commit 96a932fdaf0ff910ff09ed0604d9052170c46d86 Author: ModelHub XC Date: Tue Jul 7 19:43:19 2026 +0800 初始化项目,由ModelHub XC社区提供模型 Model: baban/QwenTranslate_English_Ukrainian 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..d07be5b --- /dev/null +++ b/README.md @@ -0,0 +1,63 @@ +--- +library_name: transformers +license: other +base_model: Qwen/Qwen2.5-3B-Instruct +tags: +- llama-factory +- full +- generated_from_trainer +model-index: +- name: MT_En_Ukrainian + results: [] +--- + + + +# MT_En_Ukrainian + +This model is a fine-tuned version of [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-Instruct) on the MT_En_Ukrainian dataset. +It achieves the following results on the evaluation set: +- Loss: 0.8353 +- Num Input Tokens Seen: 327416064 + +## Model description + +More information needed + +## Intended uses & limitations + +More information needed + +## Training and evaluation data + +More information needed + +## Training procedure + +### Training hyperparameters + +The following hyperparameters were used during training: +- learning_rate: 5e-05 +- train_batch_size: 8 +- eval_batch_size: 16 +- seed: 42 +- distributed_type: multi-GPU +- num_devices: 8 +- gradient_accumulation_steps: 16 +- total_train_batch_size: 1024 +- total_eval_batch_size: 128 +- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments +- lr_scheduler_type: inverse_sqrt +- num_epochs: 3.0 + +### Training results + + + +### Framework versions + +- Transformers 5.6.0 +- Pytorch 2.7.0+cu128 +- Datasets 3.6.0 +- Tokenizers 0.22.2 diff --git a/all_results.json b/all_results.json new file mode 100644 index 0000000..70702ab --- /dev/null +++ b/all_results.json @@ -0,0 +1,13 @@ +{ + "epoch": 3.0, + "eval_loss": 0.8352677822113037, + "eval_runtime": 1.2332, + "eval_samples_per_second": 808.48, + "eval_steps_per_second": 6.487, + "num_input_tokens_seen": 327416064, + "total_flos": 5.451032989129507e+18, + "train_loss": 0.7516615965776337, + "train_runtime": 14909.6648, + "train_samples_per_second": 201.212, + "train_steps_per_second": 0.197 +} \ No newline at end of file diff --git a/chat_template.jinja b/chat_template.jinja new file mode 100644 index 0000000..bdf7919 --- /dev/null +++ b/chat_template.jinja @@ -0,0 +1,54 @@ +{%- if tools %} + {{- '<|im_start|>system\n' }} + {%- if messages[0]['role'] == 'system' %} + {{- messages[0]['content'] }} + {%- else %} + {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }} + {%- endif %} + {{- "\n\n# 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' }} + {%- else %} + {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }} + {%- endif %} +{%- endif %} +{%- for message in messages %} + {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %} + {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }} + {%- elif message.role == "assistant" %} + {{- '<|im_start|>' + message.role }} + {%- if message.content %} + {{- '\n' + message.content }} + {%- endif %} + {%- for tool_call in message.tool_calls %} + {%- if tool_call.function is defined %} + {%- set tool_call = tool_call.function %} + {%- endif %} + {{- '\n\n{"name": "' }} + {{- tool_call.name }} + {{- '", "arguments": ' }} + {{- tool_call.arguments | tojson }} + {{- '}\n' }} + {%- endfor %} + {{- '<|im_end|>\n' }} + {%- elif message.role == "tool" %} + {%- if (loop.index0 == 0) 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' }} +{%- endif %} diff --git a/config.json b/config.json new file mode 100644 index 0000000..3722ad7 --- /dev/null +++ b/config.json @@ -0,0 +1,69 @@ +{ + "architectures": [ + "Qwen2ForCausalLM" + ], + "attention_dropout": 0.0, + "bos_token_id": null, + "dtype": "bfloat16", + "eos_token_id": 151645, + "hidden_act": "silu", + "hidden_size": 2048, + "initializer_range": 0.02, + "intermediate_size": 11008, + "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": 70, + "model_type": "qwen2", + "num_attention_heads": 16, + "num_hidden_layers": 36, + "num_key_value_heads": 2, + "pad_token_id": 151643, + "rms_norm_eps": 1e-06, + "rope_parameters": { + "rope_theta": 1000000.0, + "rope_type": "default" + }, + "sliding_window": null, + "tie_word_embeddings": true, + "transformers_version": "5.6.0", + "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..e45fdc2 --- /dev/null +++ b/eval_results.json @@ -0,0 +1,8 @@ +{ + "epoch": 3.0, + "eval_loss": 0.8352677822113037, + "eval_runtime": 1.2332, + "eval_samples_per_second": 808.48, + "eval_steps_per_second": 6.487, + "num_input_tokens_seen": 327416064 +} \ No newline at end of file diff --git a/generation_config.json b/generation_config.json new file mode 100644 index 0000000..c9eaa67 --- /dev/null +++ b/generation_config.json @@ -0,0 +1,13 @@ +{ + "do_sample": true, + "eos_token_id": [ + 151645, + 151643 + ], + "pad_token_id": 151643, + "repetition_penalty": 1.05, + "temperature": 0.7, + "top_k": 20, + "top_p": 0.8, + "transformers_version": "5.6.0" +} diff --git a/model.safetensors b/model.safetensors new file mode 100644 index 0000000..65a9aeb --- /dev/null +++ b/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8cf5b1fe7affba8d099079fd9ad1fcd48e1bc9b2b3026f47835070f4b0f62dd7 +size 6171927112 diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000..34510ff --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3fd169731d2cbde95e10bf356d66d5997fd885dd8dbb6fb4684da3f23b2585d8 +size 11421892 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000..6623da0 --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,31 @@ +{ + "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": false, + "local_files_only": false, + "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..58f03f7 --- /dev/null +++ b/train_results.json @@ -0,0 +1,9 @@ +{ + "epoch": 3.0, + "num_input_tokens_seen": 327416064, + "total_flos": 5.451032989129507e+18, + "train_loss": 0.7516615965776337, + "train_runtime": 14909.6648, + "train_samples_per_second": 201.212, + "train_steps_per_second": 0.197 +} \ No newline at end of file diff --git a/trainer_log.jsonl b/trainer_log.jsonl new file mode 100644 index 0000000..b03b3eb --- /dev/null +++ b/trainer_log.jsonl @@ -0,0 +1,295 @@ +{"current_steps": 10, "total_steps": 2931, "loss": 1.942220687866211, "lr": 4.9977515176118345e-05, "epoch": 0.01024, "percentage": 0.34, "elapsed_time": "0:00:54", "remaining_time": "4:27:44", "throughput": 20832.34, "total_tokens": 1145728} +{"current_steps": 20, "total_steps": 2931, "loss": 1.3728597640991211, "lr": 4.9952567580506e-05, "epoch": 0.02048, "percentage": 0.68, "elapsed_time": "0:01:44", "remaining_time": "4:14:12", "throughput": 21552.06, "total_tokens": 2258496} +{"current_steps": 30, "total_steps": 2931, "loss": 1.2849248886108398, "lr": 4.992765730738634e-05, "epoch": 0.03072, "percentage": 1.02, "elapsed_time": "0:02:37", "remaining_time": "4:13:21", "throughput": 21519.2, "total_tokens": 3382784} +{"current_steps": 40, "total_steps": 2931, "loss": 1.2331731796264649, "lr": 4.9902784263792476e-05, "epoch": 0.04096, "percentage": 1.36, "elapsed_time": "0:03:28", "remaining_time": "4:10:51", "throughput": 21662.67, "total_tokens": 4511424} +{"current_steps": 50, "total_steps": 2931, "loss": 1.2049736022949218, "lr": 4.987794835708133e-05, "epoch": 0.0512, "percentage": 1.71, "elapsed_time": "0:04:19", "remaining_time": "4:09:28", "throughput": 21708.19, "total_tokens": 5639296} +{"current_steps": 60, "total_steps": 2931, "loss": 1.1723523139953613, "lr": 4.985314949493234e-05, "epoch": 0.06144, "percentage": 2.05, "elapsed_time": "0:05:12", "remaining_time": "4:09:09", "throughput": 21668.63, "total_tokens": 6769920} +{"current_steps": 70, "total_steps": 2931, "loss": 1.1766769409179687, "lr": 4.982838758534584e-05, "epoch": 0.07168, "percentage": 2.39, "elapsed_time": "0:06:02", "remaining_time": "4:06:58", "throughput": 21757.46, "total_tokens": 7888320} +{"current_steps": 80, "total_steps": 2931, "loss": 1.1496356010437012, "lr": 4.980366253664179e-05, "epoch": 0.08192, "percentage": 2.73, "elapsed_time": "0:06:50", "remaining_time": "4:03:54", "throughput": 21862.35, "total_tokens": 8977472} +{"current_steps": 90, "total_steps": 2931, "loss": 1.137571430206299, "lr": 4.977897425745825e-05, "epoch": 0.09216, "percentage": 3.07, "elapsed_time": "0:07:40", "remaining_time": "4:02:09", "throughput": 21890.65, "total_tokens": 10075968} +{"current_steps": 100, "total_steps": 2931, "loss": 1.1251113891601563, "lr": 4.975432265674997e-05, "epoch": 0.1024, "percentage": 3.41, "elapsed_time": "0:08:33", "remaining_time": "4:02:04", "throughput": 21835.51, "total_tokens": 11202944} +{"current_steps": 110, "total_steps": 2931, "loss": 1.1137601852416992, "lr": 4.972970764378705e-05, "epoch": 0.11264, "percentage": 3.75, "elapsed_time": "0:09:23", "remaining_time": "4:00:49", "throughput": 21872.29, "total_tokens": 12323840} +{"current_steps": 120, "total_steps": 2931, "loss": 1.1156309127807618, "lr": 4.970512912815344e-05, "epoch": 0.12288, "percentage": 4.09, "elapsed_time": "0:10:11", "remaining_time": "3:58:41", "throughput": 21942.11, "total_tokens": 13415296} +{"current_steps": 130, "total_steps": 2931, "loss": 1.0775527954101562, "lr": 4.968058701974564e-05, "epoch": 0.13312, "percentage": 4.44, "elapsed_time": "0:11:01", "remaining_time": "3:57:26", "throughput": 21968.49, "total_tokens": 14525376} +{"current_steps": 140, "total_steps": 2931, "loss": 1.0942869186401367, "lr": 4.96560812287712e-05, "epoch": 0.14336, "percentage": 4.78, "elapsed_time": "0:11:53", "remaining_time": "3:56:59", "throughput": 21964.25, "total_tokens": 15666688} +{"current_steps": 150, "total_steps": 2931, "loss": 1.0775710105895997, "lr": 4.963161166574748e-05, "epoch": 0.1536, "percentage": 5.12, "elapsed_time": "0:12:44", "remaining_time": "3:56:17", "throughput": 21961.0, "total_tokens": 16793216} +{"current_steps": 160, "total_steps": 2931, "loss": 1.085141944885254, "lr": 4.960717824150013e-05, "epoch": 0.16384, "percentage": 5.46, "elapsed_time": "0:13:35", "remaining_time": "3:55:21", "throughput": 21969.77, "total_tokens": 17913856} +{"current_steps": 170, "total_steps": 2931, "loss": 1.0794735908508302, "lr": 4.9582780867161893e-05, "epoch": 0.17408, "percentage": 5.8, "elapsed_time": "0:14:27", "remaining_time": "3:54:45", "throughput": 21946.22, "total_tokens": 19033472} +{"current_steps": 180, "total_steps": 2931, "loss": 1.045860481262207, "lr": 4.955841945417105e-05, "epoch": 0.18432, "percentage": 6.14, "elapsed_time": "0:15:18", "remaining_time": "3:53:51", "throughput": 21944.12, "total_tokens": 20146688} +{"current_steps": 190, "total_steps": 2931, "loss": 1.0552400588989257, "lr": 4.953409391427024e-05, "epoch": 0.19456, "percentage": 6.48, "elapsed_time": "0:16:11", "remaining_time": "3:53:39", "throughput": 21911.1, "total_tokens": 21293184} +{"current_steps": 200, "total_steps": 2931, "loss": 1.0574710845947266, "lr": 4.950980415950502e-05, "epoch": 0.2048, "percentage": 6.82, "elapsed_time": "0:17:01", "remaining_time": "3:52:24", "throughput": 21940.16, "total_tokens": 22405824} +{"current_steps": 210, "total_steps": 2931, "loss": 1.047713565826416, "lr": 4.9485550102222575e-05, "epoch": 0.21504, "percentage": 7.16, "elapsed_time": "0:17:52", "remaining_time": "3:51:39", "throughput": 21931.52, "total_tokens": 23526720} +{"current_steps": 220, "total_steps": 2931, "loss": 1.037226676940918, "lr": 4.946133165507037e-05, "epoch": 0.22528, "percentage": 7.51, "elapsed_time": "0:18:44", "remaining_time": "3:50:56", "throughput": 21923.3, "total_tokens": 24651328} +{"current_steps": 230, "total_steps": 2931, "loss": 1.035720443725586, "lr": 4.943714873099483e-05, "epoch": 0.23552, "percentage": 7.85, "elapsed_time": "0:19:33", "remaining_time": "3:49:43", "throughput": 21941.83, "total_tokens": 25753088} +{"current_steps": 240, "total_steps": 2931, "loss": 1.0342046737670898, "lr": 4.9413001243240024e-05, "epoch": 0.24576, "percentage": 8.19, "elapsed_time": "0:20:26", "remaining_time": "3:49:09", "throughput": 21941.69, "total_tokens": 26905856} +{"current_steps": 250, "total_steps": 2931, "loss": 1.0358741760253907, "lr": 4.938888910534637e-05, "epoch": 0.256, "percentage": 8.53, "elapsed_time": "0:21:16", "remaining_time": "3:48:08", "throughput": 21958.46, "total_tokens": 28028864} +{"current_steps": 260, "total_steps": 2931, "loss": 1.038590431213379, "lr": 4.936481223114932e-05, "epoch": 0.26624, "percentage": 8.87, "elapsed_time": "0:22:07", "remaining_time": "3:47:14", "throughput": 21946.7, "total_tokens": 29126976} +{"current_steps": 270, "total_steps": 2931, "loss": 1.0328522682189942, "lr": 4.934077053477808e-05, "epoch": 0.27648, "percentage": 9.21, "elapsed_time": "0:22:58", "remaining_time": "3:46:30", "throughput": 21940.14, "total_tokens": 30254208} +{"current_steps": 280, "total_steps": 2931, "loss": 1.0223213195800782, "lr": 4.931676393065431e-05, "epoch": 0.28672, "percentage": 9.55, "elapsed_time": "0:23:52", "remaining_time": "3:45:58", "throughput": 21931.15, "total_tokens": 31405888} +{"current_steps": 290, "total_steps": 2931, "loss": 1.0172260284423829, "lr": 4.929279233349088e-05, "epoch": 0.29696, "percentage": 9.89, "elapsed_time": "0:24:41", "remaining_time": "3:44:52", "throughput": 21940.33, "total_tokens": 32506432} +{"current_steps": 300, "total_steps": 2931, "loss": 1.016135025024414, "lr": 4.926885565829051e-05, "epoch": 0.3072, "percentage": 10.24, "elapsed_time": "0:25:31", "remaining_time": "3:43:47", "throughput": 21952.12, "total_tokens": 33610176} +{"current_steps": 310, "total_steps": 2931, "loss": 1.0107959747314452, "lr": 4.924495382034461e-05, "epoch": 0.31744, "percentage": 10.58, "elapsed_time": "0:26:20", "remaining_time": "3:42:43", "throughput": 21959.37, "total_tokens": 34707264} +{"current_steps": 320, "total_steps": 2931, "loss": 1.002080535888672, "lr": 4.9221086735231975e-05, "epoch": 0.32768, "percentage": 10.92, "elapsed_time": "0:27:11", "remaining_time": "3:41:54", "throughput": 21952.55, "total_tokens": 35822272} +{"current_steps": 330, "total_steps": 2931, "loss": 1.0000602722167968, "lr": 4.919725431881751e-05, "epoch": 0.33792, "percentage": 11.26, "elapsed_time": "0:28:05", "remaining_time": "3:41:25", "throughput": 21926.15, "total_tokens": 36959680} +{"current_steps": 340, "total_steps": 2931, "loss": 1.0111896514892578, "lr": 4.917345648725101e-05, "epoch": 0.34816, "percentage": 11.6, "elapsed_time": "0:28:56", "remaining_time": "3:40:33", "throughput": 21934.83, "total_tokens": 38092160} +{"current_steps": 350, "total_steps": 2931, "loss": 1.010312271118164, "lr": 4.914969315696596e-05, "epoch": 0.3584, "percentage": 11.94, "elapsed_time": "0:29:45", "remaining_time": "3:39:23", "throughput": 21950.77, "total_tokens": 39182848} +{"current_steps": 360, "total_steps": 2931, "loss": 0.9806657791137695, "lr": 4.912596424467818e-05, "epoch": 0.36864, "percentage": 12.28, "elapsed_time": "0:30:38", "remaining_time": "3:38:51", "throughput": 21928.22, "total_tokens": 40318272} +{"current_steps": 370, "total_steps": 2931, "loss": 0.9908489227294922, "lr": 4.910226966738475e-05, "epoch": 0.37888, "percentage": 12.62, "elapsed_time": "0:31:28", "remaining_time": "3:37:53", "throughput": 21941.38, "total_tokens": 41443072} +{"current_steps": 380, "total_steps": 2931, "loss": 0.973599624633789, "lr": 4.9078609342362666e-05, "epoch": 0.38912, "percentage": 12.96, "elapsed_time": "0:32:24", "remaining_time": "3:37:31", "throughput": 21904.86, "total_tokens": 42587392} +{"current_steps": 390, "total_steps": 2931, "loss": 0.9856637954711914, "lr": 4.905498318716775e-05, "epoch": 0.39936, "percentage": 13.31, "elapsed_time": "0:33:14", "remaining_time": "3:36:36", "throughput": 21910.2, "total_tokens": 43704832} +{"current_steps": 400, "total_steps": 2931, "loss": 0.9691889762878418, "lr": 4.9031391119633295e-05, "epoch": 0.4096, "percentage": 13.65, "elapsed_time": "0:34:05", "remaining_time": "3:35:41", "throughput": 21918.61, "total_tokens": 44831360} +{"current_steps": 410, "total_steps": 2931, "loss": 0.9910012245178222, "lr": 4.9007833057869e-05, "epoch": 0.41984, "percentage": 13.99, "elapsed_time": "0:34:55", "remaining_time": "3:34:46", "throughput": 21921.06, "total_tokens": 45943168} +{"current_steps": 420, "total_steps": 2931, "loss": 0.9832890510559082, "lr": 4.898430892025967e-05, "epoch": 0.43008, "percentage": 14.33, "elapsed_time": "0:35:46", "remaining_time": "3:33:54", "throughput": 21922.18, "total_tokens": 47060288} +{"current_steps": 430, "total_steps": 2931, "loss": 0.9752557754516602, "lr": 4.896081862546415e-05, "epoch": 0.44032, "percentage": 14.67, "elapsed_time": "0:36:38", "remaining_time": "3:33:08", "throughput": 21914.52, "total_tokens": 48185472} +{"current_steps": 440, "total_steps": 2931, "loss": 0.9778125762939454, "lr": 4.8937362092414e-05, "epoch": 0.45056, "percentage": 15.01, "elapsed_time": "0:37:26", "remaining_time": "3:31:56", "throughput": 21929.47, "total_tokens": 49256448} +{"current_steps": 450, "total_steps": 2931, "loss": 0.9636163711547852, "lr": 4.891393924031244e-05, "epoch": 0.4608, "percentage": 15.35, "elapsed_time": "0:38:16", "remaining_time": "3:31:02", "throughput": 21934.2, "total_tokens": 50374592} +{"current_steps": 460, "total_steps": 2931, "loss": 0.9704037666320801, "lr": 4.8890549988633095e-05, "epoch": 0.47104, "percentage": 15.69, "elapsed_time": "0:39:06", "remaining_time": "3:30:06", "throughput": 21938.22, "total_tokens": 51485248} +{"current_steps": 470, "total_steps": 2931, "loss": 0.9702651977539063, "lr": 4.8867194257118907e-05, "epoch": 0.48128, "percentage": 16.04, "elapsed_time": "0:39:57", "remaining_time": "3:29:14", "throughput": 21939.31, "total_tokens": 52604160} +{"current_steps": 480, "total_steps": 2931, "loss": 0.9792324066162109, "lr": 4.884387196578093e-05, "epoch": 0.49152, "percentage": 16.38, "elapsed_time": "0:40:48", "remaining_time": "3:28:23", "throughput": 21934.6, "total_tokens": 53710144} +{"current_steps": 490, "total_steps": 2931, "loss": 0.9673162460327148, "lr": 4.882058303489718e-05, "epoch": 0.50176, "percentage": 16.72, "elapsed_time": "0:41:40", "remaining_time": "3:27:35", "throughput": 21929.53, "total_tokens": 54830272} +{"current_steps": 500, "total_steps": 2931, "loss": 0.9507923126220703, "lr": 4.8797327385011496e-05, "epoch": 0.512, "percentage": 17.06, "elapsed_time": "0:42:29", "remaining_time": "3:26:36", "throughput": 21936.2, "total_tokens": 55928384} +{"current_steps": 510, "total_steps": 2931, "loss": 0.9491652488708496, "lr": 4.8774104936932425e-05, "epoch": 0.52224, "percentage": 17.4, "elapsed_time": "0:43:17", "remaining_time": "3:25:31", "throughput": 21960.53, "total_tokens": 57048704} +{"current_steps": 520, "total_steps": 2931, "loss": 0.9639401435852051, "lr": 4.8750915611732076e-05, "epoch": 0.53248, "percentage": 17.74, "elapsed_time": "0:44:06", "remaining_time": "3:24:28", "throughput": 21972.38, "total_tokens": 58140160} +{"current_steps": 530, "total_steps": 2931, "loss": 0.9479743957519531, "lr": 4.8727759330744986e-05, "epoch": 0.54272, "percentage": 18.08, "elapsed_time": "0:44:58", "remaining_time": "3:23:44", "throughput": 21963.25, "total_tokens": 59267200} +{"current_steps": 540, "total_steps": 2931, "loss": 0.9524721145629883, "lr": 4.870463601556696e-05, "epoch": 0.55296, "percentage": 18.42, "elapsed_time": "0:45:50", "remaining_time": "3:22:58", "throughput": 21968.41, "total_tokens": 60421504} +{"current_steps": 550, "total_steps": 2931, "loss": 0.962583065032959, "lr": 4.8681545588054075e-05, "epoch": 0.5632, "percentage": 18.76, "elapsed_time": "0:46:38", "remaining_time": "3:21:52", "throughput": 21981.53, "total_tokens": 61505280} +{"current_steps": 560, "total_steps": 2931, "loss": 0.9441762924194336, "lr": 4.8658487970321404e-05, "epoch": 0.57344, "percentage": 19.11, "elapsed_time": "0:47:27", "remaining_time": "3:20:54", "throughput": 21990.72, "total_tokens": 62611136} +{"current_steps": 570, "total_steps": 2931, "loss": 0.9428766250610352, "lr": 4.863546308474209e-05, "epoch": 0.58368, "percentage": 19.45, "elapsed_time": "0:48:18", "remaining_time": "3:20:06", "throughput": 21987.17, "total_tokens": 63731712} +{"current_steps": 580, "total_steps": 2931, "loss": 0.9273331642150879, "lr": 4.86124708539461e-05, "epoch": 0.59392, "percentage": 19.79, "elapsed_time": "0:49:08", "remaining_time": "3:19:12", "throughput": 21985.05, "total_tokens": 64828608} +{"current_steps": 590, "total_steps": 2931, "loss": 0.9427030563354493, "lr": 4.8589511200819216e-05, "epoch": 0.60416, "percentage": 20.13, "elapsed_time": "0:49:56", "remaining_time": "3:18:10", "throughput": 21999.55, "total_tokens": 65924608} +{"current_steps": 600, "total_steps": 2931, "loss": 0.9379173278808594, "lr": 4.8566584048501926e-05, "epoch": 0.6144, "percentage": 20.47, "elapsed_time": "0:50:48", "remaining_time": "3:17:24", "throughput": 21990.29, "total_tokens": 67042624} +{"current_steps": 610, "total_steps": 2931, "loss": 0.9400325775146484, "lr": 4.854368932038835e-05, "epoch": 0.62464, "percentage": 20.81, "elapsed_time": "0:51:37", "remaining_time": "3:16:27", "throughput": 21998.53, "total_tokens": 68149376} +{"current_steps": 620, "total_steps": 2931, "loss": 0.9410287857055664, "lr": 4.8520826940125144e-05, "epoch": 0.63488, "percentage": 21.15, "elapsed_time": "0:52:29", "remaining_time": "3:15:40", "throughput": 21990.62, "total_tokens": 69264256} +{"current_steps": 630, "total_steps": 2931, "loss": 0.9263475418090821, "lr": 4.849799683161046e-05, "epoch": 0.64512, "percentage": 21.49, "elapsed_time": "0:53:22", "remaining_time": "3:14:56", "throughput": 21993.79, "total_tokens": 70435904} +{"current_steps": 640, "total_steps": 2931, "loss": 0.9329448699951172, "lr": 4.8475198918992835e-05, "epoch": 0.65536, "percentage": 21.84, "elapsed_time": "0:54:13", "remaining_time": "3:14:05", "throughput": 21995.11, "total_tokens": 71552896} +{"current_steps": 650, "total_steps": 2931, "loss": 0.9400642395019532, "lr": 4.845243312667023e-05, "epoch": 0.6656, "percentage": 22.18, "elapsed_time": "0:55:02", "remaining_time": "3:13:07", "throughput": 22000.14, "total_tokens": 72646400} +{"current_steps": 660, "total_steps": 2931, "loss": 0.9307914733886719, "lr": 4.842969937928884e-05, "epoch": 0.67584, "percentage": 22.52, "elapsed_time": "0:55:51", "remaining_time": "3:12:12", "throughput": 22008.4, "total_tokens": 73760192} +{"current_steps": 670, "total_steps": 2931, "loss": 0.9233476638793945, "lr": 4.840699760174217e-05, "epoch": 0.68608, "percentage": 22.86, "elapsed_time": "0:56:42", "remaining_time": "3:11:23", "throughput": 22006.03, "total_tokens": 74882496} +{"current_steps": 680, "total_steps": 2931, "loss": 0.9302739143371582, "lr": 4.8384327719169906e-05, "epoch": 0.69632, "percentage": 23.2, "elapsed_time": "0:57:35", "remaining_time": "3:10:38", "throughput": 21996.03, "total_tokens": 76004736} +{"current_steps": 690, "total_steps": 2931, "loss": 0.9305820465087891, "lr": 4.836168965695694e-05, "epoch": 0.70656, "percentage": 23.54, "elapsed_time": "0:58:23", "remaining_time": "3:09:38", "throughput": 22007.1, "total_tokens": 77100736} +{"current_steps": 700, "total_steps": 2931, "loss": 0.9203786849975586, "lr": 4.8339083340732304e-05, "epoch": 0.7168, "percentage": 23.88, "elapsed_time": "0:59:13", "remaining_time": "3:08:44", "throughput": 22013.66, "total_tokens": 78219136} +{"current_steps": 710, "total_steps": 2931, "loss": 0.9239031791687011, "lr": 4.8316508696368154e-05, "epoch": 0.72704, "percentage": 24.22, "elapsed_time": "1:00:02", "remaining_time": "3:07:47", "throughput": 22018.35, "total_tokens": 79310848} +{"current_steps": 720, "total_steps": 2931, "loss": 0.9024551391601563, "lr": 4.8293965649978714e-05, "epoch": 0.73728, "percentage": 24.56, "elapsed_time": "1:00:53", "remaining_time": "3:06:59", "throughput": 22012.68, "total_tokens": 80425728} +{"current_steps": 730, "total_steps": 2931, "loss": 0.9112434387207031, "lr": 4.8271454127919364e-05, "epoch": 0.74752, "percentage": 24.91, "elapsed_time": "1:01:39", "remaining_time": "3:05:55", "throughput": 22027.22, "total_tokens": 81496320} +{"current_steps": 740, "total_steps": 2931, "loss": 0.9149089813232422, "lr": 4.824897405678549e-05, "epoch": 0.75776, "percentage": 25.25, "elapsed_time": "1:02:33", "remaining_time": "3:05:14", "throughput": 22015.75, "total_tokens": 82645120} +{"current_steps": 750, "total_steps": 2931, "loss": 0.9063904762268067, "lr": 4.8226525363411576e-05, "epoch": 0.768, "percentage": 25.59, "elapsed_time": "1:03:26", "remaining_time": "3:04:29", "throughput": 22005.54, "total_tokens": 83764288} +{"current_steps": 760, "total_steps": 2931, "loss": 0.9081829071044922, "lr": 4.820410797487017e-05, "epoch": 0.77824, "percentage": 25.93, "elapsed_time": "1:04:19", "remaining_time": "3:03:45", "throughput": 21993.25, "total_tokens": 84887744} +{"current_steps": 770, "total_steps": 2931, "loss": 0.8958653450012207, "lr": 4.818172181847091e-05, "epoch": 0.78848, "percentage": 26.27, "elapsed_time": "1:05:11", "remaining_time": "3:02:58", "throughput": 21985.94, "total_tokens": 86004992} +{"current_steps": 780, "total_steps": 2931, "loss": 0.9148517608642578, "lr": 4.81593668217595e-05, "epoch": 0.79872, "percentage": 26.61, "elapsed_time": "1:06:01", "remaining_time": "3:02:05", "throughput": 21986.96, "total_tokens": 87105216} +{"current_steps": 790, "total_steps": 2931, "loss": 0.9078433990478516, "lr": 4.813704291251675e-05, "epoch": 0.80896, "percentage": 26.95, "elapsed_time": "1:06:53", "remaining_time": "3:01:17", "throughput": 21985.25, "total_tokens": 88239680} +{"current_steps": 800, "total_steps": 2931, "loss": 0.9124856948852539, "lr": 4.811475001875759e-05, "epoch": 0.8192, "percentage": 27.29, "elapsed_time": "1:07:42", "remaining_time": "3:00:22", "throughput": 21992.3, "total_tokens": 89349568} +{"current_steps": 810, "total_steps": 2931, "loss": 0.9061111450195313, "lr": 4.8092488068730105e-05, "epoch": 0.82944, "percentage": 27.64, "elapsed_time": "1:08:35", "remaining_time": "2:59:37", "throughput": 21981.29, "total_tokens": 90469312} +{"current_steps": 820, "total_steps": 2931, "loss": 0.8909177780151367, "lr": 4.807025699091452e-05, "epoch": 0.83968, "percentage": 27.98, "elapsed_time": "1:09:27", "remaining_time": "2:58:48", "throughput": 21977.14, "total_tokens": 91591424} +{"current_steps": 830, "total_steps": 2931, "loss": 0.8970900535583496, "lr": 4.8048056714022325e-05, "epoch": 0.84992, "percentage": 28.32, "elapsed_time": "1:10:19", "remaining_time": "2:57:59", "throughput": 21975.85, "total_tokens": 92718336} +{"current_steps": 840, "total_steps": 2931, "loss": 0.9085554122924805, "lr": 4.802588716699519e-05, "epoch": 0.86016, "percentage": 28.66, "elapsed_time": "1:11:08", "remaining_time": "2:57:06", "throughput": 21976.59, "total_tokens": 93814400} +{"current_steps": 850, "total_steps": 2931, "loss": 0.8963174819946289, "lr": 4.8003748279004156e-05, "epoch": 0.8704, "percentage": 29.0, "elapsed_time": "1:12:01", "remaining_time": "2:56:21", "throughput": 21966.58, "total_tokens": 94937408} +{"current_steps": 860, "total_steps": 2931, "loss": 0.9105907440185547, "lr": 4.798163997944854e-05, "epoch": 0.88064, "percentage": 29.34, "elapsed_time": "1:12:51", "remaining_time": "2:55:28", "throughput": 21969.72, "total_tokens": 96048512} +{"current_steps": 870, "total_steps": 2931, "loss": 0.877834415435791, "lr": 4.79595621979551e-05, "epoch": 0.89088, "percentage": 29.68, "elapsed_time": "1:13:45", "remaining_time": "2:54:44", "throughput": 21960.06, "total_tokens": 97193472} +{"current_steps": 880, "total_steps": 2931, "loss": 0.8991375923156738, "lr": 4.793751486437702e-05, "epoch": 0.90112, "percentage": 30.02, "elapsed_time": "1:14:36", "remaining_time": "2:53:53", "throughput": 21961.59, "total_tokens": 98317184} +{"current_steps": 890, "total_steps": 2931, "loss": 0.8886856079101563, "lr": 4.7915497908793064e-05, "epoch": 0.91136, "percentage": 30.37, "elapsed_time": "1:15:29", "remaining_time": "2:53:06", "throughput": 21958.45, "total_tokens": 99450688} +{"current_steps": 900, "total_steps": 2931, "loss": 0.902672004699707, "lr": 4.7893511261506516e-05, "epoch": 0.9216, "percentage": 30.71, "elapsed_time": "1:16:19", "remaining_time": "2:52:13", "throughput": 21961.06, "total_tokens": 100563840} +{"current_steps": 910, "total_steps": 2931, "loss": 0.8937115669250488, "lr": 4.787155485304435e-05, "epoch": 0.93184, "percentage": 31.05, "elapsed_time": "1:17:07", "remaining_time": "2:51:17", "throughput": 21966.22, "total_tokens": 101656640} +{"current_steps": 920, "total_steps": 2931, "loss": 0.9013708114624024, "lr": 4.784962861415629e-05, "epoch": 0.94208, "percentage": 31.39, "elapsed_time": "1:17:59", "remaining_time": "2:50:27", "throughput": 21966.1, "total_tokens": 102780288} +{"current_steps": 930, "total_steps": 2931, "loss": 0.8749137878417969, "lr": 4.7827732475813884e-05, "epoch": 0.95232, "percentage": 31.73, "elapsed_time": "1:18:52", "remaining_time": "2:49:42", "throughput": 21961.1, "total_tokens": 103933056} +{"current_steps": 940, "total_steps": 2931, "loss": 0.8961335182189941, "lr": 4.7805866369209576e-05, "epoch": 0.96256, "percentage": 32.07, "elapsed_time": "1:19:42", "remaining_time": "2:48:50", "throughput": 21961.88, "total_tokens": 105038016} +{"current_steps": 950, "total_steps": 2931, "loss": 0.874142074584961, "lr": 4.778403022575583e-05, "epoch": 0.9728, "percentage": 32.41, "elapsed_time": "1:20:36", "remaining_time": "2:48:04", "throughput": 21953.55, "total_tokens": 106174336} +{"current_steps": 960, "total_steps": 2931, "loss": 0.8942262649536132, "lr": 4.7762223977084195e-05, "epoch": 0.98304, "percentage": 32.75, "elapsed_time": "1:21:26", "remaining_time": "2:47:13", "throughput": 21952.29, "total_tokens": 107277440} +{"current_steps": 970, "total_steps": 2931, "loss": 0.8799090385437012, "lr": 4.774044755504444e-05, "epoch": 0.99328, "percentage": 33.09, "elapsed_time": "1:22:18", "remaining_time": "2:46:24", "throughput": 21948.26, "total_tokens": 108394496} +{"current_steps": 980, "total_steps": 2931, "loss": 0.8489143371582031, "lr": 4.7718700891703616e-05, "epoch": 1.003072, "percentage": 33.44, "elapsed_time": "1:23:07", "remaining_time": "2:45:28", "throughput": 21947.17, "total_tokens": 109453888} +{"current_steps": 990, "total_steps": 2931, "loss": 0.7443439483642578, "lr": 4.7696983919345215e-05, "epoch": 1.013312, "percentage": 33.78, "elapsed_time": "1:23:58", "remaining_time": "2:44:39", "throughput": 21945.12, "total_tokens": 110580736} +{"current_steps": 1000, "total_steps": 2931, "loss": 0.7479698181152343, "lr": 4.7675296570468216e-05, "epoch": 1.023552, "percentage": 34.12, "elapsed_time": "1:24:52", "remaining_time": "2:43:53", "throughput": 21939.05, "total_tokens": 111723136} +{"current_steps": 1010, "total_steps": 2931, "loss": 0.7372268676757813, "lr": 4.76536387777863e-05, "epoch": 1.033792, "percentage": 34.46, "elapsed_time": "1:25:44", "remaining_time": "2:43:04", "throughput": 21933.22, "total_tokens": 112836800} +{"current_steps": 1020, "total_steps": 2931, "loss": 0.7579574584960938, "lr": 4.7632010474226915e-05, "epoch": 1.044032, "percentage": 34.8, "elapsed_time": "1:26:35", "remaining_time": "2:42:14", "throughput": 21933.64, "total_tokens": 113961088} +{"current_steps": 1030, "total_steps": 2931, "loss": 0.7489605903625488, "lr": 4.761041159293035e-05, "epoch": 1.054272, "percentage": 35.14, "elapsed_time": "1:27:27", "remaining_time": "2:41:25", "throughput": 21932.13, "total_tokens": 115089984} +{"current_steps": 1040, "total_steps": 2931, "loss": 0.7482340812683106, "lr": 4.7588842067249e-05, "epoch": 1.064512, "percentage": 35.48, "elapsed_time": "1:28:19", "remaining_time": "2:40:36", "throughput": 21928.18, "total_tokens": 116217216} +{"current_steps": 1050, "total_steps": 2931, "loss": 0.7579518318176269, "lr": 4.756730183074637e-05, "epoch": 1.074752, "percentage": 35.82, "elapsed_time": "1:29:11", "remaining_time": "2:39:46", "throughput": 21929.78, "total_tokens": 117347648} +{"current_steps": 1060, "total_steps": 2931, "loss": 0.7438392639160156, "lr": 4.7545790817196314e-05, "epoch": 1.084992, "percentage": 36.17, "elapsed_time": "1:30:05", "remaining_time": "2:39:01", "throughput": 21920.76, "total_tokens": 118493632} +{"current_steps": 1070, "total_steps": 2931, "loss": 0.7573630332946777, "lr": 4.752430896058212e-05, "epoch": 1.095232, "percentage": 36.51, "elapsed_time": "1:30:55", "remaining_time": "2:38:07", "throughput": 21924.06, "total_tokens": 119595776} +{"current_steps": 1080, "total_steps": 2931, "loss": 0.7446182250976563, "lr": 4.750285619509567e-05, "epoch": 1.105472, "percentage": 36.85, "elapsed_time": "1:31:47", "remaining_time": "2:37:19", "throughput": 21919.83, "total_tokens": 120728576} +{"current_steps": 1090, "total_steps": 2931, "loss": 0.7489546775817871, "lr": 4.7481432455136644e-05, "epoch": 1.115712, "percentage": 37.19, "elapsed_time": "1:32:39", "remaining_time": "2:36:29", "throughput": 21919.68, "total_tokens": 121851968} +{"current_steps": 1100, "total_steps": 2931, "loss": 0.757651948928833, "lr": 4.7460037675311584e-05, "epoch": 1.125952, "percentage": 37.53, "elapsed_time": "1:33:28", "remaining_time": "2:35:35", "throughput": 21924.07, "total_tokens": 122954752} +{"current_steps": 1110, "total_steps": 2931, "loss": 0.7577178955078125, "lr": 4.7438671790433126e-05, "epoch": 1.136192, "percentage": 37.87, "elapsed_time": "1:34:17", "remaining_time": "2:34:41", "throughput": 21926.92, "total_tokens": 124052416} +{"current_steps": 1120, "total_steps": 2931, "loss": 0.7523440361022949, "lr": 4.741733473551915e-05, "epoch": 1.146432, "percentage": 38.21, "elapsed_time": "1:35:07", "remaining_time": "2:33:48", "throughput": 21930.32, "total_tokens": 125165248} +{"current_steps": 1130, "total_steps": 2931, "loss": 0.7425838470458984, "lr": 4.7396026445791966e-05, "epoch": 1.156672, "percentage": 38.55, "elapsed_time": "1:35:59", "remaining_time": "2:32:59", "throughput": 21926.7, "total_tokens": 126283840} +{"current_steps": 1140, "total_steps": 2931, "loss": 0.7449512004852294, "lr": 4.737474685667742e-05, "epoch": 1.166912, "percentage": 38.89, "elapsed_time": "1:36:51", "remaining_time": "2:32:10", "throughput": 21926.08, "total_tokens": 127421952} +{"current_steps": 1150, "total_steps": 2931, "loss": 0.761650562286377, "lr": 4.7353495903804165e-05, "epoch": 1.177152, "percentage": 39.24, "elapsed_time": "1:37:39", "remaining_time": "2:31:14", "throughput": 21932.17, "total_tokens": 128504576} +{"current_steps": 1160, "total_steps": 2931, "loss": 0.7624397277832031, "lr": 4.733227352300277e-05, "epoch": 1.187392, "percentage": 39.58, "elapsed_time": "1:38:28", "remaining_time": "2:30:21", "throughput": 21936.61, "total_tokens": 129617728} +{"current_steps": 1170, "total_steps": 2931, "loss": 0.756078052520752, "lr": 4.731107965030496e-05, "epoch": 1.197632, "percentage": 39.92, "elapsed_time": "1:39:19", "remaining_time": "2:29:29", "throughput": 21934.78, "total_tokens": 130713728} +{"current_steps": 1180, "total_steps": 2931, "loss": 0.7492989063262939, "lr": 4.728991422194278e-05, "epoch": 1.207872, "percentage": 40.26, "elapsed_time": "1:40:08", "remaining_time": "2:28:35", "throughput": 21936.77, "total_tokens": 131802368} +{"current_steps": 1190, "total_steps": 2931, "loss": 0.7438122749328613, "lr": 4.726877717434773e-05, "epoch": 1.218112, "percentage": 40.6, "elapsed_time": "1:40:57", "remaining_time": "2:27:42", "throughput": 21939.27, "total_tokens": 132905472} +{"current_steps": 1200, "total_steps": 2931, "loss": 0.7444162368774414, "lr": 4.724766844415013e-05, "epoch": 1.228352, "percentage": 40.94, "elapsed_time": "1:41:47", "remaining_time": "2:26:50", "throughput": 21943.71, "total_tokens": 134023296} +{"current_steps": 1210, "total_steps": 2931, "loss": 0.7356367588043213, "lr": 4.722658796817813e-05, "epoch": 1.238592, "percentage": 41.28, "elapsed_time": "1:42:40", "remaining_time": "2:26:01", "throughput": 21941.9, "total_tokens": 135165312} +{"current_steps": 1220, "total_steps": 2931, "loss": 0.7440855026245117, "lr": 4.7205535683457044e-05, "epoch": 1.248832, "percentage": 41.62, "elapsed_time": "1:43:34", "remaining_time": "2:25:15", "throughput": 21936.28, "total_tokens": 136319808} +{"current_steps": 1230, "total_steps": 2931, "loss": 0.7333660125732422, "lr": 4.7184511527208484e-05, "epoch": 1.259072, "percentage": 41.97, "elapsed_time": "1:44:24", "remaining_time": "2:24:23", "throughput": 21938.49, "total_tokens": 137441088} +{"current_steps": 1240, "total_steps": 2931, "loss": 0.7524197101593018, "lr": 4.7163515436849644e-05, "epoch": 1.269312, "percentage": 42.31, "elapsed_time": "1:45:13", "remaining_time": "2:23:30", "throughput": 21944.24, "total_tokens": 138549568} +{"current_steps": 1250, "total_steps": 2931, "loss": 0.7486214637756348, "lr": 4.714254734999245e-05, "epoch": 1.279552, "percentage": 42.65, "elapsed_time": "1:46:01", "remaining_time": "2:22:34", "throughput": 21950.48, "total_tokens": 139628416} +{"current_steps": 1260, "total_steps": 2931, "loss": 0.7422505378723144, "lr": 4.712160720444284e-05, "epoch": 1.289792, "percentage": 42.99, "elapsed_time": "1:46:52", "remaining_time": "2:21:43", "throughput": 21950.57, "total_tokens": 140751552} +{"current_steps": 1270, "total_steps": 2931, "loss": 0.7394045352935791, "lr": 4.710069493819992e-05, "epoch": 1.300032, "percentage": 43.33, "elapsed_time": "1:47:43", "remaining_time": "2:20:53", "throughput": 21950.97, "total_tokens": 141875392} +{"current_steps": 1280, "total_steps": 2931, "loss": 0.7524402618408204, "lr": 4.70798104894553e-05, "epoch": 1.3102719999999999, "percentage": 43.67, "elapsed_time": "1:48:32", "remaining_time": "2:20:00", "throughput": 21951.84, "total_tokens": 142968768} +{"current_steps": 1290, "total_steps": 2931, "loss": 0.73507719039917, "lr": 4.705895379659219e-05, "epoch": 1.320512, "percentage": 44.01, "elapsed_time": "1:49:23", "remaining_time": "2:19:09", "throughput": 21951.55, "total_tokens": 144088448} +{"current_steps": 1300, "total_steps": 2931, "loss": 0.744103193283081, "lr": 4.7038124798184766e-05, "epoch": 1.330752, "percentage": 44.35, "elapsed_time": "1:50:12", "remaining_time": "2:18:15", "throughput": 21954.12, "total_tokens": 145166528} +{"current_steps": 1310, "total_steps": 2931, "loss": 0.7421403884887695, "lr": 4.7017323432997304e-05, "epoch": 1.340992, "percentage": 44.69, "elapsed_time": "1:51:03", "remaining_time": "2:17:25", "throughput": 21951.31, "total_tokens": 146277888} +{"current_steps": 1320, "total_steps": 2931, "loss": 0.7422142505645752, "lr": 4.6996549639983506e-05, "epoch": 1.351232, "percentage": 45.04, "elapsed_time": "1:51:53", "remaining_time": "2:16:34", "throughput": 21954.52, "total_tokens": 147402112} +{"current_steps": 1330, "total_steps": 2931, "loss": 0.747988224029541, "lr": 4.697580335828569e-05, "epoch": 1.361472, "percentage": 45.38, "elapsed_time": "1:52:42", "remaining_time": "2:15:40", "throughput": 21958.44, "total_tokens": 148498816} +{"current_steps": 1340, "total_steps": 2931, "loss": 0.7415881633758545, "lr": 4.6955084527234076e-05, "epoch": 1.371712, "percentage": 45.72, "elapsed_time": "1:53:33", "remaining_time": "2:14:49", "throughput": 21960.88, "total_tokens": 149633024} +{"current_steps": 1350, "total_steps": 2931, "loss": 0.7278687953948975, "lr": 4.6934393086346034e-05, "epoch": 1.381952, "percentage": 46.06, "elapsed_time": "1:54:25", "remaining_time": "2:13:59", "throughput": 21957.7, "total_tokens": 150743424} +{"current_steps": 1360, "total_steps": 2931, "loss": 0.7370668411254883, "lr": 4.6913728975325324e-05, "epoch": 1.392192, "percentage": 46.4, "elapsed_time": "1:55:15", "remaining_time": "2:13:07", "throughput": 21961.68, "total_tokens": 151865536} +{"current_steps": 1370, "total_steps": 2931, "loss": 0.7397951126098633, "lr": 4.6893092134061393e-05, "epoch": 1.4024320000000001, "percentage": 46.74, "elapsed_time": "1:56:06", "remaining_time": "2:12:17", "throughput": 21962.84, "total_tokens": 152997504} +{"current_steps": 1380, "total_steps": 2931, "loss": 0.7369170188903809, "lr": 4.687248250262859e-05, "epoch": 1.412672, "percentage": 47.08, "elapsed_time": "1:56:57", "remaining_time": "2:11:27", "throughput": 21959.34, "total_tokens": 154098752} +{"current_steps": 1390, "total_steps": 2931, "loss": 0.7380311965942383, "lr": 4.685190002128548e-05, "epoch": 1.422912, "percentage": 47.42, "elapsed_time": "1:57:49", "remaining_time": "2:10:37", "throughput": 21956.97, "total_tokens": 155218880} +{"current_steps": 1400, "total_steps": 2931, "loss": 0.719688892364502, "lr": 4.6831344630474114e-05, "epoch": 1.433152, "percentage": 47.77, "elapsed_time": "1:58:41", "remaining_time": "2:09:48", "throughput": 21956.39, "total_tokens": 156365696} +{"current_steps": 1410, "total_steps": 2931, "loss": 0.7329507827758789, "lr": 4.6810816270819276e-05, "epoch": 1.443392, "percentage": 48.11, "elapsed_time": "1:59:31", "remaining_time": "2:08:56", "throughput": 21959.71, "total_tokens": 157494784} +{"current_steps": 1420, "total_steps": 2931, "loss": 0.7250077247619628, "lr": 4.679031488312777e-05, "epoch": 1.453632, "percentage": 48.45, "elapsed_time": "2:00:23", "remaining_time": "2:08:05", "throughput": 21962.12, "total_tokens": 158633216} +{"current_steps": 1430, "total_steps": 2931, "loss": 0.7379475116729737, "lr": 4.6769840408387717e-05, "epoch": 1.463872, "percentage": 48.79, "elapsed_time": "2:01:14", "remaining_time": "2:07:15", "throughput": 21960.57, "total_tokens": 159750208} +{"current_steps": 1440, "total_steps": 2931, "loss": 0.7261748313903809, "lr": 4.674939278776787e-05, "epoch": 1.4741119999999999, "percentage": 49.13, "elapsed_time": "2:02:04", "remaining_time": "2:06:23", "throughput": 21962.81, "total_tokens": 160865344} +{"current_steps": 1450, "total_steps": 2931, "loss": 0.718736219406128, "lr": 4.672897196261683e-05, "epoch": 1.484352, "percentage": 49.47, "elapsed_time": "2:02:56", "remaining_time": "2:05:34", "throughput": 21960.43, "total_tokens": 162000640} +{"current_steps": 1460, "total_steps": 2931, "loss": 0.7371679782867432, "lr": 4.670857787446238e-05, "epoch": 1.494592, "percentage": 49.81, "elapsed_time": "2:03:46", "remaining_time": "2:04:42", "throughput": 21965.72, "total_tokens": 163120704} +{"current_steps": 1470, "total_steps": 2931, "loss": 0.7367044448852539, "lr": 4.668821046501082e-05, "epoch": 1.504832, "percentage": 50.15, "elapsed_time": "2:04:34", "remaining_time": "2:03:48", "throughput": 21972.98, "total_tokens": 164228160} +{"current_steps": 1480, "total_steps": 2931, "loss": 0.7399471282958985, "lr": 4.6667869676146194e-05, "epoch": 1.515072, "percentage": 50.49, "elapsed_time": "2:05:25", "remaining_time": "2:02:58", "throughput": 21971.44, "total_tokens": 165347456} +{"current_steps": 1490, "total_steps": 2931, "loss": 0.7276147842407227, "lr": 4.6647555449929645e-05, "epoch": 1.525312, "percentage": 50.84, "elapsed_time": "2:06:15", "remaining_time": "2:02:06", "throughput": 21972.03, "total_tokens": 166452672} +{"current_steps": 1500, "total_steps": 2931, "loss": 0.7368722915649414, "lr": 4.662726772859869e-05, "epoch": 1.535552, "percentage": 51.18, "elapsed_time": "2:07:04", "remaining_time": "2:01:13", "throughput": 21975.3, "total_tokens": 167547520} +{"current_steps": 1510, "total_steps": 2931, "loss": 0.7236470222473145, "lr": 4.660700645456655e-05, "epoch": 1.545792, "percentage": 51.52, "elapsed_time": "2:07:54", "remaining_time": "2:00:22", "throughput": 21975.92, "total_tokens": 168664320} +{"current_steps": 1520, "total_steps": 2931, "loss": 0.7258675575256348, "lr": 4.658677157042149e-05, "epoch": 1.556032, "percentage": 51.86, "elapsed_time": "2:08:46", "remaining_time": "1:59:31", "throughput": 21974.48, "total_tokens": 169775168} +{"current_steps": 1530, "total_steps": 2931, "loss": 0.7226676940917969, "lr": 4.656656301892605e-05, "epoch": 1.566272, "percentage": 52.2, "elapsed_time": "2:09:37", "remaining_time": "1:58:41", "throughput": 21975.36, "total_tokens": 170910912} +{"current_steps": 1540, "total_steps": 2931, "loss": 0.7360187530517578, "lr": 4.6546380743016465e-05, "epoch": 1.5765120000000001, "percentage": 52.54, "elapsed_time": "2:10:24", "remaining_time": "1:57:47", "throughput": 21983.03, "total_tokens": 172011072} +{"current_steps": 1550, "total_steps": 2931, "loss": 0.7214366912841796, "lr": 4.652622468580193e-05, "epoch": 1.5867520000000002, "percentage": 52.88, "elapsed_time": "2:11:15", "remaining_time": "1:56:57", "throughput": 21983.34, "total_tokens": 173135616} +{"current_steps": 1560, "total_steps": 2931, "loss": 0.7311969757080078, "lr": 4.650609479056392e-05, "epoch": 1.596992, "percentage": 53.22, "elapsed_time": "2:12:03", "remaining_time": "1:56:03", "throughput": 21990.67, "total_tokens": 174239232} +{"current_steps": 1570, "total_steps": 2931, "loss": 0.7286062240600586, "lr": 4.648599100075556e-05, "epoch": 1.607232, "percentage": 53.57, "elapsed_time": "2:12:52", "remaining_time": "1:55:11", "throughput": 21992.28, "total_tokens": 175335168} +{"current_steps": 1580, "total_steps": 2931, "loss": 0.7191134929656983, "lr": 4.6465913260000945e-05, "epoch": 1.617472, "percentage": 53.91, "elapsed_time": "2:13:45", "remaining_time": "1:54:22", "throughput": 21988.21, "total_tokens": 176463680} +{"current_steps": 1590, "total_steps": 2931, "loss": 0.7198073387145996, "lr": 4.644586151209444e-05, "epoch": 1.627712, "percentage": 54.25, "elapsed_time": "2:14:36", "remaining_time": "1:53:31", "throughput": 21988.24, "total_tokens": 177578176} +{"current_steps": 1600, "total_steps": 2931, "loss": 0.7295777320861816, "lr": 4.6425835701000084e-05, "epoch": 1.6379519999999999, "percentage": 54.59, "elapsed_time": "2:15:24", "remaining_time": "1:52:38", "throughput": 21991.27, "total_tokens": 178670912} +{"current_steps": 1610, "total_steps": 2931, "loss": 0.7139708518981933, "lr": 4.640583577085084e-05, "epoch": 1.6481919999999999, "percentage": 54.93, "elapsed_time": "2:16:14", "remaining_time": "1:51:47", "throughput": 21992.63, "total_tokens": 179777600} +{"current_steps": 1620, "total_steps": 2931, "loss": 0.733104658126831, "lr": 4.638586166594806e-05, "epoch": 1.658432, "percentage": 55.27, "elapsed_time": "2:17:03", "remaining_time": "1:50:55", "throughput": 21993.85, "total_tokens": 180869504} +{"current_steps": 1630, "total_steps": 2931, "loss": 0.7275556564331055, "lr": 4.6365913330760726e-05, "epoch": 1.668672, "percentage": 55.61, "elapsed_time": "2:17:55", "remaining_time": "1:50:05", "throughput": 21993.93, "total_tokens": 182019072} +{"current_steps": 1640, "total_steps": 2931, "loss": 0.7281203269958496, "lr": 4.6345990709924855e-05, "epoch": 1.678912, "percentage": 55.95, "elapsed_time": "2:18:45", "remaining_time": "1:49:14", "throughput": 21994.98, "total_tokens": 183130112} +{"current_steps": 1650, "total_steps": 2931, "loss": 0.7167181968688965, "lr": 4.632609374824284e-05, "epoch": 1.689152, "percentage": 56.29, "elapsed_time": "2:19:39", "remaining_time": "1:48:25", "throughput": 21991.04, "total_tokens": 184265024} +{"current_steps": 1660, "total_steps": 2931, "loss": 0.7205727100372314, "lr": 4.630622239068285e-05, "epoch": 1.699392, "percentage": 56.64, "elapsed_time": "2:20:31", "remaining_time": "1:47:35", "throughput": 21989.63, "total_tokens": 185403968} +{"current_steps": 1670, "total_steps": 2931, "loss": 0.7173449039459229, "lr": 4.628637658237808e-05, "epoch": 1.709632, "percentage": 56.98, "elapsed_time": "2:21:22", "remaining_time": "1:46:45", "throughput": 21990.73, "total_tokens": 186535552} +{"current_steps": 1680, "total_steps": 2931, "loss": 0.7150672912597656, "lr": 4.626655626862625e-05, "epoch": 1.719872, "percentage": 57.32, "elapsed_time": "2:22:13", "remaining_time": "1:45:54", "throughput": 21990.36, "total_tokens": 187648448} +{"current_steps": 1690, "total_steps": 2931, "loss": 0.7177637577056885, "lr": 4.624676139488888e-05, "epoch": 1.730112, "percentage": 57.66, "elapsed_time": "2:23:02", "remaining_time": "1:45:02", "throughput": 21994.78, "total_tokens": 188770624} +{"current_steps": 1700, "total_steps": 2931, "loss": 0.7107383728027343, "lr": 4.6226991906790686e-05, "epoch": 1.7403520000000001, "percentage": 58.0, "elapsed_time": "2:23:55", "remaining_time": "1:44:13", "throughput": 21991.45, "total_tokens": 189905536} +{"current_steps": 1710, "total_steps": 2931, "loss": 0.7159247398376465, "lr": 4.620724775011897e-05, "epoch": 1.7505920000000001, "percentage": 58.34, "elapsed_time": "2:24:48", "remaining_time": "1:43:24", "throughput": 21987.88, "total_tokens": 191047360} +{"current_steps": 1720, "total_steps": 2931, "loss": 0.7231245517730713, "lr": 4.618752887082297e-05, "epoch": 1.760832, "percentage": 58.68, "elapsed_time": "2:25:39", "remaining_time": "1:42:33", "throughput": 21988.42, "total_tokens": 192176448} +{"current_steps": 1730, "total_steps": 2931, "loss": 0.7093376159667969, "lr": 4.616783521501325e-05, "epoch": 1.771072, "percentage": 59.02, "elapsed_time": "2:26:29", "remaining_time": "1:41:41", "throughput": 21989.71, "total_tokens": 193282496} +{"current_steps": 1740, "total_steps": 2931, "loss": 0.7214728832244873, "lr": 4.614816672896108e-05, "epoch": 1.781312, "percentage": 59.37, "elapsed_time": "2:27:18", "remaining_time": "1:40:50", "throughput": 21989.82, "total_tokens": 194367808} +{"current_steps": 1750, "total_steps": 2931, "loss": 0.7111657619476318, "lr": 4.612852335909782e-05, "epoch": 1.791552, "percentage": 59.71, "elapsed_time": "2:28:10", "remaining_time": "1:39:59", "throughput": 21987.15, "total_tokens": 195472320} +{"current_steps": 1760, "total_steps": 2931, "loss": 0.7161635398864746, "lr": 4.6108905052014323e-05, "epoch": 1.8017919999999998, "percentage": 60.05, "elapsed_time": "2:28:59", "remaining_time": "1:39:07", "throughput": 21988.33, "total_tokens": 196566912} +{"current_steps": 1770, "total_steps": 2931, "loss": 0.7039215087890625, "lr": 4.608931175446027e-05, "epoch": 1.8120319999999999, "percentage": 60.39, "elapsed_time": "2:29:51", "remaining_time": "1:38:17", "throughput": 21986.88, "total_tokens": 197692864} +{"current_steps": 1780, "total_steps": 2931, "loss": 0.7046633720397949, "lr": 4.606974341334367e-05, "epoch": 1.822272, "percentage": 60.73, "elapsed_time": "2:30:45", "remaining_time": "1:37:29", "throughput": 21982.75, "total_tokens": 198851136} +{"current_steps": 1790, "total_steps": 2931, "loss": 0.7108986854553223, "lr": 4.605019997573011e-05, "epoch": 1.832512, "percentage": 61.07, "elapsed_time": "2:31:38", "remaining_time": "1:36:39", "throughput": 21980.91, "total_tokens": 199991616} +{"current_steps": 1800, "total_steps": 2931, "loss": 0.7045407295227051, "lr": 4.603068138884229e-05, "epoch": 1.842752, "percentage": 61.41, "elapsed_time": "2:32:29", "remaining_time": "1:35:49", "throughput": 21981.79, "total_tokens": 201125312} +{"current_steps": 1810, "total_steps": 2931, "loss": 0.7162825584411621, "lr": 4.6011187600059345e-05, "epoch": 1.852992, "percentage": 61.75, "elapsed_time": "2:33:19", "remaining_time": "1:34:57", "throughput": 21982.81, "total_tokens": 202239936} +{"current_steps": 1820, "total_steps": 2931, "loss": 0.7169035911560059, "lr": 4.599171855691629e-05, "epoch": 1.863232, "percentage": 62.09, "elapsed_time": "2:34:07", "remaining_time": "1:34:04", "throughput": 21987.77, "total_tokens": 203321984} +{"current_steps": 1830, "total_steps": 2931, "loss": 0.6997568130493164, "lr": 4.597227420710335e-05, "epoch": 1.873472, "percentage": 62.44, "elapsed_time": "2:35:00", "remaining_time": "1:33:15", "throughput": 21982.33, "total_tokens": 204444032} +{"current_steps": 1840, "total_steps": 2931, "loss": 0.7147369861602784, "lr": 4.595285449846551e-05, "epoch": 1.883712, "percentage": 62.78, "elapsed_time": "2:35:50", "remaining_time": "1:32:24", "throughput": 21982.46, "total_tokens": 205551872} +{"current_steps": 1850, "total_steps": 2931, "loss": 0.700528621673584, "lr": 4.593345937900178e-05, "epoch": 1.893952, "percentage": 63.12, "elapsed_time": "2:36:42", "remaining_time": "1:31:33", "throughput": 21982.39, "total_tokens": 206680128} +{"current_steps": 1860, "total_steps": 2931, "loss": 0.707237434387207, "lr": 4.591408879686472e-05, "epoch": 1.904192, "percentage": 63.46, "elapsed_time": "2:37:34", "remaining_time": "1:30:43", "throughput": 21980.83, "total_tokens": 207813696} +{"current_steps": 1870, "total_steps": 2931, "loss": 0.7043401718139648, "lr": 4.5894742700359775e-05, "epoch": 1.9144320000000001, "percentage": 63.8, "elapsed_time": "2:38:22", "remaining_time": "1:29:51", "throughput": 21983.18, "total_tokens": 208895808} +{"current_steps": 1880, "total_steps": 2931, "loss": 0.7066624641418457, "lr": 4.587542103794477e-05, "epoch": 1.9246720000000002, "percentage": 64.14, "elapsed_time": "2:39:13", "remaining_time": "1:29:01", "throughput": 21983.38, "total_tokens": 210025600} +{"current_steps": 1890, "total_steps": 2931, "loss": 0.7068476676940918, "lr": 4.5856123758229247e-05, "epoch": 1.934912, "percentage": 64.48, "elapsed_time": "2:40:04", "remaining_time": "1:28:10", "throughput": 21983.22, "total_tokens": 211141184} +{"current_steps": 1900, "total_steps": 2931, "loss": 0.7084201812744141, "lr": 4.5836850809973993e-05, "epoch": 1.945152, "percentage": 64.82, "elapsed_time": "2:40:53", "remaining_time": "1:27:18", "throughput": 21986.03, "total_tokens": 212250112} +{"current_steps": 1910, "total_steps": 2931, "loss": 0.6999262809753418, "lr": 4.5817602142090385e-05, "epoch": 1.955392, "percentage": 65.17, "elapsed_time": "2:41:44", "remaining_time": "1:26:27", "throughput": 21985.87, "total_tokens": 213370240} +{"current_steps": 1920, "total_steps": 2931, "loss": 0.7026149749755859, "lr": 4.579837770363989e-05, "epoch": 1.965632, "percentage": 65.51, "elapsed_time": "2:42:37", "remaining_time": "1:25:38", "throughput": 21983.44, "total_tokens": 214511104} +{"current_steps": 1930, "total_steps": 2931, "loss": 0.7023416042327881, "lr": 4.57791774438334e-05, "epoch": 1.9758719999999999, "percentage": 65.85, "elapsed_time": "2:43:27", "remaining_time": "1:24:46", "throughput": 21987.28, "total_tokens": 215629504} +{"current_steps": 1940, "total_steps": 2931, "loss": 0.7092232704162598, "lr": 4.576000131203078e-05, "epoch": 1.9861119999999999, "percentage": 66.19, "elapsed_time": "2:44:18", "remaining_time": "1:23:56", "throughput": 21984.99, "total_tokens": 216746880} +{"current_steps": 1950, "total_steps": 2931, "loss": 0.6931307792663575, "lr": 4.574084925774023e-05, "epoch": 1.996352, "percentage": 66.53, "elapsed_time": "2:45:08", "remaining_time": "1:23:04", "throughput": 21988.05, "total_tokens": 217858496} +{"current_steps": 1960, "total_steps": 2931, "loss": 0.6160273551940918, "lr": 4.5721721230617795e-05, "epoch": 2.006144, "percentage": 66.87, "elapsed_time": "2:45:57", "remaining_time": "1:22:12", "throughput": 21985.65, "total_tokens": 218918208} +{"current_steps": 1970, "total_steps": 2931, "loss": 0.5476226806640625, "lr": 4.57026171804667e-05, "epoch": 2.016384, "percentage": 67.21, "elapsed_time": "2:46:45", "remaining_time": "1:21:20", "throughput": 21988.92, "total_tokens": 220009600} +{"current_steps": 1980, "total_steps": 2931, "loss": 0.5443649768829346, "lr": 4.568353705723692e-05, "epoch": 2.026624, "percentage": 67.55, "elapsed_time": "2:47:33", "remaining_time": "1:20:28", "throughput": 21992.8, "total_tokens": 221112448} +{"current_steps": 1990, "total_steps": 2931, "loss": 0.530079698562622, "lr": 4.566448081102455e-05, "epoch": 2.036864, "percentage": 67.89, "elapsed_time": "2:48:26", "remaining_time": "1:19:38", "throughput": 21990.03, "total_tokens": 222236352} +{"current_steps": 2000, "total_steps": 2931, "loss": 0.5314745426177978, "lr": 4.564544839207128e-05, "epoch": 2.047104, "percentage": 68.24, "elapsed_time": "2:49:17", "remaining_time": "1:18:48", "throughput": 21990.82, "total_tokens": 223371840} +{"current_steps": 2010, "total_steps": 2931, "loss": 0.536113691329956, "lr": 4.562643975076387e-05, "epoch": 2.057344, "percentage": 68.58, "elapsed_time": "2:50:10", "remaining_time": "1:17:58", "throughput": 21987.34, "total_tokens": 224505152} +{"current_steps": 2020, "total_steps": 2931, "loss": 0.537871265411377, "lr": 4.560745483763357e-05, "epoch": 2.067584, "percentage": 68.92, "elapsed_time": "2:51:00", "remaining_time": "1:17:07", "throughput": 21987.23, "total_tokens": 225609024} +{"current_steps": 2030, "total_steps": 2931, "loss": 0.5416299819946289, "lr": 4.5588493603355595e-05, "epoch": 2.077824, "percentage": 69.26, "elapsed_time": "2:51:50", "remaining_time": "1:16:16", "throughput": 21989.46, "total_tokens": 226721728} +{"current_steps": 2040, "total_steps": 2931, "loss": 0.5329193592071533, "lr": 4.556955599874859e-05, "epoch": 2.088064, "percentage": 69.6, "elapsed_time": "2:52:42", "remaining_time": "1:15:25", "throughput": 21987.73, "total_tokens": 227839680} +{"current_steps": 2050, "total_steps": 2931, "loss": 0.5448167800903321, "lr": 4.555064197477409e-05, "epoch": 2.098304, "percentage": 69.94, "elapsed_time": "2:53:31", "remaining_time": "1:14:34", "throughput": 21988.86, "total_tokens": 228945344} +{"current_steps": 2060, "total_steps": 2931, "loss": 0.5400659084320069, "lr": 4.5531751482536e-05, "epoch": 2.108544, "percentage": 70.28, "elapsed_time": "2:54:21", "remaining_time": "1:13:43", "throughput": 21991.04, "total_tokens": 230068608} +{"current_steps": 2070, "total_steps": 2931, "loss": 0.5294137954711914, "lr": 4.5512884473280024e-05, "epoch": 2.118784, "percentage": 70.62, "elapsed_time": "2:55:11", "remaining_time": "1:12:52", "throughput": 21990.79, "total_tokens": 231161728} +{"current_steps": 2080, "total_steps": 2931, "loss": 0.535043478012085, "lr": 4.549404089839322e-05, "epoch": 2.129024, "percentage": 70.97, "elapsed_time": "2:56:02", "remaining_time": "1:12:01", "throughput": 21991.68, "total_tokens": 232286656} +{"current_steps": 2090, "total_steps": 2931, "loss": 0.5406003952026367, "lr": 4.547522070940335e-05, "epoch": 2.139264, "percentage": 71.31, "elapsed_time": "2:56:54", "remaining_time": "1:11:11", "throughput": 21988.94, "total_tokens": 233403328} +{"current_steps": 2100, "total_steps": 2931, "loss": 0.5352214813232422, "lr": 4.545642385797848e-05, "epoch": 2.149504, "percentage": 71.65, "elapsed_time": "2:57:48", "remaining_time": "1:10:21", "throughput": 21985.18, "total_tokens": 234542400} +{"current_steps": 2110, "total_steps": 2931, "loss": 0.5299077987670898, "lr": 4.543765029592637e-05, "epoch": 2.159744, "percentage": 71.99, "elapsed_time": "2:58:38", "remaining_time": "1:09:30", "throughput": 21984.69, "total_tokens": 235650624} +{"current_steps": 2120, "total_steps": 2931, "loss": 0.5387727737426757, "lr": 4.541889997519403e-05, "epoch": 2.169984, "percentage": 72.33, "elapsed_time": "2:59:29", "remaining_time": "1:08:39", "throughput": 21985.68, "total_tokens": 236777600} +{"current_steps": 2130, "total_steps": 2931, "loss": 0.5402078628540039, "lr": 4.5400172847867095e-05, "epoch": 2.180224, "percentage": 72.67, "elapsed_time": "3:00:20", "remaining_time": "1:07:49", "throughput": 21985.28, "total_tokens": 237897728} +{"current_steps": 2140, "total_steps": 2931, "loss": 0.5417552947998047, "lr": 4.5381468866169466e-05, "epoch": 2.190464, "percentage": 73.01, "elapsed_time": "3:01:09", "remaining_time": "1:06:57", "throughput": 21988.15, "total_tokens": 238992704} +{"current_steps": 2150, "total_steps": 2931, "loss": 0.530327320098877, "lr": 4.5362787982462616e-05, "epoch": 2.200704, "percentage": 73.35, "elapsed_time": "3:02:02", "remaining_time": "1:06:07", "throughput": 21985.49, "total_tokens": 240146496} +{"current_steps": 2160, "total_steps": 2931, "loss": 0.5349247932434082, "lr": 4.5344130149245275e-05, "epoch": 2.210944, "percentage": 73.69, "elapsed_time": "3:02:52", "remaining_time": "1:05:16", "throughput": 21985.81, "total_tokens": 241241856} +{"current_steps": 2170, "total_steps": 2931, "loss": 0.5235861301422119, "lr": 4.5325495319152715e-05, "epoch": 2.221184, "percentage": 74.04, "elapsed_time": "3:03:45", "remaining_time": "1:04:26", "throughput": 21983.57, "total_tokens": 242377600} +{"current_steps": 2180, "total_steps": 2931, "loss": 0.5335872173309326, "lr": 4.530688344495644e-05, "epoch": 2.231424, "percentage": 74.38, "elapsed_time": "3:04:38", "remaining_time": "1:03:36", "throughput": 21982.6, "total_tokens": 243525248} +{"current_steps": 2190, "total_steps": 2931, "loss": 0.5255549430847168, "lr": 4.528829447956357e-05, "epoch": 2.241664, "percentage": 74.72, "elapsed_time": "3:05:29", "remaining_time": "1:02:45", "throughput": 21982.97, "total_tokens": 244661248} +{"current_steps": 2200, "total_steps": 2931, "loss": 0.5395485877990722, "lr": 4.526972837601633e-05, "epoch": 2.251904, "percentage": 75.06, "elapsed_time": "3:06:20", "remaining_time": "1:01:54", "throughput": 21983.41, "total_tokens": 245780800} +{"current_steps": 2210, "total_steps": 2931, "loss": 0.5427422523498535, "lr": 4.525118508749165e-05, "epoch": 2.262144, "percentage": 75.4, "elapsed_time": "3:07:09", "remaining_time": "1:01:03", "throughput": 21985.09, "total_tokens": 246870912} +{"current_steps": 2220, "total_steps": 2931, "loss": 0.5383429527282715, "lr": 4.5232664567300546e-05, "epoch": 2.272384, "percentage": 75.74, "elapsed_time": "3:07:58", "remaining_time": "1:00:12", "throughput": 21986.02, "total_tokens": 247963136} +{"current_steps": 2230, "total_steps": 2931, "loss": 0.5311496257781982, "lr": 4.521416676888773e-05, "epoch": 2.282624, "percentage": 76.08, "elapsed_time": "3:08:51", "remaining_time": "0:59:22", "throughput": 21982.96, "total_tokens": 249098240} +{"current_steps": 2240, "total_steps": 2931, "loss": 0.5386343479156495, "lr": 4.519569164583107e-05, "epoch": 2.292864, "percentage": 76.42, "elapsed_time": "3:09:42", "remaining_time": "0:58:31", "throughput": 21982.16, "total_tokens": 250219840} +{"current_steps": 2250, "total_steps": 2931, "loss": 0.5370828628540039, "lr": 4.517723915184109e-05, "epoch": 2.303104, "percentage": 76.77, "elapsed_time": "3:10:31", "remaining_time": "0:57:40", "throughput": 21983.91, "total_tokens": 251317952} +{"current_steps": 2260, "total_steps": 2931, "loss": 0.5292394638061524, "lr": 4.5158809240760506e-05, "epoch": 2.313344, "percentage": 77.11, "elapsed_time": "3:11:25", "remaining_time": "0:56:49", "throughput": 21981.95, "total_tokens": 252466688} +{"current_steps": 2270, "total_steps": 2931, "loss": 0.5335429191589356, "lr": 4.514040186656375e-05, "epoch": 2.323584, "percentage": 77.45, "elapsed_time": "3:12:18", "remaining_time": "0:55:59", "throughput": 21979.44, "total_tokens": 253612992} +{"current_steps": 2280, "total_steps": 2931, "loss": 0.5375387668609619, "lr": 4.512201698335644e-05, "epoch": 2.333824, "percentage": 77.79, "elapsed_time": "3:13:10", "remaining_time": "0:55:09", "throughput": 21978.6, "total_tokens": 254732480} +{"current_steps": 2290, "total_steps": 2931, "loss": 0.5277575492858887, "lr": 4.510365454537496e-05, "epoch": 2.344064, "percentage": 78.13, "elapsed_time": "3:14:00", "remaining_time": "0:54:18", "throughput": 21979.77, "total_tokens": 255856448} +{"current_steps": 2300, "total_steps": 2931, "loss": 0.523950719833374, "lr": 4.5085314506985945e-05, "epoch": 2.354304, "percentage": 78.47, "elapsed_time": "3:14:52", "remaining_time": "0:53:27", "throughput": 21979.49, "total_tokens": 256999168} +{"current_steps": 2310, "total_steps": 2931, "loss": 0.5421285629272461, "lr": 4.50669968226858e-05, "epoch": 2.364544, "percentage": 78.81, "elapsed_time": "3:15:45", "remaining_time": "0:52:37", "throughput": 21977.56, "total_tokens": 258137984} +{"current_steps": 2320, "total_steps": 2931, "loss": 0.5283915996551514, "lr": 4.504870144710027e-05, "epoch": 2.374784, "percentage": 79.15, "elapsed_time": "3:16:36", "remaining_time": "0:51:46", "throughput": 21978.1, "total_tokens": 259258496} +{"current_steps": 2330, "total_steps": 2931, "loss": 0.5356220245361328, "lr": 4.5030428334983884e-05, "epoch": 2.385024, "percentage": 79.5, "elapsed_time": "3:17:30", "remaining_time": "0:50:56", "throughput": 21973.81, "total_tokens": 260398080} +{"current_steps": 2340, "total_steps": 2931, "loss": 0.5268114089965821, "lr": 4.501217744121959e-05, "epoch": 2.395264, "percentage": 79.84, "elapsed_time": "3:18:23", "remaining_time": "0:50:06", "throughput": 21972.09, "total_tokens": 261541440} +{"current_steps": 2350, "total_steps": 2931, "loss": 0.5375988960266114, "lr": 4.499394872081821e-05, "epoch": 2.405504, "percentage": 80.18, "elapsed_time": "3:19:11", "remaining_time": "0:49:14", "throughput": 21974.51, "total_tokens": 262635264} +{"current_steps": 2360, "total_steps": 2931, "loss": 0.5376396656036377, "lr": 4.4975742128918e-05, "epoch": 2.415744, "percentage": 80.52, "elapsed_time": "3:20:00", "remaining_time": "0:48:23", "throughput": 21977.94, "total_tokens": 263740288} +{"current_steps": 2370, "total_steps": 2931, "loss": 0.5323172569274902, "lr": 4.495755762078418e-05, "epoch": 2.425984, "percentage": 80.86, "elapsed_time": "3:20:50", "remaining_time": "0:47:32", "throughput": 21979.4, "total_tokens": 264867968} +{"current_steps": 2380, "total_steps": 2931, "loss": 0.5391307830810547, "lr": 4.49393951518085e-05, "epoch": 2.436224, "percentage": 81.2, "elapsed_time": "3:21:39", "remaining_time": "0:46:41", "throughput": 21981.34, "total_tokens": 265965440} +{"current_steps": 2390, "total_steps": 2931, "loss": 0.5373844146728516, "lr": 4.4921254677508716e-05, "epoch": 2.446464, "percentage": 81.54, "elapsed_time": "3:22:29", "remaining_time": "0:45:50", "throughput": 21983.41, "total_tokens": 267080128} +{"current_steps": 2400, "total_steps": 2931, "loss": 0.5455498218536377, "lr": 4.490313615352821e-05, "epoch": 2.456704, "percentage": 81.88, "elapsed_time": "3:23:20", "remaining_time": "0:44:59", "throughput": 21982.19, "total_tokens": 268188096} +{"current_steps": 2410, "total_steps": 2931, "loss": 0.5363270759582519, "lr": 4.48850395356355e-05, "epoch": 2.466944, "percentage": 82.22, "elapsed_time": "3:24:14", "remaining_time": "0:44:09", "throughput": 21977.65, "total_tokens": 269321984} +{"current_steps": 2420, "total_steps": 2931, "loss": 0.5418241024017334, "lr": 4.486696477972375e-05, "epoch": 2.477184, "percentage": 82.57, "elapsed_time": "3:25:03", "remaining_time": "0:43:18", "throughput": 21978.59, "total_tokens": 270419776} +{"current_steps": 2430, "total_steps": 2931, "loss": 0.5297726631164551, "lr": 4.484891184181041e-05, "epoch": 2.487424, "percentage": 82.91, "elapsed_time": "3:25:54", "remaining_time": "0:42:27", "throughput": 21979.64, "total_tokens": 271548736} +{"current_steps": 2440, "total_steps": 2931, "loss": 0.5376968383789062, "lr": 4.483088067803662e-05, "epoch": 2.497664, "percentage": 83.25, "elapsed_time": "3:26:43", "remaining_time": "0:41:35", "throughput": 21982.63, "total_tokens": 272650816} +{"current_steps": 2450, "total_steps": 2931, "loss": 0.5285521507263183, "lr": 4.481287124466697e-05, "epoch": 2.507904, "percentage": 83.59, "elapsed_time": "3:27:31", "remaining_time": "0:40:44", "throughput": 21985.43, "total_tokens": 273744576} +{"current_steps": 2460, "total_steps": 2931, "loss": 0.5299195289611817, "lr": 4.479488349808885e-05, "epoch": 2.518144, "percentage": 83.93, "elapsed_time": "3:28:23", "remaining_time": "0:39:53", "throughput": 21982.03, "total_tokens": 274848640} +{"current_steps": 2470, "total_steps": 2931, "loss": 0.5259761810302734, "lr": 4.4776917394812114e-05, "epoch": 2.528384, "percentage": 84.27, "elapsed_time": "3:29:16", "remaining_time": "0:39:03", "throughput": 21979.02, "total_tokens": 275977600} +{"current_steps": 2480, "total_steps": 2931, "loss": 0.5346611022949219, "lr": 4.475897289146862e-05, "epoch": 2.538624, "percentage": 84.61, "elapsed_time": "3:30:07", "remaining_time": "0:38:12", "throughput": 21979.52, "total_tokens": 277115072} +{"current_steps": 2490, "total_steps": 2931, "loss": 0.5369565963745118, "lr": 4.4741049944811806e-05, "epoch": 2.548864, "percentage": 84.95, "elapsed_time": "3:30:55", "remaining_time": "0:37:21", "throughput": 21982.76, "total_tokens": 278210816} +{"current_steps": 2500, "total_steps": 2931, "loss": 0.5275493621826172, "lr": 4.472314851171621e-05, "epoch": 2.559104, "percentage": 85.3, "elapsed_time": "3:31:47", "remaining_time": "0:36:30", "throughput": 21982.52, "total_tokens": 279331968} +{"current_steps": 2510, "total_steps": 2931, "loss": 0.5220746040344239, "lr": 4.4705268549177084e-05, "epoch": 2.569344, "percentage": 85.64, "elapsed_time": "3:32:38", "remaining_time": "0:35:39", "throughput": 21982.04, "total_tokens": 280461056} +{"current_steps": 2520, "total_steps": 2931, "loss": 0.5226367950439453, "lr": 4.468741001430989e-05, "epoch": 2.579584, "percentage": 85.98, "elapsed_time": "3:33:29", "remaining_time": "0:34:49", "throughput": 21981.53, "total_tokens": 281572608} +{"current_steps": 2530, "total_steps": 2931, "loss": 0.515837574005127, "lr": 4.466957286434997e-05, "epoch": 2.589824, "percentage": 86.32, "elapsed_time": "3:34:25", "remaining_time": "0:33:59", "throughput": 21976.61, "total_tokens": 282737216} +{"current_steps": 2540, "total_steps": 2931, "loss": 0.5301560401916504, "lr": 4.4651757056652e-05, "epoch": 2.600064, "percentage": 86.66, "elapsed_time": "3:35:13", "remaining_time": "0:33:07", "throughput": 21979.4, "total_tokens": 283827008} +{"current_steps": 2550, "total_steps": 2931, "loss": 0.5253063201904297, "lr": 4.463396254868968e-05, "epoch": 2.610304, "percentage": 87.0, "elapsed_time": "3:36:07", "remaining_time": "0:32:17", "throughput": 21977.53, "total_tokens": 284983680} +{"current_steps": 2560, "total_steps": 2931, "loss": 0.5248900890350342, "lr": 4.461618929805519e-05, "epoch": 2.6205439999999998, "percentage": 87.34, "elapsed_time": "3:36:55", "remaining_time": "0:31:26", "throughput": 21979.0, "total_tokens": 286077504} +{"current_steps": 2570, "total_steps": 2931, "loss": 0.53463134765625, "lr": 4.459843726245888e-05, "epoch": 2.6307840000000002, "percentage": 87.68, "elapsed_time": "3:37:47", "remaining_time": "0:30:35", "throughput": 21979.37, "total_tokens": 287207168} +{"current_steps": 2580, "total_steps": 2931, "loss": 0.5340803146362305, "lr": 4.458070639972875e-05, "epoch": 2.641024, "percentage": 88.02, "elapsed_time": "3:38:35", "remaining_time": "0:29:44", "throughput": 21982.77, "total_tokens": 288307776} +{"current_steps": 2590, "total_steps": 2931, "loss": 0.5143898963928223, "lr": 4.456299666781007e-05, "epoch": 2.651264, "percentage": 88.37, "elapsed_time": "3:39:33", "remaining_time": "0:28:54", "throughput": 21975.89, "total_tokens": 289491392} +{"current_steps": 2600, "total_steps": 2931, "loss": 0.5279585838317871, "lr": 4.4545308024764984e-05, "epoch": 2.661504, "percentage": 88.71, "elapsed_time": "3:40:22", "remaining_time": "0:28:03", "throughput": 21977.36, "total_tokens": 290588288} +{"current_steps": 2610, "total_steps": 2931, "loss": 0.5243070602416993, "lr": 4.452764042877207e-05, "epoch": 2.671744, "percentage": 89.05, "elapsed_time": "3:41:14", "remaining_time": "0:27:12", "throughput": 21975.68, "total_tokens": 291706432} +{"current_steps": 2620, "total_steps": 2931, "loss": 0.525565767288208, "lr": 4.45099938381259e-05, "epoch": 2.681984, "percentage": 89.39, "elapsed_time": "3:42:04", "remaining_time": "0:26:21", "throughput": 21976.36, "total_tokens": 292825664} +{"current_steps": 2630, "total_steps": 2931, "loss": 0.5281280040740967, "lr": 4.449236821123667e-05, "epoch": 2.692224, "percentage": 89.73, "elapsed_time": "3:42:55", "remaining_time": "0:25:30", "throughput": 21975.31, "total_tokens": 293933504} +{"current_steps": 2640, "total_steps": 2931, "loss": 0.5201663017272949, "lr": 4.447476350662976e-05, "epoch": 2.702464, "percentage": 90.07, "elapsed_time": "3:43:44", "remaining_time": "0:24:39", "throughput": 21976.8, "total_tokens": 295024000} +{"current_steps": 2650, "total_steps": 2931, "loss": 0.5273939609527588, "lr": 4.4457179682945346e-05, "epoch": 2.712704, "percentage": 90.41, "elapsed_time": "3:44:33", "remaining_time": "0:23:48", "throughput": 21977.75, "total_tokens": 296126464} +{"current_steps": 2660, "total_steps": 2931, "loss": 0.5174345970153809, "lr": 4.443961669893798e-05, "epoch": 2.722944, "percentage": 90.75, "elapsed_time": "3:45:24", "remaining_time": "0:22:57", "throughput": 21979.45, "total_tokens": 297259712} +{"current_steps": 2670, "total_steps": 2931, "loss": 0.528529167175293, "lr": 4.4422074513476155e-05, "epoch": 2.733184, "percentage": 91.1, "elapsed_time": "3:46:14", "remaining_time": "0:22:06", "throughput": 21979.82, "total_tokens": 298367680} +{"current_steps": 2680, "total_steps": 2931, "loss": 0.5198683738708496, "lr": 4.4404553085541955e-05, "epoch": 2.743424, "percentage": 91.44, "elapsed_time": "3:47:04", "remaining_time": "0:21:16", "throughput": 21982.55, "total_tokens": 299497536} +{"current_steps": 2690, "total_steps": 2931, "loss": 0.5379956245422364, "lr": 4.438705237423063e-05, "epoch": 2.753664, "percentage": 91.78, "elapsed_time": "3:47:53", "remaining_time": "0:20:25", "throughput": 21984.43, "total_tokens": 300608512} +{"current_steps": 2700, "total_steps": 2931, "loss": 0.5276507377624512, "lr": 4.436957233875017e-05, "epoch": 2.763904, "percentage": 92.12, "elapsed_time": "3:48:42", "remaining_time": "0:19:34", "throughput": 21986.0, "total_tokens": 301702976} +{"current_steps": 2710, "total_steps": 2931, "loss": 0.5213168144226075, "lr": 4.4352112938420956e-05, "epoch": 2.774144, "percentage": 92.46, "elapsed_time": "3:49:35", "remaining_time": "0:18:43", "throughput": 21984.91, "total_tokens": 302851072} +{"current_steps": 2720, "total_steps": 2931, "loss": 0.5262949943542481, "lr": 4.433467413267529e-05, "epoch": 2.784384, "percentage": 92.8, "elapsed_time": "3:50:23", "remaining_time": "0:17:52", "throughput": 21987.16, "total_tokens": 303937408} +{"current_steps": 2730, "total_steps": 2931, "loss": 0.5267552852630615, "lr": 4.431725588105708e-05, "epoch": 2.7946239999999998, "percentage": 93.14, "elapsed_time": "3:51:16", "remaining_time": "0:17:01", "throughput": 21984.38, "total_tokens": 305070272} +{"current_steps": 2740, "total_steps": 2931, "loss": 0.5038492202758789, "lr": 4.4299858143221377e-05, "epoch": 2.8048640000000002, "percentage": 93.48, "elapsed_time": "3:52:07", "remaining_time": "0:16:10", "throughput": 21986.15, "total_tokens": 306205888} +{"current_steps": 2750, "total_steps": 2931, "loss": 0.5188664436340332, "lr": 4.4282480878934065e-05, "epoch": 2.815104, "percentage": 93.82, "elapsed_time": "3:52:58", "remaining_time": "0:15:20", "throughput": 21986.45, "total_tokens": 307332736} +{"current_steps": 2760, "total_steps": 2931, "loss": 0.5213587760925293, "lr": 4.4265124048071346e-05, "epoch": 2.825344, "percentage": 94.17, "elapsed_time": "3:53:48", "remaining_time": "0:14:29", "throughput": 21986.9, "total_tokens": 308433152} +{"current_steps": 2770, "total_steps": 2931, "loss": 0.5300797462463379, "lr": 4.4247787610619477e-05, "epoch": 2.835584, "percentage": 94.51, "elapsed_time": "3:54:36", "remaining_time": "0:13:38", "throughput": 21989.26, "total_tokens": 309529920} +{"current_steps": 2780, "total_steps": 2931, "loss": 0.5275647640228271, "lr": 4.42304715266743e-05, "epoch": 2.845824, "percentage": 94.85, "elapsed_time": "3:55:25", "remaining_time": "0:12:47", "throughput": 21989.29, "total_tokens": 310620032} +{"current_steps": 2790, "total_steps": 2931, "loss": 0.5151683807373046, "lr": 4.421317575644092e-05, "epoch": 2.856064, "percentage": 95.19, "elapsed_time": "3:56:17", "remaining_time": "0:11:56", "throughput": 21989.05, "total_tokens": 311744704} +{"current_steps": 2800, "total_steps": 2931, "loss": 0.5277010440826416, "lr": 4.419590026023325e-05, "epoch": 2.866304, "percentage": 95.53, "elapsed_time": "3:57:08", "remaining_time": "0:11:05", "throughput": 21988.09, "total_tokens": 312862848} +{"current_steps": 2810, "total_steps": 2931, "loss": 0.5130796432495117, "lr": 4.417864499847368e-05, "epoch": 2.876544, "percentage": 95.87, "elapsed_time": "3:57:57", "remaining_time": "0:10:14", "throughput": 21990.59, "total_tokens": 313974848} +{"current_steps": 2820, "total_steps": 2931, "loss": 0.5243385791778564, "lr": 4.4161409931692676e-05, "epoch": 2.886784, "percentage": 96.21, "elapsed_time": "3:58:45", "remaining_time": "0:09:23", "throughput": 21993.0, "total_tokens": 315060032} +{"current_steps": 2830, "total_steps": 2931, "loss": 0.5174403667449952, "lr": 4.414419502052841e-05, "epoch": 2.897024, "percentage": 96.55, "elapsed_time": "3:59:34", "remaining_time": "0:08:33", "throughput": 21995.36, "total_tokens": 316182976} +{"current_steps": 2840, "total_steps": 2931, "loss": 0.5114859580993653, "lr": 4.412700022572637e-05, "epoch": 2.907264, "percentage": 96.9, "elapsed_time": "4:00:23", "remaining_time": "0:07:42", "throughput": 21997.87, "total_tokens": 317286656} +{"current_steps": 2850, "total_steps": 2931, "loss": 0.5237509727478027, "lr": 4.410982550813902e-05, "epoch": 2.917504, "percentage": 97.24, "elapsed_time": "4:01:13", "remaining_time": "0:06:51", "throughput": 21997.3, "total_tokens": 318388160} +{"current_steps": 2860, "total_steps": 2931, "loss": 0.5249311447143554, "lr": 4.409267082872535e-05, "epoch": 2.927744, "percentage": 97.58, "elapsed_time": "4:02:03", "remaining_time": "0:06:00", "throughput": 21997.96, "total_tokens": 319492928} +{"current_steps": 2870, "total_steps": 2931, "loss": 0.515770959854126, "lr": 4.407553614855059e-05, "epoch": 2.937984, "percentage": 97.92, "elapsed_time": "4:02:57", "remaining_time": "0:05:09", "throughput": 21994.6, "total_tokens": 320622336} +{"current_steps": 2880, "total_steps": 2931, "loss": 0.5268994331359863, "lr": 4.405842142878579e-05, "epoch": 2.9482239999999997, "percentage": 98.26, "elapsed_time": "4:03:47", "remaining_time": "0:04:19", "throughput": 21995.82, "total_tokens": 321735488} +{"current_steps": 2890, "total_steps": 2931, "loss": 0.5111500740051269, "lr": 4.404132663070745e-05, "epoch": 2.958464, "percentage": 98.6, "elapsed_time": "4:04:38", "remaining_time": "0:03:28", "throughput": 21995.01, "total_tokens": 322857728} +{"current_steps": 2900, "total_steps": 2931, "loss": 0.5016695499420166, "lr": 4.402425171569716e-05, "epoch": 2.968704, "percentage": 98.94, "elapsed_time": "4:05:31", "remaining_time": "0:02:37", "throughput": 21993.58, "total_tokens": 323996736} +{"current_steps": 2910, "total_steps": 2931, "loss": 0.5131624221801758, "lr": 4.400719664524127e-05, "epoch": 2.9789440000000003, "percentage": 99.28, "elapsed_time": "4:06:23", "remaining_time": "0:01:46", "throughput": 21992.46, "total_tokens": 325130944} +{"current_steps": 2920, "total_steps": 2931, "loss": 0.5178108215332031, "lr": 4.399016138093044e-05, "epoch": 2.989184, "percentage": 99.62, "elapsed_time": "4:07:14", "remaining_time": "0:00:55", "throughput": 21991.46, "total_tokens": 326241984} +{"current_steps": 2930, "total_steps": 2931, "loss": 0.5120769500732422, "lr": 4.397314588445937e-05, "epoch": 2.999424, "percentage": 99.97, "elapsed_time": "4:08:05", "remaining_time": "0:00:05", "throughput": 21991.83, "total_tokens": 327359424} +{"current_steps": 2931, "total_steps": 2931, "eval_loss": 0.8352677822113037, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "4:08:09", "remaining_time": "0:00:00", "throughput": 21990.34, "total_tokens": 327416064} +{"current_steps": 2931, "total_steps": 2931, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "4:08:29", "remaining_time": "0:00:00", "throughput": 21959.99, "total_tokens": 327416064} diff --git a/trainer_state.json b/trainer_state.json new file mode 100644 index 0000000..6fcfb54 --- /dev/null +++ b/trainer_state.json @@ -0,0 +1,2983 @@ +{ + "best_global_step": 2931, + "best_metric": 0.8352677822113037, + "best_model_checkpoint": "/workspace/MT_En_Ukrainian/checkpoint-2931", + "epoch": 3.0, + "eval_steps": 5000, + "global_step": 2931, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.01024, + "grad_norm": 2.453125, + "learning_rate": 4.9977515176118345e-05, + "loss": 1.942220687866211, + "num_input_tokens_seen": 1145728, + "step": 10, + "train_runtime": 55.0006, + "train_tokens_per_second": 20831.189 + }, + { + "epoch": 0.02048, + "grad_norm": 1.859375, + "learning_rate": 4.9952567580506e-05, + "loss": 1.3728597640991211, + "num_input_tokens_seen": 2258496, + "step": 20, + "train_runtime": 104.7954, + "train_tokens_per_second": 21551.485 + }, + { + "epoch": 0.03072, + "grad_norm": 1.7734375, + "learning_rate": 4.992765730738634e-05, + "loss": 1.2849248886108398, + "num_input_tokens_seen": 3382784, + "step": 30, + "train_runtime": 157.2016, + "train_tokens_per_second": 21518.764 + }, + { + "epoch": 0.04096, + "grad_norm": 1.6953125, + "learning_rate": 4.9902784263792476e-05, + "loss": 1.2331731796264649, + "num_input_tokens_seen": 4511424, + "step": 40, + "train_runtime": 208.2613, + "train_tokens_per_second": 21662.324 + }, + { + "epoch": 0.0512, + "grad_norm": 1.703125, + "learning_rate": 4.987794835708133e-05, + "loss": 1.2049736022949218, + "num_input_tokens_seen": 5639296, + "step": 50, + "train_runtime": 259.7804, + "train_tokens_per_second": 21707.936 + }, + { + "epoch": 0.06144, + "grad_norm": 1.6328125, + "learning_rate": 4.985314949493234e-05, + "loss": 1.1723523139953613, + "num_input_tokens_seen": 6769920, + "step": 60, + "train_runtime": 312.433, + "train_tokens_per_second": 21668.387 + }, + { + "epoch": 0.07168, + "grad_norm": 1.6171875, + "learning_rate": 4.982838758534584e-05, + "loss": 1.1766769409179687, + "num_input_tokens_seen": 7888320, + "step": 70, + "train_runtime": 362.5602, + "train_tokens_per_second": 21757.267 + }, + { + "epoch": 0.08192, + "grad_norm": 1.71875, + "learning_rate": 4.980366253664179e-05, + "loss": 1.1496356010437012, + "num_input_tokens_seen": 8977472, + "step": 80, + "train_runtime": 410.6394, + "train_tokens_per_second": 21862.179 + }, + { + "epoch": 0.09216, + "grad_norm": 1.7578125, + "learning_rate": 4.977897425745825e-05, + "loss": 1.137571430206299, + "num_input_tokens_seen": 10075968, + "step": 90, + "train_runtime": 460.2896, + "train_tokens_per_second": 21890.499 + }, + { + "epoch": 0.1024, + "grad_norm": 1.7421875, + "learning_rate": 4.975432265674997e-05, + "loss": 1.1251113891601563, + "num_input_tokens_seen": 11202944, + "step": 100, + "train_runtime": 513.0638, + "train_tokens_per_second": 21835.384 + }, + { + "epoch": 0.11264, + "grad_norm": 1.8125, + "learning_rate": 4.972970764378705e-05, + "loss": 1.1137601852416992, + "num_input_tokens_seen": 12323840, + "step": 110, + "train_runtime": 563.4486, + "train_tokens_per_second": 21872.164 + }, + { + "epoch": 0.12288, + "grad_norm": 1.6875, + "learning_rate": 4.970512912815344e-05, + "loss": 1.1156309127807618, + "num_input_tokens_seen": 13415296, + "step": 120, + "train_runtime": 611.3984, + "train_tokens_per_second": 21941.987 + }, + { + "epoch": 0.13312, + "grad_norm": 1.625, + "learning_rate": 4.968058701974564e-05, + "loss": 1.0775527954101562, + "num_input_tokens_seen": 14525376, + "step": 130, + "train_runtime": 661.1947, + "train_tokens_per_second": 21968.378 + }, + { + "epoch": 0.14336, + "grad_norm": 1.5078125, + "learning_rate": 4.96560812287712e-05, + "loss": 1.0942869186401367, + "num_input_tokens_seen": 15666688, + "step": 140, + "train_runtime": 713.2845, + "train_tokens_per_second": 21964.151 + }, + { + "epoch": 0.1536, + "grad_norm": 1.6484375, + "learning_rate": 4.963161166574748e-05, + "loss": 1.0775710105895997, + "num_input_tokens_seen": 16793216, + "step": 150, + "train_runtime": 764.6867, + "train_tokens_per_second": 21960.91 + }, + { + "epoch": 0.16384, + "grad_norm": 1.609375, + "learning_rate": 4.960717824150013e-05, + "loss": 1.085141944885254, + "num_input_tokens_seen": 17913856, + "step": 160, + "train_runtime": 815.3899, + "train_tokens_per_second": 21969.68 + }, + { + "epoch": 0.17408, + "grad_norm": 1.6171875, + "learning_rate": 4.9582780867161893e-05, + "loss": 1.0794735908508302, + "num_input_tokens_seen": 19033472, + "step": 170, + "train_runtime": 867.281, + "train_tokens_per_second": 21946.141 + }, + { + "epoch": 0.18432, + "grad_norm": 1.6328125, + "learning_rate": 4.955841945417105e-05, + "loss": 1.045860481262207, + "num_input_tokens_seen": 20146688, + "step": 180, + "train_runtime": 918.094, + "train_tokens_per_second": 21944.036 + }, + { + "epoch": 0.19456, + "grad_norm": 1.6953125, + "learning_rate": 4.953409391427024e-05, + "loss": 1.0552400588989257, + "num_input_tokens_seen": 21293184, + "step": 190, + "train_runtime": 971.8022, + "train_tokens_per_second": 21911.027 + }, + { + "epoch": 0.2048, + "grad_norm": 1.6015625, + "learning_rate": 4.950980415950502e-05, + "loss": 1.0574710845947266, + "num_input_tokens_seen": 22405824, + "step": 200, + "train_runtime": 1021.2273, + "train_tokens_per_second": 21940.094 + }, + { + "epoch": 0.21504, + "grad_norm": 1.5078125, + "learning_rate": 4.9485550102222575e-05, + "loss": 1.047713565826416, + "num_input_tokens_seen": 23526720, + "step": 210, + "train_runtime": 1072.7388, + "train_tokens_per_second": 21931.452 + }, + { + "epoch": 0.22528, + "grad_norm": 1.546875, + "learning_rate": 4.946133165507037e-05, + "loss": 1.037226676940918, + "num_input_tokens_seen": 24651328, + "step": 220, + "train_runtime": 1124.4387, + "train_tokens_per_second": 21923.229 + }, + { + "epoch": 0.23552, + "grad_norm": 1.6171875, + "learning_rate": 4.943714873099483e-05, + "loss": 1.035720443725586, + "num_input_tokens_seen": 25753088, + "step": 230, + "train_runtime": 1173.7005, + "train_tokens_per_second": 21941.788 + }, + { + "epoch": 0.24576, + "grad_norm": 1.546875, + "learning_rate": 4.9413001243240024e-05, + "loss": 1.0342046737670898, + "num_input_tokens_seen": 26905856, + "step": 240, + "train_runtime": 1226.2471, + "train_tokens_per_second": 21941.627 + }, + { + "epoch": 0.256, + "grad_norm": 1.484375, + "learning_rate": 4.938888910534637e-05, + "loss": 1.0358741760253907, + "num_input_tokens_seen": 28028864, + "step": 250, + "train_runtime": 1276.4532, + "train_tokens_per_second": 21958.396 + }, + { + "epoch": 0.26624, + "grad_norm": 1.46875, + "learning_rate": 4.936481223114932e-05, + "loss": 1.038590431213379, + "num_input_tokens_seen": 29126976, + "step": 260, + "train_runtime": 1327.1722, + "train_tokens_per_second": 21946.643 + }, + { + "epoch": 0.27648, + "grad_norm": 1.59375, + "learning_rate": 4.934077053477808e-05, + "loss": 1.0328522682189942, + "num_input_tokens_seen": 30254208, + "step": 270, + "train_runtime": 1378.9466, + "train_tokens_per_second": 21940.087 + }, + { + "epoch": 0.28672, + "grad_norm": 1.59375, + "learning_rate": 4.931676393065431e-05, + "loss": 1.0223213195800782, + "num_input_tokens_seen": 31405888, + "step": 280, + "train_runtime": 1432.0252, + "train_tokens_per_second": 21931.1 + }, + { + "epoch": 0.29696, + "grad_norm": 1.4609375, + "learning_rate": 4.929279233349088e-05, + "loss": 1.0172260284423829, + "num_input_tokens_seen": 32506432, + "step": 290, + "train_runtime": 1481.5865, + "train_tokens_per_second": 21940.286 + }, + { + "epoch": 0.3072, + "grad_norm": 1.5625, + "learning_rate": 4.926885565829051e-05, + "loss": 1.016135025024414, + "num_input_tokens_seen": 33610176, + "step": 300, + "train_runtime": 1531.0711, + "train_tokens_per_second": 21952.067 + }, + { + "epoch": 0.31744, + "grad_norm": 1.4453125, + "learning_rate": 4.924495382034461e-05, + "loss": 1.0107959747314452, + "num_input_tokens_seen": 34707264, + "step": 310, + "train_runtime": 1580.5249, + "train_tokens_per_second": 21959.327 + }, + { + "epoch": 0.32768, + "grad_norm": 1.5078125, + "learning_rate": 4.9221086735231975e-05, + "loss": 1.002080535888672, + "num_input_tokens_seen": 35822272, + "step": 320, + "train_runtime": 1631.8076, + "train_tokens_per_second": 21952.51 + }, + { + "epoch": 0.33792, + "grad_norm": 1.5703125, + "learning_rate": 4.919725431881751e-05, + "loss": 1.0000602722167968, + "num_input_tokens_seen": 36959680, + "step": 330, + "train_runtime": 1685.6469, + "train_tokens_per_second": 21926.11 + }, + { + "epoch": 0.34816, + "grad_norm": 1.5390625, + "learning_rate": 4.917345648725101e-05, + "loss": 1.0111896514892578, + "num_input_tokens_seen": 38092160, + "step": 340, + "train_runtime": 1736.6097, + "train_tokens_per_second": 21934.785 + }, + { + "epoch": 0.3584, + "grad_norm": 1.4453125, + "learning_rate": 4.914969315696596e-05, + "loss": 1.010312271118164, + "num_input_tokens_seen": 39182848, + "step": 350, + "train_runtime": 1785.0368, + "train_tokens_per_second": 21950.723 + }, + { + "epoch": 0.36864, + "grad_norm": 1.5625, + "learning_rate": 4.912596424467818e-05, + "loss": 0.9806657791137695, + "num_input_tokens_seen": 40318272, + "step": 360, + "train_runtime": 1838.6514, + "train_tokens_per_second": 21928.176 + }, + { + "epoch": 0.37888, + "grad_norm": 1.390625, + "learning_rate": 4.910226966738475e-05, + "loss": 0.9908489227294922, + "num_input_tokens_seen": 41443072, + "step": 370, + "train_runtime": 1888.8124, + "train_tokens_per_second": 21941.338 + }, + { + "epoch": 0.38912, + "grad_norm": 1.421875, + "learning_rate": 4.9078609342362666e-05, + "loss": 0.973599624633789, + "num_input_tokens_seen": 42587392, + "step": 380, + "train_runtime": 1944.2016, + "train_tokens_per_second": 21904.823 + }, + { + "epoch": 0.39936, + "grad_norm": 1.5703125, + "learning_rate": 4.905498318716775e-05, + "loss": 0.9856637954711914, + "num_input_tokens_seen": 43704832, + "step": 390, + "train_runtime": 1994.7284, + "train_tokens_per_second": 21910.167 + }, + { + "epoch": 0.4096, + "grad_norm": 1.46875, + "learning_rate": 4.9031391119633295e-05, + "loss": 0.9691889762878418, + "num_input_tokens_seen": 44831360, + "step": 400, + "train_runtime": 2045.3596, + "train_tokens_per_second": 21918.571 + }, + { + "epoch": 0.41984, + "grad_norm": 1.546875, + "learning_rate": 4.9007833057869e-05, + "loss": 0.9910012245178222, + "num_input_tokens_seen": 45943168, + "step": 410, + "train_runtime": 2095.8494, + "train_tokens_per_second": 21921.025 + }, + { + "epoch": 0.43008, + "grad_norm": 1.5, + "learning_rate": 4.898430892025967e-05, + "loss": 0.9832890510559082, + "num_input_tokens_seen": 47060288, + "step": 420, + "train_runtime": 2146.7011, + "train_tokens_per_second": 21922.143 + }, + { + "epoch": 0.44032, + "grad_norm": 1.4453125, + "learning_rate": 4.896081862546415e-05, + "loss": 0.9752557754516602, + "num_input_tokens_seen": 48185472, + "step": 430, + "train_runtime": 2198.7956, + "train_tokens_per_second": 21914.485 + }, + { + "epoch": 0.45056, + "grad_norm": 1.453125, + "learning_rate": 4.8937362092414e-05, + "loss": 0.9778125762939454, + "num_input_tokens_seen": 49256448, + "step": 440, + "train_runtime": 2246.1336, + "train_tokens_per_second": 21929.438 + }, + { + "epoch": 0.4608, + "grad_norm": 1.4921875, + "learning_rate": 4.891393924031244e-05, + "loss": 0.9636163711547852, + "num_input_tokens_seen": 50374592, + "step": 450, + "train_runtime": 2296.6263, + "train_tokens_per_second": 21934.17 + }, + { + "epoch": 0.47104, + "grad_norm": 1.515625, + "learning_rate": 4.8890549988633095e-05, + "loss": 0.9704037666320801, + "num_input_tokens_seen": 51485248, + "step": 460, + "train_runtime": 2346.8324, + "train_tokens_per_second": 21938.187 + }, + { + "epoch": 0.48128, + "grad_norm": 1.453125, + "learning_rate": 4.8867194257118907e-05, + "loss": 0.9702651977539063, + "num_input_tokens_seen": 52604160, + "step": 470, + "train_runtime": 2397.7165, + "train_tokens_per_second": 21939.274 + }, + { + "epoch": 0.49152, + "grad_norm": 1.4609375, + "learning_rate": 4.884387196578093e-05, + "loss": 0.9792324066162109, + "num_input_tokens_seen": 53710144, + "step": 480, + "train_runtime": 2448.6532, + "train_tokens_per_second": 21934.566 + }, + { + "epoch": 0.50176, + "grad_norm": 1.4140625, + "learning_rate": 4.882058303489718e-05, + "loss": 0.9673162460327148, + "num_input_tokens_seen": 54830272, + "step": 490, + "train_runtime": 2500.2972, + "train_tokens_per_second": 21929.502 + }, + { + "epoch": 0.512, + "grad_norm": 1.5703125, + "learning_rate": 4.8797327385011496e-05, + "loss": 0.9507923126220703, + "num_input_tokens_seen": 55928384, + "step": 500, + "train_runtime": 2549.596, + "train_tokens_per_second": 21936.175 + }, + { + "epoch": 0.52224, + "grad_norm": 1.53125, + "learning_rate": 4.8774104936932425e-05, + "loss": 0.9491652488708496, + "num_input_tokens_seen": 57048704, + "step": 510, + "train_runtime": 2597.7871, + "train_tokens_per_second": 21960.5 + }, + { + "epoch": 0.53248, + "grad_norm": 1.4375, + "learning_rate": 4.8750915611732076e-05, + "loss": 0.9639401435852051, + "num_input_tokens_seen": 58140160, + "step": 520, + "train_runtime": 2646.0597, + "train_tokens_per_second": 21972.353 + }, + { + "epoch": 0.54272, + "grad_norm": 1.5625, + "learning_rate": 4.8727759330744986e-05, + "loss": 0.9479743957519531, + "num_input_tokens_seen": 59267200, + "step": 530, + "train_runtime": 2698.4747, + "train_tokens_per_second": 21963.222 + }, + { + "epoch": 0.55296, + "grad_norm": 1.375, + "learning_rate": 4.870463601556696e-05, + "loss": 0.9524721145629883, + "num_input_tokens_seen": 60421504, + "step": 540, + "train_runtime": 2750.3842, + "train_tokens_per_second": 21968.387 + }, + { + "epoch": 0.5632, + "grad_norm": 1.4140625, + "learning_rate": 4.8681545588054075e-05, + "loss": 0.962583065032959, + "num_input_tokens_seen": 61505280, + "step": 550, + "train_runtime": 2798.047, + "train_tokens_per_second": 21981.504 + }, + { + "epoch": 0.57344, + "grad_norm": 1.4296875, + "learning_rate": 4.8658487970321404e-05, + "loss": 0.9441762924194336, + "num_input_tokens_seen": 62611136, + "step": 560, + "train_runtime": 2847.1648, + "train_tokens_per_second": 21990.696 + }, + { + "epoch": 0.58368, + "grad_norm": 1.4453125, + "learning_rate": 4.863546308474209e-05, + "loss": 0.9428766250610352, + "num_input_tokens_seen": 63731712, + "step": 570, + "train_runtime": 2898.5894, + "train_tokens_per_second": 21987.147 + }, + { + "epoch": 0.59392, + "grad_norm": 1.4296875, + "learning_rate": 4.86124708539461e-05, + "loss": 0.9273331642150879, + "num_input_tokens_seen": 64828608, + "step": 580, + "train_runtime": 2948.7617, + "train_tokens_per_second": 21985.028 + }, + { + "epoch": 0.60416, + "grad_norm": 1.4453125, + "learning_rate": 4.8589511200819216e-05, + "loss": 0.9427030563354493, + "num_input_tokens_seen": 65924608, + "step": 590, + "train_runtime": 2996.6383, + "train_tokens_per_second": 21999.521 + }, + { + "epoch": 0.6144, + "grad_norm": 1.4375, + "learning_rate": 4.8566584048501926e-05, + "loss": 0.9379173278808594, + "num_input_tokens_seen": 67042624, + "step": 600, + "train_runtime": 3048.7405, + "train_tokens_per_second": 21990.269 + }, + { + "epoch": 0.62464, + "grad_norm": 1.34375, + "learning_rate": 4.854368932038835e-05, + "loss": 0.9400325775146484, + "num_input_tokens_seen": 68149376, + "step": 610, + "train_runtime": 3097.9096, + "train_tokens_per_second": 21998.503 + }, + { + "epoch": 0.63488, + "grad_norm": 1.5234375, + "learning_rate": 4.8520826940125144e-05, + "loss": 0.9410287857055664, + "num_input_tokens_seen": 69264256, + "step": 620, + "train_runtime": 3149.7216, + "train_tokens_per_second": 21990.596 + }, + { + "epoch": 0.64512, + "grad_norm": 1.40625, + "learning_rate": 4.849799683161046e-05, + "loss": 0.9263475418090821, + "num_input_tokens_seen": 70435904, + "step": 630, + "train_runtime": 3202.5382, + "train_tokens_per_second": 21993.775 + }, + { + "epoch": 0.65536, + "grad_norm": 1.4375, + "learning_rate": 4.8475198918992835e-05, + "loss": 0.9329448699951172, + "num_input_tokens_seen": 71552896, + "step": 640, + "train_runtime": 3253.1308, + "train_tokens_per_second": 21995.088 + }, + { + "epoch": 0.6656, + "grad_norm": 1.3203125, + "learning_rate": 4.845243312667023e-05, + "loss": 0.9400642395019532, + "num_input_tokens_seen": 72646400, + "step": 650, + "train_runtime": 3302.0912, + "train_tokens_per_second": 22000.119 + }, + { + "epoch": 0.67584, + "grad_norm": 1.4375, + "learning_rate": 4.842969937928884e-05, + "loss": 0.9307914733886719, + "num_input_tokens_seen": 73760192, + "step": 660, + "train_runtime": 3351.4596, + "train_tokens_per_second": 22008.379 + }, + { + "epoch": 0.68608, + "grad_norm": 1.421875, + "learning_rate": 4.840699760174217e-05, + "loss": 0.9233476638793945, + "num_input_tokens_seen": 74882496, + "step": 670, + "train_runtime": 3402.821, + "train_tokens_per_second": 22006.005 + }, + { + "epoch": 0.69632, + "grad_norm": 1.34375, + "learning_rate": 4.8384327719169906e-05, + "loss": 0.9302739143371582, + "num_input_tokens_seen": 76004736, + "step": 680, + "train_runtime": 3455.3873, + "train_tokens_per_second": 21996.011 + }, + { + "epoch": 0.70656, + "grad_norm": 1.40625, + "learning_rate": 4.836168965695694e-05, + "loss": 0.9305820465087891, + "num_input_tokens_seen": 77100736, + "step": 690, + "train_runtime": 3503.451, + "train_tokens_per_second": 22007.083 + }, + { + "epoch": 0.7168, + "grad_norm": 1.390625, + "learning_rate": 4.8339083340732304e-05, + "loss": 0.9203786849975586, + "num_input_tokens_seen": 78219136, + "step": 700, + "train_runtime": 3553.2121, + "train_tokens_per_second": 22013.641 + }, + { + "epoch": 0.72704, + "grad_norm": 1.484375, + "learning_rate": 4.8316508696368154e-05, + "loss": 0.9239031791687011, + "num_input_tokens_seen": 79310848, + "step": 710, + "train_runtime": 3602.0368, + "train_tokens_per_second": 22018.334 + }, + { + "epoch": 0.73728, + "grad_norm": 1.3828125, + "learning_rate": 4.8293965649978714e-05, + "loss": 0.9024551391601563, + "num_input_tokens_seen": 80425728, + "step": 720, + "train_runtime": 3653.6123, + "train_tokens_per_second": 22012.66 + }, + { + "epoch": 0.74752, + "grad_norm": 1.4453125, + "learning_rate": 4.8271454127919364e-05, + "loss": 0.9112434387207031, + "num_input_tokens_seen": 81496320, + "step": 730, + "train_runtime": 3699.8035, + "train_tokens_per_second": 22027.202 + }, + { + "epoch": 0.75776, + "grad_norm": 1.390625, + "learning_rate": 4.824897405678549e-05, + "loss": 0.9149089813232422, + "num_input_tokens_seen": 82645120, + "step": 740, + "train_runtime": 3753.9121, + "train_tokens_per_second": 22015.731 + }, + { + "epoch": 0.768, + "grad_norm": 1.359375, + "learning_rate": 4.8226525363411576e-05, + "loss": 0.9063904762268067, + "num_input_tokens_seen": 83764288, + "step": 750, + "train_runtime": 3806.5124, + "train_tokens_per_second": 22005.521 + }, + { + "epoch": 0.77824, + "grad_norm": 1.453125, + "learning_rate": 4.820410797487017e-05, + "loss": 0.9081829071044922, + "num_input_tokens_seen": 84887744, + "step": 760, + "train_runtime": 3859.7215, + "train_tokens_per_second": 21993.23 + }, + { + "epoch": 0.78848, + "grad_norm": 1.359375, + "learning_rate": 4.818172181847091e-05, + "loss": 0.8958653450012207, + "num_input_tokens_seen": 86004992, + "step": 770, + "train_runtime": 3911.8203, + "train_tokens_per_second": 21985.926 + }, + { + "epoch": 0.79872, + "grad_norm": 1.46875, + "learning_rate": 4.81593668217595e-05, + "loss": 0.9148517608642578, + "num_input_tokens_seen": 87105216, + "step": 780, + "train_runtime": 3961.6794, + "train_tokens_per_second": 21986.942 + }, + { + "epoch": 0.80896, + "grad_norm": 1.40625, + "learning_rate": 4.813704291251675e-05, + "loss": 0.9078433990478516, + "num_input_tokens_seen": 88239680, + "step": 790, + "train_runtime": 4013.5881, + "train_tokens_per_second": 21985.236 + }, + { + "epoch": 0.8192, + "grad_norm": 1.421875, + "learning_rate": 4.811475001875759e-05, + "loss": 0.9124856948852539, + "num_input_tokens_seen": 89349568, + "step": 800, + "train_runtime": 4062.7697, + "train_tokens_per_second": 21992.28 + }, + { + "epoch": 0.82944, + "grad_norm": 1.4375, + "learning_rate": 4.8092488068730105e-05, + "loss": 0.9061111450195313, + "num_input_tokens_seen": 90469312, + "step": 810, + "train_runtime": 4115.7452, + "train_tokens_per_second": 21981.271 + }, + { + "epoch": 0.83968, + "grad_norm": 1.3984375, + "learning_rate": 4.807025699091452e-05, + "loss": 0.8909177780151367, + "num_input_tokens_seen": 91591424, + "step": 820, + "train_runtime": 4167.5795, + "train_tokens_per_second": 21977.127 + }, + { + "epoch": 0.84992, + "grad_norm": 1.4609375, + "learning_rate": 4.8048056714022325e-05, + "loss": 0.8970900535583496, + "num_input_tokens_seen": 92718336, + "step": 830, + "train_runtime": 4219.1054, + "train_tokens_per_second": 21975.828 + }, + { + "epoch": 0.86016, + "grad_norm": 1.3984375, + "learning_rate": 4.802588716699519e-05, + "loss": 0.9085554122924805, + "num_input_tokens_seen": 93814400, + "step": 840, + "train_runtime": 4268.8362, + "train_tokens_per_second": 21976.575 + }, + { + "epoch": 0.8704, + "grad_norm": 1.3359375, + "learning_rate": 4.8003748279004156e-05, + "loss": 0.8963174819946289, + "num_input_tokens_seen": 94937408, + "step": 850, + "train_runtime": 4321.9061, + "train_tokens_per_second": 21966.56 + }, + { + "epoch": 0.88064, + "grad_norm": 1.4765625, + "learning_rate": 4.798163997944854e-05, + "loss": 0.9105907440185547, + "num_input_tokens_seen": 96048512, + "step": 860, + "train_runtime": 4371.8613, + "train_tokens_per_second": 21969.707 + }, + { + "epoch": 0.89088, + "grad_norm": 1.4609375, + "learning_rate": 4.79595621979551e-05, + "loss": 0.877834415435791, + "num_input_tokens_seen": 97193472, + "step": 870, + "train_runtime": 4425.9231, + "train_tokens_per_second": 21960.046 + }, + { + "epoch": 0.90112, + "grad_norm": 1.3828125, + "learning_rate": 4.793751486437702e-05, + "loss": 0.8991375923156738, + "num_input_tokens_seen": 98317184, + "step": 880, + "train_runtime": 4476.7825, + "train_tokens_per_second": 21961.573 + }, + { + "epoch": 0.91136, + "grad_norm": 1.515625, + "learning_rate": 4.7915497908793064e-05, + "loss": 0.8886856079101563, + "num_input_tokens_seen": 99450688, + "step": 890, + "train_runtime": 4529.0422, + "train_tokens_per_second": 21958.437 + }, + { + "epoch": 0.9216, + "grad_norm": 1.3515625, + "learning_rate": 4.7893511261506516e-05, + "loss": 0.902672004699707, + "num_input_tokens_seen": 100563840, + "step": 900, + "train_runtime": 4579.192, + "train_tokens_per_second": 21961.045 + }, + { + "epoch": 0.93184, + "grad_norm": 1.4453125, + "learning_rate": 4.787155485304435e-05, + "loss": 0.8937115669250488, + "num_input_tokens_seen": 101656640, + "step": 910, + "train_runtime": 4627.8649, + "train_tokens_per_second": 21966.208 + }, + { + "epoch": 0.94208, + "grad_norm": 1.375, + "learning_rate": 4.784962861415629e-05, + "loss": 0.9013708114624024, + "num_input_tokens_seen": 102780288, + "step": 920, + "train_runtime": 4679.0445, + "train_tokens_per_second": 21966.085 + }, + { + "epoch": 0.95232, + "grad_norm": 1.296875, + "learning_rate": 4.7827732475813884e-05, + "loss": 0.8749137878417969, + "num_input_tokens_seen": 103933056, + "step": 930, + "train_runtime": 4732.6015, + "train_tokens_per_second": 21961.083 + }, + { + "epoch": 0.96256, + "grad_norm": 1.3984375, + "learning_rate": 4.7805866369209576e-05, + "loss": 0.8961335182189941, + "num_input_tokens_seen": 105038016, + "step": 940, + "train_runtime": 4782.7459, + "train_tokens_per_second": 21961.864 + }, + { + "epoch": 0.9728, + "grad_norm": 1.3515625, + "learning_rate": 4.778403022575583e-05, + "loss": 0.874142074584961, + "num_input_tokens_seen": 106174336, + "step": 950, + "train_runtime": 4836.32, + "train_tokens_per_second": 21953.538 + }, + { + "epoch": 0.98304, + "grad_norm": 1.390625, + "learning_rate": 4.7762223977084195e-05, + "loss": 0.8942262649536132, + "num_input_tokens_seen": 107277440, + "step": 960, + "train_runtime": 4886.8487, + "train_tokens_per_second": 21952.273 + }, + { + "epoch": 0.99328, + "grad_norm": 1.3125, + "learning_rate": 4.774044755504444e-05, + "loss": 0.8799090385437012, + "num_input_tokens_seen": 108394496, + "step": 970, + "train_runtime": 4938.6396, + "train_tokens_per_second": 21948.25 + }, + { + "epoch": 1.003072, + "grad_norm": 1.25, + "learning_rate": 4.7718700891703616e-05, + "loss": 0.8489143371582031, + "num_input_tokens_seen": 109453888, + "step": 980, + "train_runtime": 4987.1553, + "train_tokens_per_second": 21947.158 + }, + { + "epoch": 1.013312, + "grad_norm": 1.40625, + "learning_rate": 4.7696983919345215e-05, + "loss": 0.7443439483642578, + "num_input_tokens_seen": 110580736, + "step": 990, + "train_runtime": 5038.9713, + "train_tokens_per_second": 21945.101 + }, + { + "epoch": 1.023552, + "grad_norm": 1.3671875, + "learning_rate": 4.7675296570468216e-05, + "loss": 0.7479698181152343, + "num_input_tokens_seen": 111723136, + "step": 1000, + "train_runtime": 5092.4363, + "train_tokens_per_second": 21939.034 + }, + { + "epoch": 1.033792, + "grad_norm": 1.5, + "learning_rate": 4.76536387777863e-05, + "loss": 0.7372268676757813, + "num_input_tokens_seen": 112836800, + "step": 1010, + "train_runtime": 5144.5655, + "train_tokens_per_second": 21933.203 + }, + { + "epoch": 1.044032, + "grad_norm": 1.3828125, + "learning_rate": 4.7632010474226915e-05, + "loss": 0.7579574584960938, + "num_input_tokens_seen": 113961088, + "step": 1020, + "train_runtime": 5195.725, + "train_tokens_per_second": 21933.626 + }, + { + "epoch": 1.054272, + "grad_norm": 1.59375, + "learning_rate": 4.761041159293035e-05, + "loss": 0.7489605903625488, + "num_input_tokens_seen": 115089984, + "step": 1030, + "train_runtime": 5247.5552, + "train_tokens_per_second": 21932.115 + }, + { + "epoch": 1.064512, + "grad_norm": 1.4140625, + "learning_rate": 4.7588842067249e-05, + "loss": 0.7482340812683106, + "num_input_tokens_seen": 116217216, + "step": 1040, + "train_runtime": 5299.9058, + "train_tokens_per_second": 21928.167 + }, + { + "epoch": 1.074752, + "grad_norm": 1.5078125, + "learning_rate": 4.756730183074637e-05, + "loss": 0.7579518318176269, + "num_input_tokens_seen": 117347648, + "step": 1050, + "train_runtime": 5351.066, + "train_tokens_per_second": 21929.77 + }, + { + "epoch": 1.084992, + "grad_norm": 1.4921875, + "learning_rate": 4.7545790817196314e-05, + "loss": 0.7438392639160156, + "num_input_tokens_seen": 118493632, + "step": 1060, + "train_runtime": 5405.5478, + "train_tokens_per_second": 21920.744 + }, + { + "epoch": 1.095232, + "grad_norm": 1.4296875, + "learning_rate": 4.752430896058212e-05, + "loss": 0.7573630332946777, + "num_input_tokens_seen": 119595776, + "step": 1070, + "train_runtime": 5455.0034, + "train_tokens_per_second": 21924.052 + }, + { + "epoch": 1.105472, + "grad_norm": 1.4765625, + "learning_rate": 4.750285619509567e-05, + "loss": 0.7446182250976563, + "num_input_tokens_seen": 120728576, + "step": 1080, + "train_runtime": 5507.7373, + "train_tokens_per_second": 21919.814 + }, + { + "epoch": 1.115712, + "grad_norm": 1.40625, + "learning_rate": 4.7481432455136644e-05, + "loss": 0.7489546775817871, + "num_input_tokens_seen": 121851968, + "step": 1090, + "train_runtime": 5559.0243, + "train_tokens_per_second": 21919.668 + }, + { + "epoch": 1.125952, + "grad_norm": 1.4609375, + "learning_rate": 4.7460037675311584e-05, + "loss": 0.757651948928833, + "num_input_tokens_seen": 122954752, + "step": 1100, + "train_runtime": 5608.2123, + "train_tokens_per_second": 21924.054 + }, + { + "epoch": 1.136192, + "grad_norm": 1.375, + "learning_rate": 4.7438671790433126e-05, + "loss": 0.7577178955078125, + "num_input_tokens_seen": 124052416, + "step": 1110, + "train_runtime": 5657.5424, + "train_tokens_per_second": 21926.909 + }, + { + "epoch": 1.146432, + "grad_norm": 1.4296875, + "learning_rate": 4.741733473551915e-05, + "loss": 0.7523440361022949, + "num_input_tokens_seen": 125165248, + "step": 1120, + "train_runtime": 5707.4101, + "train_tokens_per_second": 21930.306 + }, + { + "epoch": 1.156672, + "grad_norm": 1.4375, + "learning_rate": 4.7396026445791966e-05, + "loss": 0.7425838470458984, + "num_input_tokens_seen": 126283840, + "step": 1130, + "train_runtime": 5759.366, + "train_tokens_per_second": 21926.691 + }, + { + "epoch": 1.166912, + "grad_norm": 1.4609375, + "learning_rate": 4.737474685667742e-05, + "loss": 0.7449512004852294, + "num_input_tokens_seen": 127421952, + "step": 1140, + "train_runtime": 5811.4353, + "train_tokens_per_second": 21926.072 + }, + { + "epoch": 1.177152, + "grad_norm": 1.4140625, + "learning_rate": 4.7353495903804165e-05, + "loss": 0.761650562286377, + "num_input_tokens_seen": 128504576, + "step": 1150, + "train_runtime": 5859.1852, + "train_tokens_per_second": 21932.158 + }, + { + "epoch": 1.187392, + "grad_norm": 1.484375, + "learning_rate": 4.733227352300277e-05, + "loss": 0.7624397277832031, + "num_input_tokens_seen": 129617728, + "step": 1160, + "train_runtime": 5908.742, + "train_tokens_per_second": 21936.603 + }, + { + "epoch": 1.197632, + "grad_norm": 1.5234375, + "learning_rate": 4.731107965030496e-05, + "loss": 0.756078052520752, + "num_input_tokens_seen": 130713728, + "step": 1170, + "train_runtime": 5959.2042, + "train_tokens_per_second": 21934.762 + }, + { + "epoch": 1.207872, + "grad_norm": 1.3515625, + "learning_rate": 4.728991422194278e-05, + "loss": 0.7492989063262939, + "num_input_tokens_seen": 131802368, + "step": 1180, + "train_runtime": 6008.2874, + "train_tokens_per_second": 21936.762 + }, + { + "epoch": 1.218112, + "grad_norm": 1.4375, + "learning_rate": 4.726877717434773e-05, + "loss": 0.7438122749328613, + "num_input_tokens_seen": 132905472, + "step": 1190, + "train_runtime": 6057.8828, + "train_tokens_per_second": 21939.261 + }, + { + "epoch": 1.228352, + "grad_norm": 1.3671875, + "learning_rate": 4.724766844415013e-05, + "loss": 0.7444162368774414, + "num_input_tokens_seen": 134023296, + "step": 1200, + "train_runtime": 6107.5979, + "train_tokens_per_second": 21943.7 + }, + { + "epoch": 1.238592, + "grad_norm": 1.390625, + "learning_rate": 4.722658796817813e-05, + "loss": 0.7356367588043213, + "num_input_tokens_seen": 135165312, + "step": 1210, + "train_runtime": 6160.1499, + "train_tokens_per_second": 21941.887 + }, + { + "epoch": 1.248832, + "grad_norm": 1.359375, + "learning_rate": 4.7205535683457044e-05, + "loss": 0.7440855026245117, + "num_input_tokens_seen": 136319808, + "step": 1220, + "train_runtime": 6214.3559, + "train_tokens_per_second": 21936.273 + }, + { + "epoch": 1.259072, + "grad_norm": 1.453125, + "learning_rate": 4.7184511527208484e-05, + "loss": 0.7333660125732422, + "num_input_tokens_seen": 137441088, + "step": 1230, + "train_runtime": 6264.8424, + "train_tokens_per_second": 21938.475 + }, + { + "epoch": 1.269312, + "grad_norm": 1.5390625, + "learning_rate": 4.7163515436849644e-05, + "loss": 0.7524197101593018, + "num_input_tokens_seen": 138549568, + "step": 1240, + "train_runtime": 6313.7128, + "train_tokens_per_second": 21944.23 + }, + { + "epoch": 1.279552, + "grad_norm": 1.484375, + "learning_rate": 4.714254734999245e-05, + "loss": 0.7486214637756348, + "num_input_tokens_seen": 139628416, + "step": 1250, + "train_runtime": 6361.0684, + "train_tokens_per_second": 21950.466 + }, + { + "epoch": 1.289792, + "grad_norm": 1.578125, + "learning_rate": 4.712160720444284e-05, + "loss": 0.7422505378723144, + "num_input_tokens_seen": 140751552, + "step": 1260, + "train_runtime": 6412.2087, + "train_tokens_per_second": 21950.557 + }, + { + "epoch": 1.300032, + "grad_norm": 1.4921875, + "learning_rate": 4.710069493819992e-05, + "loss": 0.7394045352935791, + "num_input_tokens_seen": 141875392, + "step": 1270, + "train_runtime": 6463.2897, + "train_tokens_per_second": 21950.957 + }, + { + "epoch": 1.3102719999999999, + "grad_norm": 1.4375, + "learning_rate": 4.70798104894553e-05, + "loss": 0.7524402618408204, + "num_input_tokens_seen": 142968768, + "step": 1280, + "train_runtime": 6512.8407, + "train_tokens_per_second": 21951.829 + }, + { + "epoch": 1.320512, + "grad_norm": 1.3671875, + "learning_rate": 4.705895379659219e-05, + "loss": 0.73507719039917, + "num_input_tokens_seen": 144088448, + "step": 1290, + "train_runtime": 6563.9326, + "train_tokens_per_second": 21951.543 + }, + { + "epoch": 1.330752, + "grad_norm": 1.4140625, + "learning_rate": 4.7038124798184766e-05, + "loss": 0.744103193283081, + "num_input_tokens_seen": 145166528, + "step": 1300, + "train_runtime": 6612.2717, + "train_tokens_per_second": 21954.108 + }, + { + "epoch": 1.340992, + "grad_norm": 1.4609375, + "learning_rate": 4.7017323432997304e-05, + "loss": 0.7421403884887695, + "num_input_tokens_seen": 146277888, + "step": 1310, + "train_runtime": 6663.7474, + "train_tokens_per_second": 21951.296 + }, + { + "epoch": 1.351232, + "grad_norm": 1.4453125, + "learning_rate": 4.6996549639983506e-05, + "loss": 0.7422142505645752, + "num_input_tokens_seen": 147402112, + "step": 1320, + "train_runtime": 6713.9776, + "train_tokens_per_second": 21954.513 + }, + { + "epoch": 1.361472, + "grad_norm": 1.4140625, + "learning_rate": 4.697580335828569e-05, + "loss": 0.747988224029541, + "num_input_tokens_seen": 148498816, + "step": 1330, + "train_runtime": 6762.7236, + "train_tokens_per_second": 21958.433 + }, + { + "epoch": 1.371712, + "grad_norm": 1.4140625, + "learning_rate": 4.6955084527234076e-05, + "loss": 0.7415881633758545, + "num_input_tokens_seen": 149633024, + "step": 1340, + "train_runtime": 6813.6204, + "train_tokens_per_second": 21960.869 + }, + { + "epoch": 1.381952, + "grad_norm": 1.3984375, + "learning_rate": 4.6934393086346034e-05, + "loss": 0.7278687953948975, + "num_input_tokens_seen": 150743424, + "step": 1350, + "train_runtime": 6865.1777, + "train_tokens_per_second": 21957.687 + }, + { + "epoch": 1.392192, + "grad_norm": 1.484375, + "learning_rate": 4.6913728975325324e-05, + "loss": 0.7370668411254883, + "num_input_tokens_seen": 151865536, + "step": 1360, + "train_runtime": 6915.0272, + "train_tokens_per_second": 21961.669 + }, + { + "epoch": 1.4024320000000001, + "grad_norm": 1.4765625, + "learning_rate": 4.6893092134061393e-05, + "loss": 0.7397951126098633, + "num_input_tokens_seen": 152997504, + "step": 1370, + "train_runtime": 6966.2008, + "train_tokens_per_second": 21962.833 + }, + { + "epoch": 1.412672, + "grad_norm": 1.4453125, + "learning_rate": 4.687248250262859e-05, + "loss": 0.7369170188903809, + "num_input_tokens_seen": 154098752, + "step": 1380, + "train_runtime": 7017.463, + "train_tokens_per_second": 21959.325 + }, + { + "epoch": 1.422912, + "grad_norm": 1.4453125, + "learning_rate": 4.685190002128548e-05, + "loss": 0.7380311965942383, + "num_input_tokens_seen": 155218880, + "step": 1390, + "train_runtime": 7069.234, + "train_tokens_per_second": 21956.959 + }, + { + "epoch": 1.433152, + "grad_norm": 1.3359375, + "learning_rate": 4.6831344630474114e-05, + "loss": 0.719688892364502, + "num_input_tokens_seen": 156365696, + "step": 1400, + "train_runtime": 7121.6519, + "train_tokens_per_second": 21956.38 + }, + { + "epoch": 1.443392, + "grad_norm": 1.4765625, + "learning_rate": 4.6810816270819276e-05, + "loss": 0.7329507827758789, + "num_input_tokens_seen": 157494784, + "step": 1410, + "train_runtime": 7171.9905, + "train_tokens_per_second": 21959.703 + }, + { + "epoch": 1.453632, + "grad_norm": 1.421875, + "learning_rate": 4.679031488312777e-05, + "loss": 0.7250077247619628, + "num_input_tokens_seen": 158633216, + "step": 1420, + "train_runtime": 7223.042, + "train_tokens_per_second": 21962.106 + }, + { + "epoch": 1.463872, + "grad_norm": 1.4453125, + "learning_rate": 4.6769840408387717e-05, + "loss": 0.7379475116729737, + "num_input_tokens_seen": 159750208, + "step": 1430, + "train_runtime": 7274.4145, + "train_tokens_per_second": 21960.559 + }, + { + "epoch": 1.4741119999999999, + "grad_norm": 1.46875, + "learning_rate": 4.674939278776787e-05, + "loss": 0.7261748313903809, + "num_input_tokens_seen": 160865344, + "step": 1440, + "train_runtime": 7324.4445, + "train_tokens_per_second": 21962.805 + }, + { + "epoch": 1.484352, + "grad_norm": 1.390625, + "learning_rate": 4.672897196261683e-05, + "loss": 0.718736219406128, + "num_input_tokens_seen": 162000640, + "step": 1450, + "train_runtime": 7376.9389, + "train_tokens_per_second": 21960.415 + }, + { + "epoch": 1.494592, + "grad_norm": 1.421875, + "learning_rate": 4.670857787446238e-05, + "loss": 0.7371679782867432, + "num_input_tokens_seen": 163120704, + "step": 1460, + "train_runtime": 7426.1512, + "train_tokens_per_second": 21965.713 + }, + { + "epoch": 1.504832, + "grad_norm": 1.3984375, + "learning_rate": 4.668821046501082e-05, + "loss": 0.7367044448852539, + "num_input_tokens_seen": 164228160, + "step": 1470, + "train_runtime": 7474.0997, + "train_tokens_per_second": 21972.969 + }, + { + "epoch": 1.515072, + "grad_norm": 1.5390625, + "learning_rate": 4.6667869676146194e-05, + "loss": 0.7399471282958985, + "num_input_tokens_seen": 165347456, + "step": 1480, + "train_runtime": 7525.5655, + "train_tokens_per_second": 21971.433 + }, + { + "epoch": 1.525312, + "grad_norm": 1.390625, + "learning_rate": 4.6647555449929645e-05, + "loss": 0.7276147842407227, + "num_input_tokens_seen": 166452672, + "step": 1490, + "train_runtime": 7575.6639, + "train_tokens_per_second": 21972.024 + }, + { + "epoch": 1.535552, + "grad_norm": 1.4453125, + "learning_rate": 4.662726772859869e-05, + "loss": 0.7368722915649414, + "num_input_tokens_seen": 167547520, + "step": 1500, + "train_runtime": 7624.3614, + "train_tokens_per_second": 21975.286 + }, + { + "epoch": 1.545792, + "grad_norm": 1.3984375, + "learning_rate": 4.660700645456655e-05, + "loss": 0.7236470222473145, + "num_input_tokens_seen": 168664320, + "step": 1510, + "train_runtime": 7674.9645, + "train_tokens_per_second": 21975.909 + }, + { + "epoch": 1.556032, + "grad_norm": 1.4453125, + "learning_rate": 4.658677157042149e-05, + "loss": 0.7258675575256348, + "num_input_tokens_seen": 169775168, + "step": 1520, + "train_runtime": 7726.0187, + "train_tokens_per_second": 21974.47 + }, + { + "epoch": 1.566272, + "grad_norm": 1.5, + "learning_rate": 4.656656301892605e-05, + "loss": 0.7226676940917969, + "num_input_tokens_seen": 170910912, + "step": 1530, + "train_runtime": 7777.3904, + "train_tokens_per_second": 21975.355 + }, + { + "epoch": 1.5765120000000001, + "grad_norm": 1.5, + "learning_rate": 4.6546380743016465e-05, + "loss": 0.7360187530517578, + "num_input_tokens_seen": 172011072, + "step": 1540, + "train_runtime": 7824.7246, + "train_tokens_per_second": 21983.019 + }, + { + "epoch": 1.5867520000000002, + "grad_norm": 1.46875, + "learning_rate": 4.652622468580193e-05, + "loss": 0.7214366912841796, + "num_input_tokens_seen": 173135616, + "step": 1550, + "train_runtime": 7875.7668, + "train_tokens_per_second": 21983.335 + }, + { + "epoch": 1.596992, + "grad_norm": 1.4921875, + "learning_rate": 4.650609479056392e-05, + "loss": 0.7311969757080078, + "num_input_tokens_seen": 174239232, + "step": 1560, + "train_runtime": 7923.3292, + "train_tokens_per_second": 21990.659 + }, + { + "epoch": 1.607232, + "grad_norm": 1.4609375, + "learning_rate": 4.648599100075556e-05, + "loss": 0.7286062240600586, + "num_input_tokens_seen": 175335168, + "step": 1570, + "train_runtime": 7972.5819, + "train_tokens_per_second": 21992.269 + }, + { + "epoch": 1.617472, + "grad_norm": 1.4765625, + "learning_rate": 4.6465913260000945e-05, + "loss": 0.7191134929656983, + "num_input_tokens_seen": 176463680, + "step": 1580, + "train_runtime": 8025.3812, + "train_tokens_per_second": 21988.199 + }, + { + "epoch": 1.627712, + "grad_norm": 1.609375, + "learning_rate": 4.644586151209444e-05, + "loss": 0.7198073387145996, + "num_input_tokens_seen": 177578176, + "step": 1590, + "train_runtime": 8076.0558, + "train_tokens_per_second": 21988.23 + }, + { + "epoch": 1.6379519999999999, + "grad_norm": 1.5390625, + "learning_rate": 4.6425835701000084e-05, + "loss": 0.7295777320861816, + "num_input_tokens_seen": 178670912, + "step": 1600, + "train_runtime": 8124.634, + "train_tokens_per_second": 21991.257 + }, + { + "epoch": 1.6481919999999999, + "grad_norm": 1.4765625, + "learning_rate": 4.640583577085084e-05, + "loss": 0.7139708518981933, + "num_input_tokens_seen": 179777600, + "step": 1610, + "train_runtime": 8174.4495, + "train_tokens_per_second": 21992.625 + }, + { + "epoch": 1.658432, + "grad_norm": 1.578125, + "learning_rate": 4.638586166594806e-05, + "loss": 0.733104658126831, + "num_input_tokens_seen": 180869504, + "step": 1620, + "train_runtime": 8223.6421, + "train_tokens_per_second": 21993.844 + }, + { + "epoch": 1.668672, + "grad_norm": 1.4921875, + "learning_rate": 4.6365913330760726e-05, + "loss": 0.7275556564331055, + "num_input_tokens_seen": 182019072, + "step": 1630, + "train_runtime": 8275.8793, + "train_tokens_per_second": 21993.925 + }, + { + "epoch": 1.678912, + "grad_norm": 1.4453125, + "learning_rate": 4.6345990709924855e-05, + "loss": 0.7281203269958496, + "num_input_tokens_seen": 183130112, + "step": 1640, + "train_runtime": 8325.9976, + "train_tokens_per_second": 21994.975 + }, + { + "epoch": 1.689152, + "grad_norm": 1.46875, + "learning_rate": 4.632609374824284e-05, + "loss": 0.7167181968688965, + "num_input_tokens_seen": 184265024, + "step": 1650, + "train_runtime": 8379.0993, + "train_tokens_per_second": 21991.03 + }, + { + "epoch": 1.699392, + "grad_norm": 1.609375, + "learning_rate": 4.630622239068285e-05, + "loss": 0.7205727100372314, + "num_input_tokens_seen": 185403968, + "step": 1660, + "train_runtime": 8431.432, + "train_tokens_per_second": 21989.618 + }, + { + "epoch": 1.709632, + "grad_norm": 1.4375, + "learning_rate": 4.628637658237808e-05, + "loss": 0.7173449039459229, + "num_input_tokens_seen": 186535552, + "step": 1670, + "train_runtime": 8482.4669, + "train_tokens_per_second": 21990.72 + }, + { + "epoch": 1.719872, + "grad_norm": 1.4140625, + "learning_rate": 4.626655626862625e-05, + "loss": 0.7150672912597656, + "num_input_tokens_seen": 187648448, + "step": 1680, + "train_runtime": 8533.219, + "train_tokens_per_second": 21990.347 + }, + { + "epoch": 1.730112, + "grad_norm": 1.4375, + "learning_rate": 4.624676139488888e-05, + "loss": 0.7177637577056885, + "num_input_tokens_seen": 188770624, + "step": 1690, + "train_runtime": 8582.5231, + "train_tokens_per_second": 21994.77 + }, + { + "epoch": 1.7403520000000001, + "grad_norm": 1.4296875, + "learning_rate": 4.6226991906790686e-05, + "loss": 0.7107383728027343, + "num_input_tokens_seen": 189905536, + "step": 1700, + "train_runtime": 8635.4296, + "train_tokens_per_second": 21991.441 + }, + { + "epoch": 1.7505920000000001, + "grad_norm": 1.453125, + "learning_rate": 4.620724775011897e-05, + "loss": 0.7159247398376465, + "num_input_tokens_seen": 191047360, + "step": 1710, + "train_runtime": 8688.7592, + "train_tokens_per_second": 21987.876 + }, + { + "epoch": 1.760832, + "grad_norm": 1.5, + "learning_rate": 4.618752887082297e-05, + "loss": 0.7231245517730713, + "num_input_tokens_seen": 192176448, + "step": 1720, + "train_runtime": 8739.8956, + "train_tokens_per_second": 21988.415 + }, + { + "epoch": 1.771072, + "grad_norm": 1.6015625, + "learning_rate": 4.616783521501325e-05, + "loss": 0.7093376159667969, + "num_input_tokens_seen": 193282496, + "step": 1730, + "train_runtime": 8789.6819, + "train_tokens_per_second": 21989.703 + }, + { + "epoch": 1.781312, + "grad_norm": 1.5078125, + "learning_rate": 4.614816672896108e-05, + "loss": 0.7214728832244873, + "num_input_tokens_seen": 194367808, + "step": 1740, + "train_runtime": 8838.9917, + "train_tokens_per_second": 21989.817 + }, + { + "epoch": 1.791552, + "grad_norm": 1.484375, + "learning_rate": 4.612852335909782e-05, + "loss": 0.7111657619476318, + "num_input_tokens_seen": 195472320, + "step": 1750, + "train_runtime": 8890.2997, + "train_tokens_per_second": 21987.146 + }, + { + "epoch": 1.8017919999999998, + "grad_norm": 1.5078125, + "learning_rate": 4.6108905052014323e-05, + "loss": 0.7161635398864746, + "num_input_tokens_seen": 196566912, + "step": 1760, + "train_runtime": 8939.6068, + "train_tokens_per_second": 21988.317 + }, + { + "epoch": 1.8120319999999999, + "grad_norm": 1.453125, + "learning_rate": 4.608931175446027e-05, + "loss": 0.7039215087890625, + "num_input_tokens_seen": 197692864, + "step": 1770, + "train_runtime": 8991.4052, + "train_tokens_per_second": 21986.871 + }, + { + "epoch": 1.822272, + "grad_norm": 1.53125, + "learning_rate": 4.606974341334367e-05, + "loss": 0.7046633720397949, + "num_input_tokens_seen": 198851136, + "step": 1780, + "train_runtime": 9045.7861, + "train_tokens_per_second": 21982.737 + }, + { + "epoch": 1.832512, + "grad_norm": 1.421875, + "learning_rate": 4.605019997573011e-05, + "loss": 0.7108986854553223, + "num_input_tokens_seen": 199991616, + "step": 1790, + "train_runtime": 9098.4241, + "train_tokens_per_second": 21980.907 + }, + { + "epoch": 1.842752, + "grad_norm": 1.515625, + "learning_rate": 4.603068138884229e-05, + "loss": 0.7045407295227051, + "num_input_tokens_seen": 201125312, + "step": 1800, + "train_runtime": 9149.6357, + "train_tokens_per_second": 21981.784 + }, + { + "epoch": 1.852992, + "grad_norm": 1.46875, + "learning_rate": 4.6011187600059345e-05, + "loss": 0.7162825584411621, + "num_input_tokens_seen": 202239936, + "step": 1810, + "train_runtime": 9199.9174, + "train_tokens_per_second": 21982.799 + }, + { + "epoch": 1.863232, + "grad_norm": 1.421875, + "learning_rate": 4.599171855691629e-05, + "loss": 0.7169035911560059, + "num_input_tokens_seen": 203321984, + "step": 1820, + "train_runtime": 9247.0534, + "train_tokens_per_second": 21987.759 + }, + { + "epoch": 1.873472, + "grad_norm": 1.5390625, + "learning_rate": 4.597227420710335e-05, + "loss": 0.6997568130493164, + "num_input_tokens_seen": 204444032, + "step": 1830, + "train_runtime": 9300.3822, + "train_tokens_per_second": 21982.326 + }, + { + "epoch": 1.883712, + "grad_norm": 1.4453125, + "learning_rate": 4.595285449846551e-05, + "loss": 0.7147369861602784, + "num_input_tokens_seen": 205551872, + "step": 1840, + "train_runtime": 9350.726, + "train_tokens_per_second": 21982.451 + }, + { + "epoch": 1.893952, + "grad_norm": 1.4140625, + "learning_rate": 4.593345937900178e-05, + "loss": 0.700528621673584, + "num_input_tokens_seen": 206680128, + "step": 1850, + "train_runtime": 9402.0793, + "train_tokens_per_second": 21982.385 + }, + { + "epoch": 1.904192, + "grad_norm": 1.4140625, + "learning_rate": 4.591408879686472e-05, + "loss": 0.707237434387207, + "num_input_tokens_seen": 207813696, + "step": 1860, + "train_runtime": 9454.3191, + "train_tokens_per_second": 21980.821 + }, + { + "epoch": 1.9144320000000001, + "grad_norm": 1.5390625, + "learning_rate": 4.5894742700359775e-05, + "loss": 0.7043401718139648, + "num_input_tokens_seen": 208895808, + "step": 1870, + "train_runtime": 9502.5318, + "train_tokens_per_second": 21983.174 + }, + { + "epoch": 1.9246720000000002, + "grad_norm": 1.4140625, + "learning_rate": 4.587542103794477e-05, + "loss": 0.7066624641418457, + "num_input_tokens_seen": 210025600, + "step": 1880, + "train_runtime": 9553.8388, + "train_tokens_per_second": 21983.373 + }, + { + "epoch": 1.934912, + "grad_norm": 1.515625, + "learning_rate": 4.5856123758229247e-05, + "loss": 0.7068476676940918, + "num_input_tokens_seen": 211141184, + "step": 1890, + "train_runtime": 9604.6551, + "train_tokens_per_second": 21983.214 + }, + { + "epoch": 1.945152, + "grad_norm": 1.484375, + "learning_rate": 4.5836850809973993e-05, + "loss": 0.7084201812744141, + "num_input_tokens_seen": 212250112, + "step": 1900, + "train_runtime": 9653.8675, + "train_tokens_per_second": 21986.019 + }, + { + "epoch": 1.955392, + "grad_norm": 1.3984375, + "learning_rate": 4.5817602142090385e-05, + "loss": 0.6999262809753418, + "num_input_tokens_seen": 213370240, + "step": 1910, + "train_runtime": 9704.8849, + "train_tokens_per_second": 21985.86 + }, + { + "epoch": 1.965632, + "grad_norm": 1.4375, + "learning_rate": 4.579837770363989e-05, + "loss": 0.7026149749755859, + "num_input_tokens_seen": 214511104, + "step": 1920, + "train_runtime": 9757.8534, + "train_tokens_per_second": 21983.432 + }, + { + "epoch": 1.9758719999999999, + "grad_norm": 1.5234375, + "learning_rate": 4.57791774438334e-05, + "loss": 0.7023416042327881, + "num_input_tokens_seen": 215629504, + "step": 1930, + "train_runtime": 9807.0125, + "train_tokens_per_second": 21987.277 + }, + { + "epoch": 1.9861119999999999, + "grad_norm": 1.46875, + "learning_rate": 4.576000131203078e-05, + "loss": 0.7092232704162598, + "num_input_tokens_seen": 216746880, + "step": 1940, + "train_runtime": 9858.8618, + "train_tokens_per_second": 21984.98 + }, + { + "epoch": 1.996352, + "grad_norm": 1.5078125, + "learning_rate": 4.574084925774023e-05, + "loss": 0.6931307792663575, + "num_input_tokens_seen": 217858496, + "step": 1950, + "train_runtime": 9908.0458, + "train_tokens_per_second": 21988.039 + }, + { + "epoch": 2.006144, + "grad_norm": 2.203125, + "learning_rate": 4.5721721230617795e-05, + "loss": 0.6160273551940918, + "num_input_tokens_seen": 218918208, + "step": 1960, + "train_runtime": 9957.3271, + "train_tokens_per_second": 21985.64 + }, + { + "epoch": 2.016384, + "grad_norm": 1.5625, + "learning_rate": 4.57026171804667e-05, + "loss": 0.5476226806640625, + "num_input_tokens_seen": 220009600, + "step": 1970, + "train_runtime": 10005.4786, + "train_tokens_per_second": 21988.913 + }, + { + "epoch": 2.026624, + "grad_norm": 1.578125, + "learning_rate": 4.568353705723692e-05, + "loss": 0.5443649768829346, + "num_input_tokens_seen": 221112448, + "step": 1980, + "train_runtime": 10053.8591, + "train_tokens_per_second": 21992.794 + }, + { + "epoch": 2.036864, + "grad_norm": 1.6953125, + "learning_rate": 4.566448081102455e-05, + "loss": 0.530079698562622, + "num_input_tokens_seen": 222236352, + "step": 1990, + "train_runtime": 10106.2355, + "train_tokens_per_second": 21990.023 + }, + { + "epoch": 2.047104, + "grad_norm": 1.5859375, + "learning_rate": 4.564544839207128e-05, + "loss": 0.5314745426177978, + "num_input_tokens_seen": 223371840, + "step": 2000, + "train_runtime": 10157.507, + "train_tokens_per_second": 21990.813 + }, + { + "epoch": 2.057344, + "grad_norm": 1.703125, + "learning_rate": 4.562643975076387e-05, + "loss": 0.536113691329956, + "num_input_tokens_seen": 224505152, + "step": 2010, + "train_runtime": 10210.6596, + "train_tokens_per_second": 21987.331 + }, + { + "epoch": 2.067584, + "grad_norm": 1.6640625, + "learning_rate": 4.560745483763357e-05, + "loss": 0.537871265411377, + "num_input_tokens_seen": 225609024, + "step": 2020, + "train_runtime": 10260.9136, + "train_tokens_per_second": 21987.226 + }, + { + "epoch": 2.077824, + "grad_norm": 1.7109375, + "learning_rate": 4.5588493603355595e-05, + "loss": 0.5416299819946289, + "num_input_tokens_seen": 226721728, + "step": 2030, + "train_runtime": 10310.4748, + "train_tokens_per_second": 21989.456 + }, + { + "epoch": 2.088064, + "grad_norm": 1.5703125, + "learning_rate": 4.556955599874859e-05, + "loss": 0.5329193592071533, + "num_input_tokens_seen": 227839680, + "step": 2040, + "train_runtime": 10362.1302, + "train_tokens_per_second": 21987.726 + }, + { + "epoch": 2.098304, + "grad_norm": 1.5234375, + "learning_rate": 4.555064197477409e-05, + "loss": 0.5448167800903321, + "num_input_tokens_seen": 228945344, + "step": 2050, + "train_runtime": 10411.8833, + "train_tokens_per_second": 21988.85 + }, + { + "epoch": 2.108544, + "grad_norm": 1.6015625, + "learning_rate": 4.5531751482536e-05, + "loss": 0.5400659084320069, + "num_input_tokens_seen": 230068608, + "step": 2060, + "train_runtime": 10461.9263, + "train_tokens_per_second": 21991.037 + }, + { + "epoch": 2.118784, + "grad_norm": 1.6015625, + "learning_rate": 4.5512884473280024e-05, + "loss": 0.5294137954711914, + "num_input_tokens_seen": 231161728, + "step": 2070, + "train_runtime": 10511.7551, + "train_tokens_per_second": 21990.783 + }, + { + "epoch": 2.129024, + "grad_norm": 1.65625, + "learning_rate": 4.549404089839322e-05, + "loss": 0.535043478012085, + "num_input_tokens_seen": 232286656, + "step": 2080, + "train_runtime": 10562.4846, + "train_tokens_per_second": 21991.668 + }, + { + "epoch": 2.139264, + "grad_norm": 1.6875, + "learning_rate": 4.547522070940335e-05, + "loss": 0.5406003952026367, + "num_input_tokens_seen": 233403328, + "step": 2090, + "train_runtime": 10614.5795, + "train_tokens_per_second": 21988.938 + }, + { + "epoch": 2.149504, + "grad_norm": 1.65625, + "learning_rate": 4.545642385797848e-05, + "loss": 0.5352214813232422, + "num_input_tokens_seen": 234542400, + "step": 2100, + "train_runtime": 10668.2079, + "train_tokens_per_second": 21985.173 + }, + { + "epoch": 2.159744, + "grad_norm": 1.59375, + "learning_rate": 4.543765029592637e-05, + "loss": 0.5299077987670898, + "num_input_tokens_seen": 235650624, + "step": 2110, + "train_runtime": 10718.8559, + "train_tokens_per_second": 21984.681 + }, + { + "epoch": 2.169984, + "grad_norm": 1.6640625, + "learning_rate": 4.541889997519403e-05, + "loss": 0.5387727737426757, + "num_input_tokens_seen": 236777600, + "step": 2120, + "train_runtime": 10769.6328, + "train_tokens_per_second": 21985.671 + }, + { + "epoch": 2.180224, + "grad_norm": 1.625, + "learning_rate": 4.5400172847867095e-05, + "loss": 0.5402078628540039, + "num_input_tokens_seen": 237897728, + "step": 2130, + "train_runtime": 10820.7737, + "train_tokens_per_second": 21985.279 + }, + { + "epoch": 2.190464, + "grad_norm": 1.6484375, + "learning_rate": 4.5381468866169466e-05, + "loss": 0.5417552947998047, + "num_input_tokens_seen": 238992704, + "step": 2140, + "train_runtime": 10869.1633, + "train_tokens_per_second": 21988.142 + }, + { + "epoch": 2.200704, + "grad_norm": 1.7109375, + "learning_rate": 4.5362787982462616e-05, + "loss": 0.530327320098877, + "num_input_tokens_seen": 240146496, + "step": 2150, + "train_runtime": 10922.9566, + "train_tokens_per_second": 21985.485 + }, + { + "epoch": 2.210944, + "grad_norm": 1.671875, + "learning_rate": 4.5344130149245275e-05, + "loss": 0.5349247932434082, + "num_input_tokens_seen": 241241856, + "step": 2160, + "train_runtime": 10972.6208, + "train_tokens_per_second": 21985.801 + }, + { + "epoch": 2.221184, + "grad_norm": 1.6328125, + "learning_rate": 4.5325495319152715e-05, + "loss": 0.5235861301422119, + "num_input_tokens_seen": 242377600, + "step": 2170, + "train_runtime": 11025.3999, + "train_tokens_per_second": 21983.565 + }, + { + "epoch": 2.231424, + "grad_norm": 1.640625, + "learning_rate": 4.530688344495644e-05, + "loss": 0.5335872173309326, + "num_input_tokens_seen": 243525248, + "step": 2180, + "train_runtime": 11078.0924, + "train_tokens_per_second": 21982.598 + }, + { + "epoch": 2.241664, + "grad_norm": 1.6328125, + "learning_rate": 4.528829447956357e-05, + "loss": 0.5255549430847168, + "num_input_tokens_seen": 244661248, + "step": 2190, + "train_runtime": 11129.5816, + "train_tokens_per_second": 21982.969 + }, + { + "epoch": 2.251904, + "grad_norm": 1.734375, + "learning_rate": 4.526972837601633e-05, + "loss": 0.5395485877990722, + "num_input_tokens_seen": 245780800, + "step": 2200, + "train_runtime": 11180.2882, + "train_tokens_per_second": 21983.405 + }, + { + "epoch": 2.262144, + "grad_norm": 1.7734375, + "learning_rate": 4.525118508749165e-05, + "loss": 0.5427422523498535, + "num_input_tokens_seen": 246870912, + "step": 2210, + "train_runtime": 11229.0191, + "train_tokens_per_second": 21985.083 + }, + { + "epoch": 2.272384, + "grad_norm": 1.6796875, + "learning_rate": 4.5232664567300546e-05, + "loss": 0.5383429527282715, + "num_input_tokens_seen": 247963136, + "step": 2220, + "train_runtime": 11278.2198, + "train_tokens_per_second": 21986.017 + }, + { + "epoch": 2.282624, + "grad_norm": 1.703125, + "learning_rate": 4.521416676888773e-05, + "loss": 0.5311496257781982, + "num_input_tokens_seen": 249098240, + "step": 2230, + "train_runtime": 11331.4261, + "train_tokens_per_second": 21982.956 + }, + { + "epoch": 2.292864, + "grad_norm": 1.65625, + "learning_rate": 4.519569164583107e-05, + "loss": 0.5386343479156495, + "num_input_tokens_seen": 250219840, + "step": 2240, + "train_runtime": 11382.8605, + "train_tokens_per_second": 21982.158 + }, + { + "epoch": 2.303104, + "grad_norm": 1.671875, + "learning_rate": 4.517723915184109e-05, + "loss": 0.5370828628540039, + "num_input_tokens_seen": 251317952, + "step": 2250, + "train_runtime": 11431.9094, + "train_tokens_per_second": 21983.9 + }, + { + "epoch": 2.313344, + "grad_norm": 1.6875, + "learning_rate": 4.5158809240760506e-05, + "loss": 0.5292394638061524, + "num_input_tokens_seen": 252466688, + "step": 2260, + "train_runtime": 11485.1849, + "train_tokens_per_second": 21981.944 + }, + { + "epoch": 2.323584, + "grad_norm": 1.6640625, + "learning_rate": 4.514040186656375e-05, + "loss": 0.5335429191589356, + "num_input_tokens_seen": 253612992, + "step": 2270, + "train_runtime": 11538.6494, + "train_tokens_per_second": 21979.435 + }, + { + "epoch": 2.333824, + "grad_norm": 1.625, + "learning_rate": 4.512201698335644e-05, + "loss": 0.5375387668609619, + "num_input_tokens_seen": 254732480, + "step": 2280, + "train_runtime": 11590.0261, + "train_tokens_per_second": 21978.594 + }, + { + "epoch": 2.344064, + "grad_norm": 1.75, + "learning_rate": 4.510365454537496e-05, + "loss": 0.5277575492858887, + "num_input_tokens_seen": 255856448, + "step": 2290, + "train_runtime": 11640.5446, + "train_tokens_per_second": 21979.766 + }, + { + "epoch": 2.354304, + "grad_norm": 1.65625, + "learning_rate": 4.5085314506985945e-05, + "loss": 0.523950719833374, + "num_input_tokens_seen": 256999168, + "step": 2300, + "train_runtime": 11692.6859, + "train_tokens_per_second": 21979.481 + }, + { + "epoch": 2.364544, + "grad_norm": 1.625, + "learning_rate": 4.50669968226858e-05, + "loss": 0.5421285629272461, + "num_input_tokens_seen": 258137984, + "step": 2310, + "train_runtime": 11745.5264, + "train_tokens_per_second": 21977.558 + }, + { + "epoch": 2.374784, + "grad_norm": 1.6953125, + "learning_rate": 4.504870144710027e-05, + "loss": 0.5283915996551514, + "num_input_tokens_seen": 259258496, + "step": 2320, + "train_runtime": 11796.224, + "train_tokens_per_second": 21978.092 + }, + { + "epoch": 2.385024, + "grad_norm": 1.65625, + "learning_rate": 4.5030428334983884e-05, + "loss": 0.5356220245361328, + "num_input_tokens_seen": 260398080, + "step": 2330, + "train_runtime": 11850.3858, + "train_tokens_per_second": 21973.806 + }, + { + "epoch": 2.395264, + "grad_norm": 1.5859375, + "learning_rate": 4.501217744121959e-05, + "loss": 0.5268114089965821, + "num_input_tokens_seen": 261541440, + "step": 2340, + "train_runtime": 11903.3512, + "train_tokens_per_second": 21972.085 + }, + { + "epoch": 2.405504, + "grad_norm": 1.75, + "learning_rate": 4.499394872081821e-05, + "loss": 0.5375988960266114, + "num_input_tokens_seen": 262635264, + "step": 2350, + "train_runtime": 11951.8181, + "train_tokens_per_second": 21974.503 + }, + { + "epoch": 2.415744, + "grad_norm": 1.7578125, + "learning_rate": 4.4975742128918e-05, + "loss": 0.5376396656036377, + "num_input_tokens_seen": 263740288, + "step": 2360, + "train_runtime": 12000.2326, + "train_tokens_per_second": 21977.931 + }, + { + "epoch": 2.425984, + "grad_norm": 1.6953125, + "learning_rate": 4.495755762078418e-05, + "loss": 0.5323172569274902, + "num_input_tokens_seen": 264867968, + "step": 2370, + "train_runtime": 12050.7407, + "train_tokens_per_second": 21979.393 + }, + { + "epoch": 2.436224, + "grad_norm": 1.7578125, + "learning_rate": 4.49393951518085e-05, + "loss": 0.5391307830810547, + "num_input_tokens_seen": 265965440, + "step": 2380, + "train_runtime": 12099.6065, + "train_tokens_per_second": 21981.33 + }, + { + "epoch": 2.446464, + "grad_norm": 1.6171875, + "learning_rate": 4.4921254677508716e-05, + "loss": 0.5373844146728516, + "num_input_tokens_seen": 267080128, + "step": 2390, + "train_runtime": 12149.1708, + "train_tokens_per_second": 21983.404 + }, + { + "epoch": 2.456704, + "grad_norm": 1.7578125, + "learning_rate": 4.490313615352821e-05, + "loss": 0.5455498218536377, + "num_input_tokens_seen": 268188096, + "step": 2400, + "train_runtime": 12200.2497, + "train_tokens_per_second": 21982.181 + }, + { + "epoch": 2.466944, + "grad_norm": 1.734375, + "learning_rate": 4.48850395356355e-05, + "loss": 0.5363270759582519, + "num_input_tokens_seen": 269321984, + "step": 2410, + "train_runtime": 12254.3616, + "train_tokens_per_second": 21977.643 + }, + { + "epoch": 2.477184, + "grad_norm": 1.75, + "learning_rate": 4.486696477972375e-05, + "loss": 0.5418241024017334, + "num_input_tokens_seen": 270419776, + "step": 2420, + "train_runtime": 12303.7844, + "train_tokens_per_second": 21978.585 + }, + { + "epoch": 2.487424, + "grad_norm": 1.7890625, + "learning_rate": 4.484891184181041e-05, + "loss": 0.5297726631164551, + "num_input_tokens_seen": 271548736, + "step": 2430, + "train_runtime": 12354.5613, + "train_tokens_per_second": 21979.634 + }, + { + "epoch": 2.497664, + "grad_norm": 1.7734375, + "learning_rate": 4.483088067803662e-05, + "loss": 0.5376968383789062, + "num_input_tokens_seen": 272650816, + "step": 2440, + "train_runtime": 12403.0155, + "train_tokens_per_second": 21982.623 + }, + { + "epoch": 2.507904, + "grad_norm": 1.7734375, + "learning_rate": 4.481287124466697e-05, + "loss": 0.5285521507263183, + "num_input_tokens_seen": 273744576, + "step": 2450, + "train_runtime": 12451.1849, + "train_tokens_per_second": 21985.424 + }, + { + "epoch": 2.518144, + "grad_norm": 1.6640625, + "learning_rate": 4.479488349808885e-05, + "loss": 0.5299195289611817, + "num_input_tokens_seen": 274848640, + "step": 2460, + "train_runtime": 12503.3338, + "train_tokens_per_second": 21982.029 + }, + { + "epoch": 2.528384, + "grad_norm": 1.71875, + "learning_rate": 4.4776917394812114e-05, + "loss": 0.5259761810302734, + "num_input_tokens_seen": 275977600, + "step": 2470, + "train_runtime": 12556.4141, + "train_tokens_per_second": 21979.014 + }, + { + "epoch": 2.538624, + "grad_norm": 1.765625, + "learning_rate": 4.475897289146862e-05, + "loss": 0.5346611022949219, + "num_input_tokens_seen": 277115072, + "step": 2480, + "train_runtime": 12607.8823, + "train_tokens_per_second": 21979.51 + }, + { + "epoch": 2.548864, + "grad_norm": 1.7109375, + "learning_rate": 4.4741049944811806e-05, + "loss": 0.5369565963745118, + "num_input_tokens_seen": 278210816, + "step": 2490, + "train_runtime": 12655.8676, + "train_tokens_per_second": 21982.753 + }, + { + "epoch": 2.559104, + "grad_norm": 1.703125, + "learning_rate": 4.472314851171621e-05, + "loss": 0.5275493621826172, + "num_input_tokens_seen": 279331968, + "step": 2500, + "train_runtime": 12707.0061, + "train_tokens_per_second": 21982.516 + }, + { + "epoch": 2.569344, + "grad_norm": 1.6796875, + "learning_rate": 4.4705268549177084e-05, + "loss": 0.5220746040344239, + "num_input_tokens_seen": 280461056, + "step": 2510, + "train_runtime": 12758.6493, + "train_tokens_per_second": 21982.033 + }, + { + "epoch": 2.579584, + "grad_norm": 1.7890625, + "learning_rate": 4.468741001430989e-05, + "loss": 0.5226367950439453, + "num_input_tokens_seen": 281572608, + "step": 2520, + "train_runtime": 12809.5121, + "train_tokens_per_second": 21981.525 + }, + { + "epoch": 2.589824, + "grad_norm": 1.6328125, + "learning_rate": 4.466957286434997e-05, + "loss": 0.515837574005127, + "num_input_tokens_seen": 282737216, + "step": 2530, + "train_runtime": 12865.3707, + "train_tokens_per_second": 21976.609 + }, + { + "epoch": 2.600064, + "grad_norm": 1.8046875, + "learning_rate": 4.4651757056652e-05, + "loss": 0.5301560401916504, + "num_input_tokens_seen": 283827008, + "step": 2540, + "train_runtime": 12913.3246, + "train_tokens_per_second": 21979.391 + }, + { + "epoch": 2.610304, + "grad_norm": 1.65625, + "learning_rate": 4.463396254868968e-05, + "loss": 0.5253063201904297, + "num_input_tokens_seen": 284983680, + "step": 2550, + "train_runtime": 12967.0493, + "train_tokens_per_second": 21977.527 + }, + { + "epoch": 2.6205439999999998, + "grad_norm": 1.765625, + "learning_rate": 4.461618929805519e-05, + "loss": 0.5248900890350342, + "num_input_tokens_seen": 286077504, + "step": 2560, + "train_runtime": 13015.9506, + "train_tokens_per_second": 21978.994 + }, + { + "epoch": 2.6307840000000002, + "grad_norm": 1.5390625, + "learning_rate": 4.459843726245888e-05, + "loss": 0.53463134765625, + "num_input_tokens_seen": 287207168, + "step": 2570, + "train_runtime": 13067.1302, + "train_tokens_per_second": 21979.361 + }, + { + "epoch": 2.641024, + "grad_norm": 1.9140625, + "learning_rate": 4.458070639972875e-05, + "loss": 0.5340803146362305, + "num_input_tokens_seen": 288307776, + "step": 2580, + "train_runtime": 13115.1727, + "train_tokens_per_second": 21982.766 + }, + { + "epoch": 2.651264, + "grad_norm": 1.6875, + "learning_rate": 4.456299666781007e-05, + "loss": 0.5143898963928223, + "num_input_tokens_seen": 289491392, + "step": 2590, + "train_runtime": 13173.1381, + "train_tokens_per_second": 21975.887 + }, + { + "epoch": 2.661504, + "grad_norm": 1.7421875, + "learning_rate": 4.4545308024764984e-05, + "loss": 0.5279585838317871, + "num_input_tokens_seen": 290588288, + "step": 2600, + "train_runtime": 13222.1712, + "train_tokens_per_second": 21977.35 + }, + { + "epoch": 2.671744, + "grad_norm": 1.7890625, + "learning_rate": 4.452764042877207e-05, + "loss": 0.5243070602416993, + "num_input_tokens_seen": 291706432, + "step": 2610, + "train_runtime": 13274.0598, + "train_tokens_per_second": 21975.676 + }, + { + "epoch": 2.681984, + "grad_norm": 1.7265625, + "learning_rate": 4.45099938381259e-05, + "loss": 0.525565767288208, + "num_input_tokens_seen": 292825664, + "step": 2620, + "train_runtime": 13324.5762, + "train_tokens_per_second": 21976.359 + }, + { + "epoch": 2.692224, + "grad_norm": 1.828125, + "learning_rate": 4.449236821123667e-05, + "loss": 0.5281280040740967, + "num_input_tokens_seen": 293933504, + "step": 2630, + "train_runtime": 13375.6314, + "train_tokens_per_second": 21975.299 + }, + { + "epoch": 2.702464, + "grad_norm": 1.65625, + "learning_rate": 4.447476350662976e-05, + "loss": 0.5201663017272949, + "num_input_tokens_seen": 295024000, + "step": 2640, + "train_runtime": 13424.3426, + "train_tokens_per_second": 21976.793 + }, + { + "epoch": 2.712704, + "grad_norm": 1.859375, + "learning_rate": 4.4457179682945346e-05, + "loss": 0.5273939609527588, + "num_input_tokens_seen": 296126464, + "step": 2650, + "train_runtime": 13473.9229, + "train_tokens_per_second": 21977.747 + }, + { + "epoch": 2.722944, + "grad_norm": 1.7109375, + "learning_rate": 4.443961669893798e-05, + "loss": 0.5174345970153809, + "num_input_tokens_seen": 297259712, + "step": 2660, + "train_runtime": 13524.4385, + "train_tokens_per_second": 21979.449 + }, + { + "epoch": 2.733184, + "grad_norm": 1.8046875, + "learning_rate": 4.4422074513476155e-05, + "loss": 0.528529167175293, + "num_input_tokens_seen": 298367680, + "step": 2670, + "train_runtime": 13574.6209, + "train_tokens_per_second": 21979.817 + }, + { + "epoch": 2.743424, + "grad_norm": 1.859375, + "learning_rate": 4.4404553085541955e-05, + "loss": 0.5198683738708496, + "num_input_tokens_seen": 299497536, + "step": 2680, + "train_runtime": 13624.3324, + "train_tokens_per_second": 21982.548 + }, + { + "epoch": 2.753664, + "grad_norm": 1.71875, + "learning_rate": 4.438705237423063e-05, + "loss": 0.5379956245422364, + "num_input_tokens_seen": 300608512, + "step": 2690, + "train_runtime": 13673.7031, + "train_tokens_per_second": 21984.426 + }, + { + "epoch": 2.763904, + "grad_norm": 1.734375, + "learning_rate": 4.436957233875017e-05, + "loss": 0.5276507377624512, + "num_input_tokens_seen": 301702976, + "step": 2700, + "train_runtime": 13722.51, + "train_tokens_per_second": 21985.991 + }, + { + "epoch": 2.774144, + "grad_norm": 1.734375, + "learning_rate": 4.4352112938420956e-05, + "loss": 0.5213168144226075, + "num_input_tokens_seen": 302851072, + "step": 2710, + "train_runtime": 13775.4093, + "train_tokens_per_second": 21984.906 + }, + { + "epoch": 2.784384, + "grad_norm": 1.78125, + "learning_rate": 4.433467413267529e-05, + "loss": 0.5262949943542481, + "num_input_tokens_seen": 303937408, + "step": 2720, + "train_runtime": 13823.4103, + "train_tokens_per_second": 21987.151 + }, + { + "epoch": 2.7946239999999998, + "grad_norm": 1.8125, + "learning_rate": 4.431725588105708e-05, + "loss": 0.5267552852630615, + "num_input_tokens_seen": 305070272, + "step": 2730, + "train_runtime": 13876.6879, + "train_tokens_per_second": 21984.372 + }, + { + "epoch": 2.8048640000000002, + "grad_norm": 1.7265625, + "learning_rate": 4.4299858143221377e-05, + "loss": 0.5038492202758789, + "num_input_tokens_seen": 306205888, + "step": 2740, + "train_runtime": 13927.2204, + "train_tokens_per_second": 21986.145 + }, + { + "epoch": 2.815104, + "grad_norm": 1.8515625, + "learning_rate": 4.4282480878934065e-05, + "loss": 0.5188664436340332, + "num_input_tokens_seen": 307332736, + "step": 2750, + "train_runtime": 13978.2805, + "train_tokens_per_second": 21986.448 + }, + { + "epoch": 2.825344, + "grad_norm": 1.671875, + "learning_rate": 4.4265124048071346e-05, + "loss": 0.5213587760925293, + "num_input_tokens_seen": 308433152, + "step": 2760, + "train_runtime": 14028.0474, + "train_tokens_per_second": 21986.891 + }, + { + "epoch": 2.835584, + "grad_norm": 1.7890625, + "learning_rate": 4.4247787610619477e-05, + "loss": 0.5300797462463379, + "num_input_tokens_seen": 309529920, + "step": 2770, + "train_runtime": 14076.4147, + "train_tokens_per_second": 21989.258 + }, + { + "epoch": 2.845824, + "grad_norm": 1.859375, + "learning_rate": 4.42304715266743e-05, + "loss": 0.5275647640228271, + "num_input_tokens_seen": 310620032, + "step": 2780, + "train_runtime": 14125.9713, + "train_tokens_per_second": 21989.287 + }, + { + "epoch": 2.856064, + "grad_norm": 1.828125, + "learning_rate": 4.421317575644092e-05, + "loss": 0.5151683807373046, + "num_input_tokens_seen": 311744704, + "step": 2790, + "train_runtime": 14177.2729, + "train_tokens_per_second": 21989.046 + }, + { + "epoch": 2.866304, + "grad_norm": 1.890625, + "learning_rate": 4.419590026023325e-05, + "loss": 0.5277010440826416, + "num_input_tokens_seen": 312862848, + "step": 2800, + "train_runtime": 14228.7469, + "train_tokens_per_second": 21988.082 + }, + { + "epoch": 2.876544, + "grad_norm": 1.765625, + "learning_rate": 4.417864499847368e-05, + "loss": 0.5130796432495117, + "num_input_tokens_seen": 313974848, + "step": 2810, + "train_runtime": 14277.693, + "train_tokens_per_second": 21990.587 + }, + { + "epoch": 2.886784, + "grad_norm": 1.796875, + "learning_rate": 4.4161409931692676e-05, + "loss": 0.5243385791778564, + "num_input_tokens_seen": 315060032, + "step": 2820, + "train_runtime": 14325.4746, + "train_tokens_per_second": 21992.991 + }, + { + "epoch": 2.897024, + "grad_norm": 1.734375, + "learning_rate": 4.414419502052841e-05, + "loss": 0.5174403667449952, + "num_input_tokens_seen": 316182976, + "step": 2830, + "train_runtime": 14374.9907, + "train_tokens_per_second": 21995.352 + }, + { + "epoch": 2.907264, + "grad_norm": 1.78125, + "learning_rate": 4.412700022572637e-05, + "loss": 0.5114859580993653, + "num_input_tokens_seen": 317286656, + "step": 2840, + "train_runtime": 14423.521, + "train_tokens_per_second": 21997.864 + }, + { + "epoch": 2.917504, + "grad_norm": 1.890625, + "learning_rate": 4.410982550813902e-05, + "loss": 0.5237509727478027, + "num_input_tokens_seen": 318388160, + "step": 2850, + "train_runtime": 14473.9697, + "train_tokens_per_second": 21997.293 + }, + { + "epoch": 2.927744, + "grad_norm": 1.78125, + "learning_rate": 4.409267082872535e-05, + "loss": 0.5249311447143554, + "num_input_tokens_seen": 319492928, + "step": 2860, + "train_runtime": 14523.7571, + "train_tokens_per_second": 21997.953 + }, + { + "epoch": 2.937984, + "grad_norm": 1.828125, + "learning_rate": 4.407553614855059e-05, + "loss": 0.515770959854126, + "num_input_tokens_seen": 320622336, + "step": 2870, + "train_runtime": 14577.3266, + "train_tokens_per_second": 21994.591 + }, + { + "epoch": 2.9482239999999997, + "grad_norm": 1.796875, + "learning_rate": 4.405842142878579e-05, + "loss": 0.5268994331359863, + "num_input_tokens_seen": 321735488, + "step": 2880, + "train_runtime": 14627.1239, + "train_tokens_per_second": 21995.813 + }, + { + "epoch": 2.958464, + "grad_norm": 1.765625, + "learning_rate": 4.404132663070745e-05, + "loss": 0.5111500740051269, + "num_input_tokens_seen": 322857728, + "step": 2890, + "train_runtime": 14678.6828, + "train_tokens_per_second": 21995.007 + }, + { + "epoch": 2.968704, + "grad_norm": 1.78125, + "learning_rate": 4.402425171569716e-05, + "loss": 0.5016695499420166, + "num_input_tokens_seen": 323996736, + "step": 2900, + "train_runtime": 14731.4275, + "train_tokens_per_second": 21993.574 + }, + { + "epoch": 2.9789440000000003, + "grad_norm": 1.75, + "learning_rate": 4.400719664524127e-05, + "loss": 0.5131624221801758, + "num_input_tokens_seen": 325130944, + "step": 2910, + "train_runtime": 14783.7526, + "train_tokens_per_second": 21992.45 + }, + { + "epoch": 2.989184, + "grad_norm": 1.8203125, + "learning_rate": 4.399016138093044e-05, + "loss": 0.5178108215332031, + "num_input_tokens_seen": 326241984, + "step": 2920, + "train_runtime": 14834.9455, + "train_tokens_per_second": 21991.451 + }, + { + "epoch": 2.999424, + "grad_norm": 1.765625, + "learning_rate": 4.397314588445937e-05, + "loss": 0.5120769500732422, + "num_input_tokens_seen": 327359424, + "step": 2930, + "train_runtime": 14885.5036, + "train_tokens_per_second": 21991.827 + }, + { + "epoch": 3.0, + "eval_loss": 0.8352677822113037, + "eval_runtime": 1.2873, + "eval_samples_per_second": 774.519, + "eval_steps_per_second": 6.215, + "num_input_tokens_seen": 327416064, + "step": 2931 + }, + { + "epoch": 3.0, + "num_input_tokens_seen": 327416064, + "step": 2931, + "total_flos": 5.451032989129507e+18, + "train_loss": 0.7516615965776337, + "train_runtime": 14909.6648, + "train_samples_per_second": 201.212, + "train_steps_per_second": 0.197 + } + ], + "logging_steps": 10, + "max_steps": 2931, + "num_input_tokens_seen": 327416064, + "num_train_epochs": 3, + "save_steps": 5000, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 5.451032989129507e+18, + "train_batch_size": 8, + "trial_name": null, + "trial_params": null +} diff --git a/training_args.bin b/training_args.bin new file mode 100644 index 0000000..3436b9d --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:776eabd46fa90ef7397ccca461f1f5a0dcfe8eb44787f01e6fddf3bc4b9e22e0 +size 5905 diff --git a/training_eval_loss.png b/training_eval_loss.png new file mode 100644 index 0000000..1a911a2 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..45aad52 Binary files /dev/null and b/training_loss.png differ