commit 1c090adbdc45a844335579aa1848a725c295f7f4 Author: ModelHub XC Date: Mon Jul 13 21:23:18 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: bsq1989/qwen_4b_sql Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..52373fe --- /dev/null +++ b/.gitattributes @@ -0,0 +1,36 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt 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 +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz 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 +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl 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 +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* 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 +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/README.md b/README.md new file mode 100644 index 0000000..5c4522c --- /dev/null +++ b/README.md @@ -0,0 +1,131 @@ +--- +base_model: Qwen/Qwen3-4B-Base +library_name: transformers +pipeline_tag: text-generation +tags: + - text-to-sql + - sql + - qwen3 + - llamafactory + - spider + - spider-test-suite +--- + +# qwen_4b_sql + +`qwen_4b_sql` is a `Qwen3-4B-Base` model finetuned for text-to-SQL generation with full SFT on a cleaned split of `PipableAI/pip-txt-to-sql-spider-bird-dataset`. + +This repository tracks the stronger 4B checkpoint from our H20 single-GPU training runs. In our internal comparison, this checkpoint outperformed the corresponding `Qwen3-1.7B-Base` baseline on Spider execution accuracy. + +## Base Model + +- Base model: [`Qwen/Qwen3-4B-Base`](https://huggingface.co/Qwen/Qwen3-4B-Base) +- Finetuning framework: `LLaMA-Factory` +- Training mode: `Full SFT` +- Task: `schema + question -> SQL only` + +## Training Data + +- Primary dataset: [`PipableAI/pip-txt-to-sql-spider-bird-dataset`](https://huggingface.co/datasets/PipableAI/pip-txt-to-sql-spider-bird-dataset) +- We used a cleaned local split derived from that dataset for train/validation + +## Training Setup + +- Hardware: single `NVIDIA H20 96GB` +- Precision: `bf16` +- Context length: `2048` +- Per-device train batch size: `1` +- Gradient accumulation steps: `8` +- Effective batch size: `8` +- Learning rate: `5e-6` +- Scheduler: `cosine` +- Warmup steps: `300` +- Epochs: `4.0` +- Template: `qwen3_nothink` +- Best-checkpoint selection: `load_best_model_at_end = true` + +## Spider Benchmark + +The following numbers are from Spider dev using the official evaluation tooling: + +- Official `match` evaluation from `test-suite-sql-eval` +- Official Spider `Test Suite` execution evaluation + +### Main Results + +| Metric | Score | +| --- | ---: | +| Spider official exact match | 35.0% | +| Spider Test Suite execution accuracy | 67.6% | + +### Difficulty Breakdown + +| Difficulty | Exact Match | Test Suite Exec | +| --- | ---: | ---: | +| Easy | 64.9% | 87.5% | +| Medium | 37.4% | 72.9% | +| Hard | 16.1% | 50.0% | +| Extra | 3.6% | 42.2% | + +## Notes + +- This model is stronger under execution-based Spider evaluation than our best `Qwen3-1.7B-Base` run. +- In our experiments, exact-match metrics were often stricter than execution-based metrics because semantically valid SQL rewrites do not always match the Spider gold form exactly. +- A later 4B rerun with altered training settings underperformed this checkpoint on Spider and is not the checkpoint published here. + +## Intended Use + +This model is intended for: + +- text-to-SQL research baselines +- schema-conditioned SQL generation experiments +- single-turn SQL generation from natural language plus schema text + +It is not validated for: + +- production-grade database access control +- unrestricted execution over arbitrary enterprise schemas +- multi-turn agent workflows without extra prompting / tooling + +## Example Usage + +```python +from transformers import AutoTokenizer, AutoModelForCausalLM + +model_id = "bsq1989/qwen_4b_sql" + +tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True) +model = AutoModelForCausalLM.from_pretrained( + model_id, + torch_dtype="auto", + device_map="auto", + trust_remote_code=True, +) + +prompt = """Generate SQL from the given schema and question. Output SQL only. + +Schema: +CREATE TABLE twitter (TweetID INTEGER, UserID INTEGER, LocationID INTEGER, Lang TEXT, ...); +CREATE TABLE location (LocationID INTEGER, Country TEXT, City TEXT, ...); + +Question: +How many tweets are in English? +""" + +messages = [{"role": "user", "content": prompt}] +text = tokenizer.apply_chat_template( + messages, + tokenize=False, + add_generation_prompt=True, +) + +inputs = tokenizer(text, return_tensors="pt").to(model.device) +outputs = model.generate(**inputs, max_new_tokens=256) +print(tokenizer.decode(outputs[0], skip_special_tokens=True)) +``` + +## Limitations + +- Performance drops on more open-ended and heterogeneous SQL benchmarks than Spider. +- The model can still produce invalid column references on out-of-distribution schemas. +- Benchmark numbers here reflect our current internal setup and should be reproduced with the same evaluation pipeline for strict comparison. diff --git a/all_results.json b/all_results.json new file mode 100644 index 0000000..0363d63 --- /dev/null +++ b/all_results.json @@ -0,0 +1,13 @@ +{ + "effective_tokens_per_sec": 1153.798380212981, + "epoch": 4.0, + "eval_loss": 0.24901165068149567, + "eval_runtime": 46.8427, + "eval_samples_per_second": 18.786, + "eval_steps_per_second": 18.786, + "total_flos": 3.545203061907456e+17, + "train_loss": 0.08566030774240586, + "train_runtime": 13933.4985, + "train_samples_per_second": 3.772, + "train_steps_per_second": 0.472 +} \ No newline at end of file diff --git a/chat_template.jinja b/chat_template.jinja new file mode 100644 index 0000000..699ff8d --- /dev/null +++ b/chat_template.jinja @@ -0,0 +1,85 @@ +{%- 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 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.role == "user") or (message.role == "system" and not loop.first) %} + {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }} + {%- elif message.role == "assistant" %} + {%- set content = message.content %} + {%- set reasoning_content = '' %} + {%- if message.reasoning_content is defined and message.reasoning_content is not none %} + {%- set reasoning_content = message.reasoning_content %} + {%- else %} + {%- if '' in message.content %} + {%- set content = message.content.split('')[-1].lstrip('\n') %} + {%- set reasoning_content = message.content.split('')[0].rstrip('\n').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' }} + {{- message.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..3e55494 --- /dev/null +++ b/config.json @@ -0,0 +1,71 @@ +{ + "architectures": [ + "Qwen3ForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": null, + "dtype": "float32", + "eos_token_id": 151645, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 2560, + "initializer_range": 0.02, + "intermediate_size": 9728, + "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": 32768, + "max_window_layers": 36, + "model_type": "qwen3", + "num_attention_heads": 32, + "num_hidden_layers": 36, + "num_key_value_heads": 8, + "pad_token_id": 151643, + "rms_norm_eps": 1e-06, + "rope_parameters": { + "rope_theta": 1000000, + "rope_type": "default" + }, + "sliding_window": null, + "tie_word_embeddings": true, + "transformers_version": "5.2.0", + "use_cache": false, + "use_sliding_window": false, + "vocab_size": 151936 +} diff --git a/eval_results.json b/eval_results.json new file mode 100644 index 0000000..1443bbe --- /dev/null +++ b/eval_results.json @@ -0,0 +1,7 @@ +{ + "epoch": 4.0, + "eval_loss": 0.24901165068149567, + "eval_runtime": 46.8427, + "eval_samples_per_second": 18.786, + "eval_steps_per_second": 18.786 +} \ No newline at end of file diff --git a/generation_config.json b/generation_config.json new file mode 100644 index 0000000..c35d61c --- /dev/null +++ b/generation_config.json @@ -0,0 +1,10 @@ +{ + "do_sample": false, + "eos_token_id": [ + 151645, + 151643 + ], + "max_new_tokens": 2048, + "pad_token_id": 151643, + "transformers_version": "5.2.0" +} diff --git a/model.safetensors b/model.safetensors new file mode 100644 index 0000000..b8431bf --- /dev/null +++ b/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a04a88d8d7c026499bc570cdb42e1cbc0af368cf779b2f1060543a5d47dee1a5 +size 16089918232 diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000..c7afbed --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:be75606093db2094d7cd20f3c2f385c212750648bd6ea4fb2bf507a6a4c55506 +size 11422650 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000..271320b --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,30 @@ +{ + "add_prefix_space": false, + "backend": "tokenizers", + "bos_token": null, + "clean_up_tokenization_spaces": false, + "eos_token": "<|im_end|>", + "errors": "replace", + "extra_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|>" + ], + "is_local": true, + "model_max_length": 131072, + "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..0616efe --- /dev/null +++ b/train_results.json @@ -0,0 +1,9 @@ +{ + "effective_tokens_per_sec": 1153.798380212981, + "epoch": 4.0, + "total_flos": 3.545203061907456e+17, + "train_loss": 0.08566030774240586, + "train_runtime": 13933.4985, + "train_samples_per_second": 3.772, + "train_steps_per_second": 0.472 +} \ No newline at end of file diff --git a/trainer_log.jsonl b/trainer_log.jsonl new file mode 100644 index 0000000..4aa0fe1 --- /dev/null +++ b/trainer_log.jsonl @@ -0,0 +1,342 @@ +{"current_steps": 20, "total_steps": 6572, "loss": 1.119423007965088, "lr": 3.166666666666667e-07, "epoch": 0.0121765601217656, "percentage": 0.3, "elapsed_time": "0:00:54", "remaining_time": "4:54:53"} +{"current_steps": 40, "total_steps": 6572, "loss": 0.8098940849304199, "lr": 6.5e-07, "epoch": 0.0243531202435312, "percentage": 0.61, "elapsed_time": "0:01:37", "remaining_time": "4:25:19"} +{"current_steps": 60, "total_steps": 6572, "loss": 0.6310151576995849, "lr": 9.833333333333334e-07, "epoch": 0.0365296803652968, "percentage": 0.91, "elapsed_time": "0:02:14", "remaining_time": "4:03:37"} +{"current_steps": 80, "total_steps": 6572, "loss": 0.5526751518249512, "lr": 1.3166666666666666e-06, "epoch": 0.0487062404870624, "percentage": 1.22, "elapsed_time": "0:02:51", "remaining_time": "3:52:34"} +{"current_steps": 100, "total_steps": 6572, "loss": 0.4834024906158447, "lr": 1.6500000000000003e-06, "epoch": 0.060882800608828, "percentage": 1.52, "elapsed_time": "0:03:29", "remaining_time": "3:45:38"} +{"current_steps": 120, "total_steps": 6572, "loss": 0.44873433113098143, "lr": 1.9833333333333335e-06, "epoch": 0.0730593607305936, "percentage": 1.83, "elapsed_time": "0:04:06", "remaining_time": "3:40:39"} +{"current_steps": 140, "total_steps": 6572, "loss": 0.38286705017089845, "lr": 2.316666666666667e-06, "epoch": 0.0852359208523592, "percentage": 2.13, "elapsed_time": "0:04:44", "remaining_time": "3:37:42"} +{"current_steps": 160, "total_steps": 6572, "loss": 0.38465352058410646, "lr": 2.6500000000000005e-06, "epoch": 0.0974124809741248, "percentage": 2.43, "elapsed_time": "0:05:21", "remaining_time": "3:34:27"} +{"current_steps": 180, "total_steps": 6572, "loss": 0.3305965900421143, "lr": 2.9833333333333337e-06, "epoch": 0.1095890410958904, "percentage": 2.74, "elapsed_time": "0:05:56", "remaining_time": "3:31:11"} +{"current_steps": 200, "total_steps": 6572, "loss": 0.2875316619873047, "lr": 3.316666666666667e-06, "epoch": 0.121765601217656, "percentage": 3.04, "elapsed_time": "0:06:33", "remaining_time": "3:28:43"} +{"current_steps": 220, "total_steps": 6572, "loss": 0.3140716075897217, "lr": 3.65e-06, "epoch": 0.1339421613394216, "percentage": 3.35, "elapsed_time": "0:07:09", "remaining_time": "3:26:36"} +{"current_steps": 240, "total_steps": 6572, "loss": 0.2985520839691162, "lr": 3.983333333333334e-06, "epoch": 0.1461187214611872, "percentage": 3.65, "elapsed_time": "0:07:45", "remaining_time": "3:24:30"} +{"current_steps": 260, "total_steps": 6572, "loss": 0.2449601411819458, "lr": 4.316666666666667e-06, "epoch": 0.1582952815829528, "percentage": 3.96, "elapsed_time": "0:08:21", "remaining_time": "3:22:57"} +{"current_steps": 280, "total_steps": 6572, "loss": 0.24566495418548584, "lr": 4.65e-06, "epoch": 0.1704718417047184, "percentage": 4.26, "elapsed_time": "0:08:58", "remaining_time": "3:21:42"} +{"current_steps": 300, "total_steps": 6572, "loss": 0.20863604545593262, "lr": 4.983333333333334e-06, "epoch": 0.182648401826484, "percentage": 4.56, "elapsed_time": "0:09:34", "remaining_time": "3:20:20"} +{"current_steps": 320, "total_steps": 6572, "loss": 0.19693111181259154, "lr": 4.9998867856224845e-06, "epoch": 0.1948249619482496, "percentage": 4.87, "elapsed_time": "0:10:11", "remaining_time": "3:19:13"} +{"current_steps": 340, "total_steps": 6572, "loss": 0.21833207607269287, "lr": 4.999523005839606e-06, "epoch": 0.2070015220700152, "percentage": 5.17, "elapsed_time": "0:10:47", "remaining_time": "3:17:45"} +{"current_steps": 360, "total_steps": 6572, "loss": 0.2382877826690674, "lr": 4.998908383543311e-06, "epoch": 0.2191780821917808, "percentage": 5.48, "elapsed_time": "0:11:21", "remaining_time": "3:16:06"} +{"current_steps": 380, "total_steps": 6572, "loss": 0.21118538379669188, "lr": 4.9980429804147276e-06, "epoch": 0.2313546423135464, "percentage": 5.78, "elapsed_time": "0:11:58", "remaining_time": "3:15:01"} +{"current_steps": 400, "total_steps": 6572, "loss": 0.18367968797683715, "lr": 4.996926883302385e-06, "epoch": 0.243531202435312, "percentage": 6.09, "elapsed_time": "0:12:34", "remaining_time": "3:13:55"} +{"current_steps": 420, "total_steps": 6572, "loss": 0.22483460903167723, "lr": 4.995560204213496e-06, "epoch": 0.2557077625570776, "percentage": 6.39, "elapsed_time": "0:13:12", "remaining_time": "3:13:22"} +{"current_steps": 440, "total_steps": 6572, "loss": 0.22673561573028564, "lr": 4.993943080302715e-06, "epoch": 0.2678843226788432, "percentage": 6.7, "elapsed_time": "0:13:50", "remaining_time": "3:12:57"} +{"current_steps": 460, "total_steps": 6572, "loss": 0.185296630859375, "lr": 4.992075673858379e-06, "epoch": 0.2800608828006088, "percentage": 7.0, "elapsed_time": "0:14:26", "remaining_time": "3:11:55"} +{"current_steps": 480, "total_steps": 6572, "loss": 0.18437937498092652, "lr": 4.989958172286214e-06, "epoch": 0.2922374429223744, "percentage": 7.3, "elapsed_time": "0:15:04", "remaining_time": "3:11:23"} +{"current_steps": 500, "total_steps": 6572, "loss": 0.1850834846496582, "lr": 4.987590788090533e-06, "epoch": 0.30441400304414, "percentage": 7.61, "elapsed_time": "0:15:42", "remaining_time": "3:10:48"} +{"current_steps": 500, "total_steps": 6572, "eval_loss": 0.2562323212623596, "epoch": 0.30441400304414, "percentage": 7.61, "elapsed_time": "0:16:30", "remaining_time": "3:20:23"} +{"current_steps": 520, "total_steps": 6572, "loss": 0.16346561908721924, "lr": 4.984973758852904e-06, "epoch": 0.3165905631659056, "percentage": 7.91, "elapsed_time": "0:18:27", "remaining_time": "3:34:45"} +{"current_steps": 540, "total_steps": 6572, "loss": 0.18838067054748536, "lr": 4.982107347208317e-06, "epoch": 0.3287671232876712, "percentage": 8.22, "elapsed_time": "0:19:03", "remaining_time": "3:32:49"} +{"current_steps": 560, "total_steps": 6572, "loss": 0.177593994140625, "lr": 4.978991840818816e-06, "epoch": 0.3409436834094368, "percentage": 8.52, "elapsed_time": "0:19:40", "remaining_time": "3:31:13"} +{"current_steps": 580, "total_steps": 6572, "loss": 0.20775914192199707, "lr": 4.975627552344638e-06, "epoch": 0.3531202435312024, "percentage": 8.83, "elapsed_time": "0:20:18", "remaining_time": "3:29:49"} +{"current_steps": 600, "total_steps": 6572, "loss": 0.16498700380325318, "lr": 4.97201481941283e-06, "epoch": 0.365296803652968, "percentage": 9.13, "elapsed_time": "0:20:55", "remaining_time": "3:28:12"} +{"current_steps": 620, "total_steps": 6572, "loss": 0.17565951347351075, "lr": 4.968154004583374e-06, "epoch": 0.3774733637747336, "percentage": 9.43, "elapsed_time": "0:21:31", "remaining_time": "3:26:35"} +{"current_steps": 640, "total_steps": 6572, "loss": 0.16204673051834106, "lr": 4.964045495312794e-06, "epoch": 0.3896499238964992, "percentage": 9.74, "elapsed_time": "0:22:07", "remaining_time": "3:24:59"} +{"current_steps": 660, "total_steps": 6572, "loss": 0.17068564891815186, "lr": 4.959689703915272e-06, "epoch": 0.4018264840182648, "percentage": 10.04, "elapsed_time": "0:22:42", "remaining_time": "3:23:27"} +{"current_steps": 680, "total_steps": 6572, "loss": 0.1589680790901184, "lr": 4.95508706752128e-06, "epoch": 0.4140030441400304, "percentage": 10.35, "elapsed_time": "0:23:19", "remaining_time": "3:22:07"} +{"current_steps": 700, "total_steps": 6572, "loss": 0.17568455934524535, "lr": 4.9502380480337e-06, "epoch": 0.426179604261796, "percentage": 10.65, "elapsed_time": "0:23:56", "remaining_time": "3:20:51"} +{"current_steps": 720, "total_steps": 6572, "loss": 0.16204804182052612, "lr": 4.9451431320814715e-06, "epoch": 0.4383561643835616, "percentage": 10.96, "elapsed_time": "0:24:33", "remaining_time": "3:19:40"} +{"current_steps": 740, "total_steps": 6572, "loss": 0.16562143564224244, "lr": 4.939802830970762e-06, "epoch": 0.4505327245053272, "percentage": 11.26, "elapsed_time": "0:25:10", "remaining_time": "3:18:26"} +{"current_steps": 760, "total_steps": 6572, "loss": 0.17697544097900392, "lr": 4.934217680633646e-06, "epoch": 0.4627092846270928, "percentage": 11.56, "elapsed_time": "0:25:46", "remaining_time": "3:17:03"} +{"current_steps": 780, "total_steps": 6572, "loss": 0.1649466037750244, "lr": 4.928388241574327e-06, "epoch": 0.4748858447488584, "percentage": 11.87, "elapsed_time": "0:26:23", "remaining_time": "3:15:55"} +{"current_steps": 800, "total_steps": 6572, "loss": 0.1602837324142456, "lr": 4.922315098812883e-06, "epoch": 0.487062404870624, "percentage": 12.17, "elapsed_time": "0:26:59", "remaining_time": "3:14:43"} +{"current_steps": 820, "total_steps": 6572, "loss": 0.142719042301178, "lr": 4.9159988618265585e-06, "epoch": 0.4992389649923896, "percentage": 12.48, "elapsed_time": "0:27:36", "remaining_time": "3:13:37"} +{"current_steps": 840, "total_steps": 6572, "loss": 0.14508233070373536, "lr": 4.9094401644886e-06, "epoch": 0.5114155251141552, "percentage": 12.78, "elapsed_time": "0:28:12", "remaining_time": "3:12:31"} +{"current_steps": 860, "total_steps": 6572, "loss": 0.1821539044380188, "lr": 4.902639665004641e-06, "epoch": 0.5235920852359208, "percentage": 13.09, "elapsed_time": "0:28:49", "remaining_time": "3:11:27"} +{"current_steps": 880, "total_steps": 6572, "loss": 0.16131887435913086, "lr": 4.89559804584665e-06, "epoch": 0.5357686453576864, "percentage": 13.39, "elapsed_time": "0:29:24", "remaining_time": "3:10:16"} +{"current_steps": 900, "total_steps": 6572, "loss": 0.17404688596725465, "lr": 4.888316013684435e-06, "epoch": 0.547945205479452, "percentage": 13.69, "elapsed_time": "0:30:01", "remaining_time": "3:09:11"} +{"current_steps": 920, "total_steps": 6572, "loss": 0.14134640693664552, "lr": 4.880794299314732e-06, "epoch": 0.5601217656012176, "percentage": 14.0, "elapsed_time": "0:30:38", "remaining_time": "3:08:12"} +{"current_steps": 940, "total_steps": 6572, "loss": 0.14891813993453978, "lr": 4.87303365758786e-06, "epoch": 0.5722983257229832, "percentage": 14.3, "elapsed_time": "0:31:15", "remaining_time": "3:07:19"} +{"current_steps": 960, "total_steps": 6572, "loss": 0.1696299910545349, "lr": 4.865034867331967e-06, "epoch": 0.5844748858447488, "percentage": 14.61, "elapsed_time": "0:31:52", "remaining_time": "3:06:21"} +{"current_steps": 980, "total_steps": 6572, "loss": 0.14085158109664916, "lr": 4.856798731274874e-06, "epoch": 0.5966514459665144, "percentage": 14.91, "elapsed_time": "0:32:29", "remaining_time": "3:05:22"} +{"current_steps": 1000, "total_steps": 6572, "loss": 0.133053982257843, "lr": 4.84832607596351e-06, "epoch": 0.60882800608828, "percentage": 15.22, "elapsed_time": "0:33:04", "remaining_time": "3:04:16"} +{"current_steps": 1000, "total_steps": 6572, "eval_loss": 0.24901165068149567, "epoch": 0.60882800608828, "percentage": 15.22, "elapsed_time": "0:33:51", "remaining_time": "3:08:37"} +{"current_steps": 1020, "total_steps": 6572, "loss": 0.12680984735488893, "lr": 4.8396177516809695e-06, "epoch": 0.6210045662100456, "percentage": 15.52, "elapsed_time": "0:36:06", "remaining_time": "3:16:32"} +{"current_steps": 1040, "total_steps": 6572, "loss": 0.14880582094192504, "lr": 4.830674632361178e-06, "epoch": 0.6331811263318112, "percentage": 15.82, "elapsed_time": "0:36:43", "remaining_time": "3:15:23"} +{"current_steps": 1060, "total_steps": 6572, "loss": 0.1447562575340271, "lr": 4.821497615501186e-06, "epoch": 0.6453576864535768, "percentage": 16.13, "elapsed_time": "0:37:21", "remaining_time": "3:14:16"} +{"current_steps": 1080, "total_steps": 6572, "loss": 0.15530819892883302, "lr": 4.812087622071104e-06, "epoch": 0.6575342465753424, "percentage": 16.43, "elapsed_time": "0:37:59", "remaining_time": "3:13:10"} +{"current_steps": 1100, "total_steps": 6572, "loss": 0.14426586627960206, "lr": 4.80244559642167e-06, "epoch": 0.669710806697108, "percentage": 16.74, "elapsed_time": "0:38:36", "remaining_time": "3:12:03"} +{"current_steps": 1120, "total_steps": 6572, "loss": 0.15436025857925414, "lr": 4.792572506189489e-06, "epoch": 0.6818873668188736, "percentage": 17.04, "elapsed_time": "0:39:13", "remaining_time": "3:10:54"} +{"current_steps": 1140, "total_steps": 6572, "loss": 0.14860854148864747, "lr": 4.782469342199915e-06, "epoch": 0.6940639269406392, "percentage": 17.35, "elapsed_time": "0:39:50", "remaining_time": "3:09:48"} +{"current_steps": 1160, "total_steps": 6572, "loss": 0.1313084125518799, "lr": 4.7721371183676205e-06, "epoch": 0.7062404870624048, "percentage": 17.65, "elapsed_time": "0:40:26", "remaining_time": "3:08:40"} +{"current_steps": 1180, "total_steps": 6572, "loss": 0.150812029838562, "lr": 4.761576871594841e-06, "epoch": 0.7184170471841704, "percentage": 17.95, "elapsed_time": "0:41:02", "remaining_time": "3:07:33"} +{"current_steps": 1200, "total_steps": 6572, "loss": 0.13278884887695314, "lr": 4.750789661667318e-06, "epoch": 0.730593607305936, "percentage": 18.26, "elapsed_time": "0:41:38", "remaining_time": "3:06:24"} +{"current_steps": 1220, "total_steps": 6572, "loss": 0.1612934350967407, "lr": 4.739776571147943e-06, "epoch": 0.7427701674277016, "percentage": 18.56, "elapsed_time": "0:42:14", "remaining_time": "3:05:16"} +{"current_steps": 1240, "total_steps": 6572, "loss": 0.16211290359497071, "lr": 4.728538705268116e-06, "epoch": 0.7549467275494672, "percentage": 18.87, "elapsed_time": "0:42:49", "remaining_time": "3:04:10"} +{"current_steps": 1260, "total_steps": 6572, "loss": 0.14386119842529296, "lr": 4.717077191816824e-06, "epoch": 0.7671232876712328, "percentage": 19.17, "elapsed_time": "0:43:26", "remaining_time": "3:03:09"} +{"current_steps": 1280, "total_steps": 6572, "loss": 0.12942540645599365, "lr": 4.705393181027463e-06, "epoch": 0.7792998477929984, "percentage": 19.48, "elapsed_time": "0:44:01", "remaining_time": "3:02:02"} +{"current_steps": 1300, "total_steps": 6572, "loss": 0.14771063327789308, "lr": 4.693487845462413e-06, "epoch": 0.791476407914764, "percentage": 19.78, "elapsed_time": "0:44:38", "remaining_time": "3:01:02"} +{"current_steps": 1320, "total_steps": 6572, "loss": 0.1276724100112915, "lr": 4.681362379895349e-06, "epoch": 0.8036529680365296, "percentage": 20.09, "elapsed_time": "0:45:14", "remaining_time": "3:00:00"} +{"current_steps": 1340, "total_steps": 6572, "loss": 0.1319241166114807, "lr": 4.6690180011913524e-06, "epoch": 0.8158295281582952, "percentage": 20.39, "elapsed_time": "0:45:50", "remaining_time": "2:58:59"} +{"current_steps": 1360, "total_steps": 6572, "loss": 0.1557891011238098, "lr": 4.6564559481847795e-06, "epoch": 0.8280060882800608, "percentage": 20.69, "elapsed_time": "0:46:27", "remaining_time": "2:58:01"} +{"current_steps": 1380, "total_steps": 6572, "loss": 0.11075855493545532, "lr": 4.643677481554947e-06, "epoch": 0.8401826484018264, "percentage": 21.0, "elapsed_time": "0:47:04", "remaining_time": "2:57:07"} +{"current_steps": 1400, "total_steps": 6572, "loss": 0.1580789566040039, "lr": 4.630683883699607e-06, "epoch": 0.852359208523592, "percentage": 21.3, "elapsed_time": "0:47:40", "remaining_time": "2:56:08"} +{"current_steps": 1420, "total_steps": 6572, "loss": 0.16006500720977784, "lr": 4.6174764586062556e-06, "epoch": 0.8645357686453576, "percentage": 21.61, "elapsed_time": "0:48:18", "remaining_time": "2:55:14"} +{"current_steps": 1440, "total_steps": 6572, "loss": 0.1462727189064026, "lr": 4.6040565317212685e-06, "epoch": 0.8767123287671232, "percentage": 21.91, "elapsed_time": "0:48:54", "remaining_time": "2:54:20"} +{"current_steps": 1460, "total_steps": 6572, "loss": 0.14725338220596312, "lr": 4.59042544981688e-06, "epoch": 0.8888888888888888, "percentage": 22.22, "elapsed_time": "0:49:30", "remaining_time": "2:53:22"} +{"current_steps": 1480, "total_steps": 6572, "loss": 0.1304166793823242, "lr": 4.5765845808560334e-06, "epoch": 0.9010654490106544, "percentage": 22.52, "elapsed_time": "0:50:07", "remaining_time": "2:52:27"} +{"current_steps": 1500, "total_steps": 6572, "loss": 0.1293134570121765, "lr": 4.562535313855094e-06, "epoch": 0.91324200913242, "percentage": 22.82, "elapsed_time": "0:50:45", "remaining_time": "2:51:38"} +{"current_steps": 1500, "total_steps": 6572, "eval_loss": 0.25538697838783264, "epoch": 0.91324200913242, "percentage": 22.82, "elapsed_time": "0:51:32", "remaining_time": "2:54:17"} +{"current_steps": 1520, "total_steps": 6572, "loss": 0.11359381675720215, "lr": 4.548279058744451e-06, "epoch": 0.9254185692541856, "percentage": 23.13, "elapsed_time": "0:53:39", "remaining_time": "2:58:20"} +{"current_steps": 1540, "total_steps": 6572, "loss": 0.15145074129104613, "lr": 4.533817246227024e-06, "epoch": 0.9375951293759512, "percentage": 23.43, "elapsed_time": "0:54:15", "remaining_time": "2:57:18"} +{"current_steps": 1560, "total_steps": 6572, "loss": 0.11953675746917725, "lr": 4.519151327634685e-06, "epoch": 0.9497716894977168, "percentage": 23.74, "elapsed_time": "0:54:52", "remaining_time": "2:56:18"} +{"current_steps": 1580, "total_steps": 6572, "loss": 0.13977375030517578, "lr": 4.504282774782605e-06, "epoch": 0.9619482496194824, "percentage": 24.04, "elapsed_time": "0:55:30", "remaining_time": "2:55:21"} +{"current_steps": 1600, "total_steps": 6572, "loss": 0.1338045358657837, "lr": 4.489213079821551e-06, "epoch": 0.974124809741248, "percentage": 24.35, "elapsed_time": "0:56:07", "remaining_time": "2:54:25"} +{"current_steps": 1620, "total_steps": 6572, "loss": 0.11776142120361328, "lr": 4.4739437550881355e-06, "epoch": 0.9863013698630136, "percentage": 24.65, "elapsed_time": "0:56:43", "remaining_time": "2:53:24"} +{"current_steps": 1640, "total_steps": 6572, "loss": 0.12504475116729735, "lr": 4.458476332953051e-06, "epoch": 0.9984779299847792, "percentage": 24.95, "elapsed_time": "0:57:19", "remaining_time": "2:52:23"} +{"current_steps": 1660, "total_steps": 6572, "loss": 0.08379222154617309, "lr": 4.442812365667281e-06, "epoch": 1.0103500761035007, "percentage": 25.26, "elapsed_time": "0:57:55", "remaining_time": "2:51:23"} +{"current_steps": 1680, "total_steps": 6572, "loss": 0.08407147526741028, "lr": 4.426953425206322e-06, "epoch": 1.0225266362252663, "percentage": 25.56, "elapsed_time": "0:58:32", "remaining_time": "2:50:27"} +{"current_steps": 1700, "total_steps": 6572, "loss": 0.08041079640388489, "lr": 4.410901103112434e-06, "epoch": 1.034703196347032, "percentage": 25.87, "elapsed_time": "0:59:08", "remaining_time": "2:49:28"} +{"current_steps": 1720, "total_steps": 6572, "loss": 0.07876392006874085, "lr": 4.394657010334908e-06, "epoch": 1.0468797564687975, "percentage": 26.17, "elapsed_time": "0:59:43", "remaining_time": "2:48:29"} +{"current_steps": 1740, "total_steps": 6572, "loss": 0.10302903652191162, "lr": 4.378222777068406e-06, "epoch": 1.059056316590563, "percentage": 26.48, "elapsed_time": "1:00:19", "remaining_time": "2:47:32"} +{"current_steps": 1760, "total_steps": 6572, "loss": 0.08733606934547425, "lr": 4.361600052589358e-06, "epoch": 1.0712328767123287, "percentage": 26.78, "elapsed_time": "1:00:55", "remaining_time": "2:46:34"} +{"current_steps": 1780, "total_steps": 6572, "loss": 0.08532609939575195, "lr": 4.344790505090447e-06, "epoch": 1.0834094368340943, "percentage": 27.08, "elapsed_time": "1:01:31", "remaining_time": "2:45:38"} +{"current_steps": 1800, "total_steps": 6572, "loss": 0.08734336495399475, "lr": 4.327795821513195e-06, "epoch": 1.09558599695586, "percentage": 27.39, "elapsed_time": "1:02:09", "remaining_time": "2:44:46"} +{"current_steps": 1820, "total_steps": 6572, "loss": 0.0913870632648468, "lr": 4.3106177073786684e-06, "epoch": 1.1077625570776255, "percentage": 27.69, "elapsed_time": "1:02:46", "remaining_time": "2:43:53"} +{"current_steps": 1840, "total_steps": 6572, "loss": 0.08115516304969787, "lr": 4.293257886616318e-06, "epoch": 1.119939117199391, "percentage": 28.0, "elapsed_time": "1:03:22", "remaining_time": "2:42:59"} +{"current_steps": 1860, "total_steps": 6572, "loss": 0.08891176581382751, "lr": 4.275718101390975e-06, "epoch": 1.1321156773211567, "percentage": 28.3, "elapsed_time": "1:03:59", "remaining_time": "2:42:07"} +{"current_steps": 1880, "total_steps": 6572, "loss": 0.07950961589813232, "lr": 4.25800011192801e-06, "epoch": 1.1442922374429223, "percentage": 28.61, "elapsed_time": "1:04:37", "remaining_time": "2:41:16"} +{"current_steps": 1900, "total_steps": 6572, "loss": 0.08310645222663879, "lr": 4.240105696336687e-06, "epoch": 1.156468797564688, "percentage": 28.91, "elapsed_time": "1:05:13", "remaining_time": "2:40:22"} +{"current_steps": 1920, "total_steps": 6572, "loss": 0.07682961225509644, "lr": 4.222036650431715e-06, "epoch": 1.1686453576864535, "percentage": 29.21, "elapsed_time": "1:05:49", "remaining_time": "2:39:29"} +{"current_steps": 1940, "total_steps": 6572, "loss": 0.07520227432250977, "lr": 4.203794787553032e-06, "epoch": 1.180821917808219, "percentage": 29.52, "elapsed_time": "1:06:26", "remaining_time": "2:38:37"} +{"current_steps": 1960, "total_steps": 6572, "loss": 0.0754019558429718, "lr": 4.185381938383821e-06, "epoch": 1.1929984779299847, "percentage": 29.82, "elapsed_time": "1:07:02", "remaining_time": "2:37:44"} +{"current_steps": 1980, "total_steps": 6572, "loss": 0.08085885643959045, "lr": 4.166799950766793e-06, "epoch": 1.2051750380517503, "percentage": 30.13, "elapsed_time": "1:07:38", "remaining_time": "2:36:51"} +{"current_steps": 2000, "total_steps": 6572, "loss": 0.0884653627872467, "lr": 4.14805068951874e-06, "epoch": 1.217351598173516, "percentage": 30.43, "elapsed_time": "1:08:14", "remaining_time": "2:36:00"} +{"current_steps": 2000, "total_steps": 6572, "eval_loss": 0.2661186456680298, "epoch": 1.217351598173516, "percentage": 30.43, "elapsed_time": "1:09:01", "remaining_time": "2:37:47"} +{"current_steps": 2020, "total_steps": 6572, "loss": 0.06684748530387878, "lr": 4.1291360362433965e-06, "epoch": 1.2295281582952815, "percentage": 30.74, "elapsed_time": "1:11:15", "remaining_time": "2:40:34"} +{"current_steps": 2040, "total_steps": 6572, "loss": 0.0720324158668518, "lr": 4.110057889142601e-06, "epoch": 1.241704718417047, "percentage": 31.04, "elapsed_time": "1:11:52", "remaining_time": "2:39:40"} +{"current_steps": 2060, "total_steps": 6572, "loss": 0.08799988031387329, "lr": 4.090818162825804e-06, "epoch": 1.2538812785388127, "percentage": 31.35, "elapsed_time": "1:12:29", "remaining_time": "2:38:47"} +{"current_steps": 2080, "total_steps": 6572, "loss": 0.09275985956192016, "lr": 4.071418788117926e-06, "epoch": 1.2660578386605783, "percentage": 31.65, "elapsed_time": "1:13:07", "remaining_time": "2:37:55"} +{"current_steps": 2100, "total_steps": 6572, "loss": 0.08431113958358764, "lr": 4.0518617118655845e-06, "epoch": 1.278234398782344, "percentage": 31.95, "elapsed_time": "1:13:43", "remaining_time": "2:37:00"} +{"current_steps": 2120, "total_steps": 6572, "loss": 0.09995608925819396, "lr": 4.032148896741717e-06, "epoch": 1.2904109589041095, "percentage": 32.26, "elapsed_time": "1:14:20", "remaining_time": "2:36:07"} +{"current_steps": 2140, "total_steps": 6572, "loss": 0.07387629747390748, "lr": 4.012282321048618e-06, "epoch": 1.302587519025875, "percentage": 32.56, "elapsed_time": "1:14:57", "remaining_time": "2:35:14"} +{"current_steps": 2160, "total_steps": 6572, "loss": 0.07667248249053955, "lr": 3.992263978519398e-06, "epoch": 1.3147640791476407, "percentage": 32.87, "elapsed_time": "1:15:34", "remaining_time": "2:34:22"} +{"current_steps": 2180, "total_steps": 6572, "loss": 0.09203824400901794, "lr": 3.972095878117904e-06, "epoch": 1.3269406392694063, "percentage": 33.17, "elapsed_time": "1:16:11", "remaining_time": "2:33:30"} +{"current_steps": 2200, "total_steps": 6572, "loss": 0.07835246920585633, "lr": 3.951780043837107e-06, "epoch": 1.339117199391172, "percentage": 33.48, "elapsed_time": "1:16:48", "remaining_time": "2:32:37"} +{"current_steps": 2220, "total_steps": 6572, "loss": 0.08577624559402466, "lr": 3.9313185144959835e-06, "epoch": 1.3512937595129375, "percentage": 33.78, "elapsed_time": "1:17:25", "remaining_time": "2:31:46"} +{"current_steps": 2240, "total_steps": 6572, "loss": 0.0789969801902771, "lr": 3.9107133435349025e-06, "epoch": 1.363470319634703, "percentage": 34.08, "elapsed_time": "1:18:01", "remaining_time": "2:30:53"} +{"current_steps": 2260, "total_steps": 6572, "loss": 0.07151145935058593, "lr": 3.889966598809557e-06, "epoch": 1.3756468797564687, "percentage": 34.39, "elapsed_time": "1:18:38", "remaining_time": "2:30:02"} +{"current_steps": 2280, "total_steps": 6572, "loss": 0.09204544425010681, "lr": 3.869080362383437e-06, "epoch": 1.3878234398782343, "percentage": 34.69, "elapsed_time": "1:19:14", "remaining_time": "2:29:10"} +{"current_steps": 2300, "total_steps": 6572, "loss": 0.11079612970352173, "lr": 3.848056730318881e-06, "epoch": 1.4, "percentage": 35.0, "elapsed_time": "1:19:54", "remaining_time": "2:28:25"} +{"current_steps": 2320, "total_steps": 6572, "loss": 0.06770140528678895, "lr": 3.826897812466728e-06, "epoch": 1.4121765601217655, "percentage": 35.3, "elapsed_time": "1:20:30", "remaining_time": "2:27:32"} +{"current_steps": 2340, "total_steps": 6572, "loss": 0.08210510611534119, "lr": 3.8056057322545763e-06, "epoch": 1.4243531202435311, "percentage": 35.61, "elapsed_time": "1:21:07", "remaining_time": "2:26:42"} +{"current_steps": 2360, "total_steps": 6572, "loss": 0.09583572745323181, "lr": 3.7841826264736888e-06, "epoch": 1.4365296803652967, "percentage": 35.91, "elapsed_time": "1:21:43", "remaining_time": "2:25:51"} +{"current_steps": 2380, "total_steps": 6572, "loss": 0.09235450625419617, "lr": 3.762630645064547e-06, "epoch": 1.4487062404870623, "percentage": 36.21, "elapsed_time": "1:22:20", "remaining_time": "2:25:01"} +{"current_steps": 2400, "total_steps": 6572, "loss": 0.08658097982406616, "lr": 3.7409519509010985e-06, "epoch": 1.460882800608828, "percentage": 36.52, "elapsed_time": "1:22:56", "remaining_time": "2:24:11"} +{"current_steps": 2420, "total_steps": 6572, "loss": 0.08892765045166015, "lr": 3.7191487195736915e-06, "epoch": 1.4730593607305935, "percentage": 36.82, "elapsed_time": "1:23:32", "remaining_time": "2:23:20"} +{"current_steps": 2440, "total_steps": 6572, "loss": 0.07849371433258057, "lr": 3.697223139170748e-06, "epoch": 1.4852359208523591, "percentage": 37.13, "elapsed_time": "1:24:09", "remaining_time": "2:22:30"} +{"current_steps": 2460, "total_steps": 6572, "loss": 0.07469035387039184, "lr": 3.6751774100591716e-06, "epoch": 1.4974124809741247, "percentage": 37.43, "elapsed_time": "1:24:45", "remaining_time": "2:21:40"} +{"current_steps": 2480, "total_steps": 6572, "loss": 0.0782626211643219, "lr": 3.6530137446635265e-06, "epoch": 1.5095890410958903, "percentage": 37.74, "elapsed_time": "1:25:20", "remaining_time": "2:20:49"} +{"current_steps": 2500, "total_steps": 6572, "loss": 0.08304058909416198, "lr": 3.630734367244012e-06, "epoch": 1.521765601217656, "percentage": 38.04, "elapsed_time": "1:25:57", "remaining_time": "2:20:00"} +{"current_steps": 2500, "total_steps": 6572, "eval_loss": 0.2771773338317871, "epoch": 1.521765601217656, "percentage": 38.04, "elapsed_time": "1:26:44", "remaining_time": "2:21:17"} +{"current_steps": 2520, "total_steps": 6572, "loss": 0.08037537932395936, "lr": 3.6083415136732374e-06, "epoch": 1.5339421613394215, "percentage": 38.34, "elapsed_time": "1:28:46", "remaining_time": "2:22:44"} +{"current_steps": 2540, "total_steps": 6572, "loss": 0.08990358114242554, "lr": 3.585837431211845e-06, "epoch": 1.5461187214611871, "percentage": 38.65, "elapsed_time": "1:29:23", "remaining_time": "2:21:53"} +{"current_steps": 2560, "total_steps": 6572, "loss": 0.0773526132106781, "lr": 3.563224378282978e-06, "epoch": 1.5582952815829527, "percentage": 38.95, "elapsed_time": "1:29:59", "remaining_time": "2:21:02"} +{"current_steps": 2580, "total_steps": 6572, "loss": 0.0777865469455719, "lr": 3.5405046242456396e-06, "epoch": 1.5704718417047183, "percentage": 39.26, "elapsed_time": "1:30:36", "remaining_time": "2:20:11"} +{"current_steps": 2600, "total_steps": 6572, "loss": 0.08037815093994141, "lr": 3.517680449166943e-06, "epoch": 1.582648401826484, "percentage": 39.56, "elapsed_time": "1:31:11", "remaining_time": "2:19:18"} +{"current_steps": 2620, "total_steps": 6572, "loss": 0.08837634325027466, "lr": 3.4947541435932976e-06, "epoch": 1.5948249619482495, "percentage": 39.87, "elapsed_time": "1:31:49", "remaining_time": "2:18:30"} +{"current_steps": 2640, "total_steps": 6572, "loss": 0.08201563358306885, "lr": 3.471728008320532e-06, "epoch": 1.6070015220700151, "percentage": 40.17, "elapsed_time": "1:32:26", "remaining_time": "2:17:41"} +{"current_steps": 2660, "total_steps": 6572, "loss": 0.07379403114318847, "lr": 3.4486043541630066e-06, "epoch": 1.6191780821917807, "percentage": 40.47, "elapsed_time": "1:33:03", "remaining_time": "2:16:51"} +{"current_steps": 2680, "total_steps": 6572, "loss": 0.08871785402297974, "lr": 3.425385501721696e-06, "epoch": 1.6313546423135463, "percentage": 40.78, "elapsed_time": "1:33:39", "remaining_time": "2:16:00"} +{"current_steps": 2700, "total_steps": 6572, "loss": 0.07757498621940613, "lr": 3.4020737811513107e-06, "epoch": 1.643531202435312, "percentage": 41.08, "elapsed_time": "1:34:16", "remaining_time": "2:15:11"} +{"current_steps": 2720, "total_steps": 6572, "loss": 0.08565697669982911, "lr": 3.3786715319264483e-06, "epoch": 1.6557077625570775, "percentage": 41.39, "elapsed_time": "1:34:52", "remaining_time": "2:14:21"} +{"current_steps": 2740, "total_steps": 6572, "loss": 0.08129348754882812, "lr": 3.355181102606816e-06, "epoch": 1.6678843226788431, "percentage": 41.69, "elapsed_time": "1:35:28", "remaining_time": "2:13:31"} +{"current_steps": 2760, "total_steps": 6572, "loss": 0.07875375747680664, "lr": 3.331604850601533e-06, "epoch": 1.6800608828006087, "percentage": 42.0, "elapsed_time": "1:36:06", "remaining_time": "2:12:43"} +{"current_steps": 2780, "total_steps": 6572, "loss": 0.08989614248275757, "lr": 3.307945141932556e-06, "epoch": 1.6922374429223743, "percentage": 42.3, "elapsed_time": "1:36:41", "remaining_time": "2:11:53"} +{"current_steps": 2800, "total_steps": 6572, "loss": 0.08251298069953919, "lr": 3.2842043509972294e-06, "epoch": 1.70441400304414, "percentage": 42.6, "elapsed_time": "1:37:18", "remaining_time": "2:11:05"} +{"current_steps": 2820, "total_steps": 6572, "loss": 0.07760271430015564, "lr": 3.2603848603300026e-06, "epoch": 1.7165905631659055, "percentage": 42.91, "elapsed_time": "1:37:56", "remaining_time": "2:10:18"} +{"current_steps": 2840, "total_steps": 6572, "loss": 0.07395396232604981, "lr": 3.236489060363329e-06, "epoch": 1.7287671232876711, "percentage": 43.21, "elapsed_time": "1:38:33", "remaining_time": "2:09:31"} +{"current_steps": 2860, "total_steps": 6572, "loss": 0.07028600573539734, "lr": 3.212519349187766e-06, "epoch": 1.7409436834094367, "percentage": 43.52, "elapsed_time": "1:39:10", "remaining_time": "2:08:42"} +{"current_steps": 2880, "total_steps": 6572, "loss": 0.08469281196594239, "lr": 3.188478132311319e-06, "epoch": 1.7531202435312023, "percentage": 43.82, "elapsed_time": "1:39:46", "remaining_time": "2:07:54"} +{"current_steps": 2900, "total_steps": 6572, "loss": 0.09567424058914184, "lr": 3.164367822418029e-06, "epoch": 1.765296803652968, "percentage": 44.13, "elapsed_time": "1:40:22", "remaining_time": "2:07:05"} +{"current_steps": 2920, "total_steps": 6572, "loss": 0.07239987254142762, "lr": 3.1401908391258474e-06, "epoch": 1.7774733637747335, "percentage": 44.43, "elapsed_time": "1:41:00", "remaining_time": "2:06:19"} +{"current_steps": 2940, "total_steps": 6572, "loss": 0.0891954243183136, "lr": 3.1159496087438098e-06, "epoch": 1.7896499238964991, "percentage": 44.74, "elapsed_time": "1:41:36", "remaining_time": "2:05:31"} +{"current_steps": 2960, "total_steps": 6572, "loss": 0.07796849608421326, "lr": 3.0916465640285426e-06, "epoch": 1.8018264840182647, "percentage": 45.04, "elapsed_time": "1:42:12", "remaining_time": "2:04:43"} +{"current_steps": 2980, "total_steps": 6572, "loss": 0.08645985722541809, "lr": 3.0672841439401223e-06, "epoch": 1.8140030441400303, "percentage": 45.34, "elapsed_time": "1:42:49", "remaining_time": "2:03:55"} +{"current_steps": 3000, "total_steps": 6572, "loss": 0.08427774310111999, "lr": 3.0428647933973103e-06, "epoch": 1.826179604261796, "percentage": 45.65, "elapsed_time": "1:43:26", "remaining_time": "2:03:10"} +{"current_steps": 3000, "total_steps": 6572, "eval_loss": 0.2774485647678375, "epoch": 1.826179604261796, "percentage": 45.65, "elapsed_time": "1:44:13", "remaining_time": "2:04:05"} +{"current_steps": 3020, "total_steps": 6572, "loss": 0.07381275296211243, "lr": 3.0183909630321865e-06, "epoch": 1.8383561643835615, "percentage": 45.95, "elapsed_time": "1:46:17", "remaining_time": "2:05:00"} +{"current_steps": 3040, "total_steps": 6572, "loss": 0.07289664745330811, "lr": 2.9938651089442184e-06, "epoch": 1.8505327245053271, "percentage": 46.26, "elapsed_time": "1:46:55", "remaining_time": "2:04:14"} +{"current_steps": 3060, "total_steps": 6572, "loss": 0.07124295830726624, "lr": 2.969289692453773e-06, "epoch": 1.8627092846270927, "percentage": 46.56, "elapsed_time": "1:47:32", "remaining_time": "2:03:25"} +{"current_steps": 3080, "total_steps": 6572, "loss": 0.08125877976417542, "lr": 2.944667179855109e-06, "epoch": 1.8748858447488583, "percentage": 46.87, "elapsed_time": "1:48:09", "remaining_time": "2:02:37"} +{"current_steps": 3100, "total_steps": 6572, "loss": 0.0724608838558197, "lr": 2.920000042168871e-06, "epoch": 1.887062404870624, "percentage": 47.17, "elapsed_time": "1:48:47", "remaining_time": "2:01:51"} +{"current_steps": 3120, "total_steps": 6572, "loss": 0.07104775309562683, "lr": 2.8952907548941057e-06, "epoch": 1.8992389649923895, "percentage": 47.47, "elapsed_time": "1:49:23", "remaining_time": "2:01:01"} +{"current_steps": 3140, "total_steps": 6572, "loss": 0.0677955150604248, "lr": 2.8705417977598277e-06, "epoch": 1.9114155251141551, "percentage": 47.78, "elapsed_time": "1:50:00", "remaining_time": "2:00:14"} +{"current_steps": 3160, "total_steps": 6572, "loss": 0.07164496779441834, "lr": 2.8457556544761687e-06, "epoch": 1.9235920852359207, "percentage": 48.08, "elapsed_time": "1:50:36", "remaining_time": "1:59:25"} +{"current_steps": 3180, "total_steps": 6572, "loss": 0.071807599067688, "lr": 2.8209348124851187e-06, "epoch": 1.9357686453576863, "percentage": 48.39, "elapsed_time": "1:51:13", "remaining_time": "1:58:38"} +{"current_steps": 3200, "total_steps": 6572, "loss": 0.095755535364151, "lr": 2.7960817627108965e-06, "epoch": 1.947945205479452, "percentage": 48.69, "elapsed_time": "1:51:50", "remaining_time": "1:57:50"} +{"current_steps": 3220, "total_steps": 6572, "loss": 0.07055851817131042, "lr": 2.77119899930997e-06, "epoch": 1.9601217656012175, "percentage": 49.0, "elapsed_time": "1:52:26", "remaining_time": "1:57:03"} +{"current_steps": 3240, "total_steps": 6572, "loss": 0.07278798818588257, "lr": 2.7462890194207513e-06, "epoch": 1.9722983257229831, "percentage": 49.3, "elapsed_time": "1:53:03", "remaining_time": "1:56:16"} +{"current_steps": 3260, "total_steps": 6572, "loss": 0.07153088450431824, "lr": 2.7213543229129956e-06, "epoch": 1.9844748858447487, "percentage": 49.6, "elapsed_time": "1:53:40", "remaining_time": "1:55:28"} +{"current_steps": 3280, "total_steps": 6572, "loss": 0.07440360188484192, "lr": 2.6963974121369242e-06, "epoch": 1.9966514459665143, "percentage": 49.91, "elapsed_time": "1:54:16", "remaining_time": "1:54:41"} +{"current_steps": 3300, "total_steps": 6572, "loss": 0.0517767608165741, "lr": 2.671420791672093e-06, "epoch": 2.008523592085236, "percentage": 50.21, "elapsed_time": "1:54:51", "remaining_time": "1:53:53"} +{"current_steps": 3320, "total_steps": 6572, "loss": 0.03812239170074463, "lr": 2.646426968076052e-06, "epoch": 2.0207001522070014, "percentage": 50.52, "elapsed_time": "1:55:27", "remaining_time": "1:53:06"} +{"current_steps": 3340, "total_steps": 6572, "loss": 0.04107579588890076, "lr": 2.6214184496327865e-06, "epoch": 2.032876712328767, "percentage": 50.82, "elapsed_time": "1:56:05", "remaining_time": "1:52:19"} +{"current_steps": 3360, "total_steps": 6572, "loss": 0.04673115909099579, "lr": 2.5963977461010022e-06, "epoch": 2.0450532724505326, "percentage": 51.13, "elapsed_time": "1:56:41", "remaining_time": "1:51:33"} +{"current_steps": 3380, "total_steps": 6572, "loss": 0.03674449622631073, "lr": 2.5713673684622524e-06, "epoch": 2.057229832572298, "percentage": 51.43, "elapsed_time": "1:57:17", "remaining_time": "1:50:46"} +{"current_steps": 3400, "total_steps": 6572, "loss": 0.03422380387783051, "lr": 2.546329828668949e-06, "epoch": 2.069406392694064, "percentage": 51.73, "elapsed_time": "1:57:53", "remaining_time": "1:49:59"} +{"current_steps": 3420, "total_steps": 6572, "loss": 0.035878732800483704, "lr": 2.5212876393922657e-06, "epoch": 2.0815829528158294, "percentage": 52.04, "elapsed_time": "1:58:29", "remaining_time": "1:49:12"} +{"current_steps": 3440, "total_steps": 6572, "loss": 0.03577531576156616, "lr": 2.496243313769986e-06, "epoch": 2.093759512937595, "percentage": 52.34, "elapsed_time": "1:59:06", "remaining_time": "1:48:27"} +{"current_steps": 3460, "total_steps": 6572, "loss": 0.04281675517559051, "lr": 2.471199365154283e-06, "epoch": 2.1059360730593606, "percentage": 52.65, "elapsed_time": "1:59:42", "remaining_time": "1:47:40"} +{"current_steps": 3480, "total_steps": 6572, "loss": 0.042955422401428224, "lr": 2.4461583068595014e-06, "epoch": 2.118112633181126, "percentage": 52.95, "elapsed_time": "2:00:20", "remaining_time": "1:46:55"} +{"current_steps": 3500, "total_steps": 6572, "loss": 0.04432957172393799, "lr": 2.421122651909918e-06, "epoch": 2.130289193302892, "percentage": 53.26, "elapsed_time": "2:00:56", "remaining_time": "1:46:09"} +{"current_steps": 3500, "total_steps": 6572, "eval_loss": 0.3123805522918701, "epoch": 2.130289193302892, "percentage": 53.26, "elapsed_time": "2:01:42", "remaining_time": "1:46:49"} +{"current_steps": 3520, "total_steps": 6572, "loss": 0.03356837034225464, "lr": 2.3960949127875556e-06, "epoch": 2.1424657534246574, "percentage": 53.56, "elapsed_time": "2:03:47", "remaining_time": "1:47:19"} +{"current_steps": 3540, "total_steps": 6572, "loss": 0.036935809254646304, "lr": 2.371077601180031e-06, "epoch": 2.154642313546423, "percentage": 53.86, "elapsed_time": "2:04:23", "remaining_time": "1:46:32"} +{"current_steps": 3560, "total_steps": 6572, "loss": 0.0395690768957138, "lr": 2.3460732277284994e-06, "epoch": 2.1668188736681886, "percentage": 54.17, "elapsed_time": "2:05:01", "remaining_time": "1:45:46"} +{"current_steps": 3580, "total_steps": 6572, "loss": 0.044693085551261905, "lr": 2.321084301775689e-06, "epoch": 2.178995433789954, "percentage": 54.47, "elapsed_time": "2:05:39", "remaining_time": "1:45:01"} +{"current_steps": 3600, "total_steps": 6572, "loss": 0.03243565857410431, "lr": 2.29611333111408e-06, "epoch": 2.19117199391172, "percentage": 54.78, "elapsed_time": "2:06:17", "remaining_time": "1:44:15"} +{"current_steps": 3620, "total_steps": 6572, "loss": 0.04325798749923706, "lr": 2.271162821734225e-06, "epoch": 2.2033485540334854, "percentage": 55.08, "elapsed_time": "2:06:56", "remaining_time": "1:43:30"} +{"current_steps": 3640, "total_steps": 6572, "loss": 0.03856868743896484, "lr": 2.2462352775732653e-06, "epoch": 2.215525114155251, "percentage": 55.39, "elapsed_time": "2:07:33", "remaining_time": "1:42:45"} +{"current_steps": 3660, "total_steps": 6572, "loss": 0.041602414846420285, "lr": 2.221333200263637e-06, "epoch": 2.2277016742770166, "percentage": 55.69, "elapsed_time": "2:08:11", "remaining_time": "1:41:59"} +{"current_steps": 3680, "total_steps": 6572, "loss": 0.04286134541034699, "lr": 2.1964590888820233e-06, "epoch": 2.239878234398782, "percentage": 56.0, "elapsed_time": "2:08:48", "remaining_time": "1:41:13"} +{"current_steps": 3700, "total_steps": 6572, "loss": 0.041756758093833925, "lr": 2.1716154396985526e-06, "epoch": 2.252054794520548, "percentage": 56.3, "elapsed_time": "2:09:25", "remaining_time": "1:40:28"} +{"current_steps": 3720, "total_steps": 6572, "loss": 0.0359495222568512, "lr": 2.1468047459262882e-06, "epoch": 2.2642313546423134, "percentage": 56.6, "elapsed_time": "2:10:03", "remaining_time": "1:39:42"} +{"current_steps": 3740, "total_steps": 6572, "loss": 0.04322676360607147, "lr": 2.12202949747101e-06, "epoch": 2.276407914764079, "percentage": 56.91, "elapsed_time": "2:10:41", "remaining_time": "1:38:57"} +{"current_steps": 3760, "total_steps": 6572, "loss": 0.04191597998142242, "lr": 2.0972921806813468e-06, "epoch": 2.2885844748858446, "percentage": 57.21, "elapsed_time": "2:11:19", "remaining_time": "1:38:13"} +{"current_steps": 3780, "total_steps": 6572, "loss": 0.041278204321861266, "lr": 2.072595278099247e-06, "epoch": 2.30076103500761, "percentage": 57.52, "elapsed_time": "2:11:58", "remaining_time": "1:37:28"} +{"current_steps": 3800, "total_steps": 6572, "loss": 0.04312986135482788, "lr": 2.047941268210849e-06, "epoch": 2.312937595129376, "percentage": 57.82, "elapsed_time": "2:12:35", "remaining_time": "1:36:43"} +{"current_steps": 3820, "total_steps": 6572, "loss": 0.04236046075820923, "lr": 2.0233326251977426e-06, "epoch": 2.3251141552511414, "percentage": 58.13, "elapsed_time": "2:13:14", "remaining_time": "1:35:59"} +{"current_steps": 3840, "total_steps": 6572, "loss": 0.04433901011943817, "lr": 1.9987718186886724e-06, "epoch": 2.337290715372907, "percentage": 58.43, "elapsed_time": "2:13:52", "remaining_time": "1:35:14"} +{"current_steps": 3860, "total_steps": 6572, "loss": 0.04127628207206726, "lr": 1.9742613135116986e-06, "epoch": 2.3494672754946726, "percentage": 58.73, "elapsed_time": "2:14:29", "remaining_time": "1:34:29"} +{"current_steps": 3880, "total_steps": 6572, "loss": 0.04586326479911804, "lr": 1.949803569446828e-06, "epoch": 2.361643835616438, "percentage": 59.04, "elapsed_time": "2:15:07", "remaining_time": "1:33:45"} +{"current_steps": 3900, "total_steps": 6572, "loss": 0.03624185025691986, "lr": 1.925401040979171e-06, "epoch": 2.373820395738204, "percentage": 59.34, "elapsed_time": "2:15:44", "remaining_time": "1:33:00"} +{"current_steps": 3920, "total_steps": 6572, "loss": 0.035064518451690674, "lr": 1.9010561770526076e-06, "epoch": 2.3859969558599694, "percentage": 59.65, "elapsed_time": "2:16:21", "remaining_time": "1:32:15"} +{"current_steps": 3940, "total_steps": 6572, "loss": 0.042050021886825564, "lr": 1.8767714208240312e-06, "epoch": 2.398173515981735, "percentage": 59.95, "elapsed_time": "2:16:59", "remaining_time": "1:31:30"} +{"current_steps": 3960, "total_steps": 6572, "loss": 0.038166466355323794, "lr": 1.852549209418154e-06, "epoch": 2.4103500761035006, "percentage": 60.26, "elapsed_time": "2:17:37", "remaining_time": "1:30:46"} +{"current_steps": 3980, "total_steps": 6572, "loss": 0.040885674953460696, "lr": 1.8283919736829332e-06, "epoch": 2.422526636225266, "percentage": 60.56, "elapsed_time": "2:18:15", "remaining_time": "1:30:02"} +{"current_steps": 4000, "total_steps": 6572, "loss": 0.040162667632102966, "lr": 1.804302137945614e-06, "epoch": 2.434703196347032, "percentage": 60.86, "elapsed_time": "2:18:51", "remaining_time": "1:29:17"} +{"current_steps": 4000, "total_steps": 6572, "eval_loss": 0.31583163142204285, "epoch": 2.434703196347032, "percentage": 60.86, "elapsed_time": "2:19:38", "remaining_time": "1:29:47"} +{"current_steps": 4020, "total_steps": 6572, "loss": 0.04170995056629181, "lr": 1.7802821197694426e-06, "epoch": 2.4468797564687974, "percentage": 61.17, "elapsed_time": "2:21:49", "remaining_time": "1:30:02"} +{"current_steps": 4040, "total_steps": 6572, "loss": 0.03834344446659088, "lr": 1.7563343297110375e-06, "epoch": 2.459056316590563, "percentage": 61.47, "elapsed_time": "2:22:25", "remaining_time": "1:29:15"} +{"current_steps": 4060, "total_steps": 6572, "loss": 0.03928310573101044, "lr": 1.732461171078486e-06, "epoch": 2.4712328767123286, "percentage": 61.78, "elapsed_time": "2:23:01", "remaining_time": "1:28:29"} +{"current_steps": 4080, "total_steps": 6572, "loss": 0.03358933925628662, "lr": 1.7086650396901489e-06, "epoch": 2.483409436834094, "percentage": 62.08, "elapsed_time": "2:23:37", "remaining_time": "1:27:43"} +{"current_steps": 4100, "total_steps": 6572, "loss": 0.03547535240650177, "lr": 1.6849483236342322e-06, "epoch": 2.49558599695586, "percentage": 62.39, "elapsed_time": "2:24:14", "remaining_time": "1:26:58"} +{"current_steps": 4120, "total_steps": 6572, "loss": 0.03600102663040161, "lr": 1.6613134030291217e-06, "epoch": 2.5077625570776254, "percentage": 62.69, "elapsed_time": "2:24:51", "remaining_time": "1:26:12"} +{"current_steps": 4140, "total_steps": 6572, "loss": 0.04347077012062073, "lr": 1.6377626497845278e-06, "epoch": 2.519939117199391, "percentage": 62.99, "elapsed_time": "2:25:27", "remaining_time": "1:25:27"} +{"current_steps": 4160, "total_steps": 6572, "loss": 0.02908192276954651, "lr": 1.6142984273634505e-06, "epoch": 2.5321156773211566, "percentage": 63.3, "elapsed_time": "2:26:04", "remaining_time": "1:24:41"} +{"current_steps": 4180, "total_steps": 6572, "loss": 0.03611198365688324, "lr": 1.5909230905449846e-06, "epoch": 2.544292237442922, "percentage": 63.6, "elapsed_time": "2:26:40", "remaining_time": "1:23:56"} +{"current_steps": 4200, "total_steps": 6572, "loss": 0.03758668601512909, "lr": 1.567638985188012e-06, "epoch": 2.556468797564688, "percentage": 63.91, "elapsed_time": "2:27:17", "remaining_time": "1:23:10"} +{"current_steps": 4220, "total_steps": 6572, "loss": 0.033633843064308167, "lr": 1.544448447995773e-06, "epoch": 2.5686453576864534, "percentage": 64.21, "elapsed_time": "2:27:53", "remaining_time": "1:22:25"} +{"current_steps": 4240, "total_steps": 6572, "loss": 0.036797890067100526, "lr": 1.52135380628137e-06, "epoch": 2.580821917808219, "percentage": 64.52, "elapsed_time": "2:28:29", "remaining_time": "1:21:40"} +{"current_steps": 4260, "total_steps": 6572, "loss": 0.039166563749313356, "lr": 1.498357377734201e-06, "epoch": 2.5929984779299846, "percentage": 64.82, "elapsed_time": "2:29:07", "remaining_time": "1:20:55"} +{"current_steps": 4280, "total_steps": 6572, "loss": 0.03717599511146545, "lr": 1.4754614701873703e-06, "epoch": 2.60517503805175, "percentage": 65.12, "elapsed_time": "2:29:44", "remaining_time": "1:20:11"} +{"current_steps": 4300, "total_steps": 6572, "loss": 0.03962793946266174, "lr": 1.4526683813860792e-06, "epoch": 2.6173515981735163, "percentage": 65.43, "elapsed_time": "2:30:20", "remaining_time": "1:19:26"} +{"current_steps": 4320, "total_steps": 6572, "loss": 0.035475924611091614, "lr": 1.4299803987570396e-06, "epoch": 2.6295281582952814, "percentage": 65.73, "elapsed_time": "2:30:57", "remaining_time": "1:18:41"} +{"current_steps": 4340, "total_steps": 6572, "loss": 0.03256964683532715, "lr": 1.4073997991789078e-06, "epoch": 2.6417047184170475, "percentage": 66.04, "elapsed_time": "2:31:35", "remaining_time": "1:17:57"} +{"current_steps": 4360, "total_steps": 6572, "loss": 0.03712306022644043, "lr": 1.384928848753792e-06, "epoch": 2.6538812785388126, "percentage": 66.34, "elapsed_time": "2:32:10", "remaining_time": "1:17:12"} +{"current_steps": 4380, "total_steps": 6572, "loss": 0.041410398483276364, "lr": 1.3625698025798322e-06, "epoch": 2.6660578386605787, "percentage": 66.65, "elapsed_time": "2:32:47", "remaining_time": "1:16:27"} +{"current_steps": 4400, "total_steps": 6572, "loss": 0.03158504366874695, "lr": 1.3403249045248907e-06, "epoch": 2.678234398782344, "percentage": 66.95, "elapsed_time": "2:33:23", "remaining_time": "1:15:43"} +{"current_steps": 4420, "total_steps": 6572, "loss": 0.03525224924087524, "lr": 1.3181963870013604e-06, "epoch": 2.69041095890411, "percentage": 67.26, "elapsed_time": "2:33:59", "remaining_time": "1:14:58"} +{"current_steps": 4440, "total_steps": 6572, "loss": 0.03239959478378296, "lr": 1.2961864707421345e-06, "epoch": 2.702587519025875, "percentage": 67.56, "elapsed_time": "2:34:36", "remaining_time": "1:14:14"} +{"current_steps": 4460, "total_steps": 6572, "loss": 0.031032082438468934, "lr": 1.2742973645777394e-06, "epoch": 2.714764079147641, "percentage": 67.86, "elapsed_time": "2:35:13", "remaining_time": "1:13:30"} +{"current_steps": 4480, "total_steps": 6572, "loss": 0.030566230416297913, "lr": 1.252531265214662e-06, "epoch": 2.726940639269406, "percentage": 68.17, "elapsed_time": "2:35:48", "remaining_time": "1:12:45"} +{"current_steps": 4500, "total_steps": 6572, "loss": 0.041362547874450685, "lr": 1.2308903570149048e-06, "epoch": 2.7391171993911723, "percentage": 68.47, "elapsed_time": "2:36:23", "remaining_time": "1:12:00"} +{"current_steps": 4500, "total_steps": 6572, "eval_loss": 0.3391737937927246, "epoch": 2.7391171993911723, "percentage": 68.47, "elapsed_time": "2:37:10", "remaining_time": "1:12:22"} +{"current_steps": 4520, "total_steps": 6572, "loss": 0.0388390064239502, "lr": 1.2093768117767613e-06, "epoch": 2.7512937595129374, "percentage": 68.78, "elapsed_time": "2:39:12", "remaining_time": "1:12:16"} +{"current_steps": 4540, "total_steps": 6572, "loss": 0.032555675506591795, "lr": 1.1879927885168733e-06, "epoch": 2.7634703196347035, "percentage": 69.08, "elapsed_time": "2:39:48", "remaining_time": "1:11:31"} +{"current_steps": 4560, "total_steps": 6572, "loss": 0.03606459796428681, "lr": 1.1667404332535504e-06, "epoch": 2.7756468797564686, "percentage": 69.39, "elapsed_time": "2:40:25", "remaining_time": "1:10:47"} +{"current_steps": 4580, "total_steps": 6572, "loss": 0.032086309790611264, "lr": 1.1456218787914128e-06, "epoch": 2.7878234398782347, "percentage": 69.69, "elapsed_time": "2:41:01", "remaining_time": "1:10:02"} +{"current_steps": 4600, "total_steps": 6572, "loss": 0.033362787961959836, "lr": 1.1246392445073438e-06, "epoch": 2.8, "percentage": 69.99, "elapsed_time": "2:41:36", "remaining_time": "1:09:16"} +{"current_steps": 4620, "total_steps": 6572, "loss": 0.03638745844364166, "lr": 1.1037946361378027e-06, "epoch": 2.812176560121766, "percentage": 70.3, "elapsed_time": "2:42:12", "remaining_time": "1:08:32"} +{"current_steps": 4640, "total_steps": 6572, "loss": 0.030933958292007447, "lr": 1.0830901455674977e-06, "epoch": 2.824353120243531, "percentage": 70.6, "elapsed_time": "2:42:49", "remaining_time": "1:07:47"} +{"current_steps": 4660, "total_steps": 6572, "loss": 0.02879139482975006, "lr": 1.0625278506194538e-06, "epoch": 2.836529680365297, "percentage": 70.91, "elapsed_time": "2:43:26", "remaining_time": "1:07:03"} +{"current_steps": 4680, "total_steps": 6572, "loss": 0.03345020413398743, "lr": 1.04210981484649e-06, "epoch": 2.8487062404870622, "percentage": 71.21, "elapsed_time": "2:44:01", "remaining_time": "1:06:18"} +{"current_steps": 4700, "total_steps": 6572, "loss": 0.02593054175376892, "lr": 1.0218380873241314e-06, "epoch": 2.8608828006088283, "percentage": 71.52, "elapsed_time": "2:44:38", "remaining_time": "1:05:34"} +{"current_steps": 4720, "total_steps": 6572, "loss": 0.03906567096710205, "lr": 1.0017147024449674e-06, "epoch": 2.8730593607305934, "percentage": 71.82, "elapsed_time": "2:45:15", "remaining_time": "1:04:50"} +{"current_steps": 4740, "total_steps": 6572, "loss": 0.03371626436710358, "lr": 9.81741679714493e-07, "epoch": 2.8852359208523595, "percentage": 72.12, "elapsed_time": "2:45:52", "remaining_time": "1:04:06"} +{"current_steps": 4760, "total_steps": 6572, "loss": 0.03090968132019043, "lr": 9.619210235484333e-07, "epoch": 2.8974124809741246, "percentage": 72.43, "elapsed_time": "2:46:29", "remaining_time": "1:03:22"} +{"current_steps": 4780, "total_steps": 6572, "loss": 0.0322105199098587, "lr": 9.422547230715931e-07, "epoch": 2.9095890410958907, "percentage": 72.73, "elapsed_time": "2:47:05", "remaining_time": "1:02:38"} +{"current_steps": 4800, "total_steps": 6572, "loss": 0.035210177302360535, "lr": 9.227447519182353e-07, "epoch": 2.921765601217656, "percentage": 73.04, "elapsed_time": "2:47:42", "remaining_time": "1:01:54"} +{"current_steps": 4820, "total_steps": 6572, "loss": 0.026842504739761353, "lr": 9.033930680340097e-07, "epoch": 2.933942161339422, "percentage": 73.34, "elapsed_time": "2:48:18", "remaining_time": "1:01:10"} +{"current_steps": 4840, "total_steps": 6572, "loss": 0.03439584076404571, "lr": 8.842016134794682e-07, "epoch": 2.946118721461187, "percentage": 73.65, "elapsed_time": "2:48:55", "remaining_time": "1:00:27"} +{"current_steps": 4860, "total_steps": 6572, "loss": 0.04011322855949402, "lr": 8.651723142351603e-07, "epoch": 2.958295281582953, "percentage": 73.95, "elapsed_time": "2:49:32", "remaining_time": "0:59:43"} +{"current_steps": 4880, "total_steps": 6572, "loss": 0.03800423145294189, "lr": 8.463070800083562e-07, "epoch": 2.9704718417047182, "percentage": 74.25, "elapsed_time": "2:50:10", "remaining_time": "0:59:00"} +{"current_steps": 4900, "total_steps": 6572, "loss": 0.03839131891727447, "lr": 8.276078040413879e-07, "epoch": 2.9826484018264843, "percentage": 74.56, "elapsed_time": "2:50:47", "remaining_time": "0:58:16"} +{"current_steps": 4920, "total_steps": 6572, "loss": 0.02721840739250183, "lr": 8.090763629216589e-07, "epoch": 2.9948249619482494, "percentage": 74.86, "elapsed_time": "2:51:25", "remaining_time": "0:57:33"} +{"current_steps": 4940, "total_steps": 6572, "loss": 0.023991990089416503, "lr": 7.907146163933102e-07, "epoch": 3.006697108066971, "percentage": 75.17, "elapsed_time": "2:52:02", "remaining_time": "0:56:50"} +{"current_steps": 4960, "total_steps": 6572, "loss": 0.01451514959335327, "lr": 7.725244071705871e-07, "epoch": 3.0188736681887365, "percentage": 75.47, "elapsed_time": "2:52:40", "remaining_time": "0:56:07"} +{"current_steps": 4980, "total_steps": 6572, "loss": 0.014327619969844819, "lr": 7.545075607529104e-07, "epoch": 3.031050228310502, "percentage": 75.78, "elapsed_time": "2:53:20", "remaining_time": "0:55:24"} +{"current_steps": 5000, "total_steps": 6572, "loss": 0.017832010984420776, "lr": 7.366658852416788e-07, "epoch": 3.0432267884322677, "percentage": 76.08, "elapsed_time": "2:53:56", "remaining_time": "0:54:41"} +{"current_steps": 5000, "total_steps": 6572, "eval_loss": 0.39115142822265625, "epoch": 3.0432267884322677, "percentage": 76.08, "elapsed_time": "2:54:44", "remaining_time": "0:54:56"} +{"current_steps": 5020, "total_steps": 6572, "loss": 0.011674411594867706, "lr": 7.190011711588101e-07, "epoch": 3.0554033485540333, "percentage": 76.38, "elapsed_time": "2:56:57", "remaining_time": "0:54:42"} +{"current_steps": 5040, "total_steps": 6572, "loss": 0.013690856099128724, "lr": 7.015151912670562e-07, "epoch": 3.067579908675799, "percentage": 76.69, "elapsed_time": "2:57:34", "remaining_time": "0:53:58"} +{"current_steps": 5060, "total_steps": 6572, "loss": 0.011978642642498016, "lr": 6.842097003920903e-07, "epoch": 3.0797564687975645, "percentage": 76.99, "elapsed_time": "2:58:10", "remaining_time": "0:53:14"} +{"current_steps": 5080, "total_steps": 6572, "loss": 0.013893941044807434, "lr": 6.67086435246406e-07, "epoch": 3.09193302891933, "percentage": 77.3, "elapsed_time": "2:58:46", "remaining_time": "0:52:30"} +{"current_steps": 5100, "total_steps": 6572, "loss": 0.009910025447607041, "lr": 6.501471142550194e-07, "epoch": 3.1041095890410957, "percentage": 77.6, "elapsed_time": "2:59:23", "remaining_time": "0:51:46"} +{"current_steps": 5120, "total_steps": 6572, "loss": 0.008863755315542222, "lr": 6.333934373830222e-07, "epoch": 3.1162861491628613, "percentage": 77.91, "elapsed_time": "2:59:59", "remaining_time": "0:51:02"} +{"current_steps": 5140, "total_steps": 6572, "loss": 0.010502541810274124, "lr": 6.168270859649761e-07, "epoch": 3.128462709284627, "percentage": 78.21, "elapsed_time": "3:00:37", "remaining_time": "0:50:19"} +{"current_steps": 5160, "total_steps": 6572, "loss": 0.012096930295228958, "lr": 6.004497225361786e-07, "epoch": 3.1406392694063925, "percentage": 78.51, "elapsed_time": "3:01:13", "remaining_time": "0:49:35"} +{"current_steps": 5180, "total_steps": 6572, "loss": 0.013278065621852875, "lr": 5.842629906658226e-07, "epoch": 3.1528158295281585, "percentage": 78.82, "elapsed_time": "3:01:52", "remaining_time": "0:48:52"} +{"current_steps": 5200, "total_steps": 6572, "loss": 0.013548998534679413, "lr": 5.682685147920481e-07, "epoch": 3.1649923896499237, "percentage": 79.12, "elapsed_time": "3:02:28", "remaining_time": "0:48:08"} +{"current_steps": 5220, "total_steps": 6572, "loss": 0.013736458122730255, "lr": 5.524679000589256e-07, "epoch": 3.1771689497716897, "percentage": 79.43, "elapsed_time": "3:03:05", "remaining_time": "0:47:25"} +{"current_steps": 5240, "total_steps": 6572, "loss": 0.013177134096622467, "lr": 5.36862732155366e-07, "epoch": 3.189345509893455, "percentage": 79.73, "elapsed_time": "3:03:42", "remaining_time": "0:46:41"} +{"current_steps": 5260, "total_steps": 6572, "loss": 0.011971819400787353, "lr": 5.214545771559879e-07, "epoch": 3.201522070015221, "percentage": 80.04, "elapsed_time": "3:04:17", "remaining_time": "0:45:58"} +{"current_steps": 5280, "total_steps": 6572, "loss": 0.014422819018363953, "lr": 5.062449813639528e-07, "epoch": 3.213698630136986, "percentage": 80.34, "elapsed_time": "3:04:53", "remaining_time": "0:45:14"} +{"current_steps": 5300, "total_steps": 6572, "loss": 0.010663678497076034, "lr": 4.912354711557856e-07, "epoch": 3.225875190258752, "percentage": 80.65, "elapsed_time": "3:05:29", "remaining_time": "0:44:31"} +{"current_steps": 5320, "total_steps": 6572, "loss": 0.011400717496871948, "lr": 4.764275528281892e-07, "epoch": 3.2380517503805173, "percentage": 80.95, "elapsed_time": "3:06:07", "remaining_time": "0:43:48"} +{"current_steps": 5340, "total_steps": 6572, "loss": 0.008456526696681977, "lr": 4.6182271244688355e-07, "epoch": 3.2502283105022833, "percentage": 81.25, "elapsed_time": "3:06:44", "remaining_time": "0:43:04"} +{"current_steps": 5360, "total_steps": 6572, "loss": 0.014539115130901337, "lr": 4.4742241569746407e-07, "epoch": 3.2624048706240485, "percentage": 81.56, "elapsed_time": "3:07:22", "remaining_time": "0:42:22"} +{"current_steps": 5380, "total_steps": 6572, "loss": 0.017625690996646882, "lr": 4.332281077383177e-07, "epoch": 3.2745814307458145, "percentage": 81.86, "elapsed_time": "3:07:58", "remaining_time": "0:41:38"} +{"current_steps": 5400, "total_steps": 6572, "loss": 0.007641100138425827, "lr": 4.1924121305558563e-07, "epoch": 3.2867579908675797, "percentage": 82.17, "elapsed_time": "3:08:37", "remaining_time": "0:40:56"} +{"current_steps": 5420, "total_steps": 6572, "loss": 0.011799700558185577, "lr": 4.054631353202121e-07, "epoch": 3.2989345509893457, "percentage": 82.47, "elapsed_time": "3:09:14", "remaining_time": "0:40:13"} +{"current_steps": 5440, "total_steps": 6572, "loss": 0.011455408483743667, "lr": 3.9189525724707634e-07, "epoch": 3.311111111111111, "percentage": 82.78, "elapsed_time": "3:09:51", "remaining_time": "0:39:30"} +{"current_steps": 5460, "total_steps": 6572, "loss": 0.012499115616083144, "lr": 3.785389404562259e-07, "epoch": 3.323287671232877, "percentage": 83.08, "elapsed_time": "3:10:28", "remaining_time": "0:38:47"} +{"current_steps": 5480, "total_steps": 6572, "loss": 0.01148865669965744, "lr": 3.653955253362351e-07, "epoch": 3.335464231354642, "percentage": 83.38, "elapsed_time": "3:11:03", "remaining_time": "0:38:04"} +{"current_steps": 5500, "total_steps": 6572, "loss": 0.012819178402423859, "lr": 3.5246633090968205e-07, "epoch": 3.347640791476408, "percentage": 83.69, "elapsed_time": "3:11:39", "remaining_time": "0:37:21"} +{"current_steps": 5500, "total_steps": 6572, "eval_loss": 0.43404534459114075, "epoch": 3.347640791476408, "percentage": 83.69, "elapsed_time": "3:12:26", "remaining_time": "0:37:30"} +{"current_steps": 5520, "total_steps": 6572, "loss": 0.013325585424900055, "lr": 3.397526547007832e-07, "epoch": 3.3598173515981733, "percentage": 83.99, "elapsed_time": "3:14:26", "remaining_time": "0:37:03"} +{"current_steps": 5540, "total_steps": 6572, "loss": 0.011712662875652313, "lr": 3.2725577260517396e-07, "epoch": 3.3719939117199393, "percentage": 84.3, "elapsed_time": "3:15:02", "remaining_time": "0:36:19"} +{"current_steps": 5560, "total_steps": 6572, "loss": 0.01580573171377182, "lr": 3.14976938761867e-07, "epoch": 3.3841704718417045, "percentage": 84.6, "elapsed_time": "3:15:38", "remaining_time": "0:35:36"} +{"current_steps": 5580, "total_steps": 6572, "loss": 0.012312603741884231, "lr": 3.029173854273909e-07, "epoch": 3.3963470319634705, "percentage": 84.91, "elapsed_time": "3:16:16", "remaining_time": "0:34:53"} +{"current_steps": 5600, "total_steps": 6572, "loss": 0.011797953397035599, "lr": 2.910783228521269e-07, "epoch": 3.4085235920852357, "percentage": 85.21, "elapsed_time": "3:16:54", "remaining_time": "0:34:10"} +{"current_steps": 5620, "total_steps": 6572, "loss": 0.012182456254959107, "lr": 2.794609391588504e-07, "epoch": 3.4207001522070017, "percentage": 85.51, "elapsed_time": "3:17:32", "remaining_time": "0:33:27"} +{"current_steps": 5640, "total_steps": 6572, "loss": 0.013599888980388641, "lr": 2.6806640022349897e-07, "epoch": 3.432876712328767, "percentage": 85.82, "elapsed_time": "3:18:08", "remaining_time": "0:32:44"} +{"current_steps": 5660, "total_steps": 6572, "loss": 0.009272868931293487, "lr": 2.5689584955816497e-07, "epoch": 3.445053272450533, "percentage": 86.12, "elapsed_time": "3:18:45", "remaining_time": "0:32:01"} +{"current_steps": 5680, "total_steps": 6572, "loss": 0.008165979385375976, "lr": 2.459504081963421e-07, "epoch": 3.457229832572298, "percentage": 86.43, "elapsed_time": "3:19:21", "remaining_time": "0:31:18"} +{"current_steps": 5700, "total_steps": 6572, "loss": 0.009182130545377731, "lr": 2.3523117458041865e-07, "epoch": 3.469406392694064, "percentage": 86.73, "elapsed_time": "3:19:56", "remaining_time": "0:30:35"} +{"current_steps": 5720, "total_steps": 6572, "loss": 0.0107998326420784, "lr": 2.2473922445144485e-07, "epoch": 3.4815829528158293, "percentage": 87.04, "elapsed_time": "3:20:34", "remaining_time": "0:29:52"} +{"current_steps": 5740, "total_steps": 6572, "loss": 0.014469687640666962, "lr": 2.144756107411733e-07, "epoch": 3.4937595129375953, "percentage": 87.34, "elapsed_time": "3:21:11", "remaining_time": "0:29:09"} +{"current_steps": 5760, "total_steps": 6572, "loss": 0.0121701680123806, "lr": 2.0444136346639333e-07, "epoch": 3.5059360730593605, "percentage": 87.64, "elapsed_time": "3:21:48", "remaining_time": "0:28:26"} +{"current_steps": 5780, "total_steps": 6572, "loss": 0.014668506383895875, "lr": 1.9463748962556096e-07, "epoch": 3.5181126331811265, "percentage": 87.95, "elapsed_time": "3:22:25", "remaining_time": "0:27:44"} +{"current_steps": 5800, "total_steps": 6572, "loss": 0.010488402843475342, "lr": 1.8506497309773885e-07, "epoch": 3.5302891933028917, "percentage": 88.25, "elapsed_time": "3:23:02", "remaining_time": "0:27:01"} +{"current_steps": 5820, "total_steps": 6572, "loss": 0.010667071491479874, "lr": 1.7572477454386257e-07, "epoch": 3.5424657534246577, "percentage": 88.56, "elapsed_time": "3:23:37", "remaining_time": "0:26:18"} +{"current_steps": 5840, "total_steps": 6572, "loss": 0.011079683899879456, "lr": 1.6661783131032726e-07, "epoch": 3.554642313546423, "percentage": 88.86, "elapsed_time": "3:24:14", "remaining_time": "0:25:35"} +{"current_steps": 5860, "total_steps": 6572, "loss": 0.009308797866106033, "lr": 1.5774505733492263e-07, "epoch": 3.566818873668189, "percentage": 89.17, "elapsed_time": "3:24:50", "remaining_time": "0:24:53"} +{"current_steps": 5880, "total_steps": 6572, "loss": 0.012319787591695785, "lr": 1.49107343055111e-07, "epoch": 3.578995433789954, "percentage": 89.47, "elapsed_time": "3:25:27", "remaining_time": "0:24:10"} +{"current_steps": 5900, "total_steps": 6572, "loss": 0.00843576118350029, "lr": 1.407055553186701e-07, "epoch": 3.59117199391172, "percentage": 89.77, "elapsed_time": "3:26:04", "remaining_time": "0:23:28"} +{"current_steps": 5920, "total_steps": 6572, "loss": 0.00938587412238121, "lr": 1.3254053729669564e-07, "epoch": 3.6033485540334853, "percentage": 90.08, "elapsed_time": "3:26:41", "remaining_time": "0:22:45"} +{"current_steps": 5940, "total_steps": 6572, "loss": 0.011934128403663636, "lr": 1.2461310839898656e-07, "epoch": 3.6155251141552514, "percentage": 90.38, "elapsed_time": "3:27:17", "remaining_time": "0:22:03"} +{"current_steps": 5960, "total_steps": 6572, "loss": 0.013170333206653595, "lr": 1.169240641918104e-07, "epoch": 3.6277016742770165, "percentage": 90.69, "elapsed_time": "3:27:54", "remaining_time": "0:21:20"} +{"current_steps": 5980, "total_steps": 6572, "loss": 0.014534834027290344, "lr": 1.0947417631806539e-07, "epoch": 3.6398782343987826, "percentage": 90.99, "elapsed_time": "3:28:31", "remaining_time": "0:20:38"} +{"current_steps": 6000, "total_steps": 6572, "loss": 0.011021688580513, "lr": 1.0226419241983865e-07, "epoch": 3.6520547945205477, "percentage": 91.3, "elapsed_time": "3:29:07", "remaining_time": "0:19:56"} +{"current_steps": 6000, "total_steps": 6572, "eval_loss": 0.44063475728034973, "epoch": 3.6520547945205477, "percentage": 91.3, "elapsed_time": "3:29:53", "remaining_time": "0:20:00"} +{"current_steps": 6020, "total_steps": 6572, "loss": 0.010764393210411071, "lr": 9.529483606337902e-08, "epoch": 3.6642313546423138, "percentage": 91.6, "elapsed_time": "3:31:58", "remaining_time": "0:19:26"} +{"current_steps": 6040, "total_steps": 6572, "loss": 0.012128306180238723, "lr": 8.856680666647882e-08, "epoch": 3.676407914764079, "percentage": 91.91, "elapsed_time": "3:32:33", "remaining_time": "0:18:43"} +{"current_steps": 6060, "total_steps": 6572, "loss": 0.011729901283979416, "lr": 8.208077942828713e-08, "epoch": 3.688584474885845, "percentage": 92.21, "elapsed_time": "3:33:10", "remaining_time": "0:18:00"} +{"current_steps": 6080, "total_steps": 6572, "loss": 0.009119105339050294, "lr": 7.58374052615457e-08, "epoch": 3.70076103500761, "percentage": 92.51, "elapsed_time": "3:33:46", "remaining_time": "0:17:17"} +{"current_steps": 6100, "total_steps": 6572, "loss": 0.017101363837718965, "lr": 6.983731072726818e-08, "epoch": 3.712937595129376, "percentage": 92.82, "elapsed_time": "3:34:23", "remaining_time": "0:16:35"} +{"current_steps": 6120, "total_steps": 6572, "loss": 0.012368235737085342, "lr": 6.408109797186118e-08, "epoch": 3.7251141552511413, "percentage": 93.12, "elapsed_time": "3:34:59", "remaining_time": "0:15:52"} +{"current_steps": 6140, "total_steps": 6572, "loss": 0.008782628178596496, "lr": 5.856934466669212e-08, "epoch": 3.7372907153729074, "percentage": 93.43, "elapsed_time": "3:35:36", "remaining_time": "0:15:10"} +{"current_steps": 6160, "total_steps": 6572, "loss": 0.008994438499212266, "lr": 5.3302603950119994e-08, "epoch": 3.7494672754946725, "percentage": 93.73, "elapsed_time": "3:36:13", "remaining_time": "0:14:27"} +{"current_steps": 6180, "total_steps": 6572, "loss": 0.011286454647779465, "lr": 4.8281404371981755e-08, "epoch": 3.7616438356164386, "percentage": 94.04, "elapsed_time": "3:36:50", "remaining_time": "0:13:45"} +{"current_steps": 6200, "total_steps": 6572, "loss": 0.011785905063152313, "lr": 4.350624984055196e-08, "epoch": 3.7738203957382037, "percentage": 94.34, "elapsed_time": "3:37:26", "remaining_time": "0:13:02"} +{"current_steps": 6220, "total_steps": 6572, "loss": 0.013624191284179688, "lr": 3.897761957196877e-08, "epoch": 3.7859969558599698, "percentage": 94.64, "elapsed_time": "3:38:01", "remaining_time": "0:12:20"} +{"current_steps": 6240, "total_steps": 6572, "loss": 0.011700452119112015, "lr": 3.469596804214548e-08, "epoch": 3.798173515981735, "percentage": 94.95, "elapsed_time": "3:38:37", "remaining_time": "0:11:37"} +{"current_steps": 6260, "total_steps": 6572, "loss": 0.011029987037181855, "lr": 3.06617249411581e-08, "epoch": 3.810350076103501, "percentage": 95.25, "elapsed_time": "3:39:12", "remaining_time": "0:10:55"} +{"current_steps": 6280, "total_steps": 6572, "loss": 0.010965974628925323, "lr": 2.687529513012488e-08, "epoch": 3.822526636225266, "percentage": 95.56, "elapsed_time": "3:39:49", "remaining_time": "0:10:13"} +{"current_steps": 6300, "total_steps": 6572, "loss": 0.012378603965044022, "lr": 2.3337058600575722e-08, "epoch": 3.834703196347032, "percentage": 95.86, "elapsed_time": "3:40:25", "remaining_time": "0:09:30"} +{"current_steps": 6320, "total_steps": 6572, "loss": 0.011792077124118805, "lr": 2.0047370436317437e-08, "epoch": 3.8468797564687973, "percentage": 96.17, "elapsed_time": "3:41:00", "remaining_time": "0:08:48"} +{"current_steps": 6340, "total_steps": 6572, "loss": 0.01145942509174347, "lr": 1.7006560777798608e-08, "epoch": 3.8590563165905634, "percentage": 96.47, "elapsed_time": "3:41:35", "remaining_time": "0:08:06"} +{"current_steps": 6360, "total_steps": 6572, "loss": 0.011088228970766067, "lr": 1.421493478897945e-08, "epoch": 3.8712328767123285, "percentage": 96.77, "elapsed_time": "3:42:11", "remaining_time": "0:07:24"} +{"current_steps": 6380, "total_steps": 6572, "loss": 0.010828402638435364, "lr": 1.1672772626704909e-08, "epoch": 3.8834094368340946, "percentage": 97.08, "elapsed_time": "3:42:47", "remaining_time": "0:06:42"} +{"current_steps": 6400, "total_steps": 6572, "loss": 0.01165580153465271, "lr": 9.38032941258965e-09, "epoch": 3.8955859969558597, "percentage": 97.38, "elapsed_time": "3:43:22", "remaining_time": "0:06:00"} +{"current_steps": 6420, "total_steps": 6572, "loss": 0.010783226788043975, "lr": 7.3378352074163215e-09, "epoch": 3.9077625570776258, "percentage": 97.69, "elapsed_time": "3:43:58", "remaining_time": "0:05:18"} +{"current_steps": 6440, "total_steps": 6572, "loss": 0.011295531690120698, "lr": 5.545494988045963e-09, "epoch": 3.919939117199391, "percentage": 97.99, "elapsed_time": "3:44:33", "remaining_time": "0:04:36"} +{"current_steps": 6460, "total_steps": 6572, "loss": 0.013613662123680115, "lr": 4.003488626848073e-09, "epoch": 3.932115677321157, "percentage": 98.3, "elapsed_time": "3:45:10", "remaining_time": "0:03:54"} +{"current_steps": 6480, "total_steps": 6572, "loss": 0.011696261167526246, "lr": 2.7119708736486615e-09, "epoch": 3.944292237442922, "percentage": 98.6, "elapsed_time": "3:45:47", "remaining_time": "0:03:12"} +{"current_steps": 6500, "total_steps": 6572, "loss": 0.010873865336179733, "lr": 1.6710713402015577e-09, "epoch": 3.956468797564688, "percentage": 98.9, "elapsed_time": "3:46:23", "remaining_time": "0:02:30"} +{"current_steps": 6500, "total_steps": 6572, "eval_loss": 0.441041499376297, "epoch": 3.956468797564688, "percentage": 98.9, "elapsed_time": "3:47:09", "remaining_time": "0:02:30"} +{"current_steps": 6520, "total_steps": 6572, "loss": 0.012961818277835846, "lr": 8.80894487179651e-10, "epoch": 3.9686453576864533, "percentage": 99.21, "elapsed_time": "3:49:12", "remaining_time": "0:01:49"} +{"current_steps": 6540, "total_steps": 6572, "loss": 0.01224210560321808, "lr": 3.4151961369188745e-10, "epoch": 3.9808219178082194, "percentage": 99.51, "elapsed_time": "3:49:48", "remaining_time": "0:01:07"} +{"current_steps": 6560, "total_steps": 6572, "loss": 0.010245455056428909, "lr": 5.300084932574612e-11, "epoch": 3.9929984779299845, "percentage": 99.82, "elapsed_time": "3:50:24", "remaining_time": "0:00:25"} +{"current_steps": 6572, "total_steps": 6572, "epoch": 4.0, "percentage": 100.0, "elapsed_time": "3:52:13", "remaining_time": "0:00:00"} diff --git a/trainer_state.json b/trainer_state.json new file mode 100644 index 0000000..d44e87b --- /dev/null +++ b/trainer_state.json @@ -0,0 +1,2443 @@ +{ + "best_global_step": 1000, + "best_metric": 0.24901165068149567, + "best_model_checkpoint": "/root/workspace/finetune/checkpoints/train_run_06_qwen3_4b_formal/checkpoint-1000", + "epoch": 4.0, + "eval_steps": 500, + "global_step": 6572, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0121765601217656, + "grad_norm": 2.8322572708129883, + "learning_rate": 3.166666666666667e-07, + "loss": 1.119423007965088, + "step": 20 + }, + { + "epoch": 0.0243531202435312, + "grad_norm": 2.00534987449646, + "learning_rate": 6.5e-07, + "loss": 0.8098940849304199, + "step": 40 + }, + { + "epoch": 0.0365296803652968, + "grad_norm": 1.4319002628326416, + "learning_rate": 9.833333333333334e-07, + "loss": 0.6310151576995849, + "step": 60 + }, + { + "epoch": 0.0487062404870624, + "grad_norm": 1.3265702724456787, + "learning_rate": 1.3166666666666666e-06, + "loss": 0.5526751518249512, + "step": 80 + }, + { + "epoch": 0.060882800608828, + "grad_norm": 0.8091968297958374, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.4834024906158447, + "step": 100 + }, + { + "epoch": 0.0730593607305936, + "grad_norm": 0.8584772348403931, + "learning_rate": 1.9833333333333335e-06, + "loss": 0.44873433113098143, + "step": 120 + }, + { + "epoch": 0.0852359208523592, + "grad_norm": 0.9426456093788147, + "learning_rate": 2.316666666666667e-06, + "loss": 0.38286705017089845, + "step": 140 + }, + { + "epoch": 0.0974124809741248, + "grad_norm": 0.939085066318512, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.38465352058410646, + "step": 160 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 0.8642272353172302, + "learning_rate": 2.9833333333333337e-06, + "loss": 0.3305965900421143, + "step": 180 + }, + { + "epoch": 0.121765601217656, + "grad_norm": 0.7353930473327637, + "learning_rate": 3.316666666666667e-06, + "loss": 0.2875316619873047, + "step": 200 + }, + { + "epoch": 0.1339421613394216, + "grad_norm": 0.7686158418655396, + "learning_rate": 3.65e-06, + "loss": 0.3140716075897217, + "step": 220 + }, + { + "epoch": 0.1461187214611872, + "grad_norm": 1.153424859046936, + "learning_rate": 3.983333333333334e-06, + "loss": 0.2985520839691162, + "step": 240 + }, + { + "epoch": 0.1582952815829528, + "grad_norm": 0.7382195591926575, + "learning_rate": 4.316666666666667e-06, + "loss": 0.2449601411819458, + "step": 260 + }, + { + "epoch": 0.1704718417047184, + "grad_norm": 0.7017386555671692, + "learning_rate": 4.65e-06, + "loss": 0.24566495418548584, + "step": 280 + }, + { + "epoch": 0.182648401826484, + "grad_norm": 1.0982152223587036, + "learning_rate": 4.983333333333334e-06, + "loss": 0.20863604545593262, + "step": 300 + }, + { + "epoch": 0.1948249619482496, + "grad_norm": 0.6582516431808472, + "learning_rate": 4.9998867856224845e-06, + "loss": 0.19693111181259154, + "step": 320 + }, + { + "epoch": 0.2070015220700152, + "grad_norm": 0.6201784610748291, + "learning_rate": 4.999523005839606e-06, + "loss": 0.21833207607269287, + "step": 340 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 0.7551426887512207, + "learning_rate": 4.998908383543311e-06, + "loss": 0.2382877826690674, + "step": 360 + }, + { + "epoch": 0.2313546423135464, + "grad_norm": 0.6009605526924133, + "learning_rate": 4.9980429804147276e-06, + "loss": 0.21118538379669188, + "step": 380 + }, + { + "epoch": 0.243531202435312, + "grad_norm": 0.7086856365203857, + "learning_rate": 4.996926883302385e-06, + "loss": 0.18367968797683715, + "step": 400 + }, + { + "epoch": 0.2557077625570776, + "grad_norm": 0.964198648929596, + "learning_rate": 4.995560204213496e-06, + "loss": 0.22483460903167723, + "step": 420 + }, + { + "epoch": 0.2678843226788432, + "grad_norm": 0.6596850156784058, + "learning_rate": 4.993943080302715e-06, + "loss": 0.22673561573028564, + "step": 440 + }, + { + "epoch": 0.2800608828006088, + "grad_norm": 0.5922938585281372, + "learning_rate": 4.992075673858379e-06, + "loss": 0.185296630859375, + "step": 460 + }, + { + "epoch": 0.2922374429223744, + "grad_norm": 0.6671269536018372, + "learning_rate": 4.989958172286214e-06, + "loss": 0.18437937498092652, + "step": 480 + }, + { + "epoch": 0.30441400304414, + "grad_norm": 0.7064187526702881, + "learning_rate": 4.987590788090533e-06, + "loss": 0.1850834846496582, + "step": 500 + }, + { + "epoch": 0.30441400304414, + "eval_loss": 0.2562323212623596, + "eval_runtime": 47.3626, + "eval_samples_per_second": 18.58, + "eval_steps_per_second": 18.58, + "step": 500 + }, + { + "epoch": 0.3165905631659056, + "grad_norm": 0.7698928117752075, + "learning_rate": 4.984973758852904e-06, + "loss": 0.16346561908721924, + "step": 520 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 0.4859939515590668, + "learning_rate": 4.982107347208317e-06, + "loss": 0.18838067054748536, + "step": 540 + }, + { + "epoch": 0.3409436834094368, + "grad_norm": 0.5321416854858398, + "learning_rate": 4.978991840818816e-06, + "loss": 0.177593994140625, + "step": 560 + }, + { + "epoch": 0.3531202435312024, + "grad_norm": 0.593889594078064, + "learning_rate": 4.975627552344638e-06, + "loss": 0.20775914192199707, + "step": 580 + }, + { + "epoch": 0.365296803652968, + "grad_norm": 0.7926055192947388, + "learning_rate": 4.97201481941283e-06, + "loss": 0.16498700380325318, + "step": 600 + }, + { + "epoch": 0.3774733637747336, + "grad_norm": 0.2909524440765381, + "learning_rate": 4.968154004583374e-06, + "loss": 0.17565951347351075, + "step": 620 + }, + { + "epoch": 0.3896499238964992, + "grad_norm": 0.4575704038143158, + "learning_rate": 4.964045495312794e-06, + "loss": 0.16204673051834106, + "step": 640 + }, + { + "epoch": 0.4018264840182648, + "grad_norm": 0.5366008281707764, + "learning_rate": 4.959689703915272e-06, + "loss": 0.17068564891815186, + "step": 660 + }, + { + "epoch": 0.4140030441400304, + "grad_norm": 0.5129569172859192, + "learning_rate": 4.95508706752128e-06, + "loss": 0.1589680790901184, + "step": 680 + }, + { + "epoch": 0.426179604261796, + "grad_norm": 0.4709528684616089, + "learning_rate": 4.9502380480337e-06, + "loss": 0.17568455934524535, + "step": 700 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 0.6092886328697205, + "learning_rate": 4.9451431320814715e-06, + "loss": 0.16204804182052612, + "step": 720 + }, + { + "epoch": 0.4505327245053272, + "grad_norm": 0.5957323908805847, + "learning_rate": 4.939802830970762e-06, + "loss": 0.16562143564224244, + "step": 740 + }, + { + "epoch": 0.4627092846270928, + "grad_norm": 0.4758240580558777, + "learning_rate": 4.934217680633646e-06, + "loss": 0.17697544097900392, + "step": 760 + }, + { + "epoch": 0.4748858447488584, + "grad_norm": 0.865627646446228, + "learning_rate": 4.928388241574327e-06, + "loss": 0.1649466037750244, + "step": 780 + }, + { + "epoch": 0.487062404870624, + "grad_norm": 0.466294527053833, + "learning_rate": 4.922315098812883e-06, + "loss": 0.1602837324142456, + "step": 800 + }, + { + "epoch": 0.4992389649923896, + "grad_norm": 0.6357060670852661, + "learning_rate": 4.9159988618265585e-06, + "loss": 0.142719042301178, + "step": 820 + }, + { + "epoch": 0.5114155251141552, + "grad_norm": 0.6055647730827332, + "learning_rate": 4.9094401644886e-06, + "loss": 0.14508233070373536, + "step": 840 + }, + { + "epoch": 0.5235920852359208, + "grad_norm": 0.45214834809303284, + "learning_rate": 4.902639665004641e-06, + "loss": 0.1821539044380188, + "step": 860 + }, + { + "epoch": 0.5357686453576864, + "grad_norm": 0.5735688805580139, + "learning_rate": 4.89559804584665e-06, + "loss": 0.16131887435913086, + "step": 880 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 0.6279348731040955, + "learning_rate": 4.888316013684435e-06, + "loss": 0.17404688596725465, + "step": 900 + }, + { + "epoch": 0.5601217656012176, + "grad_norm": 0.6474089026451111, + "learning_rate": 4.880794299314732e-06, + "loss": 0.14134640693664552, + "step": 920 + }, + { + "epoch": 0.5722983257229832, + "grad_norm": 0.5808464884757996, + "learning_rate": 4.87303365758786e-06, + "loss": 0.14891813993453978, + "step": 940 + }, + { + "epoch": 0.5844748858447488, + "grad_norm": 0.5440990328788757, + "learning_rate": 4.865034867331967e-06, + "loss": 0.1696299910545349, + "step": 960 + }, + { + "epoch": 0.5966514459665144, + "grad_norm": 0.6859214901924133, + "learning_rate": 4.856798731274874e-06, + "loss": 0.14085158109664916, + "step": 980 + }, + { + "epoch": 0.60882800608828, + "grad_norm": 0.3178713619709015, + "learning_rate": 4.84832607596351e-06, + "loss": 0.133053982257843, + "step": 1000 + }, + { + "epoch": 0.60882800608828, + "eval_loss": 0.24901165068149567, + "eval_runtime": 46.6976, + "eval_samples_per_second": 18.845, + "eval_steps_per_second": 18.845, + "step": 1000 + }, + { + "epoch": 0.6210045662100456, + "grad_norm": 0.3812738358974457, + "learning_rate": 4.8396177516809695e-06, + "loss": 0.12680984735488893, + "step": 1020 + }, + { + "epoch": 0.6331811263318112, + "grad_norm": 0.5174199342727661, + "learning_rate": 4.830674632361178e-06, + "loss": 0.14880582094192504, + "step": 1040 + }, + { + "epoch": 0.6453576864535768, + "grad_norm": 0.4705193042755127, + "learning_rate": 4.821497615501186e-06, + "loss": 0.1447562575340271, + "step": 1060 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 0.42298850417137146, + "learning_rate": 4.812087622071104e-06, + "loss": 0.15530819892883302, + "step": 1080 + }, + { + "epoch": 0.669710806697108, + "grad_norm": 0.30658382177352905, + "learning_rate": 4.80244559642167e-06, + "loss": 0.14426586627960206, + "step": 1100 + }, + { + "epoch": 0.6818873668188736, + "grad_norm": 0.4838867783546448, + "learning_rate": 4.792572506189489e-06, + "loss": 0.15436025857925414, + "step": 1120 + }, + { + "epoch": 0.6940639269406392, + "grad_norm": 0.716833770275116, + "learning_rate": 4.782469342199915e-06, + "loss": 0.14860854148864747, + "step": 1140 + }, + { + "epoch": 0.7062404870624048, + "grad_norm": 0.36538004875183105, + "learning_rate": 4.7721371183676205e-06, + "loss": 0.1313084125518799, + "step": 1160 + }, + { + "epoch": 0.7184170471841704, + "grad_norm": 0.5409316420555115, + "learning_rate": 4.761576871594841e-06, + "loss": 0.150812029838562, + "step": 1180 + }, + { + "epoch": 0.730593607305936, + "grad_norm": 0.5275493264198303, + "learning_rate": 4.750789661667318e-06, + "loss": 0.13278884887695314, + "step": 1200 + }, + { + "epoch": 0.7427701674277016, + "grad_norm": 0.5485584735870361, + "learning_rate": 4.739776571147943e-06, + "loss": 0.1612934350967407, + "step": 1220 + }, + { + "epoch": 0.7549467275494672, + "grad_norm": 0.5949460864067078, + "learning_rate": 4.728538705268116e-06, + "loss": 0.16211290359497071, + "step": 1240 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 0.43323376774787903, + "learning_rate": 4.717077191816824e-06, + "loss": 0.14386119842529296, + "step": 1260 + }, + { + "epoch": 0.7792998477929984, + "grad_norm": 0.6409174799919128, + "learning_rate": 4.705393181027463e-06, + "loss": 0.12942540645599365, + "step": 1280 + }, + { + "epoch": 0.791476407914764, + "grad_norm": 0.4871342182159424, + "learning_rate": 4.693487845462413e-06, + "loss": 0.14771063327789308, + "step": 1300 + }, + { + "epoch": 0.8036529680365296, + "grad_norm": 0.5108008980751038, + "learning_rate": 4.681362379895349e-06, + "loss": 0.1276724100112915, + "step": 1320 + }, + { + "epoch": 0.8158295281582952, + "grad_norm": 0.915285587310791, + "learning_rate": 4.6690180011913524e-06, + "loss": 0.1319241166114807, + "step": 1340 + }, + { + "epoch": 0.8280060882800608, + "grad_norm": 0.5282526612281799, + "learning_rate": 4.6564559481847795e-06, + "loss": 0.1557891011238098, + "step": 1360 + }, + { + "epoch": 0.8401826484018264, + "grad_norm": 0.46745216846466064, + "learning_rate": 4.643677481554947e-06, + "loss": 0.11075855493545532, + "step": 1380 + }, + { + "epoch": 0.852359208523592, + "grad_norm": 0.40246087312698364, + "learning_rate": 4.630683883699607e-06, + "loss": 0.1580789566040039, + "step": 1400 + }, + { + "epoch": 0.8645357686453576, + "grad_norm": 0.3718211352825165, + "learning_rate": 4.6174764586062556e-06, + "loss": 0.16006500720977784, + "step": 1420 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 0.4359384775161743, + "learning_rate": 4.6040565317212685e-06, + "loss": 0.1462727189064026, + "step": 1440 + }, + { + "epoch": 0.8888888888888888, + "grad_norm": 0.3503302037715912, + "learning_rate": 4.59042544981688e-06, + "loss": 0.14725338220596312, + "step": 1460 + }, + { + "epoch": 0.9010654490106544, + "grad_norm": 0.3662220537662506, + "learning_rate": 4.5765845808560334e-06, + "loss": 0.1304166793823242, + "step": 1480 + }, + { + "epoch": 0.91324200913242, + "grad_norm": 0.45357516407966614, + "learning_rate": 4.562535313855094e-06, + "loss": 0.1293134570121765, + "step": 1500 + }, + { + "epoch": 0.91324200913242, + "eval_loss": 0.25538697838783264, + "eval_runtime": 47.0026, + "eval_samples_per_second": 18.722, + "eval_steps_per_second": 18.722, + "step": 1500 + }, + { + "epoch": 0.9254185692541856, + "grad_norm": 0.31984779238700867, + "learning_rate": 4.548279058744451e-06, + "loss": 0.11359381675720215, + "step": 1520 + }, + { + "epoch": 0.9375951293759512, + "grad_norm": 0.6947388052940369, + "learning_rate": 4.533817246227024e-06, + "loss": 0.15145074129104613, + "step": 1540 + }, + { + "epoch": 0.9497716894977168, + "grad_norm": 0.5685822367668152, + "learning_rate": 4.519151327634685e-06, + "loss": 0.11953675746917725, + "step": 1560 + }, + { + "epoch": 0.9619482496194824, + "grad_norm": 0.2805669903755188, + "learning_rate": 4.504282774782605e-06, + "loss": 0.13977375030517578, + "step": 1580 + }, + { + "epoch": 0.974124809741248, + "grad_norm": 0.333103746175766, + "learning_rate": 4.489213079821551e-06, + "loss": 0.1338045358657837, + "step": 1600 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 0.5493115782737732, + "learning_rate": 4.4739437550881355e-06, + "loss": 0.11776142120361328, + "step": 1620 + }, + { + "epoch": 0.9984779299847792, + "grad_norm": 0.4903205931186676, + "learning_rate": 4.458476332953051e-06, + "loss": 0.12504475116729735, + "step": 1640 + }, + { + "epoch": 1.0103500761035007, + "grad_norm": 0.2682284712791443, + "learning_rate": 4.442812365667281e-06, + "loss": 0.08379222154617309, + "step": 1660 + }, + { + "epoch": 1.0225266362252663, + "grad_norm": 0.34869125485420227, + "learning_rate": 4.426953425206322e-06, + "loss": 0.08407147526741028, + "step": 1680 + }, + { + "epoch": 1.034703196347032, + "grad_norm": 0.38522422313690186, + "learning_rate": 4.410901103112434e-06, + "loss": 0.08041079640388489, + "step": 1700 + }, + { + "epoch": 1.0468797564687975, + "grad_norm": 0.4700411856174469, + "learning_rate": 4.394657010334908e-06, + "loss": 0.07876392006874085, + "step": 1720 + }, + { + "epoch": 1.059056316590563, + "grad_norm": 0.3719494938850403, + "learning_rate": 4.378222777068406e-06, + "loss": 0.10302903652191162, + "step": 1740 + }, + { + "epoch": 1.0712328767123287, + "grad_norm": 0.406753808259964, + "learning_rate": 4.361600052589358e-06, + "loss": 0.08733606934547425, + "step": 1760 + }, + { + "epoch": 1.0834094368340943, + "grad_norm": 0.19822958111763, + "learning_rate": 4.344790505090447e-06, + "loss": 0.08532609939575195, + "step": 1780 + }, + { + "epoch": 1.09558599695586, + "grad_norm": 0.3569728434085846, + "learning_rate": 4.327795821513195e-06, + "loss": 0.08734336495399475, + "step": 1800 + }, + { + "epoch": 1.1077625570776255, + "grad_norm": 0.4549162685871124, + "learning_rate": 4.3106177073786684e-06, + "loss": 0.0913870632648468, + "step": 1820 + }, + { + "epoch": 1.119939117199391, + "grad_norm": 0.5802178978919983, + "learning_rate": 4.293257886616318e-06, + "loss": 0.08115516304969787, + "step": 1840 + }, + { + "epoch": 1.1321156773211567, + "grad_norm": 0.30524641275405884, + "learning_rate": 4.275718101390975e-06, + "loss": 0.08891176581382751, + "step": 1860 + }, + { + "epoch": 1.1442922374429223, + "grad_norm": 0.33804091811180115, + "learning_rate": 4.25800011192801e-06, + "loss": 0.07950961589813232, + "step": 1880 + }, + { + "epoch": 1.156468797564688, + "grad_norm": 0.33472684025764465, + "learning_rate": 4.240105696336687e-06, + "loss": 0.08310645222663879, + "step": 1900 + }, + { + "epoch": 1.1686453576864535, + "grad_norm": 0.3032575845718384, + "learning_rate": 4.222036650431715e-06, + "loss": 0.07682961225509644, + "step": 1920 + }, + { + "epoch": 1.180821917808219, + "grad_norm": 0.47542238235473633, + "learning_rate": 4.203794787553032e-06, + "loss": 0.07520227432250977, + "step": 1940 + }, + { + "epoch": 1.1929984779299847, + "grad_norm": 0.39914897084236145, + "learning_rate": 4.185381938383821e-06, + "loss": 0.0754019558429718, + "step": 1960 + }, + { + "epoch": 1.2051750380517503, + "grad_norm": 0.4697635769844055, + "learning_rate": 4.166799950766793e-06, + "loss": 0.08085885643959045, + "step": 1980 + }, + { + "epoch": 1.217351598173516, + "grad_norm": 0.28078529238700867, + "learning_rate": 4.14805068951874e-06, + "loss": 0.0884653627872467, + "step": 2000 + }, + { + "epoch": 1.217351598173516, + "eval_loss": 0.2661186456680298, + "eval_runtime": 46.548, + "eval_samples_per_second": 18.905, + "eval_steps_per_second": 18.905, + "step": 2000 + }, + { + "epoch": 1.2295281582952815, + "grad_norm": 0.2885560989379883, + "learning_rate": 4.1291360362433965e-06, + "loss": 0.06684748530387878, + "step": 2020 + }, + { + "epoch": 1.241704718417047, + "grad_norm": 0.24163688719272614, + "learning_rate": 4.110057889142601e-06, + "loss": 0.0720324158668518, + "step": 2040 + }, + { + "epoch": 1.2538812785388127, + "grad_norm": 0.52589350938797, + "learning_rate": 4.090818162825804e-06, + "loss": 0.08799988031387329, + "step": 2060 + }, + { + "epoch": 1.2660578386605783, + "grad_norm": 0.35164448618888855, + "learning_rate": 4.071418788117926e-06, + "loss": 0.09275985956192016, + "step": 2080 + }, + { + "epoch": 1.278234398782344, + "grad_norm": 0.4981421232223511, + "learning_rate": 4.0518617118655845e-06, + "loss": 0.08431113958358764, + "step": 2100 + }, + { + "epoch": 1.2904109589041095, + "grad_norm": 0.2867731750011444, + "learning_rate": 4.032148896741717e-06, + "loss": 0.09995608925819396, + "step": 2120 + }, + { + "epoch": 1.302587519025875, + "grad_norm": 0.5612165331840515, + "learning_rate": 4.012282321048618e-06, + "loss": 0.07387629747390748, + "step": 2140 + }, + { + "epoch": 1.3147640791476407, + "grad_norm": 0.4880141317844391, + "learning_rate": 3.992263978519398e-06, + "loss": 0.07667248249053955, + "step": 2160 + }, + { + "epoch": 1.3269406392694063, + "grad_norm": 0.3143049478530884, + "learning_rate": 3.972095878117904e-06, + "loss": 0.09203824400901794, + "step": 2180 + }, + { + "epoch": 1.339117199391172, + "grad_norm": 0.47859013080596924, + "learning_rate": 3.951780043837107e-06, + "loss": 0.07835246920585633, + "step": 2200 + }, + { + "epoch": 1.3512937595129375, + "grad_norm": 0.28448912501335144, + "learning_rate": 3.9313185144959835e-06, + "loss": 0.08577624559402466, + "step": 2220 + }, + { + "epoch": 1.363470319634703, + "grad_norm": 0.32139304280281067, + "learning_rate": 3.9107133435349025e-06, + "loss": 0.0789969801902771, + "step": 2240 + }, + { + "epoch": 1.3756468797564687, + "grad_norm": 0.4797567129135132, + "learning_rate": 3.889966598809557e-06, + "loss": 0.07151145935058593, + "step": 2260 + }, + { + "epoch": 1.3878234398782343, + "grad_norm": 0.2404891699552536, + "learning_rate": 3.869080362383437e-06, + "loss": 0.09204544425010681, + "step": 2280 + }, + { + "epoch": 1.4, + "grad_norm": 0.328392893075943, + "learning_rate": 3.848056730318881e-06, + "loss": 0.11079612970352173, + "step": 2300 + }, + { + "epoch": 1.4121765601217655, + "grad_norm": 0.2993980646133423, + "learning_rate": 3.826897812466728e-06, + "loss": 0.06770140528678895, + "step": 2320 + }, + { + "epoch": 1.4243531202435311, + "grad_norm": 0.47816380858421326, + "learning_rate": 3.8056057322545763e-06, + "loss": 0.08210510611534119, + "step": 2340 + }, + { + "epoch": 1.4365296803652967, + "grad_norm": 0.38082119822502136, + "learning_rate": 3.7841826264736888e-06, + "loss": 0.09583572745323181, + "step": 2360 + }, + { + "epoch": 1.4487062404870623, + "grad_norm": 0.3811774253845215, + "learning_rate": 3.762630645064547e-06, + "loss": 0.09235450625419617, + "step": 2380 + }, + { + "epoch": 1.460882800608828, + "grad_norm": 0.3869916498661041, + "learning_rate": 3.7409519509010985e-06, + "loss": 0.08658097982406616, + "step": 2400 + }, + { + "epoch": 1.4730593607305935, + "grad_norm": 0.5042543411254883, + "learning_rate": 3.7191487195736915e-06, + "loss": 0.08892765045166015, + "step": 2420 + }, + { + "epoch": 1.4852359208523591, + "grad_norm": 0.4064173996448517, + "learning_rate": 3.697223139170748e-06, + "loss": 0.07849371433258057, + "step": 2440 + }, + { + "epoch": 1.4974124809741247, + "grad_norm": 0.3958194851875305, + "learning_rate": 3.6751774100591716e-06, + "loss": 0.07469035387039184, + "step": 2460 + }, + { + "epoch": 1.5095890410958903, + "grad_norm": 0.39424121379852295, + "learning_rate": 3.6530137446635265e-06, + "loss": 0.0782626211643219, + "step": 2480 + }, + { + "epoch": 1.521765601217656, + "grad_norm": 0.46179333329200745, + "learning_rate": 3.630734367244012e-06, + "loss": 0.08304058909416198, + "step": 2500 + }, + { + "epoch": 1.521765601217656, + "eval_loss": 0.2771773338317871, + "eval_runtime": 46.9989, + "eval_samples_per_second": 18.724, + "eval_steps_per_second": 18.724, + "step": 2500 + }, + { + "epoch": 1.5339421613394215, + "grad_norm": 0.3447970151901245, + "learning_rate": 3.6083415136732374e-06, + "loss": 0.08037537932395936, + "step": 2520 + }, + { + "epoch": 1.5461187214611871, + "grad_norm": 0.544224202632904, + "learning_rate": 3.585837431211845e-06, + "loss": 0.08990358114242554, + "step": 2540 + }, + { + "epoch": 1.5582952815829527, + "grad_norm": 0.2808915376663208, + "learning_rate": 3.563224378282978e-06, + "loss": 0.0773526132106781, + "step": 2560 + }, + { + "epoch": 1.5704718417047183, + "grad_norm": 0.27800413966178894, + "learning_rate": 3.5405046242456396e-06, + "loss": 0.0777865469455719, + "step": 2580 + }, + { + "epoch": 1.582648401826484, + "grad_norm": 0.4247933328151703, + "learning_rate": 3.517680449166943e-06, + "loss": 0.08037815093994141, + "step": 2600 + }, + { + "epoch": 1.5948249619482495, + "grad_norm": 0.748912513256073, + "learning_rate": 3.4947541435932976e-06, + "loss": 0.08837634325027466, + "step": 2620 + }, + { + "epoch": 1.6070015220700151, + "grad_norm": 0.29709526896476746, + "learning_rate": 3.471728008320532e-06, + "loss": 0.08201563358306885, + "step": 2640 + }, + { + "epoch": 1.6191780821917807, + "grad_norm": 0.3287765681743622, + "learning_rate": 3.4486043541630066e-06, + "loss": 0.07379403114318847, + "step": 2660 + }, + { + "epoch": 1.6313546423135463, + "grad_norm": 0.4224306643009186, + "learning_rate": 3.425385501721696e-06, + "loss": 0.08871785402297974, + "step": 2680 + }, + { + "epoch": 1.643531202435312, + "grad_norm": 0.5213542580604553, + "learning_rate": 3.4020737811513107e-06, + "loss": 0.07757498621940613, + "step": 2700 + }, + { + "epoch": 1.6557077625570775, + "grad_norm": 0.21830426156520844, + "learning_rate": 3.3786715319264483e-06, + "loss": 0.08565697669982911, + "step": 2720 + }, + { + "epoch": 1.6678843226788431, + "grad_norm": 0.30557703971862793, + "learning_rate": 3.355181102606816e-06, + "loss": 0.08129348754882812, + "step": 2740 + }, + { + "epoch": 1.6800608828006087, + "grad_norm": 0.453703910112381, + "learning_rate": 3.331604850601533e-06, + "loss": 0.07875375747680664, + "step": 2760 + }, + { + "epoch": 1.6922374429223743, + "grad_norm": 0.33889397978782654, + "learning_rate": 3.307945141932556e-06, + "loss": 0.08989614248275757, + "step": 2780 + }, + { + "epoch": 1.70441400304414, + "grad_norm": 0.21036894619464874, + "learning_rate": 3.2842043509972294e-06, + "loss": 0.08251298069953919, + "step": 2800 + }, + { + "epoch": 1.7165905631659055, + "grad_norm": 0.2764931321144104, + "learning_rate": 3.2603848603300026e-06, + "loss": 0.07760271430015564, + "step": 2820 + }, + { + "epoch": 1.7287671232876711, + "grad_norm": 0.3566988706588745, + "learning_rate": 3.236489060363329e-06, + "loss": 0.07395396232604981, + "step": 2840 + }, + { + "epoch": 1.7409436834094367, + "grad_norm": 0.3616081476211548, + "learning_rate": 3.212519349187766e-06, + "loss": 0.07028600573539734, + "step": 2860 + }, + { + "epoch": 1.7531202435312023, + "grad_norm": 0.41925984621047974, + "learning_rate": 3.188478132311319e-06, + "loss": 0.08469281196594239, + "step": 2880 + }, + { + "epoch": 1.765296803652968, + "grad_norm": 0.4866960346698761, + "learning_rate": 3.164367822418029e-06, + "loss": 0.09567424058914184, + "step": 2900 + }, + { + "epoch": 1.7774733637747335, + "grad_norm": 0.2977049648761749, + "learning_rate": 3.1401908391258474e-06, + "loss": 0.07239987254142762, + "step": 2920 + }, + { + "epoch": 1.7896499238964991, + "grad_norm": 0.3723915219306946, + "learning_rate": 3.1159496087438098e-06, + "loss": 0.0891954243183136, + "step": 2940 + }, + { + "epoch": 1.8018264840182647, + "grad_norm": 0.22685140371322632, + "learning_rate": 3.0916465640285426e-06, + "loss": 0.07796849608421326, + "step": 2960 + }, + { + "epoch": 1.8140030441400303, + "grad_norm": 0.23534800112247467, + "learning_rate": 3.0672841439401223e-06, + "loss": 0.08645985722541809, + "step": 2980 + }, + { + "epoch": 1.826179604261796, + "grad_norm": 0.40188542008399963, + "learning_rate": 3.0428647933973103e-06, + "loss": 0.08427774310111999, + "step": 3000 + }, + { + "epoch": 1.826179604261796, + "eval_loss": 0.2774485647678375, + "eval_runtime": 46.6449, + "eval_samples_per_second": 18.866, + "eval_steps_per_second": 18.866, + "step": 3000 + }, + { + "epoch": 1.8383561643835615, + "grad_norm": 0.35732510685920715, + "learning_rate": 3.0183909630321865e-06, + "loss": 0.07381275296211243, + "step": 3020 + }, + { + "epoch": 1.8505327245053271, + "grad_norm": 0.3167949914932251, + "learning_rate": 2.9938651089442184e-06, + "loss": 0.07289664745330811, + "step": 3040 + }, + { + "epoch": 1.8627092846270927, + "grad_norm": 0.4893674850463867, + "learning_rate": 2.969289692453773e-06, + "loss": 0.07124295830726624, + "step": 3060 + }, + { + "epoch": 1.8748858447488583, + "grad_norm": 0.3017306327819824, + "learning_rate": 2.944667179855109e-06, + "loss": 0.08125877976417542, + "step": 3080 + }, + { + "epoch": 1.887062404870624, + "grad_norm": 0.3442900776863098, + "learning_rate": 2.920000042168871e-06, + "loss": 0.0724608838558197, + "step": 3100 + }, + { + "epoch": 1.8992389649923895, + "grad_norm": 0.27901750802993774, + "learning_rate": 2.8952907548941057e-06, + "loss": 0.07104775309562683, + "step": 3120 + }, + { + "epoch": 1.9114155251141551, + "grad_norm": 0.35838621854782104, + "learning_rate": 2.8705417977598277e-06, + "loss": 0.0677955150604248, + "step": 3140 + }, + { + "epoch": 1.9235920852359207, + "grad_norm": 0.3615752160549164, + "learning_rate": 2.8457556544761687e-06, + "loss": 0.07164496779441834, + "step": 3160 + }, + { + "epoch": 1.9357686453576863, + "grad_norm": 0.5117827653884888, + "learning_rate": 2.8209348124851187e-06, + "loss": 0.071807599067688, + "step": 3180 + }, + { + "epoch": 1.947945205479452, + "grad_norm": 0.36082401871681213, + "learning_rate": 2.7960817627108965e-06, + "loss": 0.095755535364151, + "step": 3200 + }, + { + "epoch": 1.9601217656012175, + "grad_norm": 0.27905145287513733, + "learning_rate": 2.77119899930997e-06, + "loss": 0.07055851817131042, + "step": 3220 + }, + { + "epoch": 1.9722983257229831, + "grad_norm": 0.5642575621604919, + "learning_rate": 2.7462890194207513e-06, + "loss": 0.07278798818588257, + "step": 3240 + }, + { + "epoch": 1.9844748858447487, + "grad_norm": 0.2286670207977295, + "learning_rate": 2.7213543229129956e-06, + "loss": 0.07153088450431824, + "step": 3260 + }, + { + "epoch": 1.9966514459665143, + "grad_norm": 0.3228706121444702, + "learning_rate": 2.6963974121369242e-06, + "loss": 0.07440360188484192, + "step": 3280 + }, + { + "epoch": 2.008523592085236, + "grad_norm": 0.21613839268684387, + "learning_rate": 2.671420791672093e-06, + "loss": 0.0517767608165741, + "step": 3300 + }, + { + "epoch": 2.0207001522070014, + "grad_norm": 0.19271881878376007, + "learning_rate": 2.646426968076052e-06, + "loss": 0.03812239170074463, + "step": 3320 + }, + { + "epoch": 2.032876712328767, + "grad_norm": 0.12863056361675262, + "learning_rate": 2.6214184496327865e-06, + "loss": 0.04107579588890076, + "step": 3340 + }, + { + "epoch": 2.0450532724505326, + "grad_norm": 0.26679477095603943, + "learning_rate": 2.5963977461010022e-06, + "loss": 0.04673115909099579, + "step": 3360 + }, + { + "epoch": 2.057229832572298, + "grad_norm": 0.2743483781814575, + "learning_rate": 2.5713673684622524e-06, + "loss": 0.03674449622631073, + "step": 3380 + }, + { + "epoch": 2.069406392694064, + "grad_norm": 0.16999909281730652, + "learning_rate": 2.546329828668949e-06, + "loss": 0.03422380387783051, + "step": 3400 + }, + { + "epoch": 2.0815829528158294, + "grad_norm": 0.2931291460990906, + "learning_rate": 2.5212876393922657e-06, + "loss": 0.035878732800483704, + "step": 3420 + }, + { + "epoch": 2.093759512937595, + "grad_norm": 0.34781521558761597, + "learning_rate": 2.496243313769986e-06, + "loss": 0.03577531576156616, + "step": 3440 + }, + { + "epoch": 2.1059360730593606, + "grad_norm": 0.472351998090744, + "learning_rate": 2.471199365154283e-06, + "loss": 0.04281675517559051, + "step": 3460 + }, + { + "epoch": 2.118112633181126, + "grad_norm": 0.38320988416671753, + "learning_rate": 2.4461583068595014e-06, + "loss": 0.042955422401428224, + "step": 3480 + }, + { + "epoch": 2.130289193302892, + "grad_norm": 0.12197626382112503, + "learning_rate": 2.421122651909918e-06, + "loss": 0.04432957172393799, + "step": 3500 + }, + { + "epoch": 2.130289193302892, + "eval_loss": 0.3123805522918701, + "eval_runtime": 46.5258, + "eval_samples_per_second": 18.914, + "eval_steps_per_second": 18.914, + "step": 3500 + }, + { + "epoch": 2.1424657534246574, + "grad_norm": 0.21291697025299072, + "learning_rate": 2.3960949127875556e-06, + "loss": 0.03356837034225464, + "step": 3520 + }, + { + "epoch": 2.154642313546423, + "grad_norm": 0.44130077958106995, + "learning_rate": 2.371077601180031e-06, + "loss": 0.036935809254646304, + "step": 3540 + }, + { + "epoch": 2.1668188736681886, + "grad_norm": 0.49069875478744507, + "learning_rate": 2.3460732277284994e-06, + "loss": 0.0395690768957138, + "step": 3560 + }, + { + "epoch": 2.178995433789954, + "grad_norm": 0.2824937701225281, + "learning_rate": 2.321084301775689e-06, + "loss": 0.044693085551261905, + "step": 3580 + }, + { + "epoch": 2.19117199391172, + "grad_norm": 0.35114097595214844, + "learning_rate": 2.29611333111408e-06, + "loss": 0.03243565857410431, + "step": 3600 + }, + { + "epoch": 2.2033485540334854, + "grad_norm": 0.47931790351867676, + "learning_rate": 2.271162821734225e-06, + "loss": 0.04325798749923706, + "step": 3620 + }, + { + "epoch": 2.215525114155251, + "grad_norm": 0.12716218829154968, + "learning_rate": 2.2462352775732653e-06, + "loss": 0.03856868743896484, + "step": 3640 + }, + { + "epoch": 2.2277016742770166, + "grad_norm": 0.3522437512874603, + "learning_rate": 2.221333200263637e-06, + "loss": 0.041602414846420285, + "step": 3660 + }, + { + "epoch": 2.239878234398782, + "grad_norm": 0.234590083360672, + "learning_rate": 2.1964590888820233e-06, + "loss": 0.04286134541034699, + "step": 3680 + }, + { + "epoch": 2.252054794520548, + "grad_norm": 0.2972453534603119, + "learning_rate": 2.1716154396985526e-06, + "loss": 0.041756758093833925, + "step": 3700 + }, + { + "epoch": 2.2642313546423134, + "grad_norm": 0.3236760199069977, + "learning_rate": 2.1468047459262882e-06, + "loss": 0.0359495222568512, + "step": 3720 + }, + { + "epoch": 2.276407914764079, + "grad_norm": 0.32255831360816956, + "learning_rate": 2.12202949747101e-06, + "loss": 0.04322676360607147, + "step": 3740 + }, + { + "epoch": 2.2885844748858446, + "grad_norm": 0.2753404378890991, + "learning_rate": 2.0972921806813468e-06, + "loss": 0.04191597998142242, + "step": 3760 + }, + { + "epoch": 2.30076103500761, + "grad_norm": 0.23182159662246704, + "learning_rate": 2.072595278099247e-06, + "loss": 0.041278204321861266, + "step": 3780 + }, + { + "epoch": 2.312937595129376, + "grad_norm": 0.25987961888313293, + "learning_rate": 2.047941268210849e-06, + "loss": 0.04312986135482788, + "step": 3800 + }, + { + "epoch": 2.3251141552511414, + "grad_norm": 0.3683331310749054, + "learning_rate": 2.0233326251977426e-06, + "loss": 0.04236046075820923, + "step": 3820 + }, + { + "epoch": 2.337290715372907, + "grad_norm": 0.2520082890987396, + "learning_rate": 1.9987718186886724e-06, + "loss": 0.04433901011943817, + "step": 3840 + }, + { + "epoch": 2.3494672754946726, + "grad_norm": 0.16200299561023712, + "learning_rate": 1.9742613135116986e-06, + "loss": 0.04127628207206726, + "step": 3860 + }, + { + "epoch": 2.361643835616438, + "grad_norm": 0.37064701318740845, + "learning_rate": 1.949803569446828e-06, + "loss": 0.04586326479911804, + "step": 3880 + }, + { + "epoch": 2.373820395738204, + "grad_norm": 0.416020005941391, + "learning_rate": 1.925401040979171e-06, + "loss": 0.03624185025691986, + "step": 3900 + }, + { + "epoch": 2.3859969558599694, + "grad_norm": 0.17131179571151733, + "learning_rate": 1.9010561770526076e-06, + "loss": 0.035064518451690674, + "step": 3920 + }, + { + "epoch": 2.398173515981735, + "grad_norm": 0.5188226103782654, + "learning_rate": 1.8767714208240312e-06, + "loss": 0.042050021886825564, + "step": 3940 + }, + { + "epoch": 2.4103500761035006, + "grad_norm": 0.3398009240627289, + "learning_rate": 1.852549209418154e-06, + "loss": 0.038166466355323794, + "step": 3960 + }, + { + "epoch": 2.422526636225266, + "grad_norm": 0.2541758418083191, + "learning_rate": 1.8283919736829332e-06, + "loss": 0.040885674953460696, + "step": 3980 + }, + { + "epoch": 2.434703196347032, + "grad_norm": 0.4074256122112274, + "learning_rate": 1.804302137945614e-06, + "loss": 0.040162667632102966, + "step": 4000 + }, + { + "epoch": 2.434703196347032, + "eval_loss": 0.31583163142204285, + "eval_runtime": 47.319, + "eval_samples_per_second": 18.597, + "eval_steps_per_second": 18.597, + "step": 4000 + }, + { + "epoch": 2.4468797564687974, + "grad_norm": 0.2724802494049072, + "learning_rate": 1.7802821197694426e-06, + "loss": 0.04170995056629181, + "step": 4020 + }, + { + "epoch": 2.459056316590563, + "grad_norm": 0.21376508474349976, + "learning_rate": 1.7563343297110375e-06, + "loss": 0.03834344446659088, + "step": 4040 + }, + { + "epoch": 2.4712328767123286, + "grad_norm": 0.2933412492275238, + "learning_rate": 1.732461171078486e-06, + "loss": 0.03928310573101044, + "step": 4060 + }, + { + "epoch": 2.483409436834094, + "grad_norm": 0.46805083751678467, + "learning_rate": 1.7086650396901489e-06, + "loss": 0.03358933925628662, + "step": 4080 + }, + { + "epoch": 2.49558599695586, + "grad_norm": 0.45667552947998047, + "learning_rate": 1.6849483236342322e-06, + "loss": 0.03547535240650177, + "step": 4100 + }, + { + "epoch": 2.5077625570776254, + "grad_norm": 0.24512450397014618, + "learning_rate": 1.6613134030291217e-06, + "loss": 0.03600102663040161, + "step": 4120 + }, + { + "epoch": 2.519939117199391, + "grad_norm": 0.20636337995529175, + "learning_rate": 1.6377626497845278e-06, + "loss": 0.04347077012062073, + "step": 4140 + }, + { + "epoch": 2.5321156773211566, + "grad_norm": 0.4584953486919403, + "learning_rate": 1.6142984273634505e-06, + "loss": 0.02908192276954651, + "step": 4160 + }, + { + "epoch": 2.544292237442922, + "grad_norm": 0.26193487644195557, + "learning_rate": 1.5909230905449846e-06, + "loss": 0.03611198365688324, + "step": 4180 + }, + { + "epoch": 2.556468797564688, + "grad_norm": 0.20813828706741333, + "learning_rate": 1.567638985188012e-06, + "loss": 0.03758668601512909, + "step": 4200 + }, + { + "epoch": 2.5686453576864534, + "grad_norm": 0.3395022749900818, + "learning_rate": 1.544448447995773e-06, + "loss": 0.033633843064308167, + "step": 4220 + }, + { + "epoch": 2.580821917808219, + "grad_norm": 0.1472434103488922, + "learning_rate": 1.52135380628137e-06, + "loss": 0.036797890067100526, + "step": 4240 + }, + { + "epoch": 2.5929984779299846, + "grad_norm": 0.5788060426712036, + "learning_rate": 1.498357377734201e-06, + "loss": 0.039166563749313356, + "step": 4260 + }, + { + "epoch": 2.60517503805175, + "grad_norm": 0.7623679637908936, + "learning_rate": 1.4754614701873703e-06, + "loss": 0.03717599511146545, + "step": 4280 + }, + { + "epoch": 2.6173515981735163, + "grad_norm": 0.16205403208732605, + "learning_rate": 1.4526683813860792e-06, + "loss": 0.03962793946266174, + "step": 4300 + }, + { + "epoch": 2.6295281582952814, + "grad_norm": 0.11986076086759567, + "learning_rate": 1.4299803987570396e-06, + "loss": 0.035475924611091614, + "step": 4320 + }, + { + "epoch": 2.6417047184170475, + "grad_norm": 0.15573006868362427, + "learning_rate": 1.4073997991789078e-06, + "loss": 0.03256964683532715, + "step": 4340 + }, + { + "epoch": 2.6538812785388126, + "grad_norm": 0.25151512026786804, + "learning_rate": 1.384928848753792e-06, + "loss": 0.03712306022644043, + "step": 4360 + }, + { + "epoch": 2.6660578386605787, + "grad_norm": 0.20408153533935547, + "learning_rate": 1.3625698025798322e-06, + "loss": 0.041410398483276364, + "step": 4380 + }, + { + "epoch": 2.678234398782344, + "grad_norm": 0.3156696856021881, + "learning_rate": 1.3403249045248907e-06, + "loss": 0.03158504366874695, + "step": 4400 + }, + { + "epoch": 2.69041095890411, + "grad_norm": 0.3835665285587311, + "learning_rate": 1.3181963870013604e-06, + "loss": 0.03525224924087524, + "step": 4420 + }, + { + "epoch": 2.702587519025875, + "grad_norm": 0.45423486828804016, + "learning_rate": 1.2961864707421345e-06, + "loss": 0.03239959478378296, + "step": 4440 + }, + { + "epoch": 2.714764079147641, + "grad_norm": 0.15982766449451447, + "learning_rate": 1.2742973645777394e-06, + "loss": 0.031032082438468934, + "step": 4460 + }, + { + "epoch": 2.726940639269406, + "grad_norm": 0.2770426869392395, + "learning_rate": 1.252531265214662e-06, + "loss": 0.030566230416297913, + "step": 4480 + }, + { + "epoch": 2.7391171993911723, + "grad_norm": 0.3693839907646179, + "learning_rate": 1.2308903570149048e-06, + "loss": 0.041362547874450685, + "step": 4500 + }, + { + "epoch": 2.7391171993911723, + "eval_loss": 0.3391737937927246, + "eval_runtime": 46.9306, + "eval_samples_per_second": 18.751, + "eval_steps_per_second": 18.751, + "step": 4500 + }, + { + "epoch": 2.7512937595129374, + "grad_norm": 0.3229275643825531, + "learning_rate": 1.2093768117767613e-06, + "loss": 0.0388390064239502, + "step": 4520 + }, + { + "epoch": 2.7634703196347035, + "grad_norm": 0.6786078214645386, + "learning_rate": 1.1879927885168733e-06, + "loss": 0.032555675506591795, + "step": 4540 + }, + { + "epoch": 2.7756468797564686, + "grad_norm": 0.32371029257774353, + "learning_rate": 1.1667404332535504e-06, + "loss": 0.03606459796428681, + "step": 4560 + }, + { + "epoch": 2.7878234398782347, + "grad_norm": 0.44066882133483887, + "learning_rate": 1.1456218787914128e-06, + "loss": 0.032086309790611264, + "step": 4580 + }, + { + "epoch": 2.8, + "grad_norm": 0.5005165338516235, + "learning_rate": 1.1246392445073438e-06, + "loss": 0.033362787961959836, + "step": 4600 + }, + { + "epoch": 2.812176560121766, + "grad_norm": 0.22586822509765625, + "learning_rate": 1.1037946361378027e-06, + "loss": 0.03638745844364166, + "step": 4620 + }, + { + "epoch": 2.824353120243531, + "grad_norm": 0.2905796766281128, + "learning_rate": 1.0830901455674977e-06, + "loss": 0.030933958292007447, + "step": 4640 + }, + { + "epoch": 2.836529680365297, + "grad_norm": 0.10796497762203217, + "learning_rate": 1.0625278506194538e-06, + "loss": 0.02879139482975006, + "step": 4660 + }, + { + "epoch": 2.8487062404870622, + "grad_norm": 0.2545916438102722, + "learning_rate": 1.04210981484649e-06, + "loss": 0.03345020413398743, + "step": 4680 + }, + { + "epoch": 2.8608828006088283, + "grad_norm": 0.2986568808555603, + "learning_rate": 1.0218380873241314e-06, + "loss": 0.02593054175376892, + "step": 4700 + }, + { + "epoch": 2.8730593607305934, + "grad_norm": 0.1755756139755249, + "learning_rate": 1.0017147024449674e-06, + "loss": 0.03906567096710205, + "step": 4720 + }, + { + "epoch": 2.8852359208523595, + "grad_norm": 0.18288888037204742, + "learning_rate": 9.81741679714493e-07, + "loss": 0.03371626436710358, + "step": 4740 + }, + { + "epoch": 2.8974124809741246, + "grad_norm": 0.368429958820343, + "learning_rate": 9.619210235484333e-07, + "loss": 0.03090968132019043, + "step": 4760 + }, + { + "epoch": 2.9095890410958907, + "grad_norm": 0.17118144035339355, + "learning_rate": 9.422547230715931e-07, + "loss": 0.0322105199098587, + "step": 4780 + }, + { + "epoch": 2.921765601217656, + "grad_norm": 0.41911277174949646, + "learning_rate": 9.227447519182353e-07, + "loss": 0.035210177302360535, + "step": 4800 + }, + { + "epoch": 2.933942161339422, + "grad_norm": 0.3521968722343445, + "learning_rate": 9.033930680340097e-07, + "loss": 0.026842504739761353, + "step": 4820 + }, + { + "epoch": 2.946118721461187, + "grad_norm": 0.20812013745307922, + "learning_rate": 8.842016134794682e-07, + "loss": 0.03439584076404571, + "step": 4840 + }, + { + "epoch": 2.958295281582953, + "grad_norm": 0.2796875834465027, + "learning_rate": 8.651723142351603e-07, + "loss": 0.04011322855949402, + "step": 4860 + }, + { + "epoch": 2.9704718417047182, + "grad_norm": 0.20960687100887299, + "learning_rate": 8.463070800083562e-07, + "loss": 0.03800423145294189, + "step": 4880 + }, + { + "epoch": 2.9826484018264843, + "grad_norm": 0.2586495876312256, + "learning_rate": 8.276078040413879e-07, + "loss": 0.03839131891727447, + "step": 4900 + }, + { + "epoch": 2.9948249619482494, + "grad_norm": 0.37137141823768616, + "learning_rate": 8.090763629216589e-07, + "loss": 0.02721840739250183, + "step": 4920 + }, + { + "epoch": 3.006697108066971, + "grad_norm": 0.3677407503128052, + "learning_rate": 7.907146163933102e-07, + "loss": 0.023991990089416503, + "step": 4940 + }, + { + "epoch": 3.0188736681887365, + "grad_norm": 0.11811063438653946, + "learning_rate": 7.725244071705871e-07, + "loss": 0.01451514959335327, + "step": 4960 + }, + { + "epoch": 3.031050228310502, + "grad_norm": 0.3449067771434784, + "learning_rate": 7.545075607529104e-07, + "loss": 0.014327619969844819, + "step": 4980 + }, + { + "epoch": 3.0432267884322677, + "grad_norm": 0.3237353265285492, + "learning_rate": 7.366658852416788e-07, + "loss": 0.017832010984420776, + "step": 5000 + }, + { + "epoch": 3.0432267884322677, + "eval_loss": 0.39115142822265625, + "eval_runtime": 47.6431, + "eval_samples_per_second": 18.471, + "eval_steps_per_second": 18.471, + "step": 5000 + }, + { + "epoch": 3.0554033485540333, + "grad_norm": 0.2226281613111496, + "learning_rate": 7.190011711588101e-07, + "loss": 0.011674411594867706, + "step": 5020 + }, + { + "epoch": 3.067579908675799, + "grad_norm": 0.08729376643896103, + "learning_rate": 7.015151912670562e-07, + "loss": 0.013690856099128724, + "step": 5040 + }, + { + "epoch": 3.0797564687975645, + "grad_norm": 0.2745465636253357, + "learning_rate": 6.842097003920903e-07, + "loss": 0.011978642642498016, + "step": 5060 + }, + { + "epoch": 3.09193302891933, + "grad_norm": 0.06905842572450638, + "learning_rate": 6.67086435246406e-07, + "loss": 0.013893941044807434, + "step": 5080 + }, + { + "epoch": 3.1041095890410957, + "grad_norm": 0.07840315997600555, + "learning_rate": 6.501471142550194e-07, + "loss": 0.009910025447607041, + "step": 5100 + }, + { + "epoch": 3.1162861491628613, + "grad_norm": 0.19672124087810516, + "learning_rate": 6.333934373830222e-07, + "loss": 0.008863755315542222, + "step": 5120 + }, + { + "epoch": 3.128462709284627, + "grad_norm": 0.37645605206489563, + "learning_rate": 6.168270859649761e-07, + "loss": 0.010502541810274124, + "step": 5140 + }, + { + "epoch": 3.1406392694063925, + "grad_norm": 0.2069159746170044, + "learning_rate": 6.004497225361786e-07, + "loss": 0.012096930295228958, + "step": 5160 + }, + { + "epoch": 3.1528158295281585, + "grad_norm": 0.2584017217159271, + "learning_rate": 5.842629906658226e-07, + "loss": 0.013278065621852875, + "step": 5180 + }, + { + "epoch": 3.1649923896499237, + "grad_norm": 0.2050527036190033, + "learning_rate": 5.682685147920481e-07, + "loss": 0.013548998534679413, + "step": 5200 + }, + { + "epoch": 3.1771689497716897, + "grad_norm": 0.13838107883930206, + "learning_rate": 5.524679000589256e-07, + "loss": 0.013736458122730255, + "step": 5220 + }, + { + "epoch": 3.189345509893455, + "grad_norm": 0.06378225982189178, + "learning_rate": 5.36862732155366e-07, + "loss": 0.013177134096622467, + "step": 5240 + }, + { + "epoch": 3.201522070015221, + "grad_norm": 0.27431613206863403, + "learning_rate": 5.214545771559879e-07, + "loss": 0.011971819400787353, + "step": 5260 + }, + { + "epoch": 3.213698630136986, + "grad_norm": 0.529901921749115, + "learning_rate": 5.062449813639528e-07, + "loss": 0.014422819018363953, + "step": 5280 + }, + { + "epoch": 3.225875190258752, + "grad_norm": 0.19417761266231537, + "learning_rate": 4.912354711557856e-07, + "loss": 0.010663678497076034, + "step": 5300 + }, + { + "epoch": 3.2380517503805173, + "grad_norm": 0.044735077768564224, + "learning_rate": 4.764275528281892e-07, + "loss": 0.011400717496871948, + "step": 5320 + }, + { + "epoch": 3.2502283105022833, + "grad_norm": 0.057179443538188934, + "learning_rate": 4.6182271244688355e-07, + "loss": 0.008456526696681977, + "step": 5340 + }, + { + "epoch": 3.2624048706240485, + "grad_norm": 0.10396906733512878, + "learning_rate": 4.4742241569746407e-07, + "loss": 0.014539115130901337, + "step": 5360 + }, + { + "epoch": 3.2745814307458145, + "grad_norm": 0.32904428243637085, + "learning_rate": 4.332281077383177e-07, + "loss": 0.017625690996646882, + "step": 5380 + }, + { + "epoch": 3.2867579908675797, + "grad_norm": 0.20823979377746582, + "learning_rate": 4.1924121305558563e-07, + "loss": 0.007641100138425827, + "step": 5400 + }, + { + "epoch": 3.2989345509893457, + "grad_norm": 0.25470009446144104, + "learning_rate": 4.054631353202121e-07, + "loss": 0.011799700558185577, + "step": 5420 + }, + { + "epoch": 3.311111111111111, + "grad_norm": 0.3968588709831238, + "learning_rate": 3.9189525724707634e-07, + "loss": 0.011455408483743667, + "step": 5440 + }, + { + "epoch": 3.323287671232877, + "grad_norm": 0.10818332433700562, + "learning_rate": 3.785389404562259e-07, + "loss": 0.012499115616083144, + "step": 5460 + }, + { + "epoch": 3.335464231354642, + "grad_norm": 0.1818460375070572, + "learning_rate": 3.653955253362351e-07, + "loss": 0.01148865669965744, + "step": 5480 + }, + { + "epoch": 3.347640791476408, + "grad_norm": 0.3504088521003723, + "learning_rate": 3.5246633090968205e-07, + "loss": 0.012819178402423859, + "step": 5500 + }, + { + "epoch": 3.347640791476408, + "eval_loss": 0.43404534459114075, + "eval_runtime": 46.4882, + "eval_samples_per_second": 18.93, + "eval_steps_per_second": 18.93, + "step": 5500 + }, + { + "epoch": 3.3598173515981733, + "grad_norm": 0.4551874101161957, + "learning_rate": 3.397526547007832e-07, + "loss": 0.013325585424900055, + "step": 5520 + }, + { + "epoch": 3.3719939117199393, + "grad_norm": 0.35187825560569763, + "learning_rate": 3.2725577260517396e-07, + "loss": 0.011712662875652313, + "step": 5540 + }, + { + "epoch": 3.3841704718417045, + "grad_norm": 0.6071529984474182, + "learning_rate": 3.14976938761867e-07, + "loss": 0.01580573171377182, + "step": 5560 + }, + { + "epoch": 3.3963470319634705, + "grad_norm": 0.18844422698020935, + "learning_rate": 3.029173854273909e-07, + "loss": 0.012312603741884231, + "step": 5580 + }, + { + "epoch": 3.4085235920852357, + "grad_norm": 0.13131535053253174, + "learning_rate": 2.910783228521269e-07, + "loss": 0.011797953397035599, + "step": 5600 + }, + { + "epoch": 3.4207001522070017, + "grad_norm": 0.4402364492416382, + "learning_rate": 2.794609391588504e-07, + "loss": 0.012182456254959107, + "step": 5620 + }, + { + "epoch": 3.432876712328767, + "grad_norm": 0.3497592508792877, + "learning_rate": 2.6806640022349897e-07, + "loss": 0.013599888980388641, + "step": 5640 + }, + { + "epoch": 3.445053272450533, + "grad_norm": 0.2316354215145111, + "learning_rate": 2.5689584955816497e-07, + "loss": 0.009272868931293487, + "step": 5660 + }, + { + "epoch": 3.457229832572298, + "grad_norm": 0.3858301341533661, + "learning_rate": 2.459504081963421e-07, + "loss": 0.008165979385375976, + "step": 5680 + }, + { + "epoch": 3.469406392694064, + "grad_norm": 0.14734333753585815, + "learning_rate": 2.3523117458041865e-07, + "loss": 0.009182130545377731, + "step": 5700 + }, + { + "epoch": 3.4815829528158293, + "grad_norm": 0.03280401974916458, + "learning_rate": 2.2473922445144485e-07, + "loss": 0.0107998326420784, + "step": 5720 + }, + { + "epoch": 3.4937595129375953, + "grad_norm": 0.1505511999130249, + "learning_rate": 2.144756107411733e-07, + "loss": 0.014469687640666962, + "step": 5740 + }, + { + "epoch": 3.5059360730593605, + "grad_norm": 0.2366904318332672, + "learning_rate": 2.0444136346639333e-07, + "loss": 0.0121701680123806, + "step": 5760 + }, + { + "epoch": 3.5181126331811265, + "grad_norm": 0.1468425989151001, + "learning_rate": 1.9463748962556096e-07, + "loss": 0.014668506383895875, + "step": 5780 + }, + { + "epoch": 3.5302891933028917, + "grad_norm": 0.14534050226211548, + "learning_rate": 1.8506497309773885e-07, + "loss": 0.010488402843475342, + "step": 5800 + }, + { + "epoch": 3.5424657534246577, + "grad_norm": 0.15501493215560913, + "learning_rate": 1.7572477454386257e-07, + "loss": 0.010667071491479874, + "step": 5820 + }, + { + "epoch": 3.554642313546423, + "grad_norm": 0.26535800099372864, + "learning_rate": 1.6661783131032726e-07, + "loss": 0.011079683899879456, + "step": 5840 + }, + { + "epoch": 3.566818873668189, + "grad_norm": 0.24390950798988342, + "learning_rate": 1.5774505733492263e-07, + "loss": 0.009308797866106033, + "step": 5860 + }, + { + "epoch": 3.578995433789954, + "grad_norm": 0.3409421443939209, + "learning_rate": 1.49107343055111e-07, + "loss": 0.012319787591695785, + "step": 5880 + }, + { + "epoch": 3.59117199391172, + "grad_norm": 0.4800300896167755, + "learning_rate": 1.407055553186701e-07, + "loss": 0.00843576118350029, + "step": 5900 + }, + { + "epoch": 3.6033485540334853, + "grad_norm": 0.11663182079792023, + "learning_rate": 1.3254053729669564e-07, + "loss": 0.00938587412238121, + "step": 5920 + }, + { + "epoch": 3.6155251141552514, + "grad_norm": 0.29512378573417664, + "learning_rate": 1.2461310839898656e-07, + "loss": 0.011934128403663636, + "step": 5940 + }, + { + "epoch": 3.6277016742770165, + "grad_norm": 0.2641650140285492, + "learning_rate": 1.169240641918104e-07, + "loss": 0.013170333206653595, + "step": 5960 + }, + { + "epoch": 3.6398782343987826, + "grad_norm": 0.47704726457595825, + "learning_rate": 1.0947417631806539e-07, + "loss": 0.014534834027290344, + "step": 5980 + }, + { + "epoch": 3.6520547945205477, + "grad_norm": 0.10114685446023941, + "learning_rate": 1.0226419241983865e-07, + "loss": 0.011021688580513, + "step": 6000 + }, + { + "epoch": 3.6520547945205477, + "eval_loss": 0.44063475728034973, + "eval_runtime": 46.1734, + "eval_samples_per_second": 19.059, + "eval_steps_per_second": 19.059, + "step": 6000 + }, + { + "epoch": 3.6642313546423138, + "grad_norm": 0.2619183659553528, + "learning_rate": 9.529483606337902e-08, + "loss": 0.010764393210411071, + "step": 6020 + }, + { + "epoch": 3.676407914764079, + "grad_norm": 0.05733129009604454, + "learning_rate": 8.856680666647882e-08, + "loss": 0.012128306180238723, + "step": 6040 + }, + { + "epoch": 3.688584474885845, + "grad_norm": 0.19483673572540283, + "learning_rate": 8.208077942828713e-08, + "loss": 0.011729901283979416, + "step": 6060 + }, + { + "epoch": 3.70076103500761, + "grad_norm": 0.2111903578042984, + "learning_rate": 7.58374052615457e-08, + "loss": 0.009119105339050294, + "step": 6080 + }, + { + "epoch": 3.712937595129376, + "grad_norm": 0.04995311424136162, + "learning_rate": 6.983731072726818e-08, + "loss": 0.017101363837718965, + "step": 6100 + }, + { + "epoch": 3.7251141552511413, + "grad_norm": 0.5839787125587463, + "learning_rate": 6.408109797186118e-08, + "loss": 0.012368235737085342, + "step": 6120 + }, + { + "epoch": 3.7372907153729074, + "grad_norm": 0.4685717523097992, + "learning_rate": 5.856934466669212e-08, + "loss": 0.008782628178596496, + "step": 6140 + }, + { + "epoch": 3.7494672754946725, + "grad_norm": 0.17204681038856506, + "learning_rate": 5.3302603950119994e-08, + "loss": 0.008994438499212266, + "step": 6160 + }, + { + "epoch": 3.7616438356164386, + "grad_norm": 0.07392167299985886, + "learning_rate": 4.8281404371981755e-08, + "loss": 0.011286454647779465, + "step": 6180 + }, + { + "epoch": 3.7738203957382037, + "grad_norm": 0.3728208541870117, + "learning_rate": 4.350624984055196e-08, + "loss": 0.011785905063152313, + "step": 6200 + }, + { + "epoch": 3.7859969558599698, + "grad_norm": 0.25468680262565613, + "learning_rate": 3.897761957196877e-08, + "loss": 0.013624191284179688, + "step": 6220 + }, + { + "epoch": 3.798173515981735, + "grad_norm": 0.09725204110145569, + "learning_rate": 3.469596804214548e-08, + "loss": 0.011700452119112015, + "step": 6240 + }, + { + "epoch": 3.810350076103501, + "grad_norm": 0.07126162946224213, + "learning_rate": 3.06617249411581e-08, + "loss": 0.011029987037181855, + "step": 6260 + }, + { + "epoch": 3.822526636225266, + "grad_norm": 0.08542267978191376, + "learning_rate": 2.687529513012488e-08, + "loss": 0.010965974628925323, + "step": 6280 + }, + { + "epoch": 3.834703196347032, + "grad_norm": 0.2627331018447876, + "learning_rate": 2.3337058600575722e-08, + "loss": 0.012378603965044022, + "step": 6300 + }, + { + "epoch": 3.8468797564687973, + "grad_norm": 0.19707690179347992, + "learning_rate": 2.0047370436317437e-08, + "loss": 0.011792077124118805, + "step": 6320 + }, + { + "epoch": 3.8590563165905634, + "grad_norm": 0.47547003626823425, + "learning_rate": 1.7006560777798608e-08, + "loss": 0.01145942509174347, + "step": 6340 + }, + { + "epoch": 3.8712328767123285, + "grad_norm": 0.3591565489768982, + "learning_rate": 1.421493478897945e-08, + "loss": 0.011088228970766067, + "step": 6360 + }, + { + "epoch": 3.8834094368340946, + "grad_norm": 0.20619548857212067, + "learning_rate": 1.1672772626704909e-08, + "loss": 0.010828402638435364, + "step": 6380 + }, + { + "epoch": 3.8955859969558597, + "grad_norm": 0.2822403311729431, + "learning_rate": 9.38032941258965e-09, + "loss": 0.01165580153465271, + "step": 6400 + }, + { + "epoch": 3.9077625570776258, + "grad_norm": 0.15682674944400787, + "learning_rate": 7.3378352074163215e-09, + "loss": 0.010783226788043975, + "step": 6420 + }, + { + "epoch": 3.919939117199391, + "grad_norm": 0.34253379702568054, + "learning_rate": 5.545494988045963e-09, + "loss": 0.011295531690120698, + "step": 6440 + }, + { + "epoch": 3.932115677321157, + "grad_norm": 0.27684664726257324, + "learning_rate": 4.003488626848073e-09, + "loss": 0.013613662123680115, + "step": 6460 + }, + { + "epoch": 3.944292237442922, + "grad_norm": 0.69688880443573, + "learning_rate": 2.7119708736486615e-09, + "loss": 0.011696261167526246, + "step": 6480 + }, + { + "epoch": 3.956468797564688, + "grad_norm": 0.2617769241333008, + "learning_rate": 1.6710713402015577e-09, + "loss": 0.010873865336179733, + "step": 6500 + }, + { + "epoch": 3.956468797564688, + "eval_loss": 0.441041499376297, + "eval_runtime": 46.1946, + "eval_samples_per_second": 19.05, + "eval_steps_per_second": 19.05, + "step": 6500 + }, + { + "epoch": 3.9686453576864533, + "grad_norm": 0.23670868575572968, + "learning_rate": 8.80894487179651e-10, + "loss": 0.012961818277835846, + "step": 6520 + }, + { + "epoch": 3.9808219178082194, + "grad_norm": 1.036125659942627, + "learning_rate": 3.4151961369188745e-10, + "loss": 0.01224210560321808, + "step": 6540 + }, + { + "epoch": 3.9929984779299845, + "grad_norm": 0.1431870311498642, + "learning_rate": 5.300084932574612e-11, + "loss": 0.010245455056428909, + "step": 6560 + }, + { + "epoch": 4.0, + "step": 6572, + "total_flos": 3.545203061907456e+17, + "train_loss": 0.08566030774240586, + "train_runtime": 13933.4985, + "train_samples_per_second": 3.772, + "train_steps_per_second": 0.472 + } + ], + "logging_steps": 20, + "max_steps": 6572, + "num_input_tokens_seen": 0, + "num_train_epochs": 4, + "save_steps": 500, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 3.545203061907456e+17, + "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..8a9cd5f --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c1c8f0bbe96b1bd3b42d30bb419fc0c08037dc014a4a8cea153b553983982dd1 +size 5585 diff --git a/training_eval_loss.png b/training_eval_loss.png new file mode 100644 index 0000000..0d78d95 Binary files /dev/null and b/training_eval_loss.png differ diff --git a/training_loss.png b/training_loss.png new file mode 100644 index 0000000..b187282 Binary files /dev/null and b/training_loss.png differ