{"current_steps": 2240, "total_steps": 2512, "loss": 0.0143, "lr": 2.9158524092894214e-06, "epoch": 7.13375796178344, "percentage": 89.17, "elapsed_time": "11:23:09", "remaining_time": "1:22:57"} +{"current_steps": 2245, "total_steps": 2512, "loss": 0.0145, "lr": 2.8110259027075258e-06, "epoch": 7.149681528662421, "percentage": 89.37, "elapsed_time": "11:24:16", "remaining_time": "1:21:22"} +{"current_steps": 2250, "total_steps": 2512, "loss": 0.0138, "lr": 2.708063827264223e-06, "epoch": 7.165605095541402, "percentage": 89.57, "elapsed_time": "11:25:25", "remaining_time": "1:19:48"} +{"current_steps": 2255, "total_steps": 2512, "loss": 0.0146, "lr": 2.6069702509796957e-06, "epoch": 7.181528662420382, "percentage": 89.77, "elapsed_time": "11:26:33", "remaining_time": "1:18:14"} +{"current_steps": 2260, "total_steps": 2512, "loss": 0.0141, "lr": 2.507749168049961e-06, "epoch": 7.197452229299363, "percentage": 89.97, "elapsed_time": "11:27:40", "remaining_time": "1:16:40"} +{"current_steps": 2265, "total_steps": 2512, "loss": 0.0152, "lr": 2.410404498688995e-06, "epoch": 7.213375796178344, "percentage": 90.17, "elapsed_time": "11:28:47", "remaining_time": "1:15:06"} +{"current_steps": 2270, "total_steps": 2512, "loss": 0.0149, "lr": 2.3149400889739157e-06, "epoch": 7.229299363057325, "percentage": 90.37, "elapsed_time": "11:29:53", "remaining_time": "1:13:32"} +{"current_steps": 2275, "total_steps": 2512, "loss": 0.0142, "lr": 2.221359710692961e-06, "epoch": 7.245222929936306, "percentage": 90.57, "elapsed_time": "11:31:03", "remaining_time": "1:11:59"} +{"current_steps": 2280, "total_steps": 2512, "loss": 0.0147, "lr": 2.1296670611965273e-06, "epoch": 7.261146496815287, "percentage": 90.76, "elapsed_time": "11:32:12", "remaining_time": "1:10:26"} +{"current_steps": 2285, "total_steps": 2512, "loss": 0.0146, "lr": 2.039865763251014e-06, "epoch": 7.277070063694268, "percentage": 90.96, "elapsed_time": "11:33:21", "remaining_time": "1:08:52"} +{"current_steps": 2290, "total_steps": 2512, "loss": 0.0148, "lr": 1.951959364895778e-06, "epoch": 7.292993630573249, "percentage": 91.16, "elapsed_time": "11:34:28", "remaining_time": "1:07:19"} +{"current_steps": 2295, "total_steps": 2512, "loss": 0.0137, "lr": 1.865951339302857e-06, "epoch": 7.308917197452229, "percentage": 91.36, "elapsed_time": "11:35:36", "remaining_time": "1:05:46"} +{"current_steps": 2300, "total_steps": 2512, "loss": 0.0148, "lr": 1.7818450846398327e-06, "epoch": 7.32484076433121, "percentage": 91.56, "elapsed_time": "11:36:44", "remaining_time": "1:04:13"} +{"current_steps": 2305, "total_steps": 2512, "loss": 0.0145, "lr": 1.6996439239354956e-06, "epoch": 7.340764331210191, "percentage": 91.76, "elapsed_time": "11:41:36", "remaining_time": "1:03:00"} +{"current_steps": 2310, "total_steps": 2512, "loss": 0.0149, "lr": 1.6193511049486076e-06, "epoch": 7.356687898089172, "percentage": 91.96, "elapsed_time": "11:42:46", "remaining_time": "1:01:27"} +{"current_steps": 2315, "total_steps": 2512, "loss": 0.0147, "lr": 1.5409698000395379e-06, "epoch": 7.372611464968153, "percentage": 92.16, "elapsed_time": "11:43:53", "remaining_time": "0:59:53"} +{"current_steps": 2320, "total_steps": 2512, "loss": 0.0149, "lr": 1.4645031060449732e-06, "epoch": 7.388535031847134, "percentage": 92.36, "elapsed_time": "11:45:01", "remaining_time": "0:58:20"} +{"current_steps": 2325, "total_steps": 2512, "loss": 0.0136, "lr": 1.389954044155517e-06, "epoch": 7.404458598726115, "percentage": 92.56, "elapsed_time": "11:46:08", "remaining_time": "0:56:47"} +{"current_steps": 2330, "total_steps": 2512, "loss": 0.0146, "lr": 1.3173255597963397e-06, "epoch": 7.420382165605096, "percentage": 92.75, "elapsed_time": "11:47:14", "remaining_time": "0:55:14"} +{"current_steps": 2335, "total_steps": 2512, "loss": 0.0142, "lr": 1.2466205225108273e-06, "epoch": 7.436305732484076, "percentage": 92.95, "elapsed_time": "11:48:21", "remaining_time": "0:53:41"} +{"current_steps": 2340, "total_steps": 2512, "loss": 0.0147, "lr": 1.1778417258471507e-06, "epoch": 7.452229299363057, "percentage": 93.15, "elapsed_time": "11:49:28", "remaining_time": "0:52:08"} +{"current_steps": 2345, "total_steps": 2512, "loss": 0.0145, "lr": 1.1109918872479642e-06, "epoch": 7.468152866242038, "percentage": 93.35, "elapsed_time": "11:50:34", "remaining_time": "0:50:36"} +{"current_steps": 2350, "total_steps": 2512, "loss": 0.0148, "lr": 1.0460736479429633e-06, "epoch": 7.484076433121019, "percentage": 93.55, "elapsed_time": "11:51:40", "remaining_time": "0:49:03"} +{"current_steps": 2355, "total_steps": 2512, "loss": 0.014, "lr": 9.830895728446143e-07, "epoch": 7.5, "percentage": 93.75, "elapsed_time": "11:52:48", "remaining_time": "0:47:31"} +{"current_steps": 2360, "total_steps": 2512, "loss": 0.0149, "lr": 9.220421504467281e-07, "epoch": 7.515923566878981, "percentage": 93.95, "elapsed_time": "11:53:58", "remaining_time": "0:45:59"} +{"current_steps": 2365, "total_steps": 2512, "loss": 0.0137, "lr": 8.629337927262171e-07, "epoch": 7.531847133757962, "percentage": 94.15, "elapsed_time": "11:55:08", "remaining_time": "0:44:27"} +{"current_steps": 2370, "total_steps": 2512, "loss": 0.0146, "lr": 8.057668350477332e-07, "epoch": 7.547770700636943, "percentage": 94.35, "elapsed_time": "11:56:12", "remaining_time": "0:42:54"} +{"current_steps": 2375, "total_steps": 2512, "loss": 0.014, "lr": 7.505435360714408e-07, "epoch": 7.563694267515924, "percentage": 94.55, "elapsed_time": "11:57:23", "remaining_time": "0:41:22"} +{"current_steps": 2380, "total_steps": 2512, "loss": 0.0143, "lr": 6.972660776637663e-07, "epoch": 7.579617834394904, "percentage": 94.75, "elapsed_time": "11:58:30", "remaining_time": "0:39:51"} +{"current_steps": 2385, "total_steps": 2512, "loss": 0.0149, "lr": 6.459365648111948e-07, "epoch": 7.595541401273885, "percentage": 94.94, "elapsed_time": "11:59:35", "remaining_time": "0:38:19"} +{"current_steps": 2390, "total_steps": 2512, "loss": 0.0142, "lr": 5.965570255370866e-07, "epoch": 7.611464968152866, "percentage": 95.14, "elapsed_time": "12:00:41", "remaining_time": "0:36:47"} +{"current_steps": 2395, "total_steps": 2512, "loss": 0.0139, "lr": 5.491294108215695e-07, "epoch": 7.627388535031847, "percentage": 95.34, "elapsed_time": "12:01:48", "remaining_time": "0:35:15"} +{"current_steps": 2400, "total_steps": 2512, "loss": 0.0147, "lr": 5.036555945244436e-07, "epoch": 7.643312101910828, "percentage": 95.54, "elapsed_time": "12:02:56", "remaining_time": "0:33:44"} +{"current_steps": 2405, "total_steps": 2512, "loss": 0.0142, "lr": 4.6013737331115913e-07, "epoch": 7.659235668789809, "percentage": 95.74, "elapsed_time": "12:05:06", "remaining_time": "0:32:15"} +{"current_steps": 2410, "total_steps": 2512, "loss": 0.0143, "lr": 4.1857646658180483e-07, "epoch": 7.67515923566879, "percentage": 95.94, "elapsed_time": "12:06:12", "remaining_time": "0:30:44"} +{"current_steps": 2415, "total_steps": 2512, "loss": 0.0139, "lr": 3.7897451640321323e-07, "epoch": 7.69108280254777, "percentage": 96.14, "elapsed_time": "12:07:18", "remaining_time": "0:29:12"} +{"current_steps": 2420, "total_steps": 2512, "loss": 0.0144, "lr": 3.413330874440401e-07, "epoch": 7.707006369426751, "percentage": 96.34, "elapsed_time": "12:08:24", "remaining_time": "0:27:41"} +{"current_steps": 2425, "total_steps": 2512, "loss": 0.015, "lr": 3.056536669129806e-07, "epoch": 7.722929936305732, "percentage": 96.54, "elapsed_time": "12:09:36", "remaining_time": "0:26:10"} +{"current_steps": 2430, "total_steps": 2512, "loss": 0.0148, "lr": 2.7193766449997717e-07, "epoch": 7.738853503184713, "percentage": 96.74, "elapsed_time": "12:10:45", "remaining_time": "0:24:39"} +{"current_steps": 2435, "total_steps": 2512, "loss": 0.0146, "lr": 2.4018641232055326e-07, "epoch": 7.754777070063694, "percentage": 96.93, "elapsed_time": "12:11:55", "remaining_time": "0:23:08"} +{"current_steps": 2440, "total_steps": 2512, "loss": 0.0143, "lr": 2.1040116486317207e-07, "epoch": 7.770700636942675, "percentage": 97.13, "elapsed_time": "12:13:04", "remaining_time": "0:21:37"} +{"current_steps": 2445, "total_steps": 2512, "loss": 0.0145, "lr": 1.8258309893965375e-07, "epoch": 7.786624203821656, "percentage": 97.33, "elapsed_time": "12:14:09", "remaining_time": "0:20:07"} +{"current_steps": 2450, "total_steps": 2512, "loss": 0.0147, "lr": 1.567333136387017e-07, "epoch": 7.802547770700637, "percentage": 97.53, "elapsed_time": "12:15:15", "remaining_time": "0:18:36"} +{"current_steps": 2455, "total_steps": 2512, "loss": 0.014, "lr": 1.3285283028247053e-07, "epoch": 7.818471337579618, "percentage": 97.73, "elapsed_time": "12:16:23", "remaining_time": "0:17:05"} +{"current_steps": 2460, "total_steps": 2512, "loss": 0.0142, "lr": 1.1094259238620952e-07, "epoch": 7.834394904458598, "percentage": 97.93, "elapsed_time": "12:17:30", "remaining_time": "0:15:35"} +{"current_steps": 2465, "total_steps": 2512, "loss": 0.0138, "lr": 9.100346562099238e-08, "epoch": 7.850318471337579, "percentage": 98.13, "elapsed_time": "12:18:38", "remaining_time": "0:14:05"} +{"current_steps": 2470, "total_steps": 2512, "loss": 0.0149, "lr": 7.30362377795113e-08, "epoch": 7.86624203821656, "percentage": 98.33, "elapsed_time": "12:19:50", "remaining_time": "0:12:34"} +{"current_steps": 2475, "total_steps": 2512, "loss": 0.0145, "lr": 5.704161874494629e-08, "epoch": 7.882165605095541, "percentage": 98.53, "elapsed_time": "12:20:57", "remaining_time": "0:11:04"} +{"current_steps": 2480, "total_steps": 2512, "loss": 0.0147, "lr": 4.302024046292097e-08, "epoch": 7.898089171974522, "percentage": 98.73, "elapsed_time": "12:22:07", "remaining_time": "0:09:34"} +{"current_steps": 2485, "total_steps": 2512, "loss": 0.0143, "lr": 3.0972656916539166e-08, "epoch": 7.914012738853503, "percentage": 98.93, "elapsed_time": "12:23:17", "remaining_time": "0:08:04"} +{"current_steps": 2490, "total_steps": 2512, "loss": 0.0144, "lr": 2.089934410449135e-08, "epoch": 7.929936305732484, "percentage": 99.12, "elapsed_time": "12:24:24", "remaining_time": "0:06:34"} +{"current_steps": 2495, "total_steps": 2512, "loss": 0.0138, "lr": 1.2800700022252975e-08, "epoch": 7.945859872611465, "percentage": 99.32, "elapsed_time": "12:25:29", "remaining_time": "0:05:04"} +{"current_steps": 2500, "total_steps": 2512, "loss": 0.0143, "lr": 6.677044646352659e-09, "epoch": 7.961783439490446, "percentage": 99.52, "elapsed_time": "12:26:38", "remaining_time": "0:03:35"} +{"current_steps": 2505, "total_steps": 2512, "loss": 0.0149, "lr": 2.528619921732256e-09, "epoch": 7.977707006369426, "percentage": 99.72, "elapsed_time": "12:28:45", "remaining_time": "0:02:05"} +{"current_steps": 2510, "total_steps": 2512, "loss": 0.0141, "lr": 3.555897522045104e-10, "epoch": 7.993630573248407, "percentage": 99.92, "elapsed_time": "12:29:52", "remaining_time": "0:00:35"} +{"current_steps": 2512, "total_steps": 2512, "epoch": 8.0, "percentage": 100.0, "elapsed_time": "12:31:49", "remaining_time": "0:00:00"} diff --git a/trainer_state.json b/trainer_state.json new file mode 100644 index 0000000..0c36582 --- /dev/null +++ b/trainer_state.json @@ -0,0 +1,5569 @@ +{ + "best_global_step": null, + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 8.0, + "eval_steps": 500, + "global_step": 2512, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.01592356687898089, + "grad_norm": 5.1627424708602865, + "learning_rate": 3.0769230769230774e-05, + "loss": 0.6402, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.5630480051040649, + "step": 5, + "valid_targets_mean": 5091.6, + "valid_targets_min": 2994 + }, + { + "epoch": 0.03184713375796178, + "grad_norm": 0.8147077629853552, + "learning_rate": 6.923076923076924e-05, + "loss": 0.4346, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.38636261224746704, + "step": 10, + "valid_targets_mean": 4814.8, + "valid_targets_min": 1220 + }, + { + "epoch": 0.04777070063694268, + "grad_norm": 0.6154406131503678, + "learning_rate": 9.999996048998592e-05, + "loss": 0.3816, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3615490794181824, + "step": 15, + "valid_targets_mean": 4980.2, + "valid_targets_min": 1527 + }, + { + "epoch": 0.06369426751592357, + "grad_norm": 0.3780803766484798, + "learning_rate": 9.999857764604897e-05, + "loss": 0.3323, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.30556270480155945, + "step": 20, + "valid_targets_mean": 4968.2, + "valid_targets_min": 2084 + }, + { + "epoch": 0.07961783439490445, + "grad_norm": 0.4181503335353106, + "learning_rate": 9.999521936384848e-05, + "loss": 0.3195, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.3232268691062927, + "step": 25, + "valid_targets_mean": 5354.2, + "valid_targets_min": 2594 + }, + { + "epoch": 0.09554140127388536, + "grad_norm": 0.36639127650079945, + "learning_rate": 9.998988577606981e-05, + "loss": 0.3016, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2855026423931122, + "step": 30, + "valid_targets_mean": 5369.9, + "valid_targets_min": 2402 + }, + { + "epoch": 0.11146496815286625, + "grad_norm": 0.3856048857076572, + "learning_rate": 9.998257709344246e-05, + "loss": 0.2895, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2963900566101074, + "step": 35, + "valid_targets_mean": 5076.0, + "valid_targets_min": 1455 + }, + { + "epoch": 0.12738853503184713, + "grad_norm": 0.3568357342827682, + "learning_rate": 9.99732936047316e-05, + "loss": 0.2816, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.27956438064575195, + "step": 40, + "valid_targets_mean": 4995.9, + "valid_targets_min": 1142 + }, + { + "epoch": 0.14331210191082802, + "grad_norm": 0.40201031518512703, + "learning_rate": 9.99620356767269e-05, + "loss": 0.2908, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.27538031339645386, + "step": 45, + "valid_targets_mean": 4971.8, + "valid_targets_min": 1301 + }, + { + "epoch": 0.1592356687898089, + "grad_norm": 0.3431144017209652, + "learning_rate": 9.994880375422783e-05, + "loss": 0.2846, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.28784871101379395, + "step": 50, + "valid_targets_mean": 5174.9, + "valid_targets_min": 1862 + }, + { + "epoch": 0.1751592356687898, + "grad_norm": 0.36286547477848036, + "learning_rate": 9.993359836002618e-05, + "loss": 0.2811, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2978650629520416, + "step": 55, + "valid_targets_mean": 5035.7, + "valid_targets_min": 2337 + }, + { + "epoch": 0.1910828025477707, + "grad_norm": 0.3371161108459167, + "learning_rate": 9.99164200948854e-05, + "loss": 0.277, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.29169130325317383, + "step": 60, + "valid_targets_mean": 5036.6, + "valid_targets_min": 2338 + }, + { + "epoch": 0.2070063694267516, + "grad_norm": 0.36824491112911706, + "learning_rate": 9.989726963751682e-05, + "loss": 0.2818, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2806537449359894, + "step": 65, + "valid_targets_mean": 5050.6, + "valid_targets_min": 2032 + }, + { + "epoch": 0.2229299363057325, + "grad_norm": 0.3350925520701665, + "learning_rate": 9.987614774455291e-05, + "loss": 0.2709, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.25298139452934265, + "step": 70, + "valid_targets_mean": 4749.6, + "valid_targets_min": 2198 + }, + { + "epoch": 0.23885350318471338, + "grad_norm": 0.30820780795004016, + "learning_rate": 9.985305525051732e-05, + "loss": 0.2791, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.25446945428848267, + "step": 75, + "valid_targets_mean": 4463.5, + "valid_targets_min": 1196 + }, + { + "epoch": 0.25477707006369427, + "grad_norm": 0.3343999813566176, + "learning_rate": 9.98279930677919e-05, + "loss": 0.2675, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.26154083013534546, + "step": 80, + "valid_targets_mean": 4928.5, + "valid_targets_min": 1309 + }, + { + "epoch": 0.27070063694267515, + "grad_norm": 0.33484829605039995, + "learning_rate": 9.980096218658075e-05, + "loss": 0.2702, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.28014805912971497, + "step": 85, + "valid_targets_mean": 4681.4, + "valid_targets_min": 1081 + }, + { + "epoch": 0.28662420382165604, + "grad_norm": 0.3584368616172035, + "learning_rate": 9.977196367487099e-05, + "loss": 0.2748, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2871628999710083, + "step": 90, + "valid_targets_mean": 4954.0, + "valid_targets_min": 2182 + }, + { + "epoch": 0.30254777070063693, + "grad_norm": 0.33726919200592925, + "learning_rate": 9.974099867839057e-05, + "loss": 0.2663, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2946035861968994, + "step": 95, + "valid_targets_mean": 5323.2, + "valid_targets_min": 3030 + }, + { + "epoch": 0.3184713375796178, + "grad_norm": 0.3353890833015672, + "learning_rate": 9.970806842056311e-05, + "loss": 0.2707, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.26885706186294556, + "step": 100, + "valid_targets_mean": 5171.5, + "valid_targets_min": 2061 + }, + { + "epoch": 0.3343949044585987, + "grad_norm": 0.3202356656812135, + "learning_rate": 9.967317420245943e-05, + "loss": 0.2688, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2668047547340393, + "step": 105, + "valid_targets_mean": 5587.8, + "valid_targets_min": 3096 + }, + { + "epoch": 0.3503184713375796, + "grad_norm": 0.3021804536694233, + "learning_rate": 9.963631740274622e-05, + "loss": 0.265, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.28386884927749634, + "step": 110, + "valid_targets_mean": 5126.0, + "valid_targets_min": 862 + }, + { + "epoch": 0.3662420382165605, + "grad_norm": 0.3334369990245805, + "learning_rate": 9.959749947763158e-05, + "loss": 0.2675, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.263800710439682, + "step": 115, + "valid_targets_mean": 4909.7, + "valid_targets_min": 1162 + }, + { + "epoch": 0.3821656050955414, + "grad_norm": 0.2949864480996826, + "learning_rate": 9.955672196080739e-05, + "loss": 0.2564, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2584902048110962, + "step": 120, + "valid_targets_mean": 5142.3, + "valid_targets_min": 1234 + }, + { + "epoch": 0.3980891719745223, + "grad_norm": 0.2914651429334071, + "learning_rate": 9.951398646338883e-05, + "loss": 0.2675, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2563759386539459, + "step": 125, + "valid_targets_mean": 5247.2, + "valid_targets_min": 1289 + }, + { + "epoch": 0.4140127388535032, + "grad_norm": 0.26880917531862386, + "learning_rate": 9.946929467385069e-05, + "loss": 0.248, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23924624919891357, + "step": 130, + "valid_targets_mean": 4959.4, + "valid_targets_min": 1226 + }, + { + "epoch": 0.4299363057324841, + "grad_norm": 0.3015826296305107, + "learning_rate": 9.94226483579606e-05, + "loss": 0.2656, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2536826729774475, + "step": 135, + "valid_targets_mean": 4916.9, + "valid_targets_min": 726 + }, + { + "epoch": 0.445859872611465, + "grad_norm": 0.3010552961474912, + "learning_rate": 9.937404935870938e-05, + "loss": 0.2458, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24484194815158844, + "step": 140, + "valid_targets_mean": 5002.2, + "valid_targets_min": 1078 + }, + { + "epoch": 0.46178343949044587, + "grad_norm": 0.2841868902626214, + "learning_rate": 9.932349959623803e-05, + "loss": 0.2523, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22377413511276245, + "step": 145, + "valid_targets_mean": 4933.1, + "valid_targets_min": 2704 + }, + { + "epoch": 0.47770700636942676, + "grad_norm": 0.29263201361908825, + "learning_rate": 9.927100106776212e-05, + "loss": 0.2608, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.25427815318107605, + "step": 150, + "valid_targets_mean": 4726.8, + "valid_targets_min": 1075 + }, + { + "epoch": 0.49363057324840764, + "grad_norm": 0.28068872626467783, + "learning_rate": 9.92165558474927e-05, + "loss": 0.265, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2538498640060425, + "step": 155, + "valid_targets_mean": 5207.2, + "valid_targets_min": 1446 + }, + { + "epoch": 0.5095541401273885, + "grad_norm": 0.2884340802705652, + "learning_rate": 9.916016608655435e-05, + "loss": 0.2487, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24591267108917236, + "step": 160, + "valid_targets_mean": 4500.6, + "valid_targets_min": 1102 + }, + { + "epoch": 0.5254777070063694, + "grad_norm": 0.2874037636407836, + "learning_rate": 9.91018340129003e-05, + "loss": 0.2566, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23591633141040802, + "step": 165, + "valid_targets_mean": 4635.1, + "valid_targets_min": 952 + }, + { + "epoch": 0.5414012738853503, + "grad_norm": 0.2730487924375769, + "learning_rate": 9.904156193122432e-05, + "loss": 0.2502, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24668367207050323, + "step": 170, + "valid_targets_mean": 5230.3, + "valid_targets_min": 1655 + }, + { + "epoch": 0.5573248407643312, + "grad_norm": 0.2785973316407628, + "learning_rate": 9.897935222286967e-05, + "loss": 0.2647, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2556725740432739, + "step": 175, + "valid_targets_mean": 5062.9, + "valid_targets_min": 1658 + }, + { + "epoch": 0.5732484076433121, + "grad_norm": 0.28238551486872815, + "learning_rate": 9.891520734573505e-05, + "loss": 0.2599, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.26440194249153137, + "step": 180, + "valid_targets_mean": 5090.6, + "valid_targets_min": 1215 + }, + { + "epoch": 0.589171974522293, + "grad_norm": 0.28261747155215106, + "learning_rate": 9.884912983417743e-05, + "loss": 0.2604, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2573757767677307, + "step": 185, + "valid_targets_mean": 5256.4, + "valid_targets_min": 1349 + }, + { + "epoch": 0.6050955414012739, + "grad_norm": 0.2798920367821444, + "learning_rate": 9.878112229891199e-05, + "loss": 0.2449, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2541036903858185, + "step": 190, + "valid_targets_mean": 5022.8, + "valid_targets_min": 2063 + }, + { + "epoch": 0.6210191082802548, + "grad_norm": 0.282425914451762, + "learning_rate": 9.87111874269089e-05, + "loss": 0.249, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.26237931847572327, + "step": 195, + "valid_targets_mean": 5177.3, + "valid_targets_min": 1598 + }, + { + "epoch": 0.6369426751592356, + "grad_norm": 0.29460873508613755, + "learning_rate": 9.86393279812872e-05, + "loss": 0.2574, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24381877481937408, + "step": 200, + "valid_targets_mean": 4988.2, + "valid_targets_min": 1304 + }, + { + "epoch": 0.6528662420382165, + "grad_norm": 0.2829320192597821, + "learning_rate": 9.856554680120565e-05, + "loss": 0.2595, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2609613537788391, + "step": 205, + "valid_targets_mean": 5108.7, + "valid_targets_min": 2200 + }, + { + "epoch": 0.6687898089171974, + "grad_norm": 0.26374886788691077, + "learning_rate": 9.848984680175049e-05, + "loss": 0.2508, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23660516738891602, + "step": 210, + "valid_targets_mean": 5015.1, + "valid_targets_min": 2166 + }, + { + "epoch": 0.6847133757961783, + "grad_norm": 0.32167959295537707, + "learning_rate": 9.841223097382029e-05, + "loss": 0.2553, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2493894398212433, + "step": 215, + "valid_targets_mean": 5226.3, + "valid_targets_min": 1195 + }, + { + "epoch": 0.7006369426751592, + "grad_norm": 0.287904583312802, + "learning_rate": 9.833270238400783e-05, + "loss": 0.2476, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2496083676815033, + "step": 220, + "valid_targets_mean": 5198.8, + "valid_targets_min": 1249 + }, + { + "epoch": 0.7165605095541401, + "grad_norm": 0.2714640083661649, + "learning_rate": 9.825126417447889e-05, + "loss": 0.2518, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.25480812788009644, + "step": 225, + "valid_targets_mean": 5315.4, + "valid_targets_min": 2701 + }, + { + "epoch": 0.732484076433121, + "grad_norm": 0.30093213872090957, + "learning_rate": 9.816791956284809e-05, + "loss": 0.2501, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24177277088165283, + "step": 230, + "valid_targets_mean": 4579.2, + "valid_targets_min": 716 + }, + { + "epoch": 0.7484076433121019, + "grad_norm": 0.28672939766760674, + "learning_rate": 9.808267184205183e-05, + "loss": 0.2544, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2554488182067871, + "step": 235, + "valid_targets_mean": 4890.7, + "valid_targets_min": 1127 + }, + { + "epoch": 0.7643312101910829, + "grad_norm": 0.28479383107443185, + "learning_rate": 9.799552438021807e-05, + "loss": 0.2522, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.25755855441093445, + "step": 240, + "valid_targets_mean": 4822.3, + "valid_targets_min": 703 + }, + { + "epoch": 0.7802547770700637, + "grad_norm": 0.24555896671998928, + "learning_rate": 9.79064806205334e-05, + "loss": 0.2466, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.25130271911621094, + "step": 245, + "valid_targets_mean": 4989.6, + "valid_targets_min": 1274 + }, + { + "epoch": 0.7961783439490446, + "grad_norm": 0.2533392245275226, + "learning_rate": 9.781554408110691e-05, + "loss": 0.2536, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.24182064831256866, + "step": 250, + "valid_targets_mean": 5014.8, + "valid_targets_min": 1430 + }, + { + "epoch": 0.8121019108280255, + "grad_norm": 0.281062255758068, + "learning_rate": 9.772271835483118e-05, + "loss": 0.2454, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.25493890047073364, + "step": 255, + "valid_targets_mean": 4887.8, + "valid_targets_min": 1051 + }, + { + "epoch": 0.8280254777070064, + "grad_norm": 0.2641955226407355, + "learning_rate": 9.762800710924038e-05, + "loss": 0.2449, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23709511756896973, + "step": 260, + "valid_targets_mean": 4766.1, + "valid_targets_min": 1023 + }, + { + "epoch": 0.8439490445859873, + "grad_norm": 0.27528984334378526, + "learning_rate": 9.753141408636532e-05, + "loss": 0.2498, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2573673129081726, + "step": 265, + "valid_targets_mean": 5194.0, + "valid_targets_min": 1395 + }, + { + "epoch": 0.8598726114649682, + "grad_norm": 0.2617199349177837, + "learning_rate": 9.743294310258564e-05, + "loss": 0.2609, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.26471590995788574, + "step": 270, + "valid_targets_mean": 4979.1, + "valid_targets_min": 660 + }, + { + "epoch": 0.8757961783439491, + "grad_norm": 0.2500164344985893, + "learning_rate": 9.733259804847901e-05, + "loss": 0.2395, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22313503921031952, + "step": 275, + "valid_targets_mean": 4646.9, + "valid_targets_min": 566 + }, + { + "epoch": 0.89171974522293, + "grad_norm": 0.29151337748052186, + "learning_rate": 9.723038288866738e-05, + "loss": 0.2567, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2850560247898102, + "step": 280, + "valid_targets_mean": 5299.6, + "valid_targets_min": 1619 + }, + { + "epoch": 0.9076433121019108, + "grad_norm": 0.25102648506956543, + "learning_rate": 9.712630166166039e-05, + "loss": 0.2417, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23329219222068787, + "step": 285, + "valid_targets_mean": 5375.3, + "valid_targets_min": 3703 + }, + { + "epoch": 0.9235668789808917, + "grad_norm": 0.27069275921174357, + "learning_rate": 9.70203584796958e-05, + "loss": 0.2544, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2480543851852417, + "step": 290, + "valid_targets_mean": 5204.4, + "valid_targets_min": 1286 + }, + { + "epoch": 0.9394904458598726, + "grad_norm": 0.2447400488725494, + "learning_rate": 9.6912557528577e-05, + "loss": 0.2438, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.26126471161842346, + "step": 295, + "valid_targets_mean": 5252.6, + "valid_targets_min": 2766 + }, + { + "epoch": 0.9554140127388535, + "grad_norm": 0.2818641601270585, + "learning_rate": 9.680290306750761e-05, + "loss": 0.2487, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2620047330856323, + "step": 300, + "valid_targets_mean": 5027.5, + "valid_targets_min": 1977 + }, + { + "epoch": 0.9713375796178344, + "grad_norm": 0.30561302087425307, + "learning_rate": 9.669139942892323e-05, + "loss": 0.242, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22884969413280487, + "step": 305, + "valid_targets_mean": 4986.8, + "valid_targets_min": 2295 + }, + { + "epoch": 0.9872611464968153, + "grad_norm": 0.24115237094564, + "learning_rate": 9.657805101832029e-05, + "loss": 0.2385, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21688750386238098, + "step": 310, + "valid_targets_mean": 4624.2, + "valid_targets_min": 1232 + }, + { + "epoch": 1.0031847133757963, + "grad_norm": 0.2845785880054518, + "learning_rate": 9.646286231408194e-05, + "loss": 0.2349, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20368462800979614, + "step": 315, + "valid_targets_mean": 4959.3, + "valid_targets_min": 1277 + }, + { + "epoch": 1.019108280254777, + "grad_norm": 0.2680394784068739, + "learning_rate": 9.63458378673011e-05, + "loss": 0.2006, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1978563666343689, + "step": 320, + "valid_targets_mean": 5383.2, + "valid_targets_min": 3672 + }, + { + "epoch": 1.035031847133758, + "grad_norm": 0.23604148080196144, + "learning_rate": 9.62269823016007e-05, + "loss": 0.2014, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2006090134382248, + "step": 325, + "valid_targets_mean": 5250.4, + "valid_targets_min": 3586 + }, + { + "epoch": 1.0509554140127388, + "grad_norm": 0.24431555149300851, + "learning_rate": 9.610630031295098e-05, + "loss": 0.2018, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20900499820709229, + "step": 330, + "valid_targets_mean": 5313.5, + "valid_targets_min": 3708 + }, + { + "epoch": 1.0668789808917198, + "grad_norm": 0.26055650549596177, + "learning_rate": 9.598379666948394e-05, + "loss": 0.2066, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2136392593383789, + "step": 335, + "valid_targets_mean": 5156.7, + "valid_targets_min": 917 + }, + { + "epoch": 1.0828025477707006, + "grad_norm": 0.24476549624254335, + "learning_rate": 9.585947621130496e-05, + "loss": 0.1968, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2064034640789032, + "step": 340, + "valid_targets_mean": 5306.4, + "valid_targets_min": 1654 + }, + { + "epoch": 1.0987261146496816, + "grad_norm": 0.2499311089114423, + "learning_rate": 9.573334385030157e-05, + "loss": 0.1962, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21572357416152954, + "step": 345, + "valid_targets_mean": 5514.9, + "valid_targets_min": 3250 + }, + { + "epoch": 1.1146496815286624, + "grad_norm": 0.23988691755478722, + "learning_rate": 9.56054045699494e-05, + "loss": 0.2086, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21599572896957397, + "step": 350, + "valid_targets_mean": 5453.1, + "valid_targets_min": 3602 + }, + { + "epoch": 1.1305732484076434, + "grad_norm": 0.2420275375960929, + "learning_rate": 9.547566342511521e-05, + "loss": 0.2018, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19987857341766357, + "step": 355, + "valid_targets_mean": 5438.3, + "valid_targets_min": 484 + }, + { + "epoch": 1.1464968152866242, + "grad_norm": 0.2531962652710879, + "learning_rate": 9.534412554185725e-05, + "loss": 0.195, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1841672658920288, + "step": 360, + "valid_targets_mean": 4668.0, + "valid_targets_min": 1333 + }, + { + "epoch": 1.1624203821656052, + "grad_norm": 0.24395094946874976, + "learning_rate": 9.521079611722277e-05, + "loss": 0.2032, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20645533502101898, + "step": 365, + "valid_targets_mean": 5059.7, + "valid_targets_min": 1401 + }, + { + "epoch": 1.178343949044586, + "grad_norm": 0.2601752111187968, + "learning_rate": 9.507568041904251e-05, + "loss": 0.2185, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21245883405208588, + "step": 370, + "valid_targets_mean": 5499.1, + "valid_targets_min": 2109 + }, + { + "epoch": 1.194267515923567, + "grad_norm": 0.23557950602226088, + "learning_rate": 9.493878378572276e-05, + "loss": 0.2057, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20772235095500946, + "step": 375, + "valid_targets_mean": 5156.8, + "valid_targets_min": 2048 + }, + { + "epoch": 1.2101910828025477, + "grad_norm": 0.23422634109351587, + "learning_rate": 9.480011162603434e-05, + "loss": 0.2051, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21189646422863007, + "step": 380, + "valid_targets_mean": 5444.8, + "valid_targets_min": 2801 + }, + { + "epoch": 1.2261146496815287, + "grad_norm": 0.24554530674289568, + "learning_rate": 9.465966941889892e-05, + "loss": 0.2024, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19627982378005981, + "step": 385, + "valid_targets_mean": 5206.1, + "valid_targets_min": 2062 + }, + { + "epoch": 1.2420382165605095, + "grad_norm": 0.2532907702684435, + "learning_rate": 9.451746271317252e-05, + "loss": 0.2004, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19581934809684753, + "step": 390, + "valid_targets_mean": 4816.7, + "valid_targets_min": 818 + }, + { + "epoch": 1.2579617834394905, + "grad_norm": 0.25284504947538317, + "learning_rate": 9.437349712742634e-05, + "loss": 0.216, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22710266709327698, + "step": 395, + "valid_targets_mean": 5238.6, + "valid_targets_min": 716 + }, + { + "epoch": 1.2738853503184713, + "grad_norm": 0.24023997999848976, + "learning_rate": 9.422777834972474e-05, + "loss": 0.2096, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20321109890937805, + "step": 400, + "valid_targets_mean": 5203.4, + "valid_targets_min": 2219 + }, + { + "epoch": 1.2898089171974523, + "grad_norm": 0.278870280422163, + "learning_rate": 9.408031213740045e-05, + "loss": 0.2018, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20658862590789795, + "step": 405, + "valid_targets_mean": 4992.2, + "valid_targets_min": 2446 + }, + { + "epoch": 1.305732484076433, + "grad_norm": 0.25264505470430226, + "learning_rate": 9.393110431682721e-05, + "loss": 0.2012, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19869352877140045, + "step": 410, + "valid_targets_mean": 4699.7, + "valid_targets_min": 1668 + }, + { + "epoch": 1.321656050955414, + "grad_norm": 0.22730971603092698, + "learning_rate": 9.378016078318947e-05, + "loss": 0.2078, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20935168862342834, + "step": 415, + "valid_targets_mean": 5081.5, + "valid_targets_min": 1814 + }, + { + "epoch": 1.3375796178343948, + "grad_norm": 0.2485975944364818, + "learning_rate": 9.362748750024954e-05, + "loss": 0.2043, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2036893665790558, + "step": 420, + "valid_targets_mean": 5002.9, + "valid_targets_min": 1779 + }, + { + "epoch": 1.3535031847133758, + "grad_norm": 0.2597841965797303, + "learning_rate": 9.347309050011187e-05, + "loss": 0.2016, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2176981270313263, + "step": 425, + "valid_targets_mean": 4902.8, + "valid_targets_min": 1108 + }, + { + "epoch": 1.3694267515923566, + "grad_norm": 0.23978756728336165, + "learning_rate": 9.331697588298487e-05, + "loss": 0.2087, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1992342621088028, + "step": 430, + "valid_targets_mean": 5103.2, + "valid_targets_min": 1142 + }, + { + "epoch": 1.3853503184713376, + "grad_norm": 0.24275987837261093, + "learning_rate": 9.315914981693974e-05, + "loss": 0.2112, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20987510681152344, + "step": 435, + "valid_targets_mean": 4843.6, + "valid_targets_min": 1202 + }, + { + "epoch": 1.4012738853503186, + "grad_norm": 0.22937721538701838, + "learning_rate": 9.29996185376669e-05, + "loss": 0.2035, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20019900798797607, + "step": 440, + "valid_targets_mean": 4867.0, + "valid_targets_min": 1708 + }, + { + "epoch": 1.4171974522292994, + "grad_norm": 0.23208736065714275, + "learning_rate": 9.283838834822948e-05, + "loss": 0.1951, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18596212565898895, + "step": 445, + "valid_targets_mean": 4923.2, + "valid_targets_min": 1619 + }, + { + "epoch": 1.4331210191082802, + "grad_norm": 0.24358434190564066, + "learning_rate": 9.267546561881442e-05, + "loss": 0.2064, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2117403745651245, + "step": 450, + "valid_targets_mean": 4836.3, + "valid_targets_min": 548 + }, + { + "epoch": 1.4490445859872612, + "grad_norm": 0.2577205056344599, + "learning_rate": 9.251085678648072e-05, + "loss": 0.2103, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21646203100681305, + "step": 455, + "valid_targets_mean": 5223.7, + "valid_targets_min": 2350 + }, + { + "epoch": 1.4649681528662422, + "grad_norm": 0.24549502094036066, + "learning_rate": 9.234456835490509e-05, + "loss": 0.2054, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20974838733673096, + "step": 460, + "valid_targets_mean": 5048.9, + "valid_targets_min": 1223 + }, + { + "epoch": 1.480891719745223, + "grad_norm": 0.22761640085028945, + "learning_rate": 9.217660689412504e-05, + "loss": 0.2003, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18818441033363342, + "step": 465, + "valid_targets_mean": 5274.9, + "valid_targets_min": 3506 + }, + { + "epoch": 1.4968152866242037, + "grad_norm": 0.25396134639384105, + "learning_rate": 9.200697904027928e-05, + "loss": 0.197, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19861872494220734, + "step": 470, + "valid_targets_mean": 4838.4, + "valid_targets_min": 1290 + }, + { + "epoch": 1.5127388535031847, + "grad_norm": 0.22936303489921805, + "learning_rate": 9.183569149534554e-05, + "loss": 0.2077, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19787994027137756, + "step": 475, + "valid_targets_mean": 4960.2, + "valid_targets_min": 2740 + }, + { + "epoch": 1.5286624203821657, + "grad_norm": 0.21150831228166767, + "learning_rate": 9.166275102687573e-05, + "loss": 0.2038, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19011111557483673, + "step": 480, + "valid_targets_mean": 5233.5, + "valid_targets_min": 2403 + }, + { + "epoch": 1.5445859872611465, + "grad_norm": 0.21699177217917037, + "learning_rate": 9.148816446772858e-05, + "loss": 0.2009, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19189313054084778, + "step": 485, + "valid_targets_mean": 5160.0, + "valid_targets_min": 2060 + }, + { + "epoch": 1.5605095541401273, + "grad_norm": 0.2425311463701338, + "learning_rate": 9.131193871579975e-05, + "loss": 0.2, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20605826377868652, + "step": 490, + "valid_targets_mean": 5393.8, + "valid_targets_min": 2515 + }, + { + "epoch": 1.5764331210191083, + "grad_norm": 0.24871001658908948, + "learning_rate": 9.113408073374913e-05, + "loss": 0.2044, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2023090273141861, + "step": 495, + "valid_targets_mean": 4494.1, + "valid_targets_min": 914 + }, + { + "epoch": 1.5923566878980893, + "grad_norm": 0.2452856486267608, + "learning_rate": 9.09545975487259e-05, + "loss": 0.2096, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1939089298248291, + "step": 500, + "valid_targets_mean": 5050.0, + "valid_targets_min": 703 + }, + { + "epoch": 1.60828025477707, + "grad_norm": 0.22823823865021425, + "learning_rate": 9.07734962520908e-05, + "loss": 0.2083, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2074129581451416, + "step": 505, + "valid_targets_mean": 5134.5, + "valid_targets_min": 3102 + }, + { + "epoch": 1.6242038216560508, + "grad_norm": 0.24364755895874365, + "learning_rate": 9.059078399913604e-05, + "loss": 0.2083, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20501920580863953, + "step": 510, + "valid_targets_mean": 4935.5, + "valid_targets_min": 635 + }, + { + "epoch": 1.6401273885350318, + "grad_norm": 0.25880710753950514, + "learning_rate": 9.040646800880244e-05, + "loss": 0.206, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20811201632022858, + "step": 515, + "valid_targets_mean": 4659.3, + "valid_targets_min": 1106 + }, + { + "epoch": 1.6560509554140128, + "grad_norm": 0.23263314164438118, + "learning_rate": 9.022055556339439e-05, + "loss": 0.21, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19428785145282745, + "step": 520, + "valid_targets_mean": 5065.1, + "valid_targets_min": 1215 + }, + { + "epoch": 1.6719745222929936, + "grad_norm": 0.234397461703385, + "learning_rate": 9.0033054008292e-05, + "loss": 0.203, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20847736299037933, + "step": 525, + "valid_targets_mean": 5439.4, + "valid_targets_min": 1968 + }, + { + "epoch": 1.6878980891719744, + "grad_norm": 0.22762995526264007, + "learning_rate": 8.984397075166095e-05, + "loss": 0.1995, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19424715638160706, + "step": 530, + "valid_targets_mean": 4820.5, + "valid_targets_min": 1514 + }, + { + "epoch": 1.7038216560509554, + "grad_norm": 0.23956784661943545, + "learning_rate": 8.965331326415976e-05, + "loss": 0.2081, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2057517021894455, + "step": 535, + "valid_targets_mean": 4696.5, + "valid_targets_min": 870 + }, + { + "epoch": 1.7197452229299364, + "grad_norm": 0.21787956908209896, + "learning_rate": 8.946108907864461e-05, + "loss": 0.2066, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19403384625911713, + "step": 540, + "valid_targets_mean": 4905.6, + "valid_targets_min": 1078 + }, + { + "epoch": 1.7356687898089171, + "grad_norm": 0.24131081747981786, + "learning_rate": 8.926730578987182e-05, + "loss": 0.2067, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19468069076538086, + "step": 545, + "valid_targets_mean": 4975.8, + "valid_targets_min": 1183 + }, + { + "epoch": 1.7515923566878981, + "grad_norm": 0.2354088036703284, + "learning_rate": 8.907197105419761e-05, + "loss": 0.2085, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1884109079837799, + "step": 550, + "valid_targets_mean": 4935.5, + "valid_targets_min": 1231 + }, + { + "epoch": 1.767515923566879, + "grad_norm": 0.2510472016670687, + "learning_rate": 8.887509258927578e-05, + "loss": 0.2035, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2117602378129959, + "step": 555, + "valid_targets_mean": 4989.0, + "valid_targets_min": 1195 + }, + { + "epoch": 1.78343949044586, + "grad_norm": 0.23400858719031709, + "learning_rate": 8.867667817375266e-05, + "loss": 0.206, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20882730185985565, + "step": 560, + "valid_targets_mean": 5361.1, + "valid_targets_min": 1232 + }, + { + "epoch": 1.799363057324841, + "grad_norm": 0.22702047242000092, + "learning_rate": 8.847673564695982e-05, + "loss": 0.2062, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2094724327325821, + "step": 565, + "valid_targets_mean": 4931.7, + "valid_targets_min": 1792 + }, + { + "epoch": 1.8152866242038217, + "grad_norm": 0.22311418142380882, + "learning_rate": 8.827527290860436e-05, + "loss": 0.2041, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20011919736862183, + "step": 570, + "valid_targets_mean": 5126.8, + "valid_targets_min": 2900 + }, + { + "epoch": 1.8312101910828025, + "grad_norm": 0.21069413909753662, + "learning_rate": 8.807229791845673e-05, + "loss": 0.203, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19907280802726746, + "step": 575, + "valid_targets_mean": 5348.9, + "valid_targets_min": 2173 + }, + { + "epoch": 1.8471337579617835, + "grad_norm": 0.2475036713282529, + "learning_rate": 8.78678186960363e-05, + "loss": 0.206, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21200208365917206, + "step": 580, + "valid_targets_mean": 4922.2, + "valid_targets_min": 1170 + }, + { + "epoch": 1.8630573248407645, + "grad_norm": 0.22294237713372475, + "learning_rate": 8.766184332029452e-05, + "loss": 0.2036, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19675835967063904, + "step": 585, + "valid_targets_mean": 5348.2, + "valid_targets_min": 2828 + }, + { + "epoch": 1.8789808917197452, + "grad_norm": 0.23642124870002312, + "learning_rate": 8.745437992929566e-05, + "loss": 0.1995, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.20921610295772552, + "step": 590, + "valid_targets_mean": 5066.7, + "valid_targets_min": 802 + }, + { + "epoch": 1.894904458598726, + "grad_norm": 0.23661026062332122, + "learning_rate": 8.72454367198953e-05, + "loss": 0.2095, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.19710376858711243, + "step": 595, + "valid_targets_mean": 5375.9, + "valid_targets_min": 1329 + }, + { + "epoch": 1.910828025477707, + "grad_norm": 0.2384171973928502, + "learning_rate": 8.703502194741648e-05, + "loss": 0.1997, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.22492662072181702, + "step": 600, + "valid_targets_mean": 5207.3, + "valid_targets_min": 2681 + }, + { + "epoch": 1.926751592356688, + "grad_norm": 0.2239397861346745, + "learning_rate": 8.68231439253236e-05, + "loss": 0.212, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.21004718542099, + "step": 605, + "valid_targets_mean": 5322.6, + "valid_targets_min": 2556 + }, + { + "epoch": 1.9426751592356688, + "grad_norm": 0.21502779221175009, + "learning_rate": 8.660981102489378e-05, + "loss": 0.2072, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2046002298593521, + "step": 610, + "valid_targets_mean": 5065.5, + "valid_targets_min": 975 + }, + { + "epoch": 1.9585987261146496, + "grad_norm": 0.22749657474977258, + "learning_rate": 8.639503167488634e-05, + "loss": 0.2029, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.2107071578502655, + "step": 615, + "valid_targets_mean": 5353.4, + "valid_targets_min": 3570 + }, + { + "epoch": 1.9745222929936306, + "grad_norm": 0.2634121088266475, + "learning_rate": 8.617881436120956e-05, + "loss": 0.2124, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.23529648780822754, + "step": 620, + "valid_targets_mean": 5355.2, + "valid_targets_min": 1274 + }, + { + "epoch": 1.9904458598726116, + "grad_norm": 0.2197949781782732, + "learning_rate": 8.596116762658554e-05, + "loss": 0.2025, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.18133550882339478, + "step": 625, + "valid_targets_mean": 5175.6, + "valid_targets_min": 583 + }, + { + "epoch": 2.0063694267515926, + "grad_norm": 0.23730900272903072, + "learning_rate": 8.57421000702127e-05, + "loss": 0.1858, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1505005657672882, + "step": 630, + "valid_targets_mean": 5301.6, + "valid_targets_min": 2921 + }, + { + "epoch": 2.022292993630573, + "grad_norm": 0.2412353096015712, + "learning_rate": 8.552162034742592e-05, + "loss": 0.1483, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1498708426952362, + "step": 635, + "valid_targets_mean": 5001.7, + "valid_targets_min": 2420 + }, + { + "epoch": 2.038216560509554, + "grad_norm": 0.2647157369891964, + "learning_rate": 8.529973716935455e-05, + "loss": 0.1526, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1452397257089615, + "step": 640, + "valid_targets_mean": 4854.9, + "valid_targets_min": 968 + }, + { + "epoch": 2.054140127388535, + "grad_norm": 0.24725529975802696, + "learning_rate": 8.507645930257846e-05, + "loss": 0.1482, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15161120891571045, + "step": 645, + "valid_targets_mean": 5191.8, + "valid_targets_min": 1284 + }, + { + "epoch": 2.070063694267516, + "grad_norm": 0.24387182568674096, + "learning_rate": 8.485179556878138e-05, + "loss": 0.1498, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15403331816196442, + "step": 650, + "valid_targets_mean": 5309.0, + "valid_targets_min": 2957 + }, + { + "epoch": 2.0859872611464967, + "grad_norm": 0.23076865902846655, + "learning_rate": 8.462575484440254e-05, + "loss": 0.1506, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14719770848751068, + "step": 655, + "valid_targets_mean": 5252.0, + "valid_targets_min": 2874 + }, + { + "epoch": 2.1019108280254777, + "grad_norm": 0.2304361956081011, + "learning_rate": 8.439834606028594e-05, + "loss": 0.1502, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14514179527759552, + "step": 660, + "valid_targets_mean": 5230.5, + "valid_targets_min": 2740 + }, + { + "epoch": 2.1178343949044587, + "grad_norm": 0.24007719119217016, + "learning_rate": 8.416957820132742e-05, + "loss": 0.1485, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14976125955581665, + "step": 665, + "valid_targets_mean": 5018.0, + "valid_targets_min": 1196 + }, + { + "epoch": 2.1337579617834397, + "grad_norm": 0.23438933057204978, + "learning_rate": 8.393946030611978e-05, + "loss": 0.1528, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14520317316055298, + "step": 670, + "valid_targets_mean": 4836.7, + "valid_targets_min": 2691 + }, + { + "epoch": 2.1496815286624202, + "grad_norm": 0.2585283629564994, + "learning_rate": 8.370800146659557e-05, + "loss": 0.1527, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15580756962299347, + "step": 675, + "valid_targets_mean": 5326.6, + "valid_targets_min": 1226 + }, + { + "epoch": 2.1656050955414012, + "grad_norm": 0.2314160104801554, + "learning_rate": 8.347521082766783e-05, + "loss": 0.151, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1576583981513977, + "step": 680, + "valid_targets_mean": 4893.4, + "valid_targets_min": 1093 + }, + { + "epoch": 2.1815286624203822, + "grad_norm": 0.2459285547012428, + "learning_rate": 8.324109758686896e-05, + "loss": 0.1591, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15383420884609222, + "step": 685, + "valid_targets_mean": 4793.5, + "valid_targets_min": 1308 + }, + { + "epoch": 2.1974522292993632, + "grad_norm": 0.24200222895829962, + "learning_rate": 8.300567099398718e-05, + "loss": 0.1583, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1608743965625763, + "step": 690, + "valid_targets_mean": 5052.7, + "valid_targets_min": 1319 + }, + { + "epoch": 2.213375796178344, + "grad_norm": 0.23316700715567448, + "learning_rate": 8.276894035070107e-05, + "loss": 0.1565, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15711182355880737, + "step": 695, + "valid_targets_mean": 5332.2, + "valid_targets_min": 2222 + }, + { + "epoch": 2.229299363057325, + "grad_norm": 0.24239733039564118, + "learning_rate": 8.25309150102121e-05, + "loss": 0.1539, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15052467584609985, + "step": 700, + "valid_targets_mean": 5452.8, + "valid_targets_min": 3940 + }, + { + "epoch": 2.245222929936306, + "grad_norm": 0.22926561634141665, + "learning_rate": 8.229160437687515e-05, + "loss": 0.1483, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14362287521362305, + "step": 705, + "valid_targets_mean": 4927.2, + "valid_targets_min": 1537 + }, + { + "epoch": 2.261146496815287, + "grad_norm": 0.2374033514007257, + "learning_rate": 8.205101790582681e-05, + "loss": 0.1571, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16457954049110413, + "step": 710, + "valid_targets_mean": 5318.2, + "valid_targets_min": 2986 + }, + { + "epoch": 2.2770700636942673, + "grad_norm": 0.23830962553798735, + "learning_rate": 8.180916510261187e-05, + "loss": 0.1497, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14998547732830048, + "step": 715, + "valid_targets_mean": 5050.0, + "valid_targets_min": 2269 + }, + { + "epoch": 2.2929936305732483, + "grad_norm": 0.23211181937948858, + "learning_rate": 8.156605552280784e-05, + "loss": 0.1461, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15161485970020294, + "step": 720, + "valid_targets_mean": 5053.8, + "valid_targets_min": 1899 + }, + { + "epoch": 2.3089171974522293, + "grad_norm": 0.2303690432107466, + "learning_rate": 8.132169877164724e-05, + "loss": 0.1502, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1520197093486786, + "step": 725, + "valid_targets_mean": 5064.9, + "valid_targets_min": 811 + }, + { + "epoch": 2.3248407643312103, + "grad_norm": 0.22797918537262526, + "learning_rate": 8.107610450363825e-05, + "loss": 0.1508, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14326338469982147, + "step": 730, + "valid_targets_mean": 4740.8, + "valid_targets_min": 870 + }, + { + "epoch": 2.340764331210191, + "grad_norm": 0.2150859669229136, + "learning_rate": 8.082928242218322e-05, + "loss": 0.1509, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1416577249765396, + "step": 735, + "valid_targets_mean": 5201.0, + "valid_targets_min": 1688 + }, + { + "epoch": 2.356687898089172, + "grad_norm": 0.23258061805313435, + "learning_rate": 8.05812422791952e-05, + "loss": 0.153, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14608557522296906, + "step": 740, + "valid_targets_mean": 4888.4, + "valid_targets_min": 2468 + }, + { + "epoch": 2.372611464968153, + "grad_norm": 0.2530429637314677, + "learning_rate": 8.033199387471277e-05, + "loss": 0.1537, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15253382921218872, + "step": 745, + "valid_targets_mean": 4900.8, + "valid_targets_min": 752 + }, + { + "epoch": 2.388535031847134, + "grad_norm": 0.24409963649157015, + "learning_rate": 8.00815470565128e-05, + "loss": 0.1554, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1619432270526886, + "step": 750, + "valid_targets_mean": 4944.6, + "valid_targets_min": 1249 + }, + { + "epoch": 2.404458598726115, + "grad_norm": 0.24250859682796513, + "learning_rate": 7.98299117197213e-05, + "loss": 0.1559, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15755924582481384, + "step": 755, + "valid_targets_mean": 4652.1, + "valid_targets_min": 800 + }, + { + "epoch": 2.4203821656050954, + "grad_norm": 0.21661029592592831, + "learning_rate": 7.957709780642258e-05, + "loss": 0.1522, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.147942453622818, + "step": 760, + "valid_targets_mean": 5154.3, + "valid_targets_min": 2223 + }, + { + "epoch": 2.4363057324840764, + "grad_norm": 0.24131852195317538, + "learning_rate": 7.932311530526631e-05, + "loss": 0.1604, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16509908437728882, + "step": 765, + "valid_targets_mean": 5092.5, + "valid_targets_min": 2788 + }, + { + "epoch": 2.4522292993630574, + "grad_norm": 0.23111984257948573, + "learning_rate": 7.9067974251073e-05, + "loss": 0.1572, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15662288665771484, + "step": 770, + "valid_targets_mean": 5262.0, + "valid_targets_min": 1192 + }, + { + "epoch": 2.468152866242038, + "grad_norm": 0.22110200587462797, + "learning_rate": 7.881168472443743e-05, + "loss": 0.1554, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15480875968933105, + "step": 775, + "valid_targets_mean": 5155.2, + "valid_targets_min": 2491 + }, + { + "epoch": 2.484076433121019, + "grad_norm": 0.21486107383900602, + "learning_rate": 7.855425685133039e-05, + "loss": 0.1565, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1386668086051941, + "step": 780, + "valid_targets_mean": 5114.8, + "valid_targets_min": 2148 + }, + { + "epoch": 2.5, + "grad_norm": 0.2501031614898536, + "learning_rate": 7.829570080269864e-05, + "loss": 0.1571, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15886273980140686, + "step": 785, + "valid_targets_mean": 5275.2, + "valid_targets_min": 1223 + }, + { + "epoch": 2.515923566878981, + "grad_norm": 0.23806768061702227, + "learning_rate": 7.8036026794063e-05, + "loss": 0.152, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14196041226387024, + "step": 790, + "valid_targets_mean": 4949.9, + "valid_targets_min": 2766 + }, + { + "epoch": 2.531847133757962, + "grad_norm": 0.24367592773568758, + "learning_rate": 7.777524508511481e-05, + "loss": 0.153, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16322669386863708, + "step": 795, + "valid_targets_mean": 5098.2, + "valid_targets_min": 1127 + }, + { + "epoch": 2.5477707006369426, + "grad_norm": 0.23909231626487618, + "learning_rate": 7.75133659793105e-05, + "loss": 0.154, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16453084349632263, + "step": 800, + "valid_targets_mean": 5428.0, + "valid_targets_min": 2597 + }, + { + "epoch": 2.5636942675159236, + "grad_norm": 0.25778743240059054, + "learning_rate": 7.725039982346448e-05, + "loss": 0.1521, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15340283513069153, + "step": 805, + "valid_targets_mean": 5038.6, + "valid_targets_min": 1260 + }, + { + "epoch": 2.5796178343949046, + "grad_norm": 0.22578388830114626, + "learning_rate": 7.698635700734048e-05, + "loss": 0.1551, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1443755328655243, + "step": 810, + "valid_targets_mean": 5271.2, + "valid_targets_min": 2295 + }, + { + "epoch": 2.595541401273885, + "grad_norm": 0.33141959752565586, + "learning_rate": 7.672124796324087e-05, + "loss": 0.1549, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15744054317474365, + "step": 815, + "valid_targets_mean": 4810.2, + "valid_targets_min": 1527 + }, + { + "epoch": 2.611464968152866, + "grad_norm": 0.2941528575710284, + "learning_rate": 7.645508316559466e-05, + "loss": 0.1522, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14850759506225586, + "step": 820, + "valid_targets_mean": 4924.2, + "valid_targets_min": 862 + }, + { + "epoch": 2.627388535031847, + "grad_norm": 0.21691189283291645, + "learning_rate": 7.61878731305435e-05, + "loss": 0.1556, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14406980574131012, + "step": 825, + "valid_targets_mean": 5178.5, + "valid_targets_min": 2668 + }, + { + "epoch": 2.643312101910828, + "grad_norm": 0.23483947467164856, + "learning_rate": 7.591962841552627e-05, + "loss": 0.1544, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15476183593273163, + "step": 830, + "valid_targets_mean": 5098.6, + "valid_targets_min": 1524 + }, + { + "epoch": 2.659235668789809, + "grad_norm": 0.24464014325931902, + "learning_rate": 7.56503596188619e-05, + "loss": 0.1619, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16757142543792725, + "step": 835, + "valid_targets_mean": 5398.4, + "valid_targets_min": 3732 + }, + { + "epoch": 2.6751592356687897, + "grad_norm": 0.24115383252435188, + "learning_rate": 7.53800773793308e-05, + "loss": 0.1542, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14994437992572784, + "step": 840, + "valid_targets_mean": 4932.7, + "valid_targets_min": 692 + }, + { + "epoch": 2.6910828025477707, + "grad_norm": 0.23454958700281958, + "learning_rate": 7.510879237575423e-05, + "loss": 0.1566, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15883654356002808, + "step": 845, + "valid_targets_mean": 5411.8, + "valid_targets_min": 2131 + }, + { + "epoch": 2.7070063694267517, + "grad_norm": 0.24662992906592007, + "learning_rate": 7.483651532657266e-05, + "loss": 0.1595, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17014352977275848, + "step": 850, + "valid_targets_mean": 4877.6, + "valid_targets_min": 1133 + }, + { + "epoch": 2.722929936305732, + "grad_norm": 0.22928830252812615, + "learning_rate": 7.456325698942214e-05, + "loss": 0.1574, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15127910673618317, + "step": 855, + "valid_targets_mean": 4945.2, + "valid_targets_min": 1116 + }, + { + "epoch": 2.738853503184713, + "grad_norm": 0.22292883224529947, + "learning_rate": 7.428902816070931e-05, + "loss": 0.1511, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15146589279174805, + "step": 860, + "valid_targets_mean": 5436.8, + "valid_targets_min": 3798 + }, + { + "epoch": 2.754777070063694, + "grad_norm": 0.28572001733373326, + "learning_rate": 7.401383967518478e-05, + "loss": 0.1599, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.17128580808639526, + "step": 865, + "valid_targets_mean": 5143.6, + "valid_targets_min": 1475 + }, + { + "epoch": 2.770700636942675, + "grad_norm": 0.22523465296680736, + "learning_rate": 7.373770240551514e-05, + "loss": 0.1581, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15213854610919952, + "step": 870, + "valid_targets_mean": 5006.6, + "valid_targets_min": 1135 + }, + { + "epoch": 2.786624203821656, + "grad_norm": 0.22480118592113466, + "learning_rate": 7.346062726185332e-05, + "loss": 0.1553, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15178827941417694, + "step": 875, + "valid_targets_mean": 4833.5, + "valid_targets_min": 730 + }, + { + "epoch": 2.802547770700637, + "grad_norm": 0.22014661615954179, + "learning_rate": 7.318262519140755e-05, + "loss": 0.1591, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14908280968666077, + "step": 880, + "valid_targets_mean": 4941.3, + "valid_targets_min": 1776 + }, + { + "epoch": 2.8184713375796178, + "grad_norm": 0.2280330702768935, + "learning_rate": 7.290370717800882e-05, + "loss": 0.1558, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14838773012161255, + "step": 885, + "valid_targets_mean": 4711.5, + "valid_targets_min": 1441 + }, + { + "epoch": 2.8343949044585988, + "grad_norm": 0.2315325512585157, + "learning_rate": 7.262388424167698e-05, + "loss": 0.1585, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.16693618893623352, + "step": 890, + "valid_targets_mean": 5120.6, + "valid_targets_min": 2166 + }, + { + "epoch": 2.8503184713375798, + "grad_norm": 0.23297535309469486, + "learning_rate": 7.23431674381852e-05, + "loss": 0.1537, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15082749724388123, + "step": 895, + "valid_targets_mean": 4803.6, + "valid_targets_min": 818 + }, + { + "epoch": 2.8662420382165603, + "grad_norm": 0.21061800602197256, + "learning_rate": 7.20615678586233e-05, + "loss": 0.1541, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.14429527521133423, + "step": 900, + "valid_targets_mean": 4958.3, + "valid_targets_min": 1514 + }, + { + "epoch": 2.8821656050955413, + "grad_norm": 0.2214504815229747, + "learning_rate": 7.177909662895949e-05, + "loss": 0.1504, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1483648717403412, + "step": 905, + "valid_targets_mean": 5231.9, + "valid_targets_min": 1494 + }, + { + "epoch": 2.8980891719745223, + "grad_norm": 0.22699090972162908, + "learning_rate": 7.149576490960073e-05, + "loss": 0.1597, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15121811628341675, + "step": 910, + "valid_targets_mean": 5019.6, + "valid_targets_min": 1730 + }, + { + "epoch": 2.9140127388535033, + "grad_norm": 0.2231867440615836, + "learning_rate": 7.121158389495186e-05, + "loss": 0.1535, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1468854546546936, + "step": 915, + "valid_targets_mean": 5194.1, + "valid_targets_min": 1954 + }, + { + "epoch": 2.9299363057324843, + "grad_norm": 0.22184044491259736, + "learning_rate": 7.092656481297332e-05, + "loss": 0.1508, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15112560987472534, + "step": 920, + "valid_targets_mean": 5167.0, + "valid_targets_min": 3480 + }, + { + "epoch": 2.945859872611465, + "grad_norm": 0.21540683565339847, + "learning_rate": 7.064071892473744e-05, + "loss": 0.1497, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.146908700466156, + "step": 925, + "valid_targets_mean": 4849.5, + "valid_targets_min": 2115 + }, + { + "epoch": 2.961783439490446, + "grad_norm": 0.21951990260266196, + "learning_rate": 7.035405752398358e-05, + "loss": 0.1493, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.13992591202259064, + "step": 930, + "valid_targets_mean": 4786.7, + "valid_targets_min": 1081 + }, + { + "epoch": 2.977707006369427, + "grad_norm": 0.22562929941154344, + "learning_rate": 7.006659193667195e-05, + "loss": 0.1614, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15292905271053314, + "step": 935, + "valid_targets_mean": 4740.9, + "valid_targets_min": 1878 + }, + { + "epoch": 2.9936305732484074, + "grad_norm": 0.22548848014881667, + "learning_rate": 6.977833352053607e-05, + "loss": 0.1559, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.15366116166114807, + "step": 940, + "valid_targets_mean": 5160.2, + "valid_targets_min": 1780 + }, + { + "epoch": 3.0095541401273884, + "grad_norm": 0.2836397536166952, + "learning_rate": 6.948929366463396e-05, + "loss": 0.1212, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10728035867214203, + "step": 945, + "valid_targets_mean": 5046.2, + "valid_targets_min": 1274 + }, + { + "epoch": 3.0254777070063694, + "grad_norm": 0.2670995379734586, + "learning_rate": 6.919948378889837e-05, + "loss": 0.1055, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10979095846414566, + "step": 950, + "valid_targets_mean": 5313.2, + "valid_targets_min": 2275 + }, + { + "epoch": 3.0414012738853504, + "grad_norm": 0.26239189784852934, + "learning_rate": 6.890891534368538e-05, + "loss": 0.1, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.09663648903369904, + "step": 955, + "valid_targets_mean": 4853.4, + "valid_targets_min": 1308 + }, + { + "epoch": 3.0573248407643314, + "grad_norm": 0.2362160524565689, + "learning_rate": 6.861759980932204e-05, + "loss": 0.1045, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.09744668006896973, + "step": 960, + "valid_targets_mean": 5021.4, + "valid_targets_min": 1146 + }, + { + "epoch": 3.073248407643312, + "grad_norm": 0.25361799165079824, + "learning_rate": 6.832554869565284e-05, + "loss": 0.1044, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10304249823093414, + "step": 965, + "valid_targets_mean": 5474.8, + "valid_targets_min": 3578 + }, + { + "epoch": 3.089171974522293, + "grad_norm": 0.2623262365900065, + "learning_rate": 6.803277354158494e-05, + "loss": 0.1064, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10771718621253967, + "step": 970, + "valid_targets_mean": 5292.8, + "valid_targets_min": 1817 + }, + { + "epoch": 3.105095541401274, + "grad_norm": 0.2675213244942131, + "learning_rate": 6.773928591463223e-05, + "loss": 0.1024, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10803531110286713, + "step": 975, + "valid_targets_mean": 5001.0, + "valid_targets_min": 1780 + }, + { + "epoch": 3.121019108280255, + "grad_norm": 0.2561637806110615, + "learning_rate": 6.744509741045835e-05, + "loss": 0.1019, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.09725865721702576, + "step": 980, + "valid_targets_mean": 5329.6, + "valid_targets_min": 1342 + }, + { + "epoch": 3.1369426751592355, + "grad_norm": 0.2820778006165851, + "learning_rate": 6.715021965241848e-05, + "loss": 0.1048, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10907495766878128, + "step": 985, + "valid_targets_mean": 4772.0, + "valid_targets_min": 703 + }, + { + "epoch": 3.1528662420382165, + "grad_norm": 0.24305057030793387, + "learning_rate": 6.68546642911002e-05, + "loss": 0.1025, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.09349951148033142, + "step": 990, + "valid_targets_mean": 4875.8, + "valid_targets_min": 1220 + }, + { + "epoch": 3.1687898089171975, + "grad_norm": 0.2593558034697089, + "learning_rate": 6.655844300386308e-05, + "loss": 0.1016, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10261768102645874, + "step": 995, + "valid_targets_mean": 4717.2, + "valid_targets_min": 1088 + }, + { + "epoch": 3.1847133757961785, + "grad_norm": 0.25177489401123127, + "learning_rate": 6.626156749437736e-05, + "loss": 0.1029, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10331865400075912, + "step": 1000, + "valid_targets_mean": 4813.5, + "valid_targets_min": 1196 + }, + { + "epoch": 3.200636942675159, + "grad_norm": 0.24155775755779882, + "learning_rate": 6.596404949216153e-05, + "loss": 0.1014, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.09425457566976547, + "step": 1005, + "valid_targets_mean": 4847.7, + "valid_targets_min": 672 + }, + { + "epoch": 3.21656050955414, + "grad_norm": 0.26238019976464894, + "learning_rate": 6.56659007521189e-05, + "loss": 0.102, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1050558015704155, + "step": 1010, + "valid_targets_mean": 5011.2, + "valid_targets_min": 2485 + }, + { + "epoch": 3.232484076433121, + "grad_norm": 0.24909665908055414, + "learning_rate": 6.536713305407314e-05, + "loss": 0.1069, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10467242449522018, + "step": 1015, + "valid_targets_mean": 5429.5, + "valid_targets_min": 1309 + }, + { + "epoch": 3.248407643312102, + "grad_norm": 0.26079758272721515, + "learning_rate": 6.506775820230295e-05, + "loss": 0.1043, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10477639734745026, + "step": 1020, + "valid_targets_mean": 5026.8, + "valid_targets_min": 1179 + }, + { + "epoch": 3.2643312101910826, + "grad_norm": 0.23916627901265192, + "learning_rate": 6.476778802507549e-05, + "loss": 0.1017, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.09870631992816925, + "step": 1025, + "valid_targets_mean": 5273.7, + "valid_targets_min": 2515 + }, + { + "epoch": 3.2802547770700636, + "grad_norm": 0.24132703015830326, + "learning_rate": 6.446723437417928e-05, + "loss": 0.1084, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10116931796073914, + "step": 1030, + "valid_targets_mean": 5046.2, + "valid_targets_min": 1921 + }, + { + "epoch": 3.2961783439490446, + "grad_norm": 0.25752656665880735, + "learning_rate": 6.416610912445574e-05, + "loss": 0.107, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10365831106901169, + "step": 1035, + "valid_targets_mean": 5141.5, + "valid_targets_min": 1078 + }, + { + "epoch": 3.3121019108280256, + "grad_norm": 0.2480360825421199, + "learning_rate": 6.386442417333011e-05, + "loss": 0.11, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10683537274599075, + "step": 1040, + "valid_targets_mean": 4802.7, + "valid_targets_min": 1190 + }, + { + "epoch": 3.328025477707006, + "grad_norm": 0.24384669321356942, + "learning_rate": 6.356219144034143e-05, + "loss": 0.1057, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10856311768293381, + "step": 1045, + "valid_targets_mean": 5358.2, + "valid_targets_min": 3682 + }, + { + "epoch": 3.343949044585987, + "grad_norm": 0.2615624842739929, + "learning_rate": 6.325942286667149e-05, + "loss": 0.1044, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1113840863108635, + "step": 1050, + "valid_targets_mean": 5181.6, + "valid_targets_min": 1216 + }, + { + "epoch": 3.359872611464968, + "grad_norm": 0.2567165445833586, + "learning_rate": 6.295613041467306e-05, + "loss": 0.1031, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10421548783779144, + "step": 1055, + "valid_targets_mean": 5056.2, + "valid_targets_min": 818 + }, + { + "epoch": 3.375796178343949, + "grad_norm": 0.25864812299068907, + "learning_rate": 6.265232606739734e-05, + "loss": 0.1028, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10285977274179459, + "step": 1060, + "valid_targets_mean": 5439.9, + "valid_targets_min": 3077 + }, + { + "epoch": 3.3917197452229297, + "grad_norm": 0.25591560597906826, + "learning_rate": 6.234802182812043e-05, + "loss": 0.1087, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11345119774341583, + "step": 1065, + "valid_targets_mean": 5100.0, + "valid_targets_min": 1143 + }, + { + "epoch": 3.4076433121019107, + "grad_norm": 0.26779916712175555, + "learning_rate": 6.20432297198691e-05, + "loss": 0.1042, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10799382627010345, + "step": 1070, + "valid_targets_mean": 4800.0, + "valid_targets_min": 1217 + }, + { + "epoch": 3.4235668789808917, + "grad_norm": 0.2623396694207434, + "learning_rate": 6.17379617849458e-05, + "loss": 0.1094, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11160917580127716, + "step": 1075, + "valid_targets_mean": 5049.7, + "valid_targets_min": 1419 + }, + { + "epoch": 3.4394904458598727, + "grad_norm": 0.24903895076990235, + "learning_rate": 6.14322300844528e-05, + "loss": 0.1062, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10769861936569214, + "step": 1080, + "valid_targets_mean": 4956.7, + "valid_targets_min": 1107 + }, + { + "epoch": 3.4554140127388537, + "grad_norm": 0.24951528237863124, + "learning_rate": 6.112604669781572e-05, + "loss": 0.1065, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10491140186786652, + "step": 1085, + "valid_targets_mean": 4922.6, + "valid_targets_min": 1444 + }, + { + "epoch": 3.4713375796178343, + "grad_norm": 0.25034291309055917, + "learning_rate": 6.0819423722306253e-05, + "loss": 0.104, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1010381430387497, + "step": 1090, + "valid_targets_mean": 5216.4, + "valid_targets_min": 2110 + }, + { + "epoch": 3.4872611464968153, + "grad_norm": 0.3390365353670313, + "learning_rate": 6.0512373272564184e-05, + "loss": 0.1097, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11959738284349442, + "step": 1095, + "valid_targets_mean": 5458.0, + "valid_targets_min": 837 + }, + { + "epoch": 3.5031847133757963, + "grad_norm": 0.24437172558260273, + "learning_rate": 6.020490748011877e-05, + "loss": 0.1041, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10164205729961395, + "step": 1100, + "valid_targets_mean": 4847.3, + "valid_targets_min": 1194 + }, + { + "epoch": 3.519108280254777, + "grad_norm": 0.2551590187858289, + "learning_rate": 5.9897038492909394e-05, + "loss": 0.1059, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10588935017585754, + "step": 1105, + "valid_targets_mean": 5376.8, + "valid_targets_min": 1114 + }, + { + "epoch": 3.535031847133758, + "grad_norm": 0.2456045285926812, + "learning_rate": 5.958877847480564e-05, + "loss": 0.106, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10410155355930328, + "step": 1110, + "valid_targets_mean": 5057.1, + "valid_targets_min": 1289 + }, + { + "epoch": 3.550955414012739, + "grad_norm": 0.2705095736783728, + "learning_rate": 5.928013960512668e-05, + "loss": 0.1085, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11431917548179626, + "step": 1115, + "valid_targets_mean": 4957.0, + "valid_targets_min": 2368 + }, + { + "epoch": 3.56687898089172, + "grad_norm": 0.2599637273643507, + "learning_rate": 5.897113407816006e-05, + "loss": 0.1076, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1085943803191185, + "step": 1120, + "valid_targets_mean": 5048.7, + "valid_targets_min": 723 + }, + { + "epoch": 3.582802547770701, + "grad_norm": 0.24504956439171913, + "learning_rate": 5.86617741026799e-05, + "loss": 0.1024, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.107721246778965, + "step": 1125, + "valid_targets_mean": 5124.7, + "valid_targets_min": 2261 + }, + { + "epoch": 3.5987261146496814, + "grad_norm": 0.24253994110152155, + "learning_rate": 5.8352071901464566e-05, + "loss": 0.1125, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1102893054485321, + "step": 1130, + "valid_targets_mean": 5029.5, + "valid_targets_min": 1075 + }, + { + "epoch": 3.6146496815286624, + "grad_norm": 0.27194849062723475, + "learning_rate": 5.804203971081371e-05, + "loss": 0.1082, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11634404212236404, + "step": 1135, + "valid_targets_mean": 4992.7, + "valid_targets_min": 1542 + }, + { + "epoch": 3.6305732484076434, + "grad_norm": 0.2516580754553014, + "learning_rate": 5.773168978006486e-05, + "loss": 0.1056, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10742191970348358, + "step": 1140, + "valid_targets_mean": 5402.7, + "valid_targets_min": 2061 + }, + { + "epoch": 3.646496815286624, + "grad_norm": 0.25068514246290896, + "learning_rate": 5.742103437110937e-05, + "loss": 0.1023, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.09810538589954376, + "step": 1145, + "valid_targets_mean": 4866.2, + "valid_targets_min": 1572 + }, + { + "epoch": 3.662420382165605, + "grad_norm": 0.24681145583948133, + "learning_rate": 5.711008575790806e-05, + "loss": 0.1071, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10473734140396118, + "step": 1150, + "valid_targets_mean": 4804.1, + "valid_targets_min": 2400 + }, + { + "epoch": 3.678343949044586, + "grad_norm": 0.262881630380637, + "learning_rate": 5.679885622600616e-05, + "loss": 0.1041, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10942289233207703, + "step": 1155, + "valid_targets_mean": 4926.2, + "valid_targets_min": 2711 + }, + { + "epoch": 3.694267515923567, + "grad_norm": 0.2831388830313044, + "learning_rate": 5.6487358072048e-05, + "loss": 0.1106, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11911377310752869, + "step": 1160, + "valid_targets_mean": 5100.8, + "valid_targets_min": 993 + }, + { + "epoch": 3.710191082802548, + "grad_norm": 0.26704618695645355, + "learning_rate": 5.617560360329114e-05, + "loss": 0.1062, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11761665344238281, + "step": 1165, + "valid_targets_mean": 5179.2, + "valid_targets_min": 1917 + }, + { + "epoch": 3.7261146496815285, + "grad_norm": 0.26325309810176767, + "learning_rate": 5.58636051371201e-05, + "loss": 0.1089, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11542294919490814, + "step": 1170, + "valid_targets_mean": 5089.2, + "valid_targets_min": 1051 + }, + { + "epoch": 3.7420382165605095, + "grad_norm": 0.2331609981220728, + "learning_rate": 5.5551375000559716e-05, + "loss": 0.1041, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.09477736055850983, + "step": 1175, + "valid_targets_mean": 4929.3, + "valid_targets_min": 1215 + }, + { + "epoch": 3.7579617834394905, + "grad_norm": 0.2626457657069755, + "learning_rate": 5.523892552978808e-05, + "loss": 0.1114, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.106768399477005, + "step": 1180, + "valid_targets_mean": 5236.1, + "valid_targets_min": 481 + }, + { + "epoch": 3.7738853503184715, + "grad_norm": 0.2633532573529452, + "learning_rate": 5.492626906964917e-05, + "loss": 0.1061, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10951051115989685, + "step": 1185, + "valid_targets_mean": 5243.7, + "valid_targets_min": 940 + }, + { + "epoch": 3.789808917197452, + "grad_norm": 0.24086117942985047, + "learning_rate": 5.4613417973165106e-05, + "loss": 0.1046, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10434311628341675, + "step": 1190, + "valid_targets_mean": 5219.0, + "valid_targets_min": 3102 + }, + { + "epoch": 3.805732484076433, + "grad_norm": 0.264859468533397, + "learning_rate": 5.430038460104803e-05, + "loss": 0.1067, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10531932860612869, + "step": 1195, + "valid_targets_mean": 4750.2, + "valid_targets_min": 1375 + }, + { + "epoch": 3.821656050955414, + "grad_norm": 0.2700805055153647, + "learning_rate": 5.39871813212118e-05, + "loss": 0.1074, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11084768176078796, + "step": 1200, + "valid_targets_mean": 5329.8, + "valid_targets_min": 1161 + }, + { + "epoch": 3.837579617834395, + "grad_norm": 0.23715443523845312, + "learning_rate": 5.36738205082833e-05, + "loss": 0.1038, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.09647548198699951, + "step": 1205, + "valid_targets_mean": 5051.8, + "valid_targets_min": 1135 + }, + { + "epoch": 3.853503184713376, + "grad_norm": 0.23915161753996375, + "learning_rate": 5.3360314543113544e-05, + "loss": 0.1106, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10543075203895569, + "step": 1210, + "valid_targets_mean": 5236.8, + "valid_targets_min": 1709 + }, + { + "epoch": 3.8694267515923566, + "grad_norm": 0.24368779650012992, + "learning_rate": 5.304667581228846e-05, + "loss": 0.1057, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10129611194133759, + "step": 1215, + "valid_targets_mean": 4842.7, + "valid_targets_min": 1457 + }, + { + "epoch": 3.8853503184713376, + "grad_norm": 0.23933972252366836, + "learning_rate": 5.2732916707639576e-05, + "loss": 0.1048, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.10172361135482788, + "step": 1220, + "valid_targets_mean": 5203.8, + "valid_targets_min": 1939 + }, + { + "epoch": 3.9012738853503186, + "grad_norm": 0.2588216069128758, + "learning_rate": 5.2419049625754346e-05, + "loss": 0.1057, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11062759160995483, + "step": 1225, + "valid_targets_mean": 4756.7, + "valid_targets_min": 484 + }, + { + "epoch": 3.917197452229299, + "grad_norm": 0.2671361073739832, + "learning_rate": 5.210508696748642e-05, + "loss": 0.1098, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11435449123382568, + "step": 1230, + "valid_targets_mean": 5069.8, + "valid_targets_min": 618 + }, + { + "epoch": 3.93312101910828, + "grad_norm": 0.2486785654428562, + "learning_rate": 5.17910411374656e-05, + "loss": 0.1034, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.1074475347995758, + "step": 1235, + "valid_targets_mean": 5156.6, + "valid_targets_min": 2701 + }, + { + "epoch": 3.949044585987261, + "grad_norm": 0.26579895266255577, + "learning_rate": 5.1476924543607884e-05, + "loss": 0.1089, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11539261043071747, + "step": 1240, + "valid_targets_mean": 5437.4, + "valid_targets_min": 2627 + }, + { + "epoch": 3.964968152866242, + "grad_norm": 0.27035345103767233, + "learning_rate": 5.116274959662508e-05, + "loss": 0.1115, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11612924933433533, + "step": 1245, + "valid_targets_mean": 5038.6, + "valid_targets_min": 1140 + }, + { + "epoch": 3.980891719745223, + "grad_norm": 0.2140692013570522, + "learning_rate": 5.0848528709534525e-05, + "loss": 0.1033, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.09687002748250961, + "step": 1250, + "valid_targets_mean": 5244.2, + "valid_targets_min": 3244 + }, + { + "epoch": 3.9968152866242037, + "grad_norm": 0.2535143333870611, + "learning_rate": 5.053427429716867e-05, + "loss": 0.1102, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.11210654675960541, + "step": 1255, + "valid_targets_mean": 5184.2, + "valid_targets_min": 1508 + }, + { + "epoch": 4.012738853503185, + "grad_norm": 0.32543565374679984, + "learning_rate": 5.021999877568452e-05, + "loss": 0.0717, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0697878897190094, + "step": 1260, + "valid_targets_mean": 4879.7, + "valid_targets_min": 1886 + }, + { + "epoch": 4.028662420382165, + "grad_norm": 0.2497859940335556, + "learning_rate": 4.990571456207311e-05, + "loss": 0.0669, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06878037005662918, + "step": 1265, + "valid_targets_mean": 5341.9, + "valid_targets_min": 2746 + }, + { + "epoch": 4.044585987261146, + "grad_norm": 0.27613213470492837, + "learning_rate": 4.959143407366891e-05, + "loss": 0.0653, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06718228757381439, + "step": 1270, + "valid_targets_mean": 5151.8, + "valid_targets_min": 2060 + }, + { + "epoch": 4.060509554140127, + "grad_norm": 0.2654860661104674, + "learning_rate": 4.927716972765918e-05, + "loss": 0.0672, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07257280498743057, + "step": 1275, + "valid_targets_mean": 5119.2, + "valid_targets_min": 1619 + }, + { + "epoch": 4.076433121019108, + "grad_norm": 0.2603303324802897, + "learning_rate": 4.8962933940593444e-05, + "loss": 0.0643, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06228986382484436, + "step": 1280, + "valid_targets_mean": 5570.8, + "valid_targets_min": 3149 + }, + { + "epoch": 4.092356687898089, + "grad_norm": 0.2518487900443168, + "learning_rate": 4.864873912789282e-05, + "loss": 0.0676, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06389397382736206, + "step": 1285, + "valid_targets_mean": 5220.3, + "valid_targets_min": 2358 + }, + { + "epoch": 4.10828025477707, + "grad_norm": 0.26677364116470986, + "learning_rate": 4.833459770335958e-05, + "loss": 0.0642, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06612858921289444, + "step": 1290, + "valid_targets_mean": 4851.1, + "valid_targets_min": 1150 + }, + { + "epoch": 4.124203821656051, + "grad_norm": 0.24599726927262444, + "learning_rate": 4.802052207868654e-05, + "loss": 0.0634, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.061971019953489304, + "step": 1295, + "valid_targets_mean": 5298.0, + "valid_targets_min": 1955 + }, + { + "epoch": 4.140127388535032, + "grad_norm": 0.23832637646511967, + "learning_rate": 4.7706524662966874e-05, + "loss": 0.0651, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.062450796365737915, + "step": 1300, + "valid_targets_mean": 4969.2, + "valid_targets_min": 1286 + }, + { + "epoch": 4.156050955414012, + "grad_norm": 0.2523903608435005, + "learning_rate": 4.739261786220368e-05, + "loss": 0.0657, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0634709820151329, + "step": 1305, + "valid_targets_mean": 5092.5, + "valid_targets_min": 2328 + }, + { + "epoch": 4.171974522292993, + "grad_norm": 0.25933448743779897, + "learning_rate": 4.707881407881988e-05, + "loss": 0.0647, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06896044313907623, + "step": 1310, + "valid_targets_mean": 4955.2, + "valid_targets_min": 2544 + }, + { + "epoch": 4.187898089171974, + "grad_norm": 0.25637900317547274, + "learning_rate": 4.67651257111682e-05, + "loss": 0.0665, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0633041262626648, + "step": 1315, + "valid_targets_mean": 5261.5, + "valid_targets_min": 2018 + }, + { + "epoch": 4.203821656050955, + "grad_norm": 0.25332929369579527, + "learning_rate": 4.6451565153041274e-05, + "loss": 0.0682, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06732308864593506, + "step": 1320, + "valid_targets_mean": 5149.8, + "valid_targets_min": 2189 + }, + { + "epoch": 4.219745222929936, + "grad_norm": 0.2631360831779554, + "learning_rate": 4.613814479318203e-05, + "loss": 0.0687, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06805001199245453, + "step": 1325, + "valid_targets_mean": 5208.4, + "valid_targets_min": 1023 + }, + { + "epoch": 4.235668789808917, + "grad_norm": 0.2668165361367289, + "learning_rate": 4.582487701479417e-05, + "loss": 0.0687, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06724009662866592, + "step": 1330, + "valid_targets_mean": 5126.9, + "valid_targets_min": 993 + }, + { + "epoch": 4.251592356687898, + "grad_norm": 0.2652760393097436, + "learning_rate": 4.551177419505288e-05, + "loss": 0.0684, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0720963403582573, + "step": 1335, + "valid_targets_mean": 4961.7, + "valid_targets_min": 1665 + }, + { + "epoch": 4.267515923566879, + "grad_norm": 0.26028868427430096, + "learning_rate": 4.5198848704615914e-05, + "loss": 0.0674, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06823834776878357, + "step": 1340, + "valid_targets_mean": 4578.2, + "valid_targets_min": 1088 + }, + { + "epoch": 4.2834394904458595, + "grad_norm": 0.2558703503727455, + "learning_rate": 4.4886112907134715e-05, + "loss": 0.0646, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06340670585632324, + "step": 1345, + "valid_targets_mean": 5014.7, + "valid_targets_min": 1002 + }, + { + "epoch": 4.2993630573248405, + "grad_norm": 0.2445891137433315, + "learning_rate": 4.457357915876599e-05, + "loss": 0.0685, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06793109327554703, + "step": 1350, + "valid_targets_mean": 5325.8, + "valid_targets_min": 2316 + }, + { + "epoch": 4.3152866242038215, + "grad_norm": 0.24716166350521976, + "learning_rate": 4.4261259807683534e-05, + "loss": 0.0664, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06318019330501556, + "step": 1355, + "valid_targets_mean": 4784.7, + "valid_targets_min": 1480 + }, + { + "epoch": 4.3312101910828025, + "grad_norm": 0.2562498787066791, + "learning_rate": 4.394916719359028e-05, + "loss": 0.069, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07023858278989792, + "step": 1360, + "valid_targets_mean": 4933.6, + "valid_targets_min": 1446 + }, + { + "epoch": 4.3471337579617835, + "grad_norm": 0.23498953082409643, + "learning_rate": 4.363731364723084e-05, + "loss": 0.065, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06365595012903214, + "step": 1365, + "valid_targets_mean": 5211.1, + "valid_targets_min": 2131 + }, + { + "epoch": 4.3630573248407645, + "grad_norm": 0.2659476073173022, + "learning_rate": 4.332571148990429e-05, + "loss": 0.0676, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0684298425912857, + "step": 1370, + "valid_targets_mean": 5160.2, + "valid_targets_min": 1814 + }, + { + "epoch": 4.3789808917197455, + "grad_norm": 0.24500356454352173, + "learning_rate": 4.3014373032977336e-05, + "loss": 0.0665, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06558418273925781, + "step": 1375, + "valid_targets_mean": 4926.4, + "valid_targets_min": 1517 + }, + { + "epoch": 4.3949044585987265, + "grad_norm": 0.25364416790466265, + "learning_rate": 4.270331057739793e-05, + "loss": 0.0672, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06497491151094437, + "step": 1380, + "valid_targets_mean": 4966.5, + "valid_targets_min": 1395 + }, + { + "epoch": 4.4108280254777075, + "grad_norm": 0.2561579738552816, + "learning_rate": 4.239253641320921e-05, + "loss": 0.0659, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06818774342536926, + "step": 1385, + "valid_targets_mean": 4865.7, + "valid_targets_min": 1691 + }, + { + "epoch": 4.426751592356688, + "grad_norm": 0.25636115860115083, + "learning_rate": 4.208206281906399e-05, + "loss": 0.0674, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06350462138652802, + "step": 1390, + "valid_targets_mean": 5171.3, + "valid_targets_min": 1413 + }, + { + "epoch": 4.442675159235669, + "grad_norm": 0.2537137269283525, + "learning_rate": 4.177190206173958e-05, + "loss": 0.0659, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0639728456735611, + "step": 1395, + "valid_targets_mean": 4515.4, + "valid_targets_min": 1241 + }, + { + "epoch": 4.45859872611465, + "grad_norm": 0.2507089998820537, + "learning_rate": 4.146206639565312e-05, + "loss": 0.0681, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06643237918615341, + "step": 1400, + "valid_targets_mean": 5098.2, + "valid_targets_min": 1246 + }, + { + "epoch": 4.474522292993631, + "grad_norm": 0.25048800523519704, + "learning_rate": 4.1152568062377474e-05, + "loss": 0.0664, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0646229088306427, + "step": 1405, + "valid_targets_mean": 5412.2, + "valid_targets_min": 2678 + }, + { + "epoch": 4.490445859872612, + "grad_norm": 0.25518840668289683, + "learning_rate": 4.084341929015749e-05, + "loss": 0.0656, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06392090767621994, + "step": 1410, + "valid_targets_mean": 4661.1, + "valid_targets_min": 1787 + }, + { + "epoch": 4.506369426751593, + "grad_norm": 0.26451003514696153, + "learning_rate": 4.05346322934269e-05, + "loss": 0.0692, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06889551132917404, + "step": 1415, + "valid_targets_mean": 5041.4, + "valid_targets_min": 1404 + }, + { + "epoch": 4.522292993630574, + "grad_norm": 0.24519959671250638, + "learning_rate": 4.022621927232577e-05, + "loss": 0.0643, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0631960928440094, + "step": 1420, + "valid_targets_mean": 4906.4, + "valid_targets_min": 1143 + }, + { + "epoch": 4.538216560509554, + "grad_norm": 0.26577047272631943, + "learning_rate": 3.991819241221835e-05, + "loss": 0.0666, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06718342006206512, + "step": 1425, + "valid_targets_mean": 4861.5, + "valid_targets_min": 1190 + }, + { + "epoch": 4.554140127388535, + "grad_norm": 0.258004292371268, + "learning_rate": 3.961056388321177e-05, + "loss": 0.0695, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0710553303360939, + "step": 1430, + "valid_targets_mean": 4989.9, + "valid_targets_min": 3250 + }, + { + "epoch": 4.570063694267516, + "grad_norm": 0.25787849659296397, + "learning_rate": 3.930334583967514e-05, + "loss": 0.0671, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06511221826076508, + "step": 1435, + "valid_targets_mean": 5124.6, + "valid_targets_min": 1170 + }, + { + "epoch": 4.585987261146497, + "grad_norm": 0.25262151176120695, + "learning_rate": 3.8996550419759315e-05, + "loss": 0.0674, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0706428587436676, + "step": 1440, + "valid_targets_mean": 4741.4, + "valid_targets_min": 576 + }, + { + "epoch": 4.601910828025478, + "grad_norm": 0.26241884175281194, + "learning_rate": 3.869018974491735e-05, + "loss": 0.0669, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06604479998350143, + "step": 1445, + "valid_targets_mean": 5243.3, + "valid_targets_min": 2277 + }, + { + "epoch": 4.617834394904459, + "grad_norm": 0.24861769697272823, + "learning_rate": 3.838427591942557e-05, + "loss": 0.0668, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06618154048919678, + "step": 1450, + "valid_targets_mean": 5323.2, + "valid_targets_min": 2472 + }, + { + "epoch": 4.63375796178344, + "grad_norm": 0.2722192591973607, + "learning_rate": 3.807882102990532e-05, + "loss": 0.0704, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07361525297164917, + "step": 1455, + "valid_targets_mean": 5013.7, + "valid_targets_min": 1792 + }, + { + "epoch": 4.649681528662421, + "grad_norm": 0.25256886749450697, + "learning_rate": 3.7773837144845455e-05, + "loss": 0.0672, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0680055022239685, + "step": 1460, + "valid_targets_mean": 5433.3, + "valid_targets_min": 3865 + }, + { + "epoch": 4.665605095541402, + "grad_norm": 0.2439164611487248, + "learning_rate": 3.746933631412548e-05, + "loss": 0.0677, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06579042971134186, + "step": 1465, + "valid_targets_mean": 4954.6, + "valid_targets_min": 1158 + }, + { + "epoch": 4.681528662420382, + "grad_norm": 0.2619080558541486, + "learning_rate": 3.716533056853948e-05, + "loss": 0.0645, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06959753483533859, + "step": 1470, + "valid_targets_mean": 4837.2, + "valid_targets_min": 889 + }, + { + "epoch": 4.697452229299363, + "grad_norm": 0.2499165064301628, + "learning_rate": 3.686183191932078e-05, + "loss": 0.068, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07037084549665451, + "step": 1475, + "valid_targets_mean": 4907.2, + "valid_targets_min": 1432 + }, + { + "epoch": 4.713375796178344, + "grad_norm": 0.269406009366861, + "learning_rate": 3.6558852357667416e-05, + "loss": 0.0685, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06931830942630768, + "step": 1480, + "valid_targets_mean": 4898.0, + "valid_targets_min": 1162 + }, + { + "epoch": 4.729299363057325, + "grad_norm": 0.2456468044189881, + "learning_rate": 3.625640385426826e-05, + "loss": 0.067, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06254740804433823, + "step": 1485, + "valid_targets_mean": 5088.6, + "valid_targets_min": 1806 + }, + { + "epoch": 4.745222929936306, + "grad_norm": 0.2743275642010713, + "learning_rate": 3.595449835883018e-05, + "loss": 0.0679, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06851696223020554, + "step": 1490, + "valid_targets_mean": 5121.9, + "valid_targets_min": 914 + }, + { + "epoch": 4.761146496815287, + "grad_norm": 0.23525141867534177, + "learning_rate": 3.5653147799605866e-05, + "loss": 0.0664, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06559790670871735, + "step": 1495, + "valid_targets_mean": 5416.6, + "valid_targets_min": 837 + }, + { + "epoch": 4.777070063694268, + "grad_norm": 0.24565853392599585, + "learning_rate": 3.535236408292253e-05, + "loss": 0.0661, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06561281532049179, + "step": 1500, + "valid_targets_mean": 4962.0, + "valid_targets_min": 1255 + }, + { + "epoch": 4.792993630573249, + "grad_norm": 0.24430716028494404, + "learning_rate": 3.505215909271149e-05, + "loss": 0.0666, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06181070953607559, + "step": 1505, + "valid_targets_mean": 5308.9, + "valid_targets_min": 2468 + }, + { + "epoch": 4.80891719745223, + "grad_norm": 0.25555301928092583, + "learning_rate": 3.4752544690038647e-05, + "loss": 0.0667, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07101315259933472, + "step": 1510, + "valid_targets_mean": 4784.2, + "valid_targets_min": 1137 + }, + { + "epoch": 4.82484076433121, + "grad_norm": 0.24376383095753595, + "learning_rate": 3.4453532712635855e-05, + "loss": 0.0653, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06604792177677155, + "step": 1515, + "valid_targets_mean": 5042.0, + "valid_targets_min": 1166 + }, + { + "epoch": 4.840764331210191, + "grad_norm": 0.2736563690710869, + "learning_rate": 3.415513497443322e-05, + "loss": 0.0675, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07168449461460114, + "step": 1520, + "valid_targets_mean": 5337.4, + "valid_targets_min": 535 + }, + { + "epoch": 4.856687898089172, + "grad_norm": 0.25049885627433904, + "learning_rate": 3.385736326509232e-05, + "loss": 0.067, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06345846503973007, + "step": 1525, + "valid_targets_mean": 4973.4, + "valid_targets_min": 1619 + }, + { + "epoch": 4.872611464968153, + "grad_norm": 0.2646714670430848, + "learning_rate": 3.3560229349540394e-05, + "loss": 0.0659, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.07117845118045807, + "step": 1530, + "valid_targets_mean": 4974.2, + "valid_targets_min": 1232 + }, + { + "epoch": 4.888535031847134, + "grad_norm": 0.24363591418875322, + "learning_rate": 3.3263744967505534e-05, + "loss": 0.0661, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06361409276723862, + "step": 1535, + "valid_targets_mean": 5185.8, + "valid_targets_min": 1292 + }, + { + "epoch": 4.904458598726115, + "grad_norm": 0.2524377419098983, + "learning_rate": 3.296792183305283e-05, + "loss": 0.0667, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06658736616373062, + "step": 1540, + "valid_targets_mean": 5193.3, + "valid_targets_min": 1524 + }, + { + "epoch": 4.920382165605096, + "grad_norm": 0.24535869980622727, + "learning_rate": 3.2672771634121543e-05, + "loss": 0.0664, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06610628962516785, + "step": 1545, + "valid_targets_mean": 5016.4, + "valid_targets_min": 1263 + }, + { + "epoch": 4.936305732484076, + "grad_norm": 0.23552879151068656, + "learning_rate": 3.237830603206333e-05, + "loss": 0.0667, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06478506326675415, + "step": 1550, + "valid_targets_mean": 5339.1, + "valid_targets_min": 3443 + }, + { + "epoch": 4.952229299363057, + "grad_norm": 0.24973997243808632, + "learning_rate": 3.208453666118151e-05, + "loss": 0.0667, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06741265952587128, + "step": 1555, + "valid_targets_mean": 5414.4, + "valid_targets_min": 1970 + }, + { + "epoch": 4.968152866242038, + "grad_norm": 0.24403940248949624, + "learning_rate": 3.1791475128271383e-05, + "loss": 0.0646, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0654819905757904, + "step": 1560, + "valid_targets_mean": 4747.5, + "valid_targets_min": 1817 + }, + { + "epoch": 4.984076433121019, + "grad_norm": 0.256246603139224, + "learning_rate": 3.149913301216164e-05, + "loss": 0.0658, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.06953182071447372, + "step": 1565, + "valid_targets_mean": 5057.5, + "valid_targets_min": 683 + }, + { + "epoch": 5.0, + "grad_norm": 0.2295144897580045, + "learning_rate": 3.1207521863256914e-05, + "loss": 0.0617, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.05196793004870415, + "step": 1570, + "valid_targets_mean": 5062.8, + "valid_targets_min": 2584 + }, + { + "epoch": 5.015923566878981, + "grad_norm": 0.2573539131061093, + "learning_rate": 3.0916653203081367e-05, + "loss": 0.0402, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03786831349134445, + "step": 1575, + "valid_targets_mean": 5213.8, + "valid_targets_min": 2079 + }, + { + "epoch": 5.031847133757962, + "grad_norm": 0.2327238193404669, + "learning_rate": 3.0626538523823565e-05, + "loss": 0.0397, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03762497007846832, + "step": 1580, + "valid_targets_mean": 5278.7, + "valid_targets_min": 2970 + }, + { + "epoch": 5.047770700636943, + "grad_norm": 0.2151122853268418, + "learning_rate": 3.033718928788233e-05, + "loss": 0.039, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.038998134434223175, + "step": 1585, + "valid_targets_mean": 5413.0, + "valid_targets_min": 3096 + }, + { + "epoch": 5.063694267515924, + "grad_norm": 0.2547252680243014, + "learning_rate": 3.004861692741391e-05, + "loss": 0.0395, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04167929291725159, + "step": 1590, + "valid_targets_mean": 5181.2, + "valid_targets_min": 1937 + }, + { + "epoch": 5.079617834394904, + "grad_norm": 0.2280434109475756, + "learning_rate": 2.976083284388031e-05, + "loss": 0.0388, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03746391087770462, + "step": 1595, + "valid_targets_mean": 5098.7, + "valid_targets_min": 1705 + }, + { + "epoch": 5.095541401273885, + "grad_norm": 0.23351952641590804, + "learning_rate": 2.947384840759878e-05, + "loss": 0.0397, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03910890221595764, + "step": 1600, + "valid_targets_mean": 4838.8, + "valid_targets_min": 1221 + }, + { + "epoch": 5.111464968152866, + "grad_norm": 0.23129001044917588, + "learning_rate": 2.918767495729261e-05, + "loss": 0.0391, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.039571650326251984, + "step": 1605, + "valid_targets_mean": 4938.4, + "valid_targets_min": 2766 + }, + { + "epoch": 5.127388535031847, + "grad_norm": 0.21390005575294557, + "learning_rate": 2.8902323799643116e-05, + "loss": 0.0384, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03702428936958313, + "step": 1610, + "valid_targets_mean": 4986.0, + "valid_targets_min": 975 + }, + { + "epoch": 5.143312101910828, + "grad_norm": 0.24171243163064313, + "learning_rate": 2.8617806208842924e-05, + "loss": 0.0389, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.039472486823797226, + "step": 1615, + "valid_targets_mean": 4969.9, + "valid_targets_min": 762 + }, + { + "epoch": 5.159235668789809, + "grad_norm": 0.22737705400402208, + "learning_rate": 2.83341334261505e-05, + "loss": 0.0387, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03793150559067726, + "step": 1620, + "valid_targets_mean": 5323.2, + "valid_targets_min": 2969 + }, + { + "epoch": 5.17515923566879, + "grad_norm": 0.23863949523790068, + "learning_rate": 2.8051316659446104e-05, + "loss": 0.0388, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03899385780096054, + "step": 1625, + "valid_targets_mean": 4808.1, + "valid_targets_min": 1114 + }, + { + "epoch": 5.191082802547771, + "grad_norm": 0.23234962061403744, + "learning_rate": 2.7769367082788812e-05, + "loss": 0.0381, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03629320114850998, + "step": 1630, + "valid_targets_mean": 5268.3, + "valid_targets_min": 2752 + }, + { + "epoch": 5.207006369426751, + "grad_norm": 0.23382670956023793, + "learning_rate": 2.74882958359752e-05, + "loss": 0.0394, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.036728255450725555, + "step": 1635, + "valid_targets_mean": 5192.7, + "valid_targets_min": 1252 + }, + { + "epoch": 5.222929936305732, + "grad_norm": 0.23909704014302888, + "learning_rate": 2.7208114024099053e-05, + "loss": 0.0389, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03943564370274544, + "step": 1640, + "valid_targets_mean": 5052.8, + "valid_targets_min": 703 + }, + { + "epoch": 5.238853503184713, + "grad_norm": 0.25146903056664593, + "learning_rate": 2.6928832717112757e-05, + "loss": 0.0407, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04068949818611145, + "step": 1645, + "valid_targets_mean": 5006.3, + "valid_targets_min": 837 + }, + { + "epoch": 5.254777070063694, + "grad_norm": 0.2306660615655428, + "learning_rate": 2.6650462949389765e-05, + "loss": 0.0397, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.037642888724803925, + "step": 1650, + "valid_targets_mean": 5171.4, + "valid_targets_min": 535 + }, + { + "epoch": 5.270700636942675, + "grad_norm": 0.2395925445552546, + "learning_rate": 2.6373015719288806e-05, + "loss": 0.0399, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04006779566407204, + "step": 1655, + "valid_targets_mean": 5110.9, + "valid_targets_min": 1598 + }, + { + "epoch": 5.286624203821656, + "grad_norm": 0.2309167233268924, + "learning_rate": 2.609650198871917e-05, + "loss": 0.0397, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03972351923584938, + "step": 1660, + "valid_targets_mean": 4968.6, + "valid_targets_min": 2374 + }, + { + "epoch": 5.302547770700637, + "grad_norm": 0.23763897355788857, + "learning_rate": 2.5820932682707732e-05, + "loss": 0.0391, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04093179851770401, + "step": 1665, + "valid_targets_mean": 5016.2, + "valid_targets_min": 544 + }, + { + "epoch": 5.318471337579618, + "grad_norm": 0.25068878457324717, + "learning_rate": 2.5546318688967254e-05, + "loss": 0.0395, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.041159652173519135, + "step": 1670, + "valid_targets_mean": 4911.2, + "valid_targets_min": 1116 + }, + { + "epoch": 5.334394904458598, + "grad_norm": 0.22753403178797857, + "learning_rate": 2.5272670857466165e-05, + "loss": 0.0396, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04052474722266197, + "step": 1675, + "valid_targets_mean": 5482.7, + "valid_targets_min": 1524 + }, + { + "epoch": 5.350318471337579, + "grad_norm": 0.22879733161534754, + "learning_rate": 2.500000000000001e-05, + "loss": 0.0403, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04112991318106651, + "step": 1680, + "valid_targets_mean": 5106.4, + "valid_targets_min": 1158 + }, + { + "epoch": 5.36624203821656, + "grad_norm": 0.2376234122880377, + "learning_rate": 2.4728316889764132e-05, + "loss": 0.0401, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04215697571635246, + "step": 1685, + "valid_targets_mean": 5061.7, + "valid_targets_min": 2297 + }, + { + "epoch": 5.382165605095541, + "grad_norm": 0.22434830031383393, + "learning_rate": 2.4457632260928142e-05, + "loss": 0.0385, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.038262754678726196, + "step": 1690, + "valid_targets_mean": 5453.3, + "valid_targets_min": 3648 + }, + { + "epoch": 5.398089171974522, + "grad_norm": 0.24309045551378064, + "learning_rate": 2.418795680821171e-05, + "loss": 0.0397, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04290362820029259, + "step": 1695, + "valid_targets_mean": 4812.7, + "valid_targets_min": 1322 + }, + { + "epoch": 5.414012738853503, + "grad_norm": 0.23465352172702575, + "learning_rate": 2.3919301186462122e-05, + "loss": 0.0388, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04129970818758011, + "step": 1700, + "valid_targets_mean": 5085.9, + "valid_targets_min": 2467 + }, + { + "epoch": 5.429936305732484, + "grad_norm": 0.2404443420309706, + "learning_rate": 2.3651676010233205e-05, + "loss": 0.0387, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04071522504091263, + "step": 1705, + "valid_targets_mean": 5138.1, + "valid_targets_min": 2789 + }, + { + "epoch": 5.445859872611465, + "grad_norm": 0.2285563859680148, + "learning_rate": 2.338509185336605e-05, + "loss": 0.0392, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0407479926943779, + "step": 1710, + "valid_targets_mean": 5241.2, + "valid_targets_min": 3393 + }, + { + "epoch": 5.461783439490446, + "grad_norm": 0.21915829436292794, + "learning_rate": 2.3119559248571128e-05, + "loss": 0.0382, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.037575460970401764, + "step": 1715, + "valid_targets_mean": 5078.1, + "valid_targets_min": 1875 + }, + { + "epoch": 5.477707006369426, + "grad_norm": 0.218354261531361, + "learning_rate": 2.2855088687012267e-05, + "loss": 0.0377, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03804899379611015, + "step": 1720, + "valid_targets_mean": 5097.3, + "valid_targets_min": 1231 + }, + { + "epoch": 5.493630573248407, + "grad_norm": 0.22967659689800157, + "learning_rate": 2.2591690617892024e-05, + "loss": 0.0382, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03832363337278366, + "step": 1725, + "valid_targets_mean": 4996.3, + "valid_targets_min": 1284 + }, + { + "epoch": 5.509554140127388, + "grad_norm": 0.2135361882215248, + "learning_rate": 2.2329375448038963e-05, + "loss": 0.038, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03654465451836586, + "step": 1730, + "valid_targets_mean": 5368.9, + "valid_targets_min": 2727 + }, + { + "epoch": 5.525477707006369, + "grad_norm": 0.219724387020262, + "learning_rate": 2.206815354149636e-05, + "loss": 0.0382, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03865468502044678, + "step": 1735, + "valid_targets_mean": 5250.5, + "valid_targets_min": 2485 + }, + { + "epoch": 5.54140127388535, + "grad_norm": 0.2280509928530714, + "learning_rate": 2.1808035219112815e-05, + "loss": 0.0383, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03821604698896408, + "step": 1740, + "valid_targets_mean": 5243.9, + "valid_targets_min": 2173 + }, + { + "epoch": 5.557324840764331, + "grad_norm": 0.22620549277434396, + "learning_rate": 2.15490307581344e-05, + "loss": 0.0377, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.036652229726314545, + "step": 1745, + "valid_targets_mean": 5087.3, + "valid_targets_min": 1078 + }, + { + "epoch": 5.573248407643312, + "grad_norm": 0.24158871777005828, + "learning_rate": 2.1291150391798702e-05, + "loss": 0.0385, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03983117267489433, + "step": 1750, + "valid_targets_mean": 4814.7, + "valid_targets_min": 1708 + }, + { + "epoch": 5.5891719745222925, + "grad_norm": 0.2144191635607289, + "learning_rate": 2.1034404308930406e-05, + "loss": 0.0386, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03703922778367996, + "step": 1755, + "valid_targets_mean": 5234.8, + "valid_targets_min": 2719 + }, + { + "epoch": 5.6050955414012735, + "grad_norm": 0.2289998905622151, + "learning_rate": 2.0778802653538825e-05, + "loss": 0.0381, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.038494180887937546, + "step": 1760, + "valid_targets_mean": 5025.2, + "valid_targets_min": 743 + }, + { + "epoch": 5.6210191082802545, + "grad_norm": 0.22707821409053633, + "learning_rate": 2.0524355524417017e-05, + "loss": 0.0393, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.038622841238975525, + "step": 1765, + "valid_targets_mean": 5387.8, + "valid_targets_min": 1387 + }, + { + "epoch": 5.6369426751592355, + "grad_norm": 0.2384539406821619, + "learning_rate": 2.0271072974742893e-05, + "loss": 0.0381, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04027251899242401, + "step": 1770, + "valid_targets_mean": 4907.1, + "valid_targets_min": 1161 + }, + { + "epoch": 5.6528662420382165, + "grad_norm": 0.22920367000624492, + "learning_rate": 2.00189650116819e-05, + "loss": 0.0388, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03986385464668274, + "step": 1775, + "valid_targets_mean": 4628.7, + "valid_targets_min": 1389 + }, + { + "epoch": 5.6687898089171975, + "grad_norm": 0.2479531372985083, + "learning_rate": 1.976804159599176e-05, + "loss": 0.0384, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.038538143038749695, + "step": 1780, + "valid_targets_mean": 4863.4, + "valid_targets_min": 481 + }, + { + "epoch": 5.6847133757961785, + "grad_norm": 0.22815175723184877, + "learning_rate": 1.9518312641628794e-05, + "loss": 0.0384, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03903673216700554, + "step": 1785, + "valid_targets_mean": 5240.6, + "valid_targets_min": 1142 + }, + { + "epoch": 5.7006369426751595, + "grad_norm": 0.2252628939974779, + "learning_rate": 1.9269788015356338e-05, + "loss": 0.0382, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0388975664973259, + "step": 1790, + "valid_targets_mean": 5302.0, + "valid_targets_min": 1508 + }, + { + "epoch": 5.7165605095541405, + "grad_norm": 0.23333066323175405, + "learning_rate": 1.902247753635486e-05, + "loss": 0.0393, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0404471755027771, + "step": 1795, + "valid_targets_mean": 4632.9, + "valid_targets_min": 1508 + }, + { + "epoch": 5.732484076433121, + "grad_norm": 0.23840989094083886, + "learning_rate": 1.8776390975833968e-05, + "loss": 0.0392, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04090998321771622, + "step": 1800, + "valid_targets_mean": 4579.7, + "valid_targets_min": 1118 + }, + { + "epoch": 5.748407643312102, + "grad_norm": 0.22837896703491764, + "learning_rate": 1.8531538056646414e-05, + "loss": 0.0393, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03896788880228996, + "step": 1805, + "valid_targets_mean": 5020.3, + "valid_targets_min": 1122 + }, + { + "epoch": 5.764331210191083, + "grad_norm": 0.222630079744137, + "learning_rate": 1.8287928452903873e-05, + "loss": 0.0381, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.039409175515174866, + "step": 1810, + "valid_targets_mean": 5114.5, + "valid_targets_min": 2084 + }, + { + "epoch": 5.780254777070064, + "grad_norm": 0.2190364710189488, + "learning_rate": 1.804557178959485e-05, + "loss": 0.0376, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03576471656560898, + "step": 1815, + "valid_targets_mean": 5217.8, + "valid_targets_min": 1527 + }, + { + "epoch": 5.796178343949045, + "grad_norm": 0.23836620702182748, + "learning_rate": 1.7804477642204222e-05, + "loss": 0.0377, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.04015360027551651, + "step": 1820, + "valid_targets_mean": 5027.3, + "valid_targets_min": 1778 + }, + { + "epoch": 5.812101910828026, + "grad_norm": 0.21946223926645111, + "learning_rate": 1.7564655536335074e-05, + "loss": 0.0388, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03854244202375412, + "step": 1825, + "valid_targets_mean": 5460.2, + "valid_targets_min": 2064 + }, + { + "epoch": 5.828025477707007, + "grad_norm": 0.22471298771133874, + "learning_rate": 1.7326114947332207e-05, + "loss": 0.0389, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03704363852739334, + "step": 1830, + "valid_targets_mean": 5051.8, + "valid_targets_min": 1170 + }, + { + "epoch": 5.843949044585988, + "grad_norm": 0.23436910547541787, + "learning_rate": 1.7088865299907898e-05, + "loss": 0.0389, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.039101142436265945, + "step": 1835, + "valid_targets_mean": 4896.2, + "valid_targets_min": 1075 + }, + { + "epoch": 5.859872611464969, + "grad_norm": 0.22806331324189577, + "learning_rate": 1.68529159677694e-05, + "loss": 0.0401, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0396120660007, + "step": 1840, + "valid_targets_mean": 5041.4, + "valid_targets_min": 1493 + }, + { + "epoch": 5.875796178343949, + "grad_norm": 0.2432542303403707, + "learning_rate": 1.6618276273248716e-05, + "loss": 0.038, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03787992149591446, + "step": 1845, + "valid_targets_mean": 4892.7, + "valid_targets_min": 1226 + }, + { + "epoch": 5.89171974522293, + "grad_norm": 0.2342506891343223, + "learning_rate": 1.6384955486934156e-05, + "loss": 0.0379, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03828079253435135, + "step": 1850, + "valid_targets_mean": 5080.2, + "valid_targets_min": 1176 + }, + { + "epoch": 5.907643312101911, + "grad_norm": 0.24498194016096428, + "learning_rate": 1.6152962827304163e-05, + "loss": 0.0385, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.038725703954696655, + "step": 1855, + "valid_targets_mean": 4935.4, + "valid_targets_min": 583 + }, + { + "epoch": 5.923566878980892, + "grad_norm": 0.21562472599952542, + "learning_rate": 1.5922307460363007e-05, + "loss": 0.0377, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.038866475224494934, + "step": 1860, + "valid_targets_mean": 5018.0, + "valid_targets_min": 2721 + }, + { + "epoch": 5.939490445859873, + "grad_norm": 0.2201135801327864, + "learning_rate": 1.5692998499278656e-05, + "loss": 0.0392, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03628136217594147, + "step": 1865, + "valid_targets_mean": 5101.0, + "valid_targets_min": 1286 + }, + { + "epoch": 5.955414012738854, + "grad_norm": 0.23346888370178664, + "learning_rate": 1.5465045004022798e-05, + "loss": 0.0377, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03842603787779808, + "step": 1870, + "valid_targets_mean": 4982.7, + "valid_targets_min": 2943 + }, + { + "epoch": 5.971337579617835, + "grad_norm": 0.21461672073811788, + "learning_rate": 1.523845598101274e-05, + "loss": 0.0394, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03725840896368027, + "step": 1875, + "valid_targets_mean": 5235.9, + "valid_targets_min": 1557 + }, + { + "epoch": 5.987261146496815, + "grad_norm": 0.2212628514414812, + "learning_rate": 1.5013240382755712e-05, + "loss": 0.0377, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.03784020245075226, + "step": 1880, + "valid_targets_mean": 5124.7, + "valid_targets_min": 2650 + }, + { + "epoch": 6.003184713375796, + "grad_norm": 0.15356338186951857, + "learning_rate": 1.4789407107495012e-05, + "loss": 0.034, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0226178877055645, + "step": 1885, + "valid_targets_mean": 5191.6, + "valid_targets_min": 530 + }, + { + "epoch": 6.019108280254777, + "grad_norm": 0.19344467500440268, + "learning_rate": 1.456696499885859e-05, + "loss": 0.0236, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022716615349054337, + "step": 1890, + "valid_targets_mean": 5122.2, + "valid_targets_min": 1364 + }, + { + "epoch": 6.035031847133758, + "grad_norm": 0.20078494885416007, + "learning_rate": 1.4345922845509473e-05, + "loss": 0.0225, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022347833961248398, + "step": 1895, + "valid_targets_mean": 5153.9, + "valid_targets_min": 1255 + }, + { + "epoch": 6.050955414012739, + "grad_norm": 0.1774796611964485, + "learning_rate": 1.412628938079869e-05, + "loss": 0.0229, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.023219004273414612, + "step": 1900, + "valid_targets_mean": 5011.7, + "valid_targets_min": 1896 + }, + { + "epoch": 6.06687898089172, + "grad_norm": 0.1761710722866455, + "learning_rate": 1.3908073282420064e-05, + "loss": 0.0229, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02335885539650917, + "step": 1905, + "valid_targets_mean": 5108.4, + "valid_targets_min": 1343 + }, + { + "epoch": 6.082802547770701, + "grad_norm": 0.21191073884572043, + "learning_rate": 1.3691283172067493e-05, + "loss": 0.023, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.024375446140766144, + "step": 1910, + "valid_targets_mean": 4903.3, + "valid_targets_min": 1106 + }, + { + "epoch": 6.098726114649682, + "grad_norm": 0.17736943743668826, + "learning_rate": 1.3475927615094198e-05, + "loss": 0.0224, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02248205430805683, + "step": 1915, + "valid_targets_mean": 5163.9, + "valid_targets_min": 1107 + }, + { + "epoch": 6.114649681528663, + "grad_norm": 0.1921715593242747, + "learning_rate": 1.3262015120174397e-05, + "loss": 0.0231, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02436264418065548, + "step": 1920, + "valid_targets_mean": 5155.4, + "valid_targets_min": 2635 + }, + { + "epoch": 6.130573248407643, + "grad_norm": 0.19424106681277256, + "learning_rate": 1.3049554138967051e-05, + "loss": 0.0219, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.023007571697235107, + "step": 1925, + "valid_targets_mean": 5242.8, + "valid_targets_min": 3103 + }, + { + "epoch": 6.146496815286624, + "grad_norm": 0.18806288625039896, + "learning_rate": 1.2838553065781994e-05, + "loss": 0.0228, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022370267659425735, + "step": 1930, + "valid_targets_mean": 5589.7, + "valid_targets_min": 3781 + }, + { + "epoch": 6.162420382165605, + "grad_norm": 0.2029449619640067, + "learning_rate": 1.262902023724824e-05, + "loss": 0.0231, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.023126302286982536, + "step": 1935, + "valid_targets_mean": 4839.7, + "valid_targets_min": 2201 + }, + { + "epoch": 6.178343949044586, + "grad_norm": 0.1895008964685314, + "learning_rate": 1.2420963931984647e-05, + "loss": 0.0227, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.021901872009038925, + "step": 1940, + "valid_targets_mean": 5206.3, + "valid_targets_min": 1221 + }, + { + "epoch": 6.194267515923567, + "grad_norm": 0.17348101805573193, + "learning_rate": 1.221439237027277e-05, + "loss": 0.0222, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022755764424800873, + "step": 1945, + "valid_targets_mean": 4914.9, + "valid_targets_min": 2277 + }, + { + "epoch": 6.210191082802548, + "grad_norm": 0.18379891797309755, + "learning_rate": 1.2009313713732146e-05, + "loss": 0.0234, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.021730592474341393, + "step": 1950, + "valid_targets_mean": 5222.5, + "valid_targets_min": 1019 + }, + { + "epoch": 6.226114649681529, + "grad_norm": 0.19268371314116128, + "learning_rate": 1.1805736064997747e-05, + "loss": 0.0223, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02193249762058258, + "step": 1955, + "valid_targets_mean": 5013.3, + "valid_targets_min": 1322 + }, + { + "epoch": 6.24203821656051, + "grad_norm": 0.18750955638441208, + "learning_rate": 1.160366746739996e-05, + "loss": 0.0225, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022195417433977127, + "step": 1960, + "valid_targets_mean": 5007.3, + "valid_targets_min": 738 + }, + { + "epoch": 6.25796178343949, + "grad_norm": 0.17549282697023869, + "learning_rate": 1.1403115904646645e-05, + "loss": 0.0218, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.021028075367212296, + "step": 1965, + "valid_targets_mean": 5085.5, + "valid_targets_min": 2035 + }, + { + "epoch": 6.273885350318471, + "grad_norm": 0.18332193208157105, + "learning_rate": 1.1204089300507847e-05, + "loss": 0.0231, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.023401284590363503, + "step": 1970, + "valid_targets_mean": 4999.0, + "valid_targets_min": 1202 + }, + { + "epoch": 6.289808917197452, + "grad_norm": 0.17715239322556667, + "learning_rate": 1.1006595518502666e-05, + "loss": 0.0225, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.021887117996811867, + "step": 1975, + "valid_targets_mean": 5229.0, + "valid_targets_min": 1309 + }, + { + "epoch": 6.305732484076433, + "grad_norm": 0.20472408132731287, + "learning_rate": 1.0810642361588519e-05, + "loss": 0.0234, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022668935358524323, + "step": 1980, + "valid_targets_mean": 4881.4, + "valid_targets_min": 1242 + }, + { + "epoch": 6.321656050955414, + "grad_norm": 0.19109469103506047, + "learning_rate": 1.0616237571852949e-05, + "loss": 0.0225, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.021523375064134598, + "step": 1985, + "valid_targets_mean": 4815.9, + "valid_targets_min": 437 + }, + { + "epoch": 6.337579617834395, + "grad_norm": 0.18784051693223566, + "learning_rate": 1.0423388830207647e-05, + "loss": 0.022, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0216823760420084, + "step": 1990, + "valid_targets_mean": 5081.6, + "valid_targets_min": 566 + }, + { + "epoch": 6.353503184713376, + "grad_norm": 0.1914676236784046, + "learning_rate": 1.0232103756085032e-05, + "loss": 0.0224, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02255776897072792, + "step": 1995, + "valid_targets_mean": 4667.8, + "valid_targets_min": 837 + }, + { + "epoch": 6.369426751592357, + "grad_norm": 0.17055422040422477, + "learning_rate": 1.0042389907137184e-05, + "loss": 0.0217, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02008126676082611, + "step": 2000, + "valid_targets_mean": 5296.8, + "valid_targets_min": 2033 + }, + { + "epoch": 6.385350318471337, + "grad_norm": 0.2203669968573359, + "learning_rate": 9.854254778937272e-06, + "loss": 0.0225, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022920632734894753, + "step": 2005, + "valid_targets_mean": 4894.8, + "valid_targets_min": 683 + }, + { + "epoch": 6.401273885350318, + "grad_norm": 0.18476841368388444, + "learning_rate": 9.667705804683347e-06, + "loss": 0.022, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022299129515886307, + "step": 2010, + "valid_targets_mean": 5404.3, + "valid_targets_min": 3378 + }, + { + "epoch": 6.417197452229299, + "grad_norm": 0.18960537620168744, + "learning_rate": 9.48275035490474e-06, + "loss": 0.0224, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022625945508480072, + "step": 2015, + "valid_targets_mean": 5167.3, + "valid_targets_min": 635 + }, + { + "epoch": 6.43312101910828, + "grad_norm": 0.1816730366904124, + "learning_rate": 9.299395737170757e-06, + "loss": 0.0226, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.021097004413604736, + "step": 2020, + "valid_targets_mean": 5405.2, + "valid_targets_min": 2358 + }, + { + "epoch": 6.449044585987261, + "grad_norm": 0.18665652868762223, + "learning_rate": 9.117649195802053e-06, + "loss": 0.0222, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.020676065236330032, + "step": 2025, + "valid_targets_mean": 5571.8, + "valid_targets_min": 3239 + }, + { + "epoch": 6.464968152866242, + "grad_norm": 0.17168469220405425, + "learning_rate": 8.93751791158432e-06, + "loss": 0.0226, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02178370952606201, + "step": 2030, + "valid_targets_mean": 5110.2, + "valid_targets_min": 823 + }, + { + "epoch": 6.480891719745223, + "grad_norm": 0.20022678741940977, + "learning_rate": 8.759009001484664e-06, + "loss": 0.0228, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.023368585854768753, + "step": 2035, + "valid_targets_mean": 5139.5, + "valid_targets_min": 1107 + }, + { + "epoch": 6.496815286624204, + "grad_norm": 0.17698077328210068, + "learning_rate": 8.582129518370325e-06, + "loss": 0.0217, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02184038981795311, + "step": 2040, + "valid_targets_mean": 4925.2, + "valid_targets_min": 693 + }, + { + "epoch": 6.512738853503185, + "grad_norm": 0.18611744135196454, + "learning_rate": 8.4068864507301e-06, + "loss": 0.0219, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02264687605202198, + "step": 2045, + "valid_targets_mean": 4918.9, + "valid_targets_min": 2328 + }, + { + "epoch": 6.528662420382165, + "grad_norm": 0.195599432513402, + "learning_rate": 8.233286722398176e-06, + "loss": 0.0226, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022832946851849556, + "step": 2050, + "valid_targets_mean": 4740.8, + "valid_targets_min": 1219 + }, + { + "epoch": 6.544585987261146, + "grad_norm": 0.19507783367638135, + "learning_rate": 8.061337192280567e-06, + "loss": 0.0226, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.024483267217874527, + "step": 2055, + "valid_targets_mean": 4811.7, + "valid_targets_min": 752 + }, + { + "epoch": 6.560509554140127, + "grad_norm": 0.19349683718825508, + "learning_rate": 7.891044654084185e-06, + "loss": 0.0219, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022460635751485825, + "step": 2060, + "valid_targets_mean": 5202.2, + "valid_targets_min": 2514 + }, + { + "epoch": 6.576433121019108, + "grad_norm": 0.19073691268685294, + "learning_rate": 7.72241583604832e-06, + "loss": 0.0221, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022780772298574448, + "step": 2065, + "valid_targets_mean": 5167.2, + "valid_targets_min": 1401 + }, + { + "epoch": 6.592356687898089, + "grad_norm": 0.1890623710324865, + "learning_rate": 7.555457400678917e-06, + "loss": 0.0227, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.020403891801834106, + "step": 2070, + "valid_targets_mean": 5422.2, + "valid_targets_min": 3294 + }, + { + "epoch": 6.60828025477707, + "grad_norm": 0.1754586533727447, + "learning_rate": 7.390175944485239e-06, + "loss": 0.0221, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022461598739027977, + "step": 2075, + "valid_targets_mean": 5009.8, + "valid_targets_min": 1619 + }, + { + "epoch": 6.624203821656051, + "grad_norm": 0.1909287332543534, + "learning_rate": 7.22657799771933e-06, + "loss": 0.0215, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.023671746253967285, + "step": 2080, + "valid_targets_mean": 5126.0, + "valid_targets_min": 1422 + }, + { + "epoch": 6.640127388535031, + "grad_norm": 0.19347376217861104, + "learning_rate": 7.06467002411792e-06, + "loss": 0.0222, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.021759046241641045, + "step": 2085, + "valid_targets_mean": 4931.0, + "valid_targets_min": 1255 + }, + { + "epoch": 6.656050955414012, + "grad_norm": 0.19898716387442925, + "learning_rate": 6.904458420647131e-06, + "loss": 0.0217, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022016607224941254, + "step": 2090, + "valid_targets_mean": 4901.8, + "valid_targets_min": 1272 + }, + { + "epoch": 6.671974522292993, + "grad_norm": 0.19668215861498958, + "learning_rate": 6.745949517249639e-06, + "loss": 0.022, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0225750133395195, + "step": 2095, + "valid_targets_mean": 5119.8, + "valid_targets_min": 1431 + }, + { + "epoch": 6.687898089171974, + "grad_norm": 0.1683752467571503, + "learning_rate": 6.589149576594683e-06, + "loss": 0.0219, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02016734890639782, + "step": 2100, + "valid_targets_mean": 5357.0, + "valid_targets_min": 2994 + }, + { + "epoch": 6.703821656050955, + "grad_norm": 0.18243637984463024, + "learning_rate": 6.43406479383053e-06, + "loss": 0.0223, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0216531865298748, + "step": 2105, + "valid_targets_mean": 4924.5, + "valid_targets_min": 2007 + }, + { + "epoch": 6.719745222929936, + "grad_norm": 0.18269497984126018, + "learning_rate": 6.280701296339775e-06, + "loss": 0.0226, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022085040807724, + "step": 2110, + "valid_targets_mean": 5075.3, + "valid_targets_min": 1179 + }, + { + "epoch": 6.735668789808917, + "grad_norm": 0.18897198497982795, + "learning_rate": 6.129065143497215e-06, + "loss": 0.0216, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.0218723863363266, + "step": 2115, + "valid_targets_mean": 5236.2, + "valid_targets_min": 3275 + }, + { + "epoch": 6.751592356687898, + "grad_norm": 0.18973674541646202, + "learning_rate": 5.979162326430438e-06, + "loss": 0.0225, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02464936301112175, + "step": 2120, + "valid_targets_mean": 4906.2, + "valid_targets_min": 2166 + }, + { + "epoch": 6.767515923566879, + "grad_norm": 0.19023332017629005, + "learning_rate": 5.830998767783135e-06, + "loss": 0.0219, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.021690942347049713, + "step": 2125, + "valid_targets_mean": 4953.8, + "valid_targets_min": 566 + }, + { + "epoch": 6.7834394904458595, + "grad_norm": 0.185624826048304, + "learning_rate": 5.6845803214811145e-06, + "loss": 0.0218, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02219153568148613, + "step": 2130, + "valid_targets_mean": 4727.1, + "valid_targets_min": 1323 + }, + { + "epoch": 6.7993630573248405, + "grad_norm": 0.172986843377289, + "learning_rate": 5.539912772500944e-06, + "loss": 0.0213, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.021420851349830627, + "step": 2135, + "valid_targets_mean": 5189.9, + "valid_targets_min": 2881 + }, + { + "epoch": 6.8152866242038215, + "grad_norm": 0.17531370450754066, + "learning_rate": 5.397001836641469e-06, + "loss": 0.0215, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02098960056900978, + "step": 2140, + "valid_targets_mean": 5334.2, + "valid_targets_min": 2943 + }, + { + "epoch": 6.8312101910828025, + "grad_norm": 0.19066329195784262, + "learning_rate": 5.255853160297947e-06, + "loss": 0.0212, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02309945598244667, + "step": 2145, + "valid_targets_mean": 5010.4, + "valid_targets_min": 2094 + }, + { + "epoch": 6.8471337579617835, + "grad_norm": 0.20771981040888132, + "learning_rate": 5.116472320238908e-06, + "loss": 0.0215, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.021487152203917503, + "step": 2150, + "valid_targets_mean": 5251.6, + "valid_targets_min": 3011 + }, + { + "epoch": 6.8630573248407645, + "grad_norm": 0.18550128047287118, + "learning_rate": 4.978864823385937e-06, + "loss": 0.0215, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.023166440427303314, + "step": 2155, + "valid_targets_mean": 4948.3, + "valid_targets_min": 2552 + }, + { + "epoch": 6.8789808917197455, + "grad_norm": 0.18009162395329947, + "learning_rate": 4.843036106595961e-06, + "loss": 0.0207, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02180762216448784, + "step": 2160, + "valid_targets_mean": 5176.1, + "valid_targets_min": 1068 + }, + { + "epoch": 6.8949044585987265, + "grad_norm": 0.1962482096952318, + "learning_rate": 4.708991536446556e-06, + "loss": 0.0222, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022379565984010696, + "step": 2165, + "valid_targets_mean": 4881.7, + "valid_targets_min": 1480 + }, + { + "epoch": 6.9108280254777075, + "grad_norm": 0.17755483906859504, + "learning_rate": 4.576736409023813e-06, + "loss": 0.0215, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.020522532984614372, + "step": 2170, + "valid_targets_mean": 5519.9, + "valid_targets_min": 2337 + }, + { + "epoch": 6.926751592356688, + "grad_norm": 0.1844666144625461, + "learning_rate": 4.446275949713186e-06, + "loss": 0.0223, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02262505143880844, + "step": 2175, + "valid_targets_mean": 4835.6, + "valid_targets_min": 1162 + }, + { + "epoch": 6.942675159235669, + "grad_norm": 0.177784796020438, + "learning_rate": 4.317615312992962e-06, + "loss": 0.0211, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.02181645669043064, + "step": 2180, + "valid_targets_mean": 4841.8, + "valid_targets_min": 815 + }, + { + "epoch": 6.95859872611465, + "grad_norm": 0.19003337507749202, + "learning_rate": 4.190759582230641e-06, + "loss": 0.0219, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.023830631747841835, + "step": 2185, + "valid_targets_mean": 4834.2, + "valid_targets_min": 1375 + }, + { + "epoch": 6.974522292993631, + "grad_norm": 0.18485013401312161, + "learning_rate": 4.065713769482082e-06, + "loss": 0.0211, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.022234294563531876, + "step": 2190, + "valid_targets_mean": 4922.1, + "valid_targets_min": 1196 + }, + { + "epoch": 6.990445859872612, + "grad_norm": 0.19125998780810116, + "learning_rate": 3.942482815293513e-06, + "loss": 0.0218, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.021418333053588867, + "step": 2195, + "valid_targets_mean": 4613.4, + "valid_targets_min": 1036 + }, + { + "epoch": 7.006369426751593, + "grad_norm": 0.1173952766800128, + "learning_rate": 3.8210715885062585e-06, + "loss": 0.0185, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014751125127077103, + "step": 2200, + "valid_targets_mean": 5112.5, + "valid_targets_min": 2055 + }, + { + "epoch": 7.022292993630574, + "grad_norm": 0.11761236389240744, + "learning_rate": 3.7014848860644656e-06, + "loss": 0.015, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.015286756679415703, + "step": 2205, + "valid_targets_mean": 5028.2, + "valid_targets_min": 2767 + }, + { + "epoch": 7.038216560509555, + "grad_norm": 0.14847874242698741, + "learning_rate": 3.5837274328254735e-06, + "loss": 0.0147, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014214769005775452, + "step": 2210, + "valid_targets_mean": 4584.9, + "valid_targets_min": 762 + }, + { + "epoch": 7.054140127388535, + "grad_norm": 0.15251329412888048, + "learning_rate": 3.467803881373233e-06, + "loss": 0.0152, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014835171401500702, + "step": 2215, + "valid_targets_mean": 4937.8, + "valid_targets_min": 1510 + }, + { + "epoch": 7.070063694267516, + "grad_norm": 0.13608132360664227, + "learning_rate": 3.3537188118343998e-06, + "loss": 0.0149, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01441236212849617, + "step": 2220, + "valid_targets_mean": 5183.4, + "valid_targets_min": 2473 + }, + { + "epoch": 7.085987261146497, + "grad_norm": 0.12805306418161266, + "learning_rate": 3.2414767316974514e-06, + "loss": 0.015, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01483713649213314, + "step": 2225, + "valid_targets_mean": 5160.8, + "valid_targets_min": 2420 + }, + { + "epoch": 7.101910828025478, + "grad_norm": 0.13354317728810114, + "learning_rate": 3.131082075634523e-06, + "loss": 0.0145, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01506053190678358, + "step": 2230, + "valid_targets_mean": 4750.2, + "valid_targets_min": 1457 + }, + { + "epoch": 7.117834394904459, + "grad_norm": 0.14056574708790995, + "learning_rate": 3.022539205326258e-06, + "loss": 0.0152, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01632022112607956, + "step": 2235, + "valid_targets_mean": 4974.7, + "valid_targets_min": 2056 + }, + { + "epoch": 7.13375796178344, + "grad_norm": 0.1508010920200611, + "learning_rate": 2.9158524092894214e-06, + "loss": 0.0143, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.015547826886177063, + "step": 2240, + "valid_targets_mean": 4667.0, + "valid_targets_min": 914 + }, + { + "epoch": 7.149681528662421, + "grad_norm": 0.14457832903928367, + "learning_rate": 2.8110259027075258e-06, + "loss": 0.0145, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.015580195002257824, + "step": 2245, + "valid_targets_mean": 5049.9, + "valid_targets_min": 1390 + }, + { + "epoch": 7.165605095541402, + "grad_norm": 0.13896829957456167, + "learning_rate": 2.708063827264223e-06, + "loss": 0.0138, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.012920582666993141, + "step": 2250, + "valid_targets_mean": 4977.2, + "valid_targets_min": 1128 + }, + { + "epoch": 7.181528662420382, + "grad_norm": 0.14314875798019044, + "learning_rate": 2.6069702509796957e-06, + "loss": 0.0146, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01511191576719284, + "step": 2255, + "valid_targets_mean": 5053.6, + "valid_targets_min": 1088 + }, + { + "epoch": 7.197452229299363, + "grad_norm": 0.12792261173248898, + "learning_rate": 2.507749168049961e-06, + "loss": 0.0141, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014054584316909313, + "step": 2260, + "valid_targets_mean": 5186.3, + "valid_targets_min": 1654 + }, + { + "epoch": 7.213375796178344, + "grad_norm": 0.1470283549316241, + "learning_rate": 2.410404498688995e-06, + "loss": 0.0152, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.016185037791728973, + "step": 2265, + "valid_targets_mean": 4499.9, + "valid_targets_min": 1202 + }, + { + "epoch": 7.229299363057325, + "grad_norm": 0.13968778084568456, + "learning_rate": 2.3149400889739157e-06, + "loss": 0.0149, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.015277128666639328, + "step": 2270, + "valid_targets_mean": 4866.9, + "valid_targets_min": 2874 + }, + { + "epoch": 7.245222929936306, + "grad_norm": 0.13781604597249436, + "learning_rate": 2.221359710692961e-06, + "loss": 0.0142, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013797796331346035, + "step": 2275, + "valid_targets_mean": 5091.4, + "valid_targets_min": 1508 + }, + { + "epoch": 7.261146496815287, + "grad_norm": 0.13683845508769282, + "learning_rate": 2.1296670611965273e-06, + "loss": 0.0147, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01435169018805027, + "step": 2280, + "valid_targets_mean": 5275.8, + "valid_targets_min": 3924 + }, + { + "epoch": 7.277070063694268, + "grad_norm": 0.14296823227878563, + "learning_rate": 2.039865763251014e-06, + "loss": 0.0146, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014473952353000641, + "step": 2285, + "valid_targets_mean": 4800.2, + "valid_targets_min": 566 + }, + { + "epoch": 7.292993630573249, + "grad_norm": 0.1575107336940688, + "learning_rate": 1.951959364895778e-06, + "loss": 0.0148, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.015667583793401718, + "step": 2290, + "valid_targets_mean": 4805.3, + "valid_targets_min": 1319 + }, + { + "epoch": 7.308917197452229, + "grad_norm": 0.1349565160583637, + "learning_rate": 1.865951339302857e-06, + "loss": 0.0137, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014135218225419521, + "step": 2295, + "valid_targets_mean": 5276.8, + "valid_targets_min": 2436 + }, + { + "epoch": 7.32484076433121, + "grad_norm": 0.14090268676822532, + "learning_rate": 1.7818450846398327e-06, + "loss": 0.0148, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014909759163856506, + "step": 2300, + "valid_targets_mean": 5077.6, + "valid_targets_min": 2532 + }, + { + "epoch": 7.340764331210191, + "grad_norm": 0.13743756479166525, + "learning_rate": 1.6996439239354956e-06, + "loss": 0.0145, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013377371244132519, + "step": 2305, + "valid_targets_mean": 5274.4, + "valid_targets_min": 2945 + }, + { + "epoch": 7.356687898089172, + "grad_norm": 0.13831095894721585, + "learning_rate": 1.6193511049486076e-06, + "loss": 0.0149, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013829350471496582, + "step": 2310, + "valid_targets_mean": 4999.8, + "valid_targets_min": 544 + }, + { + "epoch": 7.372611464968153, + "grad_norm": 0.14536010961036458, + "learning_rate": 1.5409698000395379e-06, + "loss": 0.0147, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.015978479757905006, + "step": 2315, + "valid_targets_mean": 4870.9, + "valid_targets_min": 1115 + }, + { + "epoch": 7.388535031847134, + "grad_norm": 0.149330776493363, + "learning_rate": 1.4645031060449732e-06, + "loss": 0.0149, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.015622340142726898, + "step": 2320, + "valid_targets_mean": 4621.8, + "valid_targets_min": 1352 + }, + { + "epoch": 7.404458598726115, + "grad_norm": 0.14360555016001797, + "learning_rate": 1.389954044155517e-06, + "loss": 0.0136, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013295633718371391, + "step": 2325, + "valid_targets_mean": 5079.9, + "valid_targets_min": 1036 + }, + { + "epoch": 7.420382165605096, + "grad_norm": 0.13342161850333914, + "learning_rate": 1.3173255597963397e-06, + "loss": 0.0146, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013683140277862549, + "step": 2330, + "valid_targets_mean": 4955.8, + "valid_targets_min": 1106 + }, + { + "epoch": 7.436305732484076, + "grad_norm": 0.130440591206871, + "learning_rate": 1.2466205225108273e-06, + "loss": 0.0142, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013634752482175827, + "step": 2335, + "valid_targets_mean": 5453.8, + "valid_targets_min": 2635 + }, + { + "epoch": 7.452229299363057, + "grad_norm": 0.13188356406662988, + "learning_rate": 1.1778417258471507e-06, + "loss": 0.0147, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.015135698020458221, + "step": 2340, + "valid_targets_mean": 5107.4, + "valid_targets_min": 2691 + }, + { + "epoch": 7.468152866242038, + "grad_norm": 0.14878509428839537, + "learning_rate": 1.1109918872479642e-06, + "loss": 0.0145, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01526675932109356, + "step": 2345, + "valid_targets_mean": 5059.0, + "valid_targets_min": 2043 + }, + { + "epoch": 7.484076433121019, + "grad_norm": 0.13426782382164532, + "learning_rate": 1.0460736479429633e-06, + "loss": 0.0148, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01532701589167118, + "step": 2350, + "valid_targets_mean": 5099.2, + "valid_targets_min": 1391 + }, + { + "epoch": 7.5, + "grad_norm": 0.1514130635607868, + "learning_rate": 9.830895728446143e-07, + "loss": 0.014, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01387743465602398, + "step": 2355, + "valid_targets_mean": 4820.6, + "valid_targets_min": 1019 + }, + { + "epoch": 7.515923566878981, + "grad_norm": 0.13987108759280026, + "learning_rate": 9.220421504467281e-07, + "loss": 0.0149, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013630231842398643, + "step": 2360, + "valid_targets_mean": 5016.3, + "valid_targets_min": 1992 + }, + { + "epoch": 7.531847133757962, + "grad_norm": 0.14122538050337716, + "learning_rate": 8.629337927262171e-07, + "loss": 0.0137, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014416727237403393, + "step": 2365, + "valid_targets_mean": 5320.2, + "valid_targets_min": 3998 + }, + { + "epoch": 7.547770700636943, + "grad_norm": 0.13590071943914508, + "learning_rate": 8.057668350477332e-07, + "loss": 0.0146, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014858937822282314, + "step": 2370, + "valid_targets_mean": 5052.1, + "valid_targets_min": 2215 + }, + { + "epoch": 7.563694267515924, + "grad_norm": 0.12796395658198087, + "learning_rate": 7.505435360714408e-07, + "loss": 0.014, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013363628648221493, + "step": 2375, + "valid_targets_mean": 5254.8, + "valid_targets_min": 2131 + }, + { + "epoch": 7.579617834394904, + "grad_norm": 0.13625143162000777, + "learning_rate": 6.972660776637663e-07, + "loss": 0.0143, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014759352430701256, + "step": 2380, + "valid_targets_mean": 4653.7, + "valid_targets_min": 1389 + }, + { + "epoch": 7.595541401273885, + "grad_norm": 0.14158463511416328, + "learning_rate": 6.459365648111948e-07, + "loss": 0.0149, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014833863824605942, + "step": 2385, + "valid_targets_mean": 4894.2, + "valid_targets_min": 481 + }, + { + "epoch": 7.611464968152866, + "grad_norm": 0.1368611661640719, + "learning_rate": 5.965570255370866e-07, + "loss": 0.0142, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014796525239944458, + "step": 2390, + "valid_targets_mean": 5288.2, + "valid_targets_min": 3954 + }, + { + "epoch": 7.627388535031847, + "grad_norm": 0.12992144110563536, + "learning_rate": 5.491294108215695e-07, + "loss": 0.0139, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01267107855528593, + "step": 2395, + "valid_targets_mean": 5069.5, + "valid_targets_min": 1349 + }, + { + "epoch": 7.643312101910828, + "grad_norm": 0.13296240180023966, + "learning_rate": 5.036555945244436e-07, + "loss": 0.0147, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014034183695912361, + "step": 2400, + "valid_targets_mean": 5043.5, + "valid_targets_min": 1196 + }, + { + "epoch": 7.659235668789809, + "grad_norm": 0.14206452453996443, + "learning_rate": 4.6013737331115913e-07, + "loss": 0.0142, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01428112480789423, + "step": 2405, + "valid_targets_mean": 5075.5, + "valid_targets_min": 1359 + }, + { + "epoch": 7.67515923566879, + "grad_norm": 0.11870198802782875, + "learning_rate": 4.1857646658180483e-07, + "loss": 0.0143, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.012596381828188896, + "step": 2410, + "valid_targets_mean": 5705.8, + "valid_targets_min": 1705 + }, + { + "epoch": 7.69108280254777, + "grad_norm": 0.15736513778602965, + "learning_rate": 3.7897451640321323e-07, + "loss": 0.0139, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01441703550517559, + "step": 2415, + "valid_targets_mean": 4747.2, + "valid_targets_min": 1180 + }, + { + "epoch": 7.707006369426751, + "grad_norm": 0.13906832758441198, + "learning_rate": 3.413330874440401e-07, + "loss": 0.0144, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.015470558777451515, + "step": 2420, + "valid_targets_mean": 4900.7, + "valid_targets_min": 2395 + }, + { + "epoch": 7.722929936305732, + "grad_norm": 0.12962605003550104, + "learning_rate": 3.056536669129806e-07, + "loss": 0.015, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013977251015603542, + "step": 2425, + "valid_targets_mean": 5074.6, + "valid_targets_min": 3108 + }, + { + "epoch": 7.738853503184713, + "grad_norm": 0.14054159968864988, + "learning_rate": 2.7193766449997717e-07, + "loss": 0.0148, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01572270132601261, + "step": 2430, + "valid_targets_mean": 4632.7, + "valid_targets_min": 618 + }, + { + "epoch": 7.754777070063694, + "grad_norm": 0.13870055158990865, + "learning_rate": 2.4018641232055326e-07, + "loss": 0.0146, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014715202152729034, + "step": 2435, + "valid_targets_mean": 5131.9, + "valid_targets_min": 2214 + }, + { + "epoch": 7.770700636942675, + "grad_norm": 0.13264633703287376, + "learning_rate": 2.1040116486317207e-07, + "loss": 0.0143, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013027691282331944, + "step": 2440, + "valid_targets_mean": 5244.4, + "valid_targets_min": 3246 + }, + { + "epoch": 7.786624203821656, + "grad_norm": 0.14502255964044747, + "learning_rate": 1.8258309893965375e-07, + "loss": 0.0145, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.015363199636340141, + "step": 2445, + "valid_targets_mean": 5094.7, + "valid_targets_min": 1606 + }, + { + "epoch": 7.802547770700637, + "grad_norm": 0.13990224186200548, + "learning_rate": 1.567333136387017e-07, + "loss": 0.0147, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.01517225056886673, + "step": 2450, + "valid_targets_mean": 4948.6, + "valid_targets_min": 1572 + }, + { + "epoch": 7.818471337579618, + "grad_norm": 0.14240511938677056, + "learning_rate": 1.3285283028247053e-07, + "loss": 0.014, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013703295029699802, + "step": 2455, + "valid_targets_mean": 5039.0, + "valid_targets_min": 1348 + }, + { + "epoch": 7.834394904458598, + "grad_norm": 0.12685045401716433, + "learning_rate": 1.1094259238620952e-07, + "loss": 0.0142, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014196784235537052, + "step": 2460, + "valid_targets_mean": 5497.5, + "valid_targets_min": 3848 + }, + { + "epoch": 7.850318471337579, + "grad_norm": 0.1449642144197079, + "learning_rate": 9.100346562099238e-08, + "loss": 0.0138, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013698741793632507, + "step": 2465, + "valid_targets_mean": 5246.8, + "valid_targets_min": 3522 + }, + { + "epoch": 7.86624203821656, + "grad_norm": 0.14027521679236274, + "learning_rate": 7.30362377795113e-08, + "loss": 0.0149, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013331707566976547, + "step": 2470, + "valid_targets_mean": 5441.8, + "valid_targets_min": 2542 + }, + { + "epoch": 7.882165605095541, + "grad_norm": 0.14688871780142937, + "learning_rate": 5.704161874494629e-08, + "loss": 0.0145, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014748896472156048, + "step": 2475, + "valid_targets_mean": 5267.1, + "valid_targets_min": 2112 + }, + { + "epoch": 7.898089171974522, + "grad_norm": 0.1253158283649276, + "learning_rate": 4.302024046292097e-08, + "loss": 0.0147, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013722356408834457, + "step": 2480, + "valid_targets_mean": 5239.2, + "valid_targets_min": 1600 + }, + { + "epoch": 7.914012738853503, + "grad_norm": 0.13562658789731333, + "learning_rate": 3.0972656916539166e-08, + "loss": 0.0143, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013688988983631134, + "step": 2485, + "valid_targets_mean": 5195.9, + "valid_targets_min": 3039 + }, + { + "epoch": 7.929936305732484, + "grad_norm": 0.15386821457808864, + "learning_rate": 2.089934410449135e-08, + "loss": 0.0144, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.016422059386968613, + "step": 2490, + "valid_targets_mean": 5058.2, + "valid_targets_min": 1292 + }, + { + "epoch": 7.945859872611465, + "grad_norm": 0.12178232365607229, + "learning_rate": 1.2800700022252975e-08, + "loss": 0.0138, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.012545240111649036, + "step": 2495, + "valid_targets_mean": 5131.6, + "valid_targets_min": 1244 + }, + { + "epoch": 7.961783439490446, + "grad_norm": 0.13036833645836712, + "learning_rate": 6.677044646352659e-09, + "loss": 0.0143, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.014231288805603981, + "step": 2500, + "valid_targets_mean": 5133.1, + "valid_targets_min": 837 + }, + { + "epoch": 7.977707006369426, + "grad_norm": 0.14011532682948652, + "learning_rate": 2.528619921732256e-09, + "loss": 0.0149, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.015756778419017792, + "step": 2505, + "valid_targets_mean": 4749.8, + "valid_targets_min": 2603 + }, + { + "epoch": 7.993630573248407, + "grad_norm": 0.14211784778788789, + "learning_rate": 3.555897522045104e-10, + "loss": 0.0141, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.013298088684678078, + "step": 2510, + "valid_targets_mean": 5067.2, + "valid_targets_min": 1078 + }, + { + "epoch": 8.0, + "loss_nan_ranks": 0, + "loss_rank_avg": 0.012045512907207012, + "step": 2512, + "total_flos": 2887866713112576.0, + "train_loss": 0.1099344994546549, + "train_runtime": 45112.6253, + "train_samples_per_second": 1.778, + "train_steps_per_second": 0.056, + "valid_targets_mean": 4840.4, + "valid_targets_min": 726 + } + ], + "logging_steps": 5, + "max_steps": 2512, + "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": 2887866713112576.0, + "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..c3a59bf --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f3448f909cad432a0d81f3c2741af959eb2551ace00d955e357219d51e50702e +size 8721 diff --git a/training_loss.png b/training_loss.png new file mode 100644 index 0000000..03e7163 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