commit 3c1043c33d65b66d2a9c0e045d598ddc549ab636 Author: ModelHub XC Date: Tue Jul 7 19:44:18 2026 +0800 初始化项目,由ModelHub XC社区提供模型 Model: baban/QwenTranslate_English_German 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..b5ccef1 --- /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_German + results: [] +--- + + + +# MT_En_German + +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_German dataset. +It achieves the following results on the evaluation set: +- Loss: 0.7843 +- Num Input Tokens Seen: 571898240 + +## 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..bf7cbd5 --- /dev/null +++ b/all_results.json @@ -0,0 +1,13 @@ +{ + "epoch": 3.0, + "eval_loss": 0.7842721939086914, + "eval_runtime": 0.9863, + "eval_samples_per_second": 1010.809, + "eval_steps_per_second": 8.111, + "num_input_tokens_seen": 571898240, + "total_flos": 9.521329376839336e+18, + "train_loss": 0.6782078807039174, + "train_runtime": 17469.688, + "train_samples_per_second": 171.726, + "train_steps_per_second": 0.168 +} \ 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..b52e9ba --- /dev/null +++ b/eval_results.json @@ -0,0 +1,8 @@ +{ + "epoch": 3.0, + "eval_loss": 0.7842721939086914, + "eval_runtime": 0.9863, + "eval_samples_per_second": 1010.809, + "eval_steps_per_second": 8.111, + "num_input_tokens_seen": 571898240 +} \ 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..753fe82 --- /dev/null +++ b/model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cd12798e3ca46a5f53bedb75413c0943c257304c70e8face0119672ac292bb8f +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..87fa0a4 --- /dev/null +++ b/train_results.json @@ -0,0 +1,9 @@ +{ + "epoch": 3.0, + "num_input_tokens_seen": 571898240, + "total_flos": 9.521329376839336e+18, + "train_loss": 0.6782078807039174, + "train_runtime": 17469.688, + "train_samples_per_second": 171.726, + "train_steps_per_second": 0.168 +} \ No newline at end of file diff --git a/trainer_log.jsonl b/trainer_log.jsonl new file mode 100644 index 0000000..74108f7 --- /dev/null +++ b/trainer_log.jsonl @@ -0,0 +1,295 @@ +{"current_steps": 10, "total_steps": 2931, "loss": 1.1904024124145507, "lr": 4.9977515176118345e-05, "epoch": 0.01024, "percentage": 0.34, "elapsed_time": "0:01:02", "remaining_time": "5:02:44", "throughput": 31657.35, "total_tokens": 1968640} +{"current_steps": 20, "total_steps": 2931, "loss": 0.9794286727905274, "lr": 4.9952567580506e-05, "epoch": 0.02048, "percentage": 0.68, "elapsed_time": "0:02:02", "remaining_time": "4:57:43", "throughput": 31951.53, "total_tokens": 3921472} +{"current_steps": 30, "total_steps": 2931, "loss": 0.9411283493041992, "lr": 4.992765730738634e-05, "epoch": 0.03072, "percentage": 1.02, "elapsed_time": "0:03:02", "remaining_time": "4:54:10", "throughput": 32300.11, "total_tokens": 5895808} +{"current_steps": 40, "total_steps": 2931, "loss": 0.9153067588806152, "lr": 4.9902784263792476e-05, "epoch": 0.04096, "percentage": 1.36, "elapsed_time": "0:04:01", "remaining_time": "4:50:36", "throughput": 32453.01, "total_tokens": 7829440} +{"current_steps": 50, "total_steps": 2931, "loss": 0.9085355758666992, "lr": 4.987794835708133e-05, "epoch": 0.0512, "percentage": 1.71, "elapsed_time": "0:05:00", "remaining_time": "4:48:24", "throughput": 32568.35, "total_tokens": 9781120} +{"current_steps": 60, "total_steps": 2931, "loss": 0.882135009765625, "lr": 4.985314949493234e-05, "epoch": 0.06144, "percentage": 2.05, "elapsed_time": "0:05:59", "remaining_time": "4:47:02", "throughput": 32562.58, "total_tokens": 11720320} +{"current_steps": 70, "total_steps": 2931, "loss": 0.8795860290527344, "lr": 4.982838758534584e-05, "epoch": 0.07168, "percentage": 2.39, "elapsed_time": "0:06:59", "remaining_time": "4:45:42", "throughput": 32615.2, "total_tokens": 13679552} +{"current_steps": 80, "total_steps": 2931, "loss": 0.8663822174072265, "lr": 4.980366253664179e-05, "epoch": 0.08192, "percentage": 2.73, "elapsed_time": "0:07:58", "remaining_time": "4:44:00", "throughput": 32677.81, "total_tokens": 15625216} +{"current_steps": 90, "total_steps": 2931, "loss": 0.8581996917724609, "lr": 4.977897425745825e-05, "epoch": 0.09216, "percentage": 3.07, "elapsed_time": "0:08:57", "remaining_time": "4:42:58", "throughput": 32679.22, "total_tokens": 17576768} +{"current_steps": 100, "total_steps": 2931, "loss": 0.8557216644287109, "lr": 4.975432265674997e-05, "epoch": 0.1024, "percentage": 3.41, "elapsed_time": "0:09:57", "remaining_time": "4:41:50", "throughput": 32692.3, "total_tokens": 19528128} +{"current_steps": 110, "total_steps": 2931, "loss": 0.8523456573486328, "lr": 4.972970764378705e-05, "epoch": 0.11264, "percentage": 3.75, "elapsed_time": "0:10:57", "remaining_time": "4:41:07", "throughput": 32683.07, "total_tokens": 21496192} +{"current_steps": 120, "total_steps": 2931, "loss": 0.8425041198730469, "lr": 4.970512912815344e-05, "epoch": 0.12288, "percentage": 4.09, "elapsed_time": "0:11:58", "remaining_time": "4:40:42", "throughput": 32628.18, "total_tokens": 23459072} +{"current_steps": 130, "total_steps": 2931, "loss": 0.8385295867919922, "lr": 4.968058701974564e-05, "epoch": 0.13312, "percentage": 4.44, "elapsed_time": "0:12:58", "remaining_time": "4:39:44", "throughput": 32625.31, "total_tokens": 25415040} +{"current_steps": 140, "total_steps": 2931, "loss": 0.8344181060791016, "lr": 4.96560812287712e-05, "epoch": 0.14336, "percentage": 4.78, "elapsed_time": "0:13:57", "remaining_time": "4:38:21", "throughput": 32642.78, "total_tokens": 27346432} +{"current_steps": 150, "total_steps": 2931, "loss": 0.8336277961730957, "lr": 4.963161166574748e-05, "epoch": 0.1536, "percentage": 5.12, "elapsed_time": "0:14:57", "remaining_time": "4:37:16", "throughput": 32654.06, "total_tokens": 29300736} +{"current_steps": 160, "total_steps": 2931, "loss": 0.8235919952392579, "lr": 4.960717824150013e-05, "epoch": 0.16384, "percentage": 5.46, "elapsed_time": "0:15:55", "remaining_time": "4:35:45", "throughput": 32703.34, "total_tokens": 31242688} +{"current_steps": 170, "total_steps": 2931, "loss": 0.8172248840332031, "lr": 4.9582780867161893e-05, "epoch": 0.17408, "percentage": 5.8, "elapsed_time": "0:16:55", "remaining_time": "4:34:50", "throughput": 32695.25, "total_tokens": 33196160} +{"current_steps": 180, "total_steps": 2931, "loss": 0.8182367324829102, "lr": 4.955841945417105e-05, "epoch": 0.18432, "percentage": 6.14, "elapsed_time": "0:17:54", "remaining_time": "4:33:37", "throughput": 32703.91, "total_tokens": 35131136} +{"current_steps": 190, "total_steps": 2931, "loss": 0.8155080795288085, "lr": 4.953409391427024e-05, "epoch": 0.19456, "percentage": 6.48, "elapsed_time": "0:18:54", "remaining_time": "4:32:44", "throughput": 32689.34, "total_tokens": 37082176} +{"current_steps": 200, "total_steps": 2931, "loss": 0.8091486930847168, "lr": 4.950980415950502e-05, "epoch": 0.2048, "percentage": 6.82, "elapsed_time": "0:19:55", "remaining_time": "4:31:58", "throughput": 32685.03, "total_tokens": 39060608} +{"current_steps": 210, "total_steps": 2931, "loss": 0.8105871200561523, "lr": 4.9485550102222575e-05, "epoch": 0.21504, "percentage": 7.16, "elapsed_time": "0:20:53", "remaining_time": "4:30:37", "throughput": 32715.38, "total_tokens": 40996800} +{"current_steps": 220, "total_steps": 2931, "loss": 0.8007206916809082, "lr": 4.946133165507037e-05, "epoch": 0.22528, "percentage": 7.51, "elapsed_time": "0:21:51", "remaining_time": "4:29:20", "throughput": 32734.44, "total_tokens": 42929728} +{"current_steps": 230, "total_steps": 2931, "loss": 0.7916994571685791, "lr": 4.943714873099483e-05, "epoch": 0.23552, "percentage": 7.85, "elapsed_time": "0:22:50", "remaining_time": "4:28:14", "throughput": 32756.79, "total_tokens": 44892288} +{"current_steps": 240, "total_steps": 2931, "loss": 0.7964269638061523, "lr": 4.9413001243240024e-05, "epoch": 0.24576, "percentage": 8.19, "elapsed_time": "0:23:51", "remaining_time": "4:27:35", "throughput": 32705.25, "total_tokens": 46831872} +{"current_steps": 250, "total_steps": 2931, "loss": 0.7994860172271728, "lr": 4.938888910534637e-05, "epoch": 0.256, "percentage": 8.53, "elapsed_time": "0:24:50", "remaining_time": "4:26:29", "throughput": 32726.06, "total_tokens": 48793472} +{"current_steps": 260, "total_steps": 2931, "loss": 0.7954794883728027, "lr": 4.936481223114932e-05, "epoch": 0.26624, "percentage": 8.87, "elapsed_time": "0:25:52", "remaining_time": "4:25:49", "throughput": 32697.23, "total_tokens": 50764800} +{"current_steps": 270, "total_steps": 2931, "loss": 0.7914528369903564, "lr": 4.934077053477808e-05, "epoch": 0.27648, "percentage": 9.21, "elapsed_time": "0:26:52", "remaining_time": "4:24:50", "throughput": 32704.29, "total_tokens": 52729152} +{"current_steps": 280, "total_steps": 2931, "loss": 0.7908175468444825, "lr": 4.931676393065431e-05, "epoch": 0.28672, "percentage": 9.55, "elapsed_time": "0:27:52", "remaining_time": "4:23:55", "throughput": 32701.85, "total_tokens": 54695872} +{"current_steps": 290, "total_steps": 2931, "loss": 0.7946252822875977, "lr": 4.929279233349088e-05, "epoch": 0.29696, "percentage": 9.89, "elapsed_time": "0:28:52", "remaining_time": "4:22:53", "throughput": 32707.92, "total_tokens": 56652544} +{"current_steps": 300, "total_steps": 2931, "loss": 0.7933988571166992, "lr": 4.926885565829051e-05, "epoch": 0.3072, "percentage": 10.24, "elapsed_time": "0:29:51", "remaining_time": "4:21:47", "throughput": 32724.61, "total_tokens": 58611264} +{"current_steps": 310, "total_steps": 2931, "loss": 0.7868727684020996, "lr": 4.924495382034461e-05, "epoch": 0.31744, "percentage": 10.58, "elapsed_time": "0:30:49", "remaining_time": "4:20:39", "throughput": 32727.57, "total_tokens": 60538816} +{"current_steps": 320, "total_steps": 2931, "loss": 0.7884368896484375, "lr": 4.9221086735231975e-05, "epoch": 0.32768, "percentage": 10.92, "elapsed_time": "0:31:48", "remaining_time": "4:19:32", "throughput": 32741.62, "total_tokens": 62490560} +{"current_steps": 330, "total_steps": 2931, "loss": 0.7792404174804688, "lr": 4.919725431881751e-05, "epoch": 0.33792, "percentage": 11.26, "elapsed_time": "0:32:49", "remaining_time": "4:18:47", "throughput": 32720.75, "total_tokens": 64459392} +{"current_steps": 340, "total_steps": 2931, "loss": 0.7801454067230225, "lr": 4.917345648725101e-05, "epoch": 0.34816, "percentage": 11.6, "elapsed_time": "0:33:50", "remaining_time": "4:17:53", "throughput": 32710.34, "total_tokens": 66416320} +{"current_steps": 350, "total_steps": 2931, "loss": 0.7781849861145019, "lr": 4.914969315696596e-05, "epoch": 0.3584, "percentage": 11.94, "elapsed_time": "0:34:49", "remaining_time": "4:16:51", "throughput": 32719.73, "total_tokens": 68379008} +{"current_steps": 360, "total_steps": 2931, "loss": 0.7765862941741943, "lr": 4.912596424467818e-05, "epoch": 0.36864, "percentage": 12.28, "elapsed_time": "0:35:49", "remaining_time": "4:15:51", "throughput": 32717.83, "total_tokens": 70328768} +{"current_steps": 370, "total_steps": 2931, "loss": 0.7737765312194824, "lr": 4.910226966738475e-05, "epoch": 0.37888, "percentage": 12.62, "elapsed_time": "0:36:49", "remaining_time": "4:14:52", "throughput": 32715.46, "total_tokens": 72280640} +{"current_steps": 380, "total_steps": 2931, "loss": 0.777079963684082, "lr": 4.9078609342362666e-05, "epoch": 0.38912, "percentage": 12.96, "elapsed_time": "0:37:48", "remaining_time": "4:13:46", "throughput": 32720.5, "total_tokens": 74217408} +{"current_steps": 390, "total_steps": 2931, "loss": 0.7765676498413085, "lr": 4.905498318716775e-05, "epoch": 0.39936, "percentage": 13.31, "elapsed_time": "0:38:47", "remaining_time": "4:12:42", "throughput": 32731.94, "total_tokens": 76174784} +{"current_steps": 400, "total_steps": 2931, "loss": 0.7736555099487304, "lr": 4.9031391119633295e-05, "epoch": 0.4096, "percentage": 13.65, "elapsed_time": "0:39:48", "remaining_time": "4:11:53", "throughput": 32723.02, "total_tokens": 78158016} +{"current_steps": 410, "total_steps": 2931, "loss": 0.7716507911682129, "lr": 4.9007833057869e-05, "epoch": 0.41984, "percentage": 13.99, "elapsed_time": "0:40:48", "remaining_time": "4:10:56", "throughput": 32716.45, "total_tokens": 80110400} +{"current_steps": 420, "total_steps": 2931, "loss": 0.7716224193572998, "lr": 4.898430892025967e-05, "epoch": 0.43008, "percentage": 14.33, "elapsed_time": "0:41:47", "remaining_time": "4:09:50", "throughput": 32727.18, "total_tokens": 82056960} +{"current_steps": 430, "total_steps": 2931, "loss": 0.7601117134094239, "lr": 4.896081862546415e-05, "epoch": 0.44032, "percentage": 14.67, "elapsed_time": "0:42:46", "remaining_time": "4:08:46", "throughput": 32730.85, "total_tokens": 84000832} +{"current_steps": 440, "total_steps": 2931, "loss": 0.7689449310302734, "lr": 4.8937362092414e-05, "epoch": 0.45056, "percentage": 15.01, "elapsed_time": "0:43:46", "remaining_time": "4:07:49", "throughput": 32721.78, "total_tokens": 85945600} +{"current_steps": 450, "total_steps": 2931, "loss": 0.7667889595031738, "lr": 4.891393924031244e-05, "epoch": 0.4608, "percentage": 15.35, "elapsed_time": "0:44:44", "remaining_time": "4:06:42", "throughput": 32733.75, "total_tokens": 87886144} +{"current_steps": 460, "total_steps": 2931, "loss": 0.7657638549804687, "lr": 4.8890549988633095e-05, "epoch": 0.47104, "percentage": 15.69, "elapsed_time": "0:45:44", "remaining_time": "4:05:44", "throughput": 32731.36, "total_tokens": 89839744} +{"current_steps": 470, "total_steps": 2931, "loss": 0.7696039199829101, "lr": 4.8867194257118907e-05, "epoch": 0.48128, "percentage": 16.04, "elapsed_time": "0:46:44", "remaining_time": "4:04:46", "throughput": 32732.07, "total_tokens": 91805440} +{"current_steps": 480, "total_steps": 2931, "loss": 0.7688661575317383, "lr": 4.884387196578093e-05, "epoch": 0.49152, "percentage": 16.38, "elapsed_time": "0:47:42", "remaining_time": "4:03:38", "throughput": 32754.1, "total_tokens": 93770240} +{"current_steps": 490, "total_steps": 2931, "loss": 0.7624212265014648, "lr": 4.882058303489718e-05, "epoch": 0.50176, "percentage": 16.72, "elapsed_time": "0:48:41", "remaining_time": "4:02:32", "throughput": 32763.06, "total_tokens": 95708096} +{"current_steps": 500, "total_steps": 2931, "loss": 0.7568993091583252, "lr": 4.8797327385011496e-05, "epoch": 0.512, "percentage": 17.06, "elapsed_time": "0:49:42", "remaining_time": "4:01:39", "throughput": 32754.27, "total_tokens": 97676928} +{"current_steps": 510, "total_steps": 2931, "loss": 0.7545355796813965, "lr": 4.8774104936932425e-05, "epoch": 0.52224, "percentage": 17.4, "elapsed_time": "0:50:41", "remaining_time": "4:00:38", "throughput": 32755.02, "total_tokens": 99627776} +{"current_steps": 520, "total_steps": 2931, "loss": 0.7593323707580566, "lr": 4.8750915611732076e-05, "epoch": 0.53248, "percentage": 17.74, "elapsed_time": "0:51:41", "remaining_time": "3:59:42", "throughput": 32742.95, "total_tokens": 101565696} +{"current_steps": 530, "total_steps": 2931, "loss": 0.760127067565918, "lr": 4.8727759330744986e-05, "epoch": 0.54272, "percentage": 18.08, "elapsed_time": "0:52:41", "remaining_time": "3:58:42", "throughput": 32740.02, "total_tokens": 103508992} +{"current_steps": 540, "total_steps": 2931, "loss": 0.7561708450317383, "lr": 4.870463601556696e-05, "epoch": 0.55296, "percentage": 18.42, "elapsed_time": "0:53:40", "remaining_time": "3:57:41", "throughput": 32741.67, "total_tokens": 105456832} +{"current_steps": 550, "total_steps": 2931, "loss": 0.7509193420410156, "lr": 4.8681545588054075e-05, "epoch": 0.5632, "percentage": 18.76, "elapsed_time": "0:54:40", "remaining_time": "3:56:43", "throughput": 32739.41, "total_tokens": 107417344} +{"current_steps": 560, "total_steps": 2931, "loss": 0.7538685321807861, "lr": 4.8658487970321404e-05, "epoch": 0.57344, "percentage": 19.11, "elapsed_time": "0:55:41", "remaining_time": "3:55:47", "throughput": 32733.02, "total_tokens": 109377984} +{"current_steps": 570, "total_steps": 2931, "loss": 0.7577058792114257, "lr": 4.863546308474209e-05, "epoch": 0.58368, "percentage": 19.45, "elapsed_time": "0:56:41", "remaining_time": "3:54:50", "throughput": 32736.08, "total_tokens": 111360832} +{"current_steps": 580, "total_steps": 2931, "loss": 0.7539366722106934, "lr": 4.86124708539461e-05, "epoch": 0.59392, "percentage": 19.79, "elapsed_time": "0:57:42", "remaining_time": "3:53:56", "throughput": 32728.66, "total_tokens": 113337536} +{"current_steps": 590, "total_steps": 2931, "loss": 0.7477821826934814, "lr": 4.8589511200819216e-05, "epoch": 0.60416, "percentage": 20.13, "elapsed_time": "0:58:41", "remaining_time": "3:52:51", "throughput": 32735.21, "total_tokens": 115270336} +{"current_steps": 600, "total_steps": 2931, "loss": 0.7609129905700683, "lr": 4.8566584048501926e-05, "epoch": 0.6144, "percentage": 20.47, "elapsed_time": "0:59:40", "remaining_time": "3:51:50", "throughput": 32737.93, "total_tokens": 117224512} +{"current_steps": 610, "total_steps": 2931, "loss": 0.7430164337158203, "lr": 4.854368932038835e-05, "epoch": 0.62464, "percentage": 20.81, "elapsed_time": "1:00:40", "remaining_time": "3:50:52", "throughput": 32731.88, "total_tokens": 119167552} +{"current_steps": 620, "total_steps": 2931, "loss": 0.7435213088989258, "lr": 4.8520826940125144e-05, "epoch": 0.63488, "percentage": 21.15, "elapsed_time": "1:01:40", "remaining_time": "3:49:53", "throughput": 32731.59, "total_tokens": 121128896} +{"current_steps": 630, "total_steps": 2931, "loss": 0.7433882713317871, "lr": 4.849799683161046e-05, "epoch": 0.64512, "percentage": 21.49, "elapsed_time": "1:02:40", "remaining_time": "3:48:54", "throughput": 32731.55, "total_tokens": 123088256} +{"current_steps": 640, "total_steps": 2931, "loss": 0.7475492477416992, "lr": 4.8475198918992835e-05, "epoch": 0.65536, "percentage": 21.84, "elapsed_time": "1:03:39", "remaining_time": "3:47:54", "throughput": 32733.16, "total_tokens": 125038592} +{"current_steps": 650, "total_steps": 2931, "loss": 0.7479681968688965, "lr": 4.845243312667023e-05, "epoch": 0.6656, "percentage": 22.18, "elapsed_time": "1:04:39", "remaining_time": "3:46:54", "throughput": 32729.64, "total_tokens": 126982208} +{"current_steps": 660, "total_steps": 2931, "loss": 0.7469013214111329, "lr": 4.842969937928884e-05, "epoch": 0.67584, "percentage": 22.52, "elapsed_time": "1:05:36", "remaining_time": "3:45:46", "throughput": 32743.97, "total_tokens": 128906880} +{"current_steps": 670, "total_steps": 2931, "loss": 0.7478948593139648, "lr": 4.840699760174217e-05, "epoch": 0.68608, "percentage": 22.86, "elapsed_time": "1:06:36", "remaining_time": "3:44:47", "throughput": 32746.97, "total_tokens": 130882496} +{"current_steps": 680, "total_steps": 2931, "loss": 0.7461132049560547, "lr": 4.8384327719169906e-05, "epoch": 0.69632, "percentage": 23.2, "elapsed_time": "1:07:35", "remaining_time": "3:43:43", "throughput": 32755.53, "total_tokens": 132826752} +{"current_steps": 690, "total_steps": 2931, "loss": 0.7483083724975585, "lr": 4.836168965695694e-05, "epoch": 0.70656, "percentage": 23.54, "elapsed_time": "1:08:35", "remaining_time": "3:42:45", "throughput": 32751.96, "total_tokens": 134780160} +{"current_steps": 700, "total_steps": 2931, "loss": 0.7420180320739747, "lr": 4.8339083340732304e-05, "epoch": 0.7168, "percentage": 23.88, "elapsed_time": "1:09:37", "remaining_time": "3:41:52", "throughput": 32739.94, "total_tokens": 136754944} +{"current_steps": 710, "total_steps": 2931, "loss": 0.7432499885559082, "lr": 4.8316508696368154e-05, "epoch": 0.72704, "percentage": 24.22, "elapsed_time": "1:10:35", "remaining_time": "3:40:48", "throughput": 32747.86, "total_tokens": 138696768} +{"current_steps": 720, "total_steps": 2931, "loss": 0.7435853481292725, "lr": 4.8293965649978714e-05, "epoch": 0.73728, "percentage": 24.56, "elapsed_time": "1:11:35", "remaining_time": "3:39:50", "throughput": 32745.72, "total_tokens": 140653120} +{"current_steps": 730, "total_steps": 2931, "loss": 0.7337657928466796, "lr": 4.8271454127919364e-05, "epoch": 0.74752, "percentage": 24.91, "elapsed_time": "1:12:34", "remaining_time": "3:38:49", "throughput": 32749.49, "total_tokens": 142610880} +{"current_steps": 740, "total_steps": 2931, "loss": 0.7439666748046875, "lr": 4.824897405678549e-05, "epoch": 0.75776, "percentage": 25.25, "elapsed_time": "1:13:34", "remaining_time": "3:37:51", "throughput": 32744.47, "total_tokens": 144560448} +{"current_steps": 750, "total_steps": 2931, "loss": 0.7391902923583984, "lr": 4.8226525363411576e-05, "epoch": 0.768, "percentage": 25.59, "elapsed_time": "1:14:35", "remaining_time": "3:36:55", "throughput": 32736.96, "total_tokens": 146522432} +{"current_steps": 760, "total_steps": 2931, "loss": 0.7402773857116699, "lr": 4.820410797487017e-05, "epoch": 0.77824, "percentage": 25.93, "elapsed_time": "1:15:35", "remaining_time": "3:35:55", "throughput": 32734.06, "total_tokens": 148465152} +{"current_steps": 770, "total_steps": 2931, "loss": 0.7375380039215088, "lr": 4.818172181847091e-05, "epoch": 0.78848, "percentage": 26.27, "elapsed_time": "1:16:34", "remaining_time": "3:34:53", "throughput": 32737.54, "total_tokens": 150407040} +{"current_steps": 780, "total_steps": 2931, "loss": 0.7325362205505371, "lr": 4.81593668217595e-05, "epoch": 0.79872, "percentage": 26.61, "elapsed_time": "1:17:34", "remaining_time": "3:33:55", "throughput": 32731.39, "total_tokens": 152340608} +{"current_steps": 790, "total_steps": 2931, "loss": 0.7388778686523437, "lr": 4.813704291251675e-05, "epoch": 0.80896, "percentage": 26.95, "elapsed_time": "1:18:33", "remaining_time": "3:32:53", "throughput": 32735.11, "total_tokens": 154283968} +{"current_steps": 800, "total_steps": 2931, "loss": 0.734052324295044, "lr": 4.811475001875759e-05, "epoch": 0.8192, "percentage": 27.29, "elapsed_time": "1:19:30", "remaining_time": "3:31:47", "throughput": 32740.61, "total_tokens": 156195392} +{"current_steps": 810, "total_steps": 2931, "loss": 0.7394464015960693, "lr": 4.8092488068730105e-05, "epoch": 0.82944, "percentage": 27.64, "elapsed_time": "1:20:29", "remaining_time": "3:30:45", "throughput": 32746.11, "total_tokens": 158142080} +{"current_steps": 820, "total_steps": 2931, "loss": 0.7330001831054688, "lr": 4.807025699091452e-05, "epoch": 0.83968, "percentage": 27.98, "elapsed_time": "1:21:29", "remaining_time": "3:29:47", "throughput": 32742.72, "total_tokens": 160100224} +{"current_steps": 830, "total_steps": 2931, "loss": 0.7365771293640136, "lr": 4.8048056714022325e-05, "epoch": 0.84992, "percentage": 28.32, "elapsed_time": "1:22:26", "remaining_time": "3:28:41", "throughput": 32751.38, "total_tokens": 162014592} +{"current_steps": 840, "total_steps": 2931, "loss": 0.7319395065307617, "lr": 4.802588716699519e-05, "epoch": 0.86016, "percentage": 28.66, "elapsed_time": "1:23:26", "remaining_time": "3:27:43", "throughput": 32749.08, "total_tokens": 163967296} +{"current_steps": 850, "total_steps": 2931, "loss": 0.736358642578125, "lr": 4.8003748279004156e-05, "epoch": 0.8704, "percentage": 29.0, "elapsed_time": "1:24:25", "remaining_time": "3:26:40", "throughput": 32750.45, "total_tokens": 165888128} +{"current_steps": 860, "total_steps": 2931, "loss": 0.728258752822876, "lr": 4.798163997944854e-05, "epoch": 0.88064, "percentage": 29.34, "elapsed_time": "1:25:25", "remaining_time": "3:25:41", "throughput": 32748.75, "total_tokens": 167838720} +{"current_steps": 870, "total_steps": 2931, "loss": 0.7310397624969482, "lr": 4.79595621979551e-05, "epoch": 0.89088, "percentage": 29.68, "elapsed_time": "1:26:24", "remaining_time": "3:24:42", "throughput": 32749.93, "total_tokens": 169801024} +{"current_steps": 880, "total_steps": 2931, "loss": 0.7326676845550537, "lr": 4.793751486437702e-05, "epoch": 0.90112, "percentage": 30.02, "elapsed_time": "1:27:23", "remaining_time": "3:23:41", "throughput": 32753.15, "total_tokens": 171754432} +{"current_steps": 890, "total_steps": 2931, "loss": 0.7294478416442871, "lr": 4.7915497908793064e-05, "epoch": 0.91136, "percentage": 30.37, "elapsed_time": "1:28:23", "remaining_time": "3:22:42", "throughput": 32752.75, "total_tokens": 173709952} +{"current_steps": 900, "total_steps": 2931, "loss": 0.7317845344543457, "lr": 4.7893511261506516e-05, "epoch": 0.9216, "percentage": 30.71, "elapsed_time": "1:29:22", "remaining_time": "3:21:42", "throughput": 32755.64, "total_tokens": 175665088} +{"current_steps": 910, "total_steps": 2931, "loss": 0.7250267028808594, "lr": 4.787155485304435e-05, "epoch": 0.93184, "percentage": 31.05, "elapsed_time": "1:30:21", "remaining_time": "3:20:41", "throughput": 32756.14, "total_tokens": 177597184} +{"current_steps": 920, "total_steps": 2931, "loss": 0.7249749183654786, "lr": 4.784962861415629e-05, "epoch": 0.94208, "percentage": 31.39, "elapsed_time": "1:31:21", "remaining_time": "3:19:41", "throughput": 32756.99, "total_tokens": 179556736} +{"current_steps": 930, "total_steps": 2931, "loss": 0.724465274810791, "lr": 4.7827732475813884e-05, "epoch": 0.95232, "percentage": 31.73, "elapsed_time": "1:32:23", "remaining_time": "3:18:47", "throughput": 32744.9, "total_tokens": 181523520} +{"current_steps": 940, "total_steps": 2931, "loss": 0.7229520797729492, "lr": 4.7805866369209576e-05, "epoch": 0.96256, "percentage": 32.07, "elapsed_time": "1:33:22", "remaining_time": "3:17:46", "throughput": 32745.65, "total_tokens": 183456256} +{"current_steps": 950, "total_steps": 2931, "loss": 0.727871322631836, "lr": 4.778403022575583e-05, "epoch": 0.9728, "percentage": 32.41, "elapsed_time": "1:34:23", "remaining_time": "3:16:49", "throughput": 32740.51, "total_tokens": 185424896} +{"current_steps": 960, "total_steps": 2931, "loss": 0.7236977577209472, "lr": 4.7762223977084195e-05, "epoch": 0.98304, "percentage": 32.75, "elapsed_time": "1:35:23", "remaining_time": "3:15:50", "throughput": 32740.81, "total_tokens": 187388480} +{"current_steps": 970, "total_steps": 2931, "loss": 0.7206878185272216, "lr": 4.774044755504444e-05, "epoch": 0.99328, "percentage": 33.09, "elapsed_time": "1:36:22", "remaining_time": "3:14:49", "throughput": 32744.66, "total_tokens": 189339648} +{"current_steps": 980, "total_steps": 2931, "loss": 0.710568618774414, "lr": 4.7718700891703616e-05, "epoch": 1.003072, "percentage": 33.44, "elapsed_time": "1:37:19", "remaining_time": "3:13:45", "throughput": 32742.54, "total_tokens": 191208640} +{"current_steps": 990, "total_steps": 2931, "loss": 0.6669343948364258, "lr": 4.7696983919345215e-05, "epoch": 1.013312, "percentage": 33.78, "elapsed_time": "1:38:19", "remaining_time": "3:12:46", "throughput": 32743.6, "total_tokens": 193176512} +{"current_steps": 1000, "total_steps": 2931, "loss": 0.6645569801330566, "lr": 4.7675296570468216e-05, "epoch": 1.023552, "percentage": 34.12, "elapsed_time": "1:39:20", "remaining_time": "3:11:49", "throughput": 32740.26, "total_tokens": 195141440} +{"current_steps": 1010, "total_steps": 2931, "loss": 0.6675229549407959, "lr": 4.76536387777863e-05, "epoch": 1.033792, "percentage": 34.46, "elapsed_time": "1:40:19", "remaining_time": "3:10:49", "throughput": 32741.32, "total_tokens": 197092032} +{"current_steps": 1020, "total_steps": 2931, "loss": 0.6588812828063965, "lr": 4.7632010474226915e-05, "epoch": 1.044032, "percentage": 34.8, "elapsed_time": "1:41:19", "remaining_time": "3:09:49", "throughput": 32744.95, "total_tokens": 199066240} +{"current_steps": 1030, "total_steps": 2931, "loss": 0.6624051094055176, "lr": 4.761041159293035e-05, "epoch": 1.054272, "percentage": 35.14, "elapsed_time": "1:42:19", "remaining_time": "3:08:50", "throughput": 32746.45, "total_tokens": 201031808} +{"current_steps": 1040, "total_steps": 2931, "loss": 0.6641523361206054, "lr": 4.7588842067249e-05, "epoch": 1.064512, "percentage": 35.48, "elapsed_time": "1:43:20", "remaining_time": "3:07:54", "throughput": 32735.48, "total_tokens": 202982784} +{"current_steps": 1050, "total_steps": 2931, "loss": 0.6600203037261962, "lr": 4.756730183074637e-05, "epoch": 1.074752, "percentage": 35.82, "elapsed_time": "1:44:21", "remaining_time": "3:06:56", "throughput": 32731.64, "total_tokens": 204943872} +{"current_steps": 1060, "total_steps": 2931, "loss": 0.6627418041229248, "lr": 4.7545790817196314e-05, "epoch": 1.084992, "percentage": 36.17, "elapsed_time": "1:45:21", "remaining_time": "3:05:57", "throughput": 32731.34, "total_tokens": 206907648} +{"current_steps": 1070, "total_steps": 2931, "loss": 0.665987491607666, "lr": 4.752430896058212e-05, "epoch": 1.095232, "percentage": 36.51, "elapsed_time": "1:46:20", "remaining_time": "3:04:58", "throughput": 32732.65, "total_tokens": 208866176} +{"current_steps": 1080, "total_steps": 2931, "loss": 0.6638533115386963, "lr": 4.750285619509567e-05, "epoch": 1.105472, "percentage": 36.85, "elapsed_time": "1:47:21", "remaining_time": "3:04:00", "throughput": 32727.77, "total_tokens": 210821184} +{"current_steps": 1090, "total_steps": 2931, "loss": 0.6705752372741699, "lr": 4.7481432455136644e-05, "epoch": 1.115712, "percentage": 37.19, "elapsed_time": "1:48:19", "remaining_time": "3:02:58", "throughput": 32735.24, "total_tokens": 212772928} +{"current_steps": 1100, "total_steps": 2931, "loss": 0.6647239208221436, "lr": 4.7460037675311584e-05, "epoch": 1.125952, "percentage": 37.53, "elapsed_time": "1:49:18", "remaining_time": "3:01:57", "throughput": 32737.85, "total_tokens": 214713728} +{"current_steps": 1110, "total_steps": 2931, "loss": 0.6651959419250488, "lr": 4.7438671790433126e-05, "epoch": 1.136192, "percentage": 37.87, "elapsed_time": "1:50:17", "remaining_time": "3:00:57", "throughput": 32739.6, "total_tokens": 216668864} +{"current_steps": 1120, "total_steps": 2931, "loss": 0.6629802703857421, "lr": 4.741733473551915e-05, "epoch": 1.146432, "percentage": 38.21, "elapsed_time": "1:51:17", "remaining_time": "2:59:57", "throughput": 32741.24, "total_tokens": 218629056} +{"current_steps": 1130, "total_steps": 2931, "loss": 0.6651780605316162, "lr": 4.7396026445791966e-05, "epoch": 1.156672, "percentage": 38.55, "elapsed_time": "1:52:17", "remaining_time": "2:58:57", "throughput": 32743.03, "total_tokens": 220590464} +{"current_steps": 1140, "total_steps": 2931, "loss": 0.6648248672485352, "lr": 4.737474685667742e-05, "epoch": 1.166912, "percentage": 38.89, "elapsed_time": "1:53:16", "remaining_time": "2:57:58", "throughput": 32745.04, "total_tokens": 222559552} +{"current_steps": 1150, "total_steps": 2931, "loss": 0.6667553901672363, "lr": 4.7353495903804165e-05, "epoch": 1.177152, "percentage": 39.24, "elapsed_time": "1:54:16", "remaining_time": "2:56:58", "throughput": 32743.94, "total_tokens": 224513216} +{"current_steps": 1160, "total_steps": 2931, "loss": 0.6689923286437989, "lr": 4.733227352300277e-05, "epoch": 1.187392, "percentage": 39.58, "elapsed_time": "1:55:15", "remaining_time": "2:55:57", "throughput": 32747.14, "total_tokens": 226447168} +{"current_steps": 1170, "total_steps": 2931, "loss": 0.6637516021728516, "lr": 4.731107965030496e-05, "epoch": 1.197632, "percentage": 39.92, "elapsed_time": "1:56:13", "remaining_time": "2:54:56", "throughput": 32749.67, "total_tokens": 228394368} +{"current_steps": 1180, "total_steps": 2931, "loss": 0.6736277580261231, "lr": 4.728991422194278e-05, "epoch": 1.207872, "percentage": 40.26, "elapsed_time": "1:57:12", "remaining_time": "2:53:55", "throughput": 32754.17, "total_tokens": 230336512} +{"current_steps": 1190, "total_steps": 2931, "loss": 0.6644347190856934, "lr": 4.726877717434773e-05, "epoch": 1.218112, "percentage": 40.6, "elapsed_time": "1:58:11", "remaining_time": "2:52:55", "throughput": 32753.91, "total_tokens": 232273792} +{"current_steps": 1200, "total_steps": 2931, "loss": 0.662445068359375, "lr": 4.724766844415013e-05, "epoch": 1.228352, "percentage": 40.94, "elapsed_time": "1:59:11", "remaining_time": "2:51:55", "throughput": 32754.75, "total_tokens": 234234368} +{"current_steps": 1210, "total_steps": 2931, "loss": 0.6602650165557862, "lr": 4.722658796817813e-05, "epoch": 1.238592, "percentage": 41.28, "elapsed_time": "2:00:11", "remaining_time": "2:50:57", "throughput": 32750.37, "total_tokens": 236184128} +{"current_steps": 1220, "total_steps": 2931, "loss": 0.6661070823669434, "lr": 4.7205535683457044e-05, "epoch": 1.248832, "percentage": 41.62, "elapsed_time": "2:01:12", "remaining_time": "2:49:58", "throughput": 32748.33, "total_tokens": 238148992} +{"current_steps": 1230, "total_steps": 2931, "loss": 0.6634117603302002, "lr": 4.7184511527208484e-05, "epoch": 1.259072, "percentage": 41.97, "elapsed_time": "2:02:12", "remaining_time": "2:48:59", "throughput": 32747.49, "total_tokens": 240113216} +{"current_steps": 1240, "total_steps": 2931, "loss": 0.659183406829834, "lr": 4.7163515436849644e-05, "epoch": 1.269312, "percentage": 42.31, "elapsed_time": "2:03:11", "remaining_time": "2:48:00", "throughput": 32746.03, "total_tokens": 242056192} +{"current_steps": 1250, "total_steps": 2931, "loss": 0.6668022632598877, "lr": 4.714254734999245e-05, "epoch": 1.279552, "percentage": 42.65, "elapsed_time": "2:04:10", "remaining_time": "2:46:59", "throughput": 32748.53, "total_tokens": 244001472} +{"current_steps": 1260, "total_steps": 2931, "loss": 0.6643959999084472, "lr": 4.712160720444284e-05, "epoch": 1.289792, "percentage": 42.99, "elapsed_time": "2:05:10", "remaining_time": "2:46:00", "throughput": 32748.18, "total_tokens": 245958592} +{"current_steps": 1270, "total_steps": 2931, "loss": 0.6648179054260254, "lr": 4.710069493819992e-05, "epoch": 1.300032, "percentage": 43.33, "elapsed_time": "2:06:10", "remaining_time": "2:45:00", "throughput": 32748.86, "total_tokens": 247917248} +{"current_steps": 1280, "total_steps": 2931, "loss": 0.6611180305480957, "lr": 4.70798104894553e-05, "epoch": 1.3102719999999999, "percentage": 43.67, "elapsed_time": "2:07:09", "remaining_time": "2:44:01", "throughput": 32747.92, "total_tokens": 249863104} +{"current_steps": 1290, "total_steps": 2931, "loss": 0.663090705871582, "lr": 4.705895379659219e-05, "epoch": 1.320512, "percentage": 44.01, "elapsed_time": "2:08:10", "remaining_time": "2:43:02", "throughput": 32745.46, "total_tokens": 251816832} +{"current_steps": 1300, "total_steps": 2931, "loss": 0.6652997016906739, "lr": 4.7038124798184766e-05, "epoch": 1.330752, "percentage": 44.35, "elapsed_time": "2:09:10", "remaining_time": "2:42:04", "throughput": 32739.97, "total_tokens": 253758592} +{"current_steps": 1310, "total_steps": 2931, "loss": 0.6616836547851562, "lr": 4.7017323432997304e-05, "epoch": 1.340992, "percentage": 44.69, "elapsed_time": "2:10:09", "remaining_time": "2:41:03", "throughput": 32739.2, "total_tokens": 255687360} +{"current_steps": 1320, "total_steps": 2931, "loss": 0.6657767295837402, "lr": 4.6996549639983506e-05, "epoch": 1.351232, "percentage": 45.04, "elapsed_time": "2:11:07", "remaining_time": "2:40:02", "throughput": 32745.35, "total_tokens": 257640064} +{"current_steps": 1330, "total_steps": 2931, "loss": 0.6621747970581054, "lr": 4.697580335828569e-05, "epoch": 1.361472, "percentage": 45.38, "elapsed_time": "2:12:07", "remaining_time": "2:39:03", "throughput": 32744.45, "total_tokens": 259595136} +{"current_steps": 1340, "total_steps": 2931, "loss": 0.6663068771362305, "lr": 4.6955084527234076e-05, "epoch": 1.371712, "percentage": 45.72, "elapsed_time": "2:13:08", "remaining_time": "2:38:04", "throughput": 32743.75, "total_tokens": 261559488} +{"current_steps": 1350, "total_steps": 2931, "loss": 0.6628549098968506, "lr": 4.6934393086346034e-05, "epoch": 1.381952, "percentage": 46.06, "elapsed_time": "2:14:07", "remaining_time": "2:37:04", "throughput": 32743.99, "total_tokens": 263504960} +{"current_steps": 1360, "total_steps": 2931, "loss": 0.6543691158294678, "lr": 4.6913728975325324e-05, "epoch": 1.392192, "percentage": 46.4, "elapsed_time": "2:15:08", "remaining_time": "2:36:06", "throughput": 32739.47, "total_tokens": 265475904} +{"current_steps": 1370, "total_steps": 2931, "loss": 0.6563886165618896, "lr": 4.6893092134061393e-05, "epoch": 1.4024320000000001, "percentage": 46.74, "elapsed_time": "2:16:07", "remaining_time": "2:35:06", "throughput": 32740.99, "total_tokens": 267417152} +{"current_steps": 1380, "total_steps": 2931, "loss": 0.6647259712219238, "lr": 4.687248250262859e-05, "epoch": 1.412672, "percentage": 47.08, "elapsed_time": "2:17:06", "remaining_time": "2:34:05", "throughput": 32743.69, "total_tokens": 269353152} +{"current_steps": 1390, "total_steps": 2931, "loss": 0.6596537113189698, "lr": 4.685190002128548e-05, "epoch": 1.422912, "percentage": 47.42, "elapsed_time": "2:18:05", "remaining_time": "2:33:05", "throughput": 32744.66, "total_tokens": 271311872} +{"current_steps": 1400, "total_steps": 2931, "loss": 0.6623442649841309, "lr": 4.6831344630474114e-05, "epoch": 1.433152, "percentage": 47.77, "elapsed_time": "2:19:05", "remaining_time": "2:32:06", "throughput": 32745.27, "total_tokens": 273284288} +{"current_steps": 1410, "total_steps": 2931, "loss": 0.6568885803222656, "lr": 4.6810816270819276e-05, "epoch": 1.443392, "percentage": 48.11, "elapsed_time": "2:20:04", "remaining_time": "2:31:06", "throughput": 32749.24, "total_tokens": 275251328} +{"current_steps": 1420, "total_steps": 2931, "loss": 0.6608157157897949, "lr": 4.679031488312777e-05, "epoch": 1.453632, "percentage": 48.45, "elapsed_time": "2:21:05", "remaining_time": "2:30:07", "throughput": 32746.69, "total_tokens": 277202496} +{"current_steps": 1430, "total_steps": 2931, "loss": 0.6612616539001465, "lr": 4.6769840408387717e-05, "epoch": 1.463872, "percentage": 48.79, "elapsed_time": "2:22:06", "remaining_time": "2:29:09", "throughput": 32742.38, "total_tokens": 279162944} +{"current_steps": 1440, "total_steps": 2931, "loss": 0.6639165878295898, "lr": 4.674939278776787e-05, "epoch": 1.4741119999999999, "percentage": 49.13, "elapsed_time": "2:23:05", "remaining_time": "2:28:09", "throughput": 32743.75, "total_tokens": 281109696} +{"current_steps": 1450, "total_steps": 2931, "loss": 0.6613688945770264, "lr": 4.672897196261683e-05, "epoch": 1.484352, "percentage": 49.47, "elapsed_time": "2:24:03", "remaining_time": "2:27:08", "throughput": 32746.38, "total_tokens": 283056064} +{"current_steps": 1460, "total_steps": 2931, "loss": 0.6640623569488525, "lr": 4.670857787446238e-05, "epoch": 1.494592, "percentage": 49.81, "elapsed_time": "2:25:03", "remaining_time": "2:26:09", "throughput": 32745.59, "total_tokens": 285002496} +{"current_steps": 1470, "total_steps": 2931, "loss": 0.6640773773193359, "lr": 4.668821046501082e-05, "epoch": 1.504832, "percentage": 50.15, "elapsed_time": "2:26:03", "remaining_time": "2:25:10", "throughput": 32743.2, "total_tokens": 286959616} +{"current_steps": 1480, "total_steps": 2931, "loss": 0.6624667167663574, "lr": 4.6667869676146194e-05, "epoch": 1.515072, "percentage": 50.49, "elapsed_time": "2:27:02", "remaining_time": "2:24:09", "throughput": 32746.53, "total_tokens": 288905280} +{"current_steps": 1490, "total_steps": 2931, "loss": 0.6601789474487305, "lr": 4.6647555449929645e-05, "epoch": 1.525312, "percentage": 50.84, "elapsed_time": "2:28:01", "remaining_time": "2:23:09", "throughput": 32746.4, "total_tokens": 290846400} +{"current_steps": 1500, "total_steps": 2931, "loss": 0.6581646919250488, "lr": 4.662726772859869e-05, "epoch": 1.535552, "percentage": 51.18, "elapsed_time": "2:28:59", "remaining_time": "2:22:08", "throughput": 32749.28, "total_tokens": 292768064} +{"current_steps": 1510, "total_steps": 2931, "loss": 0.6602959632873535, "lr": 4.660700645456655e-05, "epoch": 1.545792, "percentage": 51.52, "elapsed_time": "2:29:59", "remaining_time": "2:21:08", "throughput": 32749.87, "total_tokens": 294723392} +{"current_steps": 1520, "total_steps": 2931, "loss": 0.6587230682373046, "lr": 4.658677157042149e-05, "epoch": 1.556032, "percentage": 51.86, "elapsed_time": "2:30:57", "remaining_time": "2:20:08", "throughput": 32752.93, "total_tokens": 296664896} +{"current_steps": 1530, "total_steps": 2931, "loss": 0.6597561836242676, "lr": 4.656656301892605e-05, "epoch": 1.566272, "percentage": 52.2, "elapsed_time": "2:31:56", "remaining_time": "2:19:08", "throughput": 32753.62, "total_tokens": 298606592} +{"current_steps": 1540, "total_steps": 2931, "loss": 0.6614864349365235, "lr": 4.6546380743016465e-05, "epoch": 1.5765120000000001, "percentage": 52.54, "elapsed_time": "2:32:57", "remaining_time": "2:18:09", "throughput": 32751.56, "total_tokens": 300562496} +{"current_steps": 1550, "total_steps": 2931, "loss": 0.6568780422210694, "lr": 4.652622468580193e-05, "epoch": 1.5867520000000002, "percentage": 52.88, "elapsed_time": "2:33:56", "remaining_time": "2:17:09", "throughput": 32752.19, "total_tokens": 302519104} +{"current_steps": 1560, "total_steps": 2931, "loss": 0.6616767883300781, "lr": 4.650609479056392e-05, "epoch": 1.596992, "percentage": 53.22, "elapsed_time": "2:34:56", "remaining_time": "2:16:09", "throughput": 32749.71, "total_tokens": 304449152} +{"current_steps": 1570, "total_steps": 2931, "loss": 0.6637643814086914, "lr": 4.648599100075556e-05, "epoch": 1.607232, "percentage": 53.57, "elapsed_time": "2:35:55", "remaining_time": "2:15:10", "throughput": 32748.65, "total_tokens": 306388224} +{"current_steps": 1580, "total_steps": 2931, "loss": 0.6638627052307129, "lr": 4.6465913260000945e-05, "epoch": 1.617472, "percentage": 53.91, "elapsed_time": "2:36:54", "remaining_time": "2:14:10", "throughput": 32750.61, "total_tokens": 308336512} +{"current_steps": 1590, "total_steps": 2931, "loss": 0.661181640625, "lr": 4.644586151209444e-05, "epoch": 1.627712, "percentage": 54.25, "elapsed_time": "2:37:56", "remaining_time": "2:13:12", "throughput": 32746.61, "total_tokens": 310314752} +{"current_steps": 1600, "total_steps": 2931, "loss": 0.6550232887268066, "lr": 4.6425835701000084e-05, "epoch": 1.6379519999999999, "percentage": 54.59, "elapsed_time": "2:38:57", "remaining_time": "2:12:13", "throughput": 32743.73, "total_tokens": 312281088} +{"current_steps": 1610, "total_steps": 2931, "loss": 0.6569297790527344, "lr": 4.640583577085084e-05, "epoch": 1.6481919999999999, "percentage": 54.93, "elapsed_time": "2:39:57", "remaining_time": "2:11:14", "throughput": 32742.23, "total_tokens": 314245824} +{"current_steps": 1620, "total_steps": 2931, "loss": 0.652769947052002, "lr": 4.638586166594806e-05, "epoch": 1.658432, "percentage": 55.27, "elapsed_time": "2:40:58", "remaining_time": "2:10:16", "throughput": 32740.22, "total_tokens": 316216768} +{"current_steps": 1630, "total_steps": 2931, "loss": 0.6508107662200928, "lr": 4.6365913330760726e-05, "epoch": 1.668672, "percentage": 55.61, "elapsed_time": "2:41:56", "remaining_time": "2:09:15", "throughput": 32743.76, "total_tokens": 318168448} +{"current_steps": 1640, "total_steps": 2931, "loss": 0.6558364868164063, "lr": 4.6345990709924855e-05, "epoch": 1.678912, "percentage": 55.95, "elapsed_time": "2:42:56", "remaining_time": "2:08:16", "throughput": 32745.43, "total_tokens": 320136128} +{"current_steps": 1650, "total_steps": 2931, "loss": 0.6574318885803223, "lr": 4.632609374824284e-05, "epoch": 1.689152, "percentage": 56.29, "elapsed_time": "2:43:56", "remaining_time": "2:07:16", "throughput": 32745.4, "total_tokens": 322099456} +{"current_steps": 1660, "total_steps": 2931, "loss": 0.6548179626464844, "lr": 4.630622239068285e-05, "epoch": 1.699392, "percentage": 56.64, "elapsed_time": "2:44:55", "remaining_time": "2:06:16", "throughput": 32747.67, "total_tokens": 324049408} +{"current_steps": 1670, "total_steps": 2931, "loss": 0.6559741497039795, "lr": 4.628637658237808e-05, "epoch": 1.709632, "percentage": 56.98, "elapsed_time": "2:45:53", "remaining_time": "2:05:15", "throughput": 32750.76, "total_tokens": 325990016} +{"current_steps": 1680, "total_steps": 2931, "loss": 0.6587584495544434, "lr": 4.626655626862625e-05, "epoch": 1.719872, "percentage": 57.32, "elapsed_time": "2:46:52", "remaining_time": "2:04:15", "throughput": 32752.08, "total_tokens": 327937472} +{"current_steps": 1690, "total_steps": 2931, "loss": 0.656014347076416, "lr": 4.624676139488888e-05, "epoch": 1.730112, "percentage": 57.66, "elapsed_time": "2:47:52", "remaining_time": "2:03:16", "throughput": 32752.02, "total_tokens": 329908352} +{"current_steps": 1700, "total_steps": 2931, "loss": 0.658781623840332, "lr": 4.6226991906790686e-05, "epoch": 1.7403520000000001, "percentage": 58.0, "elapsed_time": "2:48:51", "remaining_time": "2:02:16", "throughput": 32754.05, "total_tokens": 331853696} +{"current_steps": 1710, "total_steps": 2931, "loss": 0.6597333908081054, "lr": 4.620724775011897e-05, "epoch": 1.7505920000000001, "percentage": 58.34, "elapsed_time": "2:49:50", "remaining_time": "2:01:16", "throughput": 32755.32, "total_tokens": 333782656} +{"current_steps": 1720, "total_steps": 2931, "loss": 0.6641686439514161, "lr": 4.618752887082297e-05, "epoch": 1.760832, "percentage": 58.68, "elapsed_time": "2:50:48", "remaining_time": "2:00:15", "throughput": 32758.95, "total_tokens": 335741568} +{"current_steps": 1730, "total_steps": 2931, "loss": 0.6582834720611572, "lr": 4.616783521501325e-05, "epoch": 1.771072, "percentage": 59.02, "elapsed_time": "2:51:48", "remaining_time": "1:59:16", "throughput": 32757.62, "total_tokens": 337673856} +{"current_steps": 1740, "total_steps": 2931, "loss": 0.6613027572631835, "lr": 4.614816672896108e-05, "epoch": 1.781312, "percentage": 59.37, "elapsed_time": "2:52:48", "remaining_time": "1:58:16", "throughput": 32756.77, "total_tokens": 339633088} +{"current_steps": 1750, "total_steps": 2931, "loss": 0.6542655467987061, "lr": 4.612852335909782e-05, "epoch": 1.791552, "percentage": 59.71, "elapsed_time": "2:53:48", "remaining_time": "1:57:17", "throughput": 32754.3, "total_tokens": 341581696} +{"current_steps": 1760, "total_steps": 2931, "loss": 0.6533420562744141, "lr": 4.6108905052014323e-05, "epoch": 1.8017919999999998, "percentage": 60.05, "elapsed_time": "2:54:48", "remaining_time": "1:56:18", "throughput": 32754.92, "total_tokens": 343534720} +{"current_steps": 1770, "total_steps": 2931, "loss": 0.6449230194091797, "lr": 4.608931175446027e-05, "epoch": 1.8120319999999999, "percentage": 60.39, "elapsed_time": "2:55:47", "remaining_time": "1:55:18", "throughput": 32755.04, "total_tokens": 345495296} +{"current_steps": 1780, "total_steps": 2931, "loss": 0.6538689613342286, "lr": 4.606974341334367e-05, "epoch": 1.822272, "percentage": 60.73, "elapsed_time": "2:56:46", "remaining_time": "1:54:18", "throughput": 32756.26, "total_tokens": 347431552} +{"current_steps": 1790, "total_steps": 2931, "loss": 0.6490047454833985, "lr": 4.605019997573011e-05, "epoch": 1.832512, "percentage": 61.07, "elapsed_time": "2:57:46", "remaining_time": "1:53:19", "throughput": 32755.65, "total_tokens": 349382976} +{"current_steps": 1800, "total_steps": 2931, "loss": 0.6522298812866211, "lr": 4.603068138884229e-05, "epoch": 1.842752, "percentage": 61.41, "elapsed_time": "2:58:46", "remaining_time": "1:52:19", "throughput": 32756.24, "total_tokens": 351343680} +{"current_steps": 1810, "total_steps": 2931, "loss": 0.6543211936950684, "lr": 4.6011187600059345e-05, "epoch": 1.852992, "percentage": 61.75, "elapsed_time": "2:59:45", "remaining_time": "1:51:20", "throughput": 32754.18, "total_tokens": 353284096} +{"current_steps": 1820, "total_steps": 2931, "loss": 0.6490991115570068, "lr": 4.599171855691629e-05, "epoch": 1.863232, "percentage": 62.09, "elapsed_time": "3:00:45", "remaining_time": "1:50:20", "throughput": 32752.64, "total_tokens": 355207296} +{"current_steps": 1830, "total_steps": 2931, "loss": 0.6518874168395996, "lr": 4.597227420710335e-05, "epoch": 1.873472, "percentage": 62.44, "elapsed_time": "3:01:44", "remaining_time": "1:49:20", "throughput": 32751.33, "total_tokens": 357148288} +{"current_steps": 1840, "total_steps": 2931, "loss": 0.6608992099761963, "lr": 4.595285449846551e-05, "epoch": 1.883712, "percentage": 62.78, "elapsed_time": "3:02:44", "remaining_time": "1:48:21", "throughput": 32751.12, "total_tokens": 359101824} +{"current_steps": 1850, "total_steps": 2931, "loss": 0.6517756462097168, "lr": 4.593345937900178e-05, "epoch": 1.893952, "percentage": 63.12, "elapsed_time": "3:03:43", "remaining_time": "1:47:21", "throughput": 32752.9, "total_tokens": 361063168} +{"current_steps": 1860, "total_steps": 2931, "loss": 0.6512045860290527, "lr": 4.591408879686472e-05, "epoch": 1.904192, "percentage": 63.46, "elapsed_time": "3:04:41", "remaining_time": "1:46:20", "throughput": 32756.39, "total_tokens": 362991872} +{"current_steps": 1870, "total_steps": 2931, "loss": 0.6591498374938964, "lr": 4.5894742700359775e-05, "epoch": 1.9144320000000001, "percentage": 63.8, "elapsed_time": "3:05:40", "remaining_time": "1:45:20", "throughput": 32759.79, "total_tokens": 364946688} +{"current_steps": 1880, "total_steps": 2931, "loss": 0.6532726287841797, "lr": 4.587542103794477e-05, "epoch": 1.9246720000000002, "percentage": 64.14, "elapsed_time": "3:06:38", "remaining_time": "1:44:20", "throughput": 32761.22, "total_tokens": 366891072} +{"current_steps": 1890, "total_steps": 2931, "loss": 0.6575124740600586, "lr": 4.5856123758229247e-05, "epoch": 1.934912, "percentage": 64.48, "elapsed_time": "3:07:38", "remaining_time": "1:43:21", "throughput": 32760.88, "total_tokens": 368845632} +{"current_steps": 1900, "total_steps": 2931, "loss": 0.6555071830749511, "lr": 4.5836850809973993e-05, "epoch": 1.945152, "percentage": 64.82, "elapsed_time": "3:08:38", "remaining_time": "1:42:21", "throughput": 32760.79, "total_tokens": 370797568} +{"current_steps": 1910, "total_steps": 2931, "loss": 0.6574622631072998, "lr": 4.5817602142090385e-05, "epoch": 1.955392, "percentage": 65.17, "elapsed_time": "3:09:36", "remaining_time": "1:41:21", "throughput": 32763.99, "total_tokens": 372745728} +{"current_steps": 1920, "total_steps": 2931, "loss": 0.6541840553283691, "lr": 4.579837770363989e-05, "epoch": 1.965632, "percentage": 65.51, "elapsed_time": "3:10:35", "remaining_time": "1:40:21", "throughput": 32765.8, "total_tokens": 374707904} +{"current_steps": 1930, "total_steps": 2931, "loss": 0.6511831283569336, "lr": 4.57791774438334e-05, "epoch": 1.9758719999999999, "percentage": 65.85, "elapsed_time": "3:11:35", "remaining_time": "1:39:22", "throughput": 32766.91, "total_tokens": 376672064} +{"current_steps": 1940, "total_steps": 2931, "loss": 0.653932523727417, "lr": 4.576000131203078e-05, "epoch": 1.9861119999999999, "percentage": 66.19, "elapsed_time": "3:12:35", "remaining_time": "1:38:22", "throughput": 32765.64, "total_tokens": 378613568} +{"current_steps": 1950, "total_steps": 2931, "loss": 0.6487459659576416, "lr": 4.574084925774023e-05, "epoch": 1.996352, "percentage": 66.53, "elapsed_time": "3:13:33", "remaining_time": "1:37:22", "throughput": 32766.32, "total_tokens": 380541952} +{"current_steps": 1960, "total_steps": 2931, "loss": 0.6155517578125, "lr": 4.5721721230617795e-05, "epoch": 2.006144, "percentage": 66.87, "elapsed_time": "3:14:30", "remaining_time": "1:36:21", "throughput": 32767.62, "total_tokens": 382413824} +{"current_steps": 1970, "total_steps": 2931, "loss": 0.5954423427581788, "lr": 4.57026171804667e-05, "epoch": 2.016384, "percentage": 67.21, "elapsed_time": "3:15:29", "remaining_time": "1:35:21", "throughput": 32768.3, "total_tokens": 384361152} +{"current_steps": 1980, "total_steps": 2931, "loss": 0.5828520774841308, "lr": 4.568353705723692e-05, "epoch": 2.026624, "percentage": 67.55, "elapsed_time": "3:16:28", "remaining_time": "1:34:21", "throughput": 32769.12, "total_tokens": 386292032} +{"current_steps": 1990, "total_steps": 2931, "loss": 0.5893785953521729, "lr": 4.566448081102455e-05, "epoch": 2.036864, "percentage": 67.89, "elapsed_time": "3:17:25", "remaining_time": "1:33:21", "throughput": 32772.9, "total_tokens": 388221376} +{"current_steps": 2000, "total_steps": 2931, "loss": 0.5838380813598633, "lr": 4.564544839207128e-05, "epoch": 2.047104, "percentage": 68.24, "elapsed_time": "3:18:27", "remaining_time": "1:32:22", "throughput": 32768.8, "total_tokens": 390196224} +{"current_steps": 2010, "total_steps": 2931, "loss": 0.5901473999023438, "lr": 4.562643975076387e-05, "epoch": 2.057344, "percentage": 68.58, "elapsed_time": "3:19:25", "remaining_time": "1:31:22", "throughput": 32770.32, "total_tokens": 392126976} +{"current_steps": 2020, "total_steps": 2931, "loss": 0.5894758224487304, "lr": 4.560745483763357e-05, "epoch": 2.067584, "percentage": 68.92, "elapsed_time": "3:20:26", "remaining_time": "1:30:23", "throughput": 32769.73, "total_tokens": 394095104} +{"current_steps": 2030, "total_steps": 2931, "loss": 0.5862543106079101, "lr": 4.5588493603355595e-05, "epoch": 2.077824, "percentage": 69.26, "elapsed_time": "3:21:25", "remaining_time": "1:29:23", "throughput": 32770.48, "total_tokens": 396042368} +{"current_steps": 2040, "total_steps": 2931, "loss": 0.5956791400909424, "lr": 4.556955599874859e-05, "epoch": 2.088064, "percentage": 69.6, "elapsed_time": "3:22:24", "remaining_time": "1:28:24", "throughput": 32771.8, "total_tokens": 397981952} +{"current_steps": 2050, "total_steps": 2931, "loss": 0.5897563934326172, "lr": 4.555064197477409e-05, "epoch": 2.098304, "percentage": 69.94, "elapsed_time": "3:23:21", "remaining_time": "1:27:23", "throughput": 32773.95, "total_tokens": 399907072} +{"current_steps": 2060, "total_steps": 2931, "loss": 0.5853563785552979, "lr": 4.5531751482536e-05, "epoch": 2.108544, "percentage": 70.28, "elapsed_time": "3:24:22", "remaining_time": "1:26:24", "throughput": 32771.47, "total_tokens": 401862848} +{"current_steps": 2070, "total_steps": 2931, "loss": 0.5935741424560547, "lr": 4.5512884473280024e-05, "epoch": 2.118784, "percentage": 70.62, "elapsed_time": "3:25:22", "remaining_time": "1:25:25", "throughput": 32771.04, "total_tokens": 403810240} +{"current_steps": 2080, "total_steps": 2931, "loss": 0.59158935546875, "lr": 4.549404089839322e-05, "epoch": 2.129024, "percentage": 70.97, "elapsed_time": "3:26:23", "remaining_time": "1:24:26", "throughput": 32768.49, "total_tokens": 405776256} +{"current_steps": 2090, "total_steps": 2931, "loss": 0.5877013206481934, "lr": 4.547522070940335e-05, "epoch": 2.139264, "percentage": 71.31, "elapsed_time": "3:27:23", "remaining_time": "1:23:27", "throughput": 32767.57, "total_tokens": 407735744} +{"current_steps": 2100, "total_steps": 2931, "loss": 0.5920863628387452, "lr": 4.545642385797848e-05, "epoch": 2.149504, "percentage": 71.65, "elapsed_time": "3:28:23", "remaining_time": "1:22:27", "throughput": 32766.44, "total_tokens": 409688064} +{"current_steps": 2110, "total_steps": 2931, "loss": 0.5942326545715332, "lr": 4.543765029592637e-05, "epoch": 2.159744, "percentage": 71.99, "elapsed_time": "3:29:21", "remaining_time": "1:21:27", "throughput": 32768.65, "total_tokens": 411607680} +{"current_steps": 2120, "total_steps": 2931, "loss": 0.5903902053833008, "lr": 4.541889997519403e-05, "epoch": 2.169984, "percentage": 72.33, "elapsed_time": "3:30:20", "remaining_time": "1:20:27", "throughput": 32768.77, "total_tokens": 413559232} +{"current_steps": 2130, "total_steps": 2931, "loss": 0.5925062656402588, "lr": 4.5400172847867095e-05, "epoch": 2.180224, "percentage": 72.67, "elapsed_time": "3:31:19", "remaining_time": "1:19:28", "throughput": 32770.16, "total_tokens": 415502272} +{"current_steps": 2140, "total_steps": 2931, "loss": 0.5988312244415284, "lr": 4.5381468866169466e-05, "epoch": 2.190464, "percentage": 73.01, "elapsed_time": "3:32:18", "remaining_time": "1:18:28", "throughput": 32772.3, "total_tokens": 417454208} +{"current_steps": 2150, "total_steps": 2931, "loss": 0.5885556221008301, "lr": 4.5362787982462616e-05, "epoch": 2.200704, "percentage": 73.35, "elapsed_time": "3:33:18", "remaining_time": "1:17:28", "throughput": 32771.2, "total_tokens": 419408384} +{"current_steps": 2160, "total_steps": 2931, "loss": 0.5948306560516358, "lr": 4.5344130149245275e-05, "epoch": 2.210944, "percentage": 73.69, "elapsed_time": "3:34:17", "remaining_time": "1:16:29", "throughput": 32770.6, "total_tokens": 421340992} +{"current_steps": 2170, "total_steps": 2931, "loss": 0.595841646194458, "lr": 4.5325495319152715e-05, "epoch": 2.221184, "percentage": 74.04, "elapsed_time": "3:35:15", "remaining_time": "1:15:29", "throughput": 32773.26, "total_tokens": 423294080} +{"current_steps": 2180, "total_steps": 2931, "loss": 0.5914995193481445, "lr": 4.530688344495644e-05, "epoch": 2.231424, "percentage": 74.38, "elapsed_time": "3:36:16", "remaining_time": "1:14:30", "throughput": 32771.43, "total_tokens": 425247872} +{"current_steps": 2190, "total_steps": 2931, "loss": 0.5963719367980957, "lr": 4.528829447956357e-05, "epoch": 2.241664, "percentage": 74.72, "elapsed_time": "3:37:14", "remaining_time": "1:13:30", "throughput": 32774.21, "total_tokens": 427203328} +{"current_steps": 2200, "total_steps": 2931, "loss": 0.5922994613647461, "lr": 4.526972837601633e-05, "epoch": 2.251904, "percentage": 75.06, "elapsed_time": "3:38:14", "remaining_time": "1:12:30", "throughput": 32775.6, "total_tokens": 429173888} +{"current_steps": 2210, "total_steps": 2931, "loss": 0.5888791561126709, "lr": 4.525118508749165e-05, "epoch": 2.262144, "percentage": 75.4, "elapsed_time": "3:39:14", "remaining_time": "1:11:31", "throughput": 32773.1, "total_tokens": 431118720} +{"current_steps": 2220, "total_steps": 2931, "loss": 0.5937693119049072, "lr": 4.5232664567300546e-05, "epoch": 2.272384, "percentage": 75.74, "elapsed_time": "3:40:13", "remaining_time": "1:10:31", "throughput": 32774.62, "total_tokens": 433071488} +{"current_steps": 2230, "total_steps": 2931, "loss": 0.5958268165588378, "lr": 4.521416676888773e-05, "epoch": 2.282624, "percentage": 76.08, "elapsed_time": "3:41:13", "remaining_time": "1:09:32", "throughput": 32774.3, "total_tokens": 435030464} +{"current_steps": 2240, "total_steps": 2931, "loss": 0.5926674842834473, "lr": 4.519569164583107e-05, "epoch": 2.292864, "percentage": 76.42, "elapsed_time": "3:42:13", "remaining_time": "1:08:33", "throughput": 32772.33, "total_tokens": 436981504} +{"current_steps": 2250, "total_steps": 2931, "loss": 0.5918882846832275, "lr": 4.517723915184109e-05, "epoch": 2.303104, "percentage": 76.77, "elapsed_time": "3:43:14", "remaining_time": "1:07:34", "throughput": 32771.2, "total_tokens": 438963648} +{"current_steps": 2260, "total_steps": 2931, "loss": 0.5919432163238525, "lr": 4.5158809240760506e-05, "epoch": 2.313344, "percentage": 77.11, "elapsed_time": "3:44:14", "remaining_time": "1:06:34", "throughput": 32770.16, "total_tokens": 440906176} +{"current_steps": 2270, "total_steps": 2931, "loss": 0.5900529384613037, "lr": 4.514040186656375e-05, "epoch": 2.323584, "percentage": 77.45, "elapsed_time": "3:45:15", "remaining_time": "1:05:35", "throughput": 32769.75, "total_tokens": 442885504} +{"current_steps": 2280, "total_steps": 2931, "loss": 0.5980159759521484, "lr": 4.512201698335644e-05, "epoch": 2.333824, "percentage": 77.79, "elapsed_time": "3:46:15", "remaining_time": "1:04:36", "throughput": 32767.46, "total_tokens": 444823488} +{"current_steps": 2290, "total_steps": 2931, "loss": 0.5917848587036133, "lr": 4.510365454537496e-05, "epoch": 2.344064, "percentage": 78.13, "elapsed_time": "3:47:14", "remaining_time": "1:03:36", "throughput": 32768.8, "total_tokens": 446776320} +{"current_steps": 2300, "total_steps": 2931, "loss": 0.6014077186584472, "lr": 4.5085314506985945e-05, "epoch": 2.354304, "percentage": 78.47, "elapsed_time": "3:48:13", "remaining_time": "1:02:36", "throughput": 32770.26, "total_tokens": 448734784} +{"current_steps": 2310, "total_steps": 2931, "loss": 0.5941033363342285, "lr": 4.50669968226858e-05, "epoch": 2.364544, "percentage": 78.81, "elapsed_time": "3:49:11", "remaining_time": "1:01:36", "throughput": 32772.33, "total_tokens": 450674688} +{"current_steps": 2320, "total_steps": 2931, "loss": 0.5903738021850586, "lr": 4.504870144710027e-05, "epoch": 2.374784, "percentage": 79.15, "elapsed_time": "3:50:10", "remaining_time": "1:00:37", "throughput": 32774.66, "total_tokens": 452620544} +{"current_steps": 2330, "total_steps": 2931, "loss": 0.59536771774292, "lr": 4.5030428334983884e-05, "epoch": 2.385024, "percentage": 79.5, "elapsed_time": "3:51:10", "remaining_time": "0:59:37", "throughput": 32773.72, "total_tokens": 454586752} +{"current_steps": 2340, "total_steps": 2931, "loss": 0.5988545417785645, "lr": 4.501217744121959e-05, "epoch": 2.395264, "percentage": 79.84, "elapsed_time": "3:52:11", "remaining_time": "0:58:38", "throughput": 32771.03, "total_tokens": 456539968} +{"current_steps": 2350, "total_steps": 2931, "loss": 0.5976475715637207, "lr": 4.499394872081821e-05, "epoch": 2.405504, "percentage": 80.18, "elapsed_time": "3:53:11", "remaining_time": "0:57:39", "throughput": 32769.83, "total_tokens": 458487936} +{"current_steps": 2360, "total_steps": 2931, "loss": 0.5921101570129395, "lr": 4.4975742128918e-05, "epoch": 2.415744, "percentage": 80.52, "elapsed_time": "3:54:11", "remaining_time": "0:56:39", "throughput": 32768.93, "total_tokens": 460446336} +{"current_steps": 2370, "total_steps": 2931, "loss": 0.5958994388580322, "lr": 4.495755762078418e-05, "epoch": 2.425984, "percentage": 80.86, "elapsed_time": "3:55:12", "remaining_time": "0:55:40", "throughput": 32767.04, "total_tokens": 462432640} +{"current_steps": 2380, "total_steps": 2931, "loss": 0.5889082908630371, "lr": 4.49393951518085e-05, "epoch": 2.436224, "percentage": 81.2, "elapsed_time": "3:56:12", "remaining_time": "0:54:41", "throughput": 32766.3, "total_tokens": 464370368} +{"current_steps": 2390, "total_steps": 2931, "loss": 0.5928831100463867, "lr": 4.4921254677508716e-05, "epoch": 2.446464, "percentage": 81.54, "elapsed_time": "3:57:12", "remaining_time": "0:53:41", "throughput": 32763.63, "total_tokens": 466318912} +{"current_steps": 2400, "total_steps": 2931, "loss": 0.5945481777191162, "lr": 4.490313615352821e-05, "epoch": 2.456704, "percentage": 81.88, "elapsed_time": "3:58:12", "remaining_time": "0:52:42", "throughput": 32763.29, "total_tokens": 468270848} +{"current_steps": 2410, "total_steps": 2931, "loss": 0.5897410392761231, "lr": 4.48850395356355e-05, "epoch": 2.466944, "percentage": 82.22, "elapsed_time": "3:59:13", "remaining_time": "0:51:43", "throughput": 32762.27, "total_tokens": 470260224} +{"current_steps": 2420, "total_steps": 2931, "loss": 0.5936931133270263, "lr": 4.486696477972375e-05, "epoch": 2.477184, "percentage": 82.57, "elapsed_time": "4:00:11", "remaining_time": "0:50:42", "throughput": 32765.7, "total_tokens": 472188416} +{"current_steps": 2430, "total_steps": 2931, "loss": 0.5995767116546631, "lr": 4.484891184181041e-05, "epoch": 2.487424, "percentage": 82.91, "elapsed_time": "4:01:09", "remaining_time": "0:49:43", "throughput": 32766.65, "total_tokens": 474117824} +{"current_steps": 2440, "total_steps": 2931, "loss": 0.6011977195739746, "lr": 4.483088067803662e-05, "epoch": 2.497664, "percentage": 83.25, "elapsed_time": "4:02:08", "remaining_time": "0:48:43", "throughput": 32768.16, "total_tokens": 476069504} +{"current_steps": 2450, "total_steps": 2931, "loss": 0.5965440273284912, "lr": 4.481287124466697e-05, "epoch": 2.507904, "percentage": 83.59, "elapsed_time": "4:03:08", "remaining_time": "0:47:44", "throughput": 32767.56, "total_tokens": 478029056} +{"current_steps": 2460, "total_steps": 2931, "loss": 0.5970401763916016, "lr": 4.479488349808885e-05, "epoch": 2.518144, "percentage": 83.93, "elapsed_time": "4:04:08", "remaining_time": "0:46:44", "throughput": 32766.71, "total_tokens": 479992832} +{"current_steps": 2470, "total_steps": 2931, "loss": 0.5875445365905761, "lr": 4.4776917394812114e-05, "epoch": 2.528384, "percentage": 84.27, "elapsed_time": "4:05:09", "remaining_time": "0:45:45", "throughput": 32765.35, "total_tokens": 481965248} +{"current_steps": 2480, "total_steps": 2931, "loss": 0.5980213165283204, "lr": 4.475897289146862e-05, "epoch": 2.538624, "percentage": 84.61, "elapsed_time": "4:06:08", "remaining_time": "0:44:45", "throughput": 32765.5, "total_tokens": 483911872} +{"current_steps": 2490, "total_steps": 2931, "loss": 0.5927760601043701, "lr": 4.4741049944811806e-05, "epoch": 2.548864, "percentage": 84.95, "elapsed_time": "4:07:08", "remaining_time": "0:43:46", "throughput": 32765.98, "total_tokens": 485858560} +{"current_steps": 2500, "total_steps": 2931, "loss": 0.5933934211730957, "lr": 4.472314851171621e-05, "epoch": 2.559104, "percentage": 85.3, "elapsed_time": "4:08:06", "remaining_time": "0:42:46", "throughput": 32768.08, "total_tokens": 487808832} +{"current_steps": 2510, "total_steps": 2931, "loss": 0.5965798854827881, "lr": 4.4705268549177084e-05, "epoch": 2.569344, "percentage": 85.64, "elapsed_time": "4:09:04", "remaining_time": "0:41:46", "throughput": 32769.33, "total_tokens": 489736832} +{"current_steps": 2520, "total_steps": 2931, "loss": 0.5946637153625488, "lr": 4.468741001430989e-05, "epoch": 2.579584, "percentage": 85.98, "elapsed_time": "4:10:05", "remaining_time": "0:40:47", "throughput": 32768.43, "total_tokens": 491707968} +{"current_steps": 2530, "total_steps": 2931, "loss": 0.5971454620361328, "lr": 4.466957286434997e-05, "epoch": 2.589824, "percentage": 86.32, "elapsed_time": "4:11:05", "remaining_time": "0:39:47", "throughput": 32768.65, "total_tokens": 493661504} +{"current_steps": 2540, "total_steps": 2931, "loss": 0.6013389587402344, "lr": 4.4651757056652e-05, "epoch": 2.600064, "percentage": 86.66, "elapsed_time": "4:12:04", "remaining_time": "0:38:48", "throughput": 32769.05, "total_tokens": 495603392} +{"current_steps": 2550, "total_steps": 2931, "loss": 0.5951766014099121, "lr": 4.463396254868968e-05, "epoch": 2.610304, "percentage": 87.0, "elapsed_time": "4:13:03", "remaining_time": "0:37:48", "throughput": 32768.57, "total_tokens": 497553920} +{"current_steps": 2560, "total_steps": 2931, "loss": 0.5954748153686523, "lr": 4.461618929805519e-05, "epoch": 2.6205439999999998, "percentage": 87.34, "elapsed_time": "4:14:03", "remaining_time": "0:36:49", "throughput": 32768.59, "total_tokens": 499495872} +{"current_steps": 2570, "total_steps": 2931, "loss": 0.586335563659668, "lr": 4.459843726245888e-05, "epoch": 2.6307840000000002, "percentage": 87.68, "elapsed_time": "4:15:02", "remaining_time": "0:35:49", "throughput": 32769.47, "total_tokens": 501455808} +{"current_steps": 2580, "total_steps": 2931, "loss": 0.5931864738464355, "lr": 4.458070639972875e-05, "epoch": 2.641024, "percentage": 88.02, "elapsed_time": "4:16:01", "remaining_time": "0:34:49", "throughput": 32769.25, "total_tokens": 503391232} +{"current_steps": 2590, "total_steps": 2931, "loss": 0.5972167015075683, "lr": 4.456299666781007e-05, "epoch": 2.651264, "percentage": 88.37, "elapsed_time": "4:17:02", "remaining_time": "0:33:50", "throughput": 32767.35, "total_tokens": 505354624} +{"current_steps": 2600, "total_steps": 2931, "loss": 0.597212266921997, "lr": 4.4545308024764984e-05, "epoch": 2.661504, "percentage": 88.71, "elapsed_time": "4:18:02", "remaining_time": "0:32:50", "throughput": 32766.26, "total_tokens": 507291008} +{"current_steps": 2610, "total_steps": 2931, "loss": 0.59915771484375, "lr": 4.452764042877207e-05, "epoch": 2.671744, "percentage": 89.05, "elapsed_time": "4:19:01", "remaining_time": "0:31:51", "throughput": 32766.83, "total_tokens": 509231104} +{"current_steps": 2620, "total_steps": 2931, "loss": 0.5966706275939941, "lr": 4.45099938381259e-05, "epoch": 2.681984, "percentage": 89.39, "elapsed_time": "4:20:01", "remaining_time": "0:30:51", "throughput": 32764.71, "total_tokens": 511185152} +{"current_steps": 2630, "total_steps": 2931, "loss": 0.5944993019104003, "lr": 4.449236821123667e-05, "epoch": 2.692224, "percentage": 89.73, "elapsed_time": "4:21:00", "remaining_time": "0:29:52", "throughput": 32766.32, "total_tokens": 513121152} +{"current_steps": 2640, "total_steps": 2931, "loss": 0.5976039409637451, "lr": 4.447476350662976e-05, "epoch": 2.702464, "percentage": 90.07, "elapsed_time": "4:21:59", "remaining_time": "0:28:52", "throughput": 32767.46, "total_tokens": 515076928} +{"current_steps": 2650, "total_steps": 2931, "loss": 0.5955704689025879, "lr": 4.4457179682945346e-05, "epoch": 2.712704, "percentage": 90.41, "elapsed_time": "4:22:58", "remaining_time": "0:27:53", "throughput": 32767.67, "total_tokens": 517031488} +{"current_steps": 2660, "total_steps": 2931, "loss": 0.5922414779663085, "lr": 4.443961669893798e-05, "epoch": 2.722944, "percentage": 90.75, "elapsed_time": "4:23:58", "remaining_time": "0:26:53", "throughput": 32766.39, "total_tokens": 518978880} +{"current_steps": 2670, "total_steps": 2931, "loss": 0.5948915481567383, "lr": 4.4422074513476155e-05, "epoch": 2.733184, "percentage": 91.1, "elapsed_time": "4:24:58", "remaining_time": "0:25:54", "throughput": 32766.68, "total_tokens": 520932416} +{"current_steps": 2680, "total_steps": 2931, "loss": 0.603402853012085, "lr": 4.4404553085541955e-05, "epoch": 2.743424, "percentage": 91.44, "elapsed_time": "4:25:56", "remaining_time": "0:24:54", "throughput": 32767.31, "total_tokens": 522866176} +{"current_steps": 2690, "total_steps": 2931, "loss": 0.5967090606689454, "lr": 4.438705237423063e-05, "epoch": 2.753664, "percentage": 91.78, "elapsed_time": "4:26:56", "remaining_time": "0:23:54", "throughput": 32768.03, "total_tokens": 524830528} +{"current_steps": 2700, "total_steps": 2931, "loss": 0.5950541496276855, "lr": 4.436957233875017e-05, "epoch": 2.763904, "percentage": 92.12, "elapsed_time": "4:27:54", "remaining_time": "0:22:55", "throughput": 32769.45, "total_tokens": 526763648} +{"current_steps": 2710, "total_steps": 2931, "loss": 0.5917387962341308, "lr": 4.4352112938420956e-05, "epoch": 2.774144, "percentage": 92.46, "elapsed_time": "4:28:54", "remaining_time": "0:21:55", "throughput": 32769.86, "total_tokens": 528714368} +{"current_steps": 2720, "total_steps": 2931, "loss": 0.592765474319458, "lr": 4.433467413267529e-05, "epoch": 2.784384, "percentage": 92.8, "elapsed_time": "4:29:52", "remaining_time": "0:20:56", "throughput": 32770.57, "total_tokens": 530652416} +{"current_steps": 2730, "total_steps": 2931, "loss": 0.5953330993652344, "lr": 4.431725588105708e-05, "epoch": 2.7946239999999998, "percentage": 93.14, "elapsed_time": "4:30:51", "remaining_time": "0:19:56", "throughput": 32771.57, "total_tokens": 532596928} +{"current_steps": 2740, "total_steps": 2931, "loss": 0.5934776306152344, "lr": 4.4299858143221377e-05, "epoch": 2.8048640000000002, "percentage": 93.48, "elapsed_time": "4:31:51", "remaining_time": "0:18:57", "throughput": 32772.79, "total_tokens": 534571776} +{"current_steps": 2750, "total_steps": 2931, "loss": 0.5929346561431885, "lr": 4.4282480878934065e-05, "epoch": 2.815104, "percentage": 93.82, "elapsed_time": "4:32:50", "remaining_time": "0:17:57", "throughput": 32773.17, "total_tokens": 536513472} +{"current_steps": 2760, "total_steps": 2931, "loss": 0.5918947219848633, "lr": 4.4265124048071346e-05, "epoch": 2.825344, "percentage": 94.17, "elapsed_time": "4:33:50", "remaining_time": "0:16:57", "throughput": 32773.4, "total_tokens": 538483968} +{"current_steps": 2770, "total_steps": 2931, "loss": 0.594475269317627, "lr": 4.4247787610619477e-05, "epoch": 2.835584, "percentage": 94.51, "elapsed_time": "4:34:49", "remaining_time": "0:15:58", "throughput": 32774.23, "total_tokens": 540438592} +{"current_steps": 2780, "total_steps": 2931, "loss": 0.5964062690734864, "lr": 4.42304715266743e-05, "epoch": 2.845824, "percentage": 94.85, "elapsed_time": "4:35:49", "remaining_time": "0:14:58", "throughput": 32773.09, "total_tokens": 542392128} +{"current_steps": 2790, "total_steps": 2931, "loss": 0.5967954635620117, "lr": 4.421317575644092e-05, "epoch": 2.856064, "percentage": 95.19, "elapsed_time": "4:36:50", "remaining_time": "0:13:59", "throughput": 32774.07, "total_tokens": 544377664} +{"current_steps": 2800, "total_steps": 2931, "loss": 0.5921245098114014, "lr": 4.419590026023325e-05, "epoch": 2.866304, "percentage": 95.53, "elapsed_time": "4:37:48", "remaining_time": "0:12:59", "throughput": 32776.15, "total_tokens": 546344448} +{"current_steps": 2810, "total_steps": 2931, "loss": 0.5939229965209961, "lr": 4.417864499847368e-05, "epoch": 2.876544, "percentage": 95.87, "elapsed_time": "4:38:48", "remaining_time": "0:12:00", "throughput": 32775.59, "total_tokens": 548293568} +{"current_steps": 2820, "total_steps": 2931, "loss": 0.6008837699890137, "lr": 4.4161409931692676e-05, "epoch": 2.886784, "percentage": 96.21, "elapsed_time": "4:39:48", "remaining_time": "0:11:00", "throughput": 32776.41, "total_tokens": 550250688} +{"current_steps": 2830, "total_steps": 2931, "loss": 0.5984635353088379, "lr": 4.414419502052841e-05, "epoch": 2.897024, "percentage": 96.55, "elapsed_time": "4:40:46", "remaining_time": "0:10:01", "throughput": 32778.48, "total_tokens": 552207104} +{"current_steps": 2840, "total_steps": 2931, "loss": 0.5894341945648194, "lr": 4.412700022572637e-05, "epoch": 2.907264, "percentage": 96.9, "elapsed_time": "4:41:46", "remaining_time": "0:09:01", "throughput": 32776.59, "total_tokens": 554150528} +{"current_steps": 2850, "total_steps": 2931, "loss": 0.5969189643859864, "lr": 4.410982550813902e-05, "epoch": 2.917504, "percentage": 97.24, "elapsed_time": "4:42:46", "remaining_time": "0:08:02", "throughput": 32777.2, "total_tokens": 556108544} +{"current_steps": 2860, "total_steps": 2931, "loss": 0.5878054618835449, "lr": 4.409267082872535e-05, "epoch": 2.927744, "percentage": 97.58, "elapsed_time": "4:43:45", "remaining_time": "0:07:02", "throughput": 32778.19, "total_tokens": 558057216} +{"current_steps": 2870, "total_steps": 2931, "loss": 0.5897252082824707, "lr": 4.407553614855059e-05, "epoch": 2.937984, "percentage": 97.92, "elapsed_time": "4:44:44", "remaining_time": "0:06:03", "throughput": 32779.25, "total_tokens": 560011008} +{"current_steps": 2880, "total_steps": 2931, "loss": 0.5927898406982421, "lr": 4.405842142878579e-05, "epoch": 2.9482239999999997, "percentage": 98.26, "elapsed_time": "4:45:44", "remaining_time": "0:05:03", "throughput": 32779.15, "total_tokens": 561967104} +{"current_steps": 2890, "total_steps": 2931, "loss": 0.5948840141296386, "lr": 4.404132663070745e-05, "epoch": 2.958464, "percentage": 98.6, "elapsed_time": "4:46:43", "remaining_time": "0:04:04", "throughput": 32779.5, "total_tokens": 563924032} +{"current_steps": 2900, "total_steps": 2931, "loss": 0.5974791526794434, "lr": 4.402425171569716e-05, "epoch": 2.968704, "percentage": 98.94, "elapsed_time": "4:47:44", "remaining_time": "0:03:04", "throughput": 32775.41, "total_tokens": 565866240} +{"current_steps": 2910, "total_steps": 2931, "loss": 0.5980208396911622, "lr": 4.400719664524127e-05, "epoch": 2.9789440000000003, "percentage": 99.28, "elapsed_time": "4:48:44", "remaining_time": "0:02:05", "throughput": 32776.63, "total_tokens": 567839616} +{"current_steps": 2920, "total_steps": 2931, "loss": 0.5934501647949219, "lr": 4.399016138093044e-05, "epoch": 2.989184, "percentage": 99.62, "elapsed_time": "4:49:44", "remaining_time": "0:01:05", "throughput": 32776.7, "total_tokens": 569804928} +{"current_steps": 2930, "total_steps": 2931, "loss": 0.5971685409545898, "lr": 4.397314588445937e-05, "epoch": 2.999424, "percentage": 99.97, "elapsed_time": "4:50:44", "remaining_time": "0:00:05", "throughput": 32777.6, "total_tokens": 571794176} +{"current_steps": 2931, "total_steps": 2931, "eval_loss": 0.7842721939086914, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "4:50:48", "remaining_time": "0:00:00", "throughput": 32775.49, "total_tokens": 571898240} +{"current_steps": 2931, "total_steps": 2931, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "4:51:09", "remaining_time": "0:00:00", "throughput": 32736.61, "total_tokens": 571898240} diff --git a/trainer_state.json b/trainer_state.json new file mode 100644 index 0000000..38050ab --- /dev/null +++ b/trainer_state.json @@ -0,0 +1,2983 @@ +{ + "best_global_step": 2931, + "best_metric": 0.7842721939086914, + "best_model_checkpoint": "/workspace/MT_En_German/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": 1.265625, + "learning_rate": 4.9977515176118345e-05, + "loss": 1.1904024124145507, + "num_input_tokens_seen": 1968640, + "step": 10, + "train_runtime": 62.1897, + "train_tokens_per_second": 31655.416 + }, + { + "epoch": 0.02048, + "grad_norm": 1.0, + "learning_rate": 4.9952567580506e-05, + "loss": 0.9794286727905274, + "num_input_tokens_seen": 3921472, + "step": 20, + "train_runtime": 122.7357, + "train_tokens_per_second": 31950.542 + }, + { + "epoch": 0.03072, + "grad_norm": 0.96484375, + "learning_rate": 4.992765730738634e-05, + "loss": 0.9411283493041992, + "num_input_tokens_seen": 5895808, + "step": 30, + "train_runtime": 182.5361, + "train_tokens_per_second": 32299.411 + }, + { + "epoch": 0.04096, + "grad_norm": 0.890625, + "learning_rate": 4.9902784263792476e-05, + "loss": 0.9153067588806152, + "num_input_tokens_seen": 7829440, + "step": 40, + "train_runtime": 241.2587, + "train_tokens_per_second": 32452.462 + }, + { + "epoch": 0.0512, + "grad_norm": 0.9609375, + "learning_rate": 4.987794835708133e-05, + "loss": 0.9085355758666992, + "num_input_tokens_seen": 9781120, + "step": 50, + "train_runtime": 300.3299, + "train_tokens_per_second": 32567.92 + }, + { + "epoch": 0.06144, + "grad_norm": 0.97265625, + "learning_rate": 4.985314949493234e-05, + "loss": 0.882135009765625, + "num_input_tokens_seen": 11720320, + "step": 60, + "train_runtime": 359.9362, + "train_tokens_per_second": 32562.216 + }, + { + "epoch": 0.07168, + "grad_norm": 0.90625, + "learning_rate": 4.982838758534584e-05, + "loss": 0.8795860290527344, + "num_input_tokens_seen": 13679552, + "step": 70, + "train_runtime": 419.4261, + "train_tokens_per_second": 32614.926 + }, + { + "epoch": 0.08192, + "grad_norm": 0.87109375, + "learning_rate": 4.980366253664179e-05, + "loss": 0.8663822174072265, + "num_input_tokens_seen": 15625216, + "step": 80, + "train_runtime": 478.1635, + "train_tokens_per_second": 32677.56 + }, + { + "epoch": 0.09216, + "grad_norm": 0.87890625, + "learning_rate": 4.977897425745825e-05, + "loss": 0.8581996917724609, + "num_input_tokens_seen": 17576768, + "step": 90, + "train_runtime": 537.8615, + "train_tokens_per_second": 32678.982 + }, + { + "epoch": 0.1024, + "grad_norm": 0.8828125, + "learning_rate": 4.975432265674997e-05, + "loss": 0.8557216644287109, + "num_input_tokens_seen": 19528128, + "step": 100, + "train_runtime": 597.3347, + "train_tokens_per_second": 32692.105 + }, + { + "epoch": 0.11264, + "grad_norm": 0.94140625, + "learning_rate": 4.972970764378705e-05, + "loss": 0.8523456573486328, + "num_input_tokens_seen": 21496192, + "step": 110, + "train_runtime": 657.7204, + "train_tokens_per_second": 32682.873 + }, + { + "epoch": 0.12288, + "grad_norm": 0.8984375, + "learning_rate": 4.970512912815344e-05, + "loss": 0.8425041198730469, + "num_input_tokens_seen": 23459072, + "step": 120, + "train_runtime": 718.986, + "train_tokens_per_second": 32627.995 + }, + { + "epoch": 0.13312, + "grad_norm": 0.96484375, + "learning_rate": 4.968058701974564e-05, + "loss": 0.8385295867919922, + "num_input_tokens_seen": 25415040, + "step": 130, + "train_runtime": 779.0018, + "train_tokens_per_second": 32625.137 + }, + { + "epoch": 0.14336, + "grad_norm": 0.890625, + "learning_rate": 4.96560812287712e-05, + "loss": 0.8344181060791016, + "num_input_tokens_seen": 27346432, + "step": 140, + "train_runtime": 837.7523, + "train_tokens_per_second": 32642.623 + }, + { + "epoch": 0.1536, + "grad_norm": 0.875, + "learning_rate": 4.963161166574748e-05, + "loss": 0.8336277961730957, + "num_input_tokens_seen": 29300736, + "step": 150, + "train_runtime": 897.3115, + "train_tokens_per_second": 32653.917 + }, + { + "epoch": 0.16384, + "grad_norm": 0.86328125, + "learning_rate": 4.960717824150013e-05, + "loss": 0.8235919952392579, + "num_input_tokens_seen": 31242688, + "step": 160, + "train_runtime": 955.3403, + "train_tokens_per_second": 32703.204 + }, + { + "epoch": 0.17408, + "grad_norm": 0.83203125, + "learning_rate": 4.9582780867161893e-05, + "loss": 0.8172248840332031, + "num_input_tokens_seen": 33196160, + "step": 170, + "train_runtime": 1015.3247, + "train_tokens_per_second": 32695.119 + }, + { + "epoch": 0.18432, + "grad_norm": 0.85546875, + "learning_rate": 4.955841945417105e-05, + "loss": 0.8182367324829102, + "num_input_tokens_seen": 35131136, + "step": 180, + "train_runtime": 1074.222, + "train_tokens_per_second": 32703.794 + }, + { + "epoch": 0.19456, + "grad_norm": 0.8828125, + "learning_rate": 4.953409391427024e-05, + "loss": 0.8155080795288085, + "num_input_tokens_seen": 37082176, + "step": 190, + "train_runtime": 1134.3854, + "train_tokens_per_second": 32689.222 + }, + { + "epoch": 0.2048, + "grad_norm": 0.828125, + "learning_rate": 4.950980415950502e-05, + "loss": 0.8091486930847168, + "num_input_tokens_seen": 39060608, + "step": 200, + "train_runtime": 1195.0649, + "train_tokens_per_second": 32684.926 + }, + { + "epoch": 0.21504, + "grad_norm": 0.796875, + "learning_rate": 4.9485550102222575e-05, + "loss": 0.8105871200561523, + "num_input_tokens_seen": 40996800, + "step": 210, + "train_runtime": 1253.1394, + "train_tokens_per_second": 32715.274 + }, + { + "epoch": 0.22528, + "grad_norm": 0.81640625, + "learning_rate": 4.946133165507037e-05, + "loss": 0.8007206916809082, + "num_input_tokens_seen": 42929728, + "step": 220, + "train_runtime": 1311.4585, + "train_tokens_per_second": 32734.339 + }, + { + "epoch": 0.23552, + "grad_norm": 0.84765625, + "learning_rate": 4.943714873099483e-05, + "loss": 0.7916994571685791, + "num_input_tokens_seen": 44892288, + "step": 230, + "train_runtime": 1370.4772, + "train_tokens_per_second": 32756.683 + }, + { + "epoch": 0.24576, + "grad_norm": 0.859375, + "learning_rate": 4.9413001243240024e-05, + "loss": 0.7964269638061523, + "num_input_tokens_seen": 46831872, + "step": 240, + "train_runtime": 1431.941, + "train_tokens_per_second": 32705.169 + }, + { + "epoch": 0.256, + "grad_norm": 0.8203125, + "learning_rate": 4.938888910534637e-05, + "loss": 0.7994860172271728, + "num_input_tokens_seen": 48793472, + "step": 250, + "train_runtime": 1490.9713, + "train_tokens_per_second": 32725.963 + }, + { + "epoch": 0.26624, + "grad_norm": 0.8203125, + "learning_rate": 4.936481223114932e-05, + "loss": 0.7954794883728027, + "num_input_tokens_seen": 50764800, + "step": 260, + "train_runtime": 1552.5757, + "train_tokens_per_second": 32697.15 + }, + { + "epoch": 0.27648, + "grad_norm": 0.890625, + "learning_rate": 4.934077053477808e-05, + "loss": 0.7914528369903564, + "num_input_tokens_seen": 52729152, + "step": 270, + "train_runtime": 1612.3049, + "train_tokens_per_second": 32704.207 + }, + { + "epoch": 0.28672, + "grad_norm": 0.875, + "learning_rate": 4.931676393065431e-05, + "loss": 0.7908175468444825, + "num_input_tokens_seen": 54695872, + "step": 280, + "train_runtime": 1672.5661, + "train_tokens_per_second": 32701.771 + }, + { + "epoch": 0.29696, + "grad_norm": 0.88671875, + "learning_rate": 4.929279233349088e-05, + "loss": 0.7946252822875977, + "num_input_tokens_seen": 56652544, + "step": 290, + "train_runtime": 1732.0781, + "train_tokens_per_second": 32707.847 + }, + { + "epoch": 0.3072, + "grad_norm": 0.8046875, + "learning_rate": 4.926885565829051e-05, + "loss": 0.7933988571166992, + "num_input_tokens_seen": 58611264, + "step": 300, + "train_runtime": 1791.0492, + "train_tokens_per_second": 32724.541 + }, + { + "epoch": 0.31744, + "grad_norm": 0.828125, + "learning_rate": 4.924495382034461e-05, + "loss": 0.7868727684020996, + "num_input_tokens_seen": 60538816, + "step": 310, + "train_runtime": 1849.7842, + "train_tokens_per_second": 32727.503 + }, + { + "epoch": 0.32768, + "grad_norm": 0.8359375, + "learning_rate": 4.9221086735231975e-05, + "loss": 0.7884368896484375, + "num_input_tokens_seen": 62490560, + "step": 320, + "train_runtime": 1908.601, + "train_tokens_per_second": 32741.553 + }, + { + "epoch": 0.33792, + "grad_norm": 0.828125, + "learning_rate": 4.919725431881751e-05, + "loss": 0.7792404174804688, + "num_input_tokens_seen": 64459392, + "step": 330, + "train_runtime": 1969.989, + "train_tokens_per_second": 32720.686 + }, + { + "epoch": 0.34816, + "grad_norm": 0.8515625, + "learning_rate": 4.917345648725101e-05, + "loss": 0.7801454067230225, + "num_input_tokens_seen": 66416320, + "step": 340, + "train_runtime": 2030.4427, + "train_tokens_per_second": 32710.266 + }, + { + "epoch": 0.3584, + "grad_norm": 0.80859375, + "learning_rate": 4.914969315696596e-05, + "loss": 0.7781849861145019, + "num_input_tokens_seen": 68379008, + "step": 350, + "train_runtime": 2089.8446, + "train_tokens_per_second": 32719.661 + }, + { + "epoch": 0.36864, + "grad_norm": 0.8203125, + "learning_rate": 4.912596424467818e-05, + "loss": 0.7765862941741943, + "num_input_tokens_seen": 70328768, + "step": 360, + "train_runtime": 2149.5587, + "train_tokens_per_second": 32717.77 + }, + { + "epoch": 0.37888, + "grad_norm": 0.79296875, + "learning_rate": 4.910226966738475e-05, + "loss": 0.7737765312194824, + "num_input_tokens_seen": 72280640, + "step": 370, + "train_runtime": 2209.3771, + "train_tokens_per_second": 32715.394 + }, + { + "epoch": 0.38912, + "grad_norm": 0.796875, + "learning_rate": 4.9078609342362666e-05, + "loss": 0.777079963684082, + "num_input_tokens_seen": 74217408, + "step": 380, + "train_runtime": 2268.2278, + "train_tokens_per_second": 32720.438 + }, + { + "epoch": 0.39936, + "grad_norm": 0.80078125, + "learning_rate": 4.905498318716775e-05, + "loss": 0.7765676498413085, + "num_input_tokens_seen": 76174784, + "step": 390, + "train_runtime": 2327.2347, + "train_tokens_per_second": 32731.888 + }, + { + "epoch": 0.4096, + "grad_norm": 0.83203125, + "learning_rate": 4.9031391119633295e-05, + "loss": 0.7736555099487304, + "num_input_tokens_seen": 78158016, + "step": 400, + "train_runtime": 2388.4766, + "train_tokens_per_second": 32722.957 + }, + { + "epoch": 0.41984, + "grad_norm": 0.83984375, + "learning_rate": 4.9007833057869e-05, + "loss": 0.7716507911682129, + "num_input_tokens_seen": 80110400, + "step": 410, + "train_runtime": 2448.6314, + "train_tokens_per_second": 32716.398 + }, + { + "epoch": 0.43008, + "grad_norm": 0.84375, + "learning_rate": 4.898430892025967e-05, + "loss": 0.7716224193572998, + "num_input_tokens_seen": 82056960, + "step": 420, + "train_runtime": 2507.3071, + "train_tokens_per_second": 32727.128 + }, + { + "epoch": 0.44032, + "grad_norm": 0.796875, + "learning_rate": 4.896081862546415e-05, + "loss": 0.7601117134094239, + "num_input_tokens_seen": 84000832, + "step": 430, + "train_runtime": 2566.4159, + "train_tokens_per_second": 32730.795 + }, + { + "epoch": 0.45056, + "grad_norm": 0.80078125, + "learning_rate": 4.8937362092414e-05, + "loss": 0.7689449310302734, + "num_input_tokens_seen": 85945600, + "step": 440, + "train_runtime": 2626.5603, + "train_tokens_per_second": 32721.731 + }, + { + "epoch": 0.4608, + "grad_norm": 0.84765625, + "learning_rate": 4.891393924031244e-05, + "loss": 0.7667889595031738, + "num_input_tokens_seen": 87886144, + "step": 450, + "train_runtime": 2684.8831, + "train_tokens_per_second": 32733.695 + }, + { + "epoch": 0.47104, + "grad_norm": 0.84375, + "learning_rate": 4.8890549988633095e-05, + "loss": 0.7657638549804687, + "num_input_tokens_seen": 89839744, + "step": 460, + "train_runtime": 2744.7649, + "train_tokens_per_second": 32731.308 + }, + { + "epoch": 0.48128, + "grad_norm": 0.90234375, + "learning_rate": 4.8867194257118907e-05, + "loss": 0.7696039199829101, + "num_input_tokens_seen": 91805440, + "step": 470, + "train_runtime": 2804.7588, + "train_tokens_per_second": 32732.027 + }, + { + "epoch": 0.49152, + "grad_norm": 0.84765625, + "learning_rate": 4.884387196578093e-05, + "loss": 0.7688661575317383, + "num_input_tokens_seen": 93770240, + "step": 480, + "train_runtime": 2862.8593, + "train_tokens_per_second": 32754.052 + }, + { + "epoch": 0.50176, + "grad_norm": 0.81640625, + "learning_rate": 4.882058303489718e-05, + "loss": 0.7624212265014648, + "num_input_tokens_seen": 95708096, + "step": 490, + "train_runtime": 2921.2238, + "train_tokens_per_second": 32763.014 + }, + { + "epoch": 0.512, + "grad_norm": 0.80859375, + "learning_rate": 4.8797327385011496e-05, + "loss": 0.7568993091583252, + "num_input_tokens_seen": 97676928, + "step": 500, + "train_runtime": 2982.1166, + "train_tokens_per_second": 32754.228 + }, + { + "epoch": 0.52224, + "grad_norm": 0.8046875, + "learning_rate": 4.8774104936932425e-05, + "loss": 0.7545355796813965, + "num_input_tokens_seen": 99627776, + "step": 510, + "train_runtime": 3041.6074, + "train_tokens_per_second": 32754.976 + }, + { + "epoch": 0.53248, + "grad_norm": 0.85546875, + "learning_rate": 4.8750915611732076e-05, + "loss": 0.7593323707580566, + "num_input_tokens_seen": 101565696, + "step": 520, + "train_runtime": 3101.9144, + "train_tokens_per_second": 32742.908 + }, + { + "epoch": 0.54272, + "grad_norm": 0.80078125, + "learning_rate": 4.8727759330744986e-05, + "loss": 0.760127067565918, + "num_input_tokens_seen": 103508992, + "step": 530, + "train_runtime": 3161.5479, + "train_tokens_per_second": 32739.972 + }, + { + "epoch": 0.55296, + "grad_norm": 0.78515625, + "learning_rate": 4.870463601556696e-05, + "loss": 0.7561708450317383, + "num_input_tokens_seen": 105456832, + "step": 540, + "train_runtime": 3220.8796, + "train_tokens_per_second": 32741.625 + }, + { + "epoch": 0.5632, + "grad_norm": 0.84375, + "learning_rate": 4.8681545588054075e-05, + "loss": 0.7509193420410156, + "num_input_tokens_seen": 107417344, + "step": 550, + "train_runtime": 3280.9845, + "train_tokens_per_second": 32739.363 + }, + { + "epoch": 0.57344, + "grad_norm": 0.8203125, + "learning_rate": 4.8658487970321404e-05, + "loss": 0.7538685321807861, + "num_input_tokens_seen": 109377984, + "step": 560, + "train_runtime": 3341.5225, + "train_tokens_per_second": 32732.978 + }, + { + "epoch": 0.58368, + "grad_norm": 0.86328125, + "learning_rate": 4.863546308474209e-05, + "loss": 0.7577058792114257, + "num_input_tokens_seen": 111360832, + "step": 570, + "train_runtime": 3401.7812, + "train_tokens_per_second": 32736.036 + }, + { + "epoch": 0.59392, + "grad_norm": 0.76171875, + "learning_rate": 4.86124708539461e-05, + "loss": 0.7539366722106934, + "num_input_tokens_seen": 113337536, + "step": 580, + "train_runtime": 3462.9489, + "train_tokens_per_second": 32728.619 + }, + { + "epoch": 0.60416, + "grad_norm": 0.82421875, + "learning_rate": 4.8589511200819216e-05, + "loss": 0.7477821826934814, + "num_input_tokens_seen": 115270336, + "step": 590, + "train_runtime": 3521.2998, + "train_tokens_per_second": 32735.167 + }, + { + "epoch": 0.6144, + "grad_norm": 0.796875, + "learning_rate": 4.8566584048501926e-05, + "loss": 0.7609129905700683, + "num_input_tokens_seen": 117224512, + "step": 600, + "train_runtime": 3580.6982, + "train_tokens_per_second": 32737.892 + }, + { + "epoch": 0.62464, + "grad_norm": 0.80859375, + "learning_rate": 4.854368932038835e-05, + "loss": 0.7430164337158203, + "num_input_tokens_seen": 119167552, + "step": 610, + "train_runtime": 3640.7229, + "train_tokens_per_second": 32731.838 + }, + { + "epoch": 0.63488, + "grad_norm": 0.796875, + "learning_rate": 4.8520826940125144e-05, + "loss": 0.7435213088989258, + "num_input_tokens_seen": 121128896, + "step": 620, + "train_runtime": 3700.6762, + "train_tokens_per_second": 32731.558 + }, + { + "epoch": 0.64512, + "grad_norm": 0.80078125, + "learning_rate": 4.849799683161046e-05, + "loss": 0.7433882713317871, + "num_input_tokens_seen": 123088256, + "step": 630, + "train_runtime": 3760.5434, + "train_tokens_per_second": 32731.508 + }, + { + "epoch": 0.65536, + "grad_norm": 0.828125, + "learning_rate": 4.8475198918992835e-05, + "loss": 0.7475492477416992, + "num_input_tokens_seen": 125038592, + "step": 640, + "train_runtime": 3819.9405, + "train_tokens_per_second": 32733.126 + }, + { + "epoch": 0.6656, + "grad_norm": 0.78125, + "learning_rate": 4.845243312667023e-05, + "loss": 0.7479681968688965, + "num_input_tokens_seen": 126982208, + "step": 650, + "train_runtime": 3879.7352, + "train_tokens_per_second": 32729.607 + }, + { + "epoch": 0.67584, + "grad_norm": 0.86328125, + "learning_rate": 4.842969937928884e-05, + "loss": 0.7469013214111329, + "num_input_tokens_seen": 128906880, + "step": 660, + "train_runtime": 3936.8159, + "train_tokens_per_second": 32743.944 + }, + { + "epoch": 0.68608, + "grad_norm": 0.8203125, + "learning_rate": 4.840699760174217e-05, + "loss": 0.7478948593139648, + "num_input_tokens_seen": 130882496, + "step": 670, + "train_runtime": 3996.7869, + "train_tokens_per_second": 32746.929 + }, + { + "epoch": 0.69632, + "grad_norm": 0.83203125, + "learning_rate": 4.8384327719169906e-05, + "loss": 0.7461132049560547, + "num_input_tokens_seen": 132826752, + "step": 680, + "train_runtime": 4055.0975, + "train_tokens_per_second": 32755.502 + }, + { + "epoch": 0.70656, + "grad_norm": 0.81640625, + "learning_rate": 4.836168965695694e-05, + "loss": 0.7483083724975585, + "num_input_tokens_seen": 134780160, + "step": 690, + "train_runtime": 4115.1829, + "train_tokens_per_second": 32751.925 + }, + { + "epoch": 0.7168, + "grad_norm": 0.77734375, + "learning_rate": 4.8339083340732304e-05, + "loss": 0.7420180320739747, + "num_input_tokens_seen": 136754944, + "step": 700, + "train_runtime": 4177.0113, + "train_tokens_per_second": 32739.903 + }, + { + "epoch": 0.72704, + "grad_norm": 0.796875, + "learning_rate": 4.8316508696368154e-05, + "loss": 0.7432499885559082, + "num_input_tokens_seen": 138696768, + "step": 710, + "train_runtime": 4235.2965, + "train_tokens_per_second": 32747.83 + }, + { + "epoch": 0.73728, + "grad_norm": 0.7578125, + "learning_rate": 4.8293965649978714e-05, + "loss": 0.7435853481292725, + "num_input_tokens_seen": 140653120, + "step": 720, + "train_runtime": 4295.3175, + "train_tokens_per_second": 32745.687 + }, + { + "epoch": 0.74752, + "grad_norm": 0.8515625, + "learning_rate": 4.8271454127919364e-05, + "loss": 0.7337657928466796, + "num_input_tokens_seen": 142610880, + "step": 730, + "train_runtime": 4354.6033, + "train_tokens_per_second": 32749.454 + }, + { + "epoch": 0.75776, + "grad_norm": 0.84765625, + "learning_rate": 4.824897405678549e-05, + "loss": 0.7439666748046875, + "num_input_tokens_seen": 144560448, + "step": 740, + "train_runtime": 4414.809, + "train_tokens_per_second": 32744.44 + }, + { + "epoch": 0.768, + "grad_norm": 0.78515625, + "learning_rate": 4.8226525363411576e-05, + "loss": 0.7391902923583984, + "num_input_tokens_seen": 146522432, + "step": 750, + "train_runtime": 4475.754, + "train_tokens_per_second": 32736.927 + }, + { + "epoch": 0.77824, + "grad_norm": 0.7890625, + "learning_rate": 4.820410797487017e-05, + "loss": 0.7402773857116699, + "num_input_tokens_seen": 148465152, + "step": 760, + "train_runtime": 4535.4981, + "train_tokens_per_second": 32734.035 + }, + { + "epoch": 0.78848, + "grad_norm": 0.7890625, + "learning_rate": 4.818172181847091e-05, + "loss": 0.7375380039215088, + "num_input_tokens_seen": 150407040, + "step": 770, + "train_runtime": 4594.3335, + "train_tokens_per_second": 32737.51 + }, + { + "epoch": 0.79872, + "grad_norm": 0.7890625, + "learning_rate": 4.81593668217595e-05, + "loss": 0.7325362205505371, + "num_input_tokens_seen": 152340608, + "step": 780, + "train_runtime": 4654.271, + "train_tokens_per_second": 32731.357 + }, + { + "epoch": 0.80896, + "grad_norm": 0.7890625, + "learning_rate": 4.813704291251675e-05, + "loss": 0.7388778686523437, + "num_input_tokens_seen": 154283968, + "step": 790, + "train_runtime": 4713.1072, + "train_tokens_per_second": 32735.086 + }, + { + "epoch": 0.8192, + "grad_norm": 0.77734375, + "learning_rate": 4.811475001875759e-05, + "loss": 0.734052324295044, + "num_input_tokens_seen": 156195392, + "step": 800, + "train_runtime": 4770.6972, + "train_tokens_per_second": 32740.58 + }, + { + "epoch": 0.82944, + "grad_norm": 0.76953125, + "learning_rate": 4.8092488068730105e-05, + "loss": 0.7394464015960693, + "num_input_tokens_seen": 158142080, + "step": 810, + "train_runtime": 4829.3439, + "train_tokens_per_second": 32746.08 + }, + { + "epoch": 0.83968, + "grad_norm": 0.76953125, + "learning_rate": 4.807025699091452e-05, + "loss": 0.7330001831054688, + "num_input_tokens_seen": 160100224, + "step": 820, + "train_runtime": 4889.6473, + "train_tokens_per_second": 32742.694 + }, + { + "epoch": 0.84992, + "grad_norm": 0.80859375, + "learning_rate": 4.8048056714022325e-05, + "loss": 0.7365771293640136, + "num_input_tokens_seen": 162014592, + "step": 830, + "train_runtime": 4946.8066, + "train_tokens_per_second": 32751.35 + }, + { + "epoch": 0.86016, + "grad_norm": 0.80078125, + "learning_rate": 4.802588716699519e-05, + "loss": 0.7319395065307617, + "num_input_tokens_seen": 163967296, + "step": 840, + "train_runtime": 5006.7801, + "train_tokens_per_second": 32749.051 + }, + { + "epoch": 0.8704, + "grad_norm": 0.77734375, + "learning_rate": 4.8003748279004156e-05, + "loss": 0.736358642578125, + "num_input_tokens_seen": 165888128, + "step": 850, + "train_runtime": 5065.2211, + "train_tokens_per_second": 32750.422 + }, + { + "epoch": 0.88064, + "grad_norm": 0.77734375, + "learning_rate": 4.798163997944854e-05, + "loss": 0.728258752822876, + "num_input_tokens_seen": 167838720, + "step": 860, + "train_runtime": 5125.0456, + "train_tokens_per_second": 32748.727 + }, + { + "epoch": 0.89088, + "grad_norm": 0.8203125, + "learning_rate": 4.79595621979551e-05, + "loss": 0.7310397624969482, + "num_input_tokens_seen": 169801024, + "step": 870, + "train_runtime": 5184.7795, + "train_tokens_per_second": 32749.903 + }, + { + "epoch": 0.90112, + "grad_norm": 0.76953125, + "learning_rate": 4.793751486437702e-05, + "loss": 0.7326676845550537, + "num_input_tokens_seen": 171754432, + "step": 880, + "train_runtime": 5243.9099, + "train_tokens_per_second": 32753.124 + }, + { + "epoch": 0.91136, + "grad_norm": 0.765625, + "learning_rate": 4.7915497908793064e-05, + "loss": 0.7294478416442871, + "num_input_tokens_seen": 173709952, + "step": 890, + "train_runtime": 5303.6797, + "train_tokens_per_second": 32752.723 + }, + { + "epoch": 0.9216, + "grad_norm": 0.76171875, + "learning_rate": 4.7893511261506516e-05, + "loss": 0.7317845344543457, + "num_input_tokens_seen": 175665088, + "step": 900, + "train_runtime": 5362.9004, + "train_tokens_per_second": 32755.613 + }, + { + "epoch": 0.93184, + "grad_norm": 0.82421875, + "learning_rate": 4.787155485304435e-05, + "loss": 0.7250267028808594, + "num_input_tokens_seen": 177597184, + "step": 910, + "train_runtime": 5421.8024, + "train_tokens_per_second": 32756.115 + }, + { + "epoch": 0.94208, + "grad_norm": 0.80859375, + "learning_rate": 4.784962861415629e-05, + "loss": 0.7249749183654786, + "num_input_tokens_seen": 179556736, + "step": 920, + "train_runtime": 5481.4833, + "train_tokens_per_second": 32756.961 + }, + { + "epoch": 0.95232, + "grad_norm": 0.8046875, + "learning_rate": 4.7827732475813884e-05, + "loss": 0.724465274810791, + "num_input_tokens_seen": 181523520, + "step": 930, + "train_runtime": 5543.5703, + "train_tokens_per_second": 32744.876 + }, + { + "epoch": 0.96256, + "grad_norm": 0.76953125, + "learning_rate": 4.7805866369209576e-05, + "loss": 0.7229520797729492, + "num_input_tokens_seen": 183456256, + "step": 940, + "train_runtime": 5602.4662, + "train_tokens_per_second": 32745.624 + }, + { + "epoch": 0.9728, + "grad_norm": 0.796875, + "learning_rate": 4.778403022575583e-05, + "loss": 0.727871322631836, + "num_input_tokens_seen": 185424896, + "step": 950, + "train_runtime": 5663.4738, + "train_tokens_per_second": 32740.488 + }, + { + "epoch": 0.98304, + "grad_norm": 0.84375, + "learning_rate": 4.7762223977084195e-05, + "loss": 0.7236977577209472, + "num_input_tokens_seen": 187388480, + "step": 960, + "train_runtime": 5723.3955, + "train_tokens_per_second": 32740.788 + }, + { + "epoch": 0.99328, + "grad_norm": 0.84765625, + "learning_rate": 4.774044755504444e-05, + "loss": 0.7206878185272216, + "num_input_tokens_seen": 189339648, + "step": 970, + "train_runtime": 5782.3102, + "train_tokens_per_second": 32744.637 + }, + { + "epoch": 1.003072, + "grad_norm": 0.7421875, + "learning_rate": 4.7718700891703616e-05, + "loss": 0.710568618774414, + "num_input_tokens_seen": 191208640, + "step": 980, + "train_runtime": 5839.7659, + "train_tokens_per_second": 32742.518 + }, + { + "epoch": 1.013312, + "grad_norm": 0.8125, + "learning_rate": 4.7696983919345215e-05, + "loss": 0.6669343948364258, + "num_input_tokens_seen": 193176512, + "step": 990, + "train_runtime": 5899.6765, + "train_tokens_per_second": 32743.577 + }, + { + "epoch": 1.023552, + "grad_norm": 0.75390625, + "learning_rate": 4.7675296570468216e-05, + "loss": 0.6645569801330566, + "num_input_tokens_seen": 195141440, + "step": 1000, + "train_runtime": 5960.2946, + "train_tokens_per_second": 32740.234 + }, + { + "epoch": 1.033792, + "grad_norm": 0.765625, + "learning_rate": 4.76536387777863e-05, + "loss": 0.6675229549407959, + "num_input_tokens_seen": 197092032, + "step": 1010, + "train_runtime": 6019.6761, + "train_tokens_per_second": 32741.302 + }, + { + "epoch": 1.044032, + "grad_norm": 0.80859375, + "learning_rate": 4.7632010474226915e-05, + "loss": 0.6588812828063965, + "num_input_tokens_seen": 199066240, + "step": 1020, + "train_runtime": 6079.3007, + "train_tokens_per_second": 32744.924 + }, + { + "epoch": 1.054272, + "grad_norm": 0.73828125, + "learning_rate": 4.761041159293035e-05, + "loss": 0.6624051094055176, + "num_input_tokens_seen": 201031808, + "step": 1030, + "train_runtime": 6139.0449, + "train_tokens_per_second": 32746.431 + }, + { + "epoch": 1.064512, + "grad_norm": 0.78515625, + "learning_rate": 4.7588842067249e-05, + "loss": 0.6641523361206054, + "num_input_tokens_seen": 202982784, + "step": 1040, + "train_runtime": 6200.7013, + "train_tokens_per_second": 32735.456 + }, + { + "epoch": 1.074752, + "grad_norm": 0.79296875, + "learning_rate": 4.756730183074637e-05, + "loss": 0.6600203037261962, + "num_input_tokens_seen": 204943872, + "step": 1050, + "train_runtime": 6261.3422, + "train_tokens_per_second": 32731.62 + }, + { + "epoch": 1.084992, + "grad_norm": 0.796875, + "learning_rate": 4.7545790817196314e-05, + "loss": 0.6627418041229248, + "num_input_tokens_seen": 206907648, + "step": 1060, + "train_runtime": 6321.396, + "train_tokens_per_second": 32731.322 + }, + { + "epoch": 1.095232, + "grad_norm": 0.765625, + "learning_rate": 4.752430896058212e-05, + "loss": 0.665987491607666, + "num_input_tokens_seen": 208866176, + "step": 1070, + "train_runtime": 6380.9787, + "train_tokens_per_second": 32732.624 + }, + { + "epoch": 1.105472, + "grad_norm": 0.828125, + "learning_rate": 4.750285619509567e-05, + "loss": 0.6638533115386963, + "num_input_tokens_seen": 210821184, + "step": 1080, + "train_runtime": 6441.6646, + "train_tokens_per_second": 32727.749 + }, + { + "epoch": 1.115712, + "grad_norm": 0.78125, + "learning_rate": 4.7481432455136644e-05, + "loss": 0.6705752372741699, + "num_input_tokens_seen": 212772928, + "step": 1090, + "train_runtime": 6499.8174, + "train_tokens_per_second": 32735.216 + }, + { + "epoch": 1.125952, + "grad_norm": 0.8203125, + "learning_rate": 4.7460037675311584e-05, + "loss": 0.6647239208221436, + "num_input_tokens_seen": 214713728, + "step": 1100, + "train_runtime": 6558.5813, + "train_tokens_per_second": 32737.831 + }, + { + "epoch": 1.136192, + "grad_norm": 0.796875, + "learning_rate": 4.7438671790433126e-05, + "loss": 0.6651959419250488, + "num_input_tokens_seen": 216668864, + "step": 1110, + "train_runtime": 6617.9492, + "train_tokens_per_second": 32739.578 + }, + { + "epoch": 1.146432, + "grad_norm": 0.8515625, + "learning_rate": 4.741733473551915e-05, + "loss": 0.6629802703857421, + "num_input_tokens_seen": 218629056, + "step": 1120, + "train_runtime": 6677.486, + "train_tokens_per_second": 32741.223 + }, + { + "epoch": 1.156672, + "grad_norm": 0.80078125, + "learning_rate": 4.7396026445791966e-05, + "loss": 0.6651780605316162, + "num_input_tokens_seen": 220590464, + "step": 1130, + "train_runtime": 6737.0253, + "train_tokens_per_second": 32743.007 + }, + { + "epoch": 1.166912, + "grad_norm": 0.84765625, + "learning_rate": 4.737474685667742e-05, + "loss": 0.6648248672485352, + "num_input_tokens_seen": 222559552, + "step": 1140, + "train_runtime": 6796.7448, + "train_tokens_per_second": 32745.021 + }, + { + "epoch": 1.177152, + "grad_norm": 0.80078125, + "learning_rate": 4.7353495903804165e-05, + "loss": 0.6667553901672363, + "num_input_tokens_seen": 224513216, + "step": 1150, + "train_runtime": 6856.6383, + "train_tokens_per_second": 32743.92 + }, + { + "epoch": 1.187392, + "grad_norm": 0.78515625, + "learning_rate": 4.733227352300277e-05, + "loss": 0.6689923286437989, + "num_input_tokens_seen": 226447168, + "step": 1160, + "train_runtime": 6915.0248, + "train_tokens_per_second": 32747.123 + }, + { + "epoch": 1.197632, + "grad_norm": 0.796875, + "learning_rate": 4.731107965030496e-05, + "loss": 0.6637516021728516, + "num_input_tokens_seen": 228394368, + "step": 1170, + "train_runtime": 6973.9487, + "train_tokens_per_second": 32749.649 + }, + { + "epoch": 1.207872, + "grad_norm": 0.796875, + "learning_rate": 4.728991422194278e-05, + "loss": 0.6736277580261231, + "num_input_tokens_seen": 230336512, + "step": 1180, + "train_runtime": 7032.2836, + "train_tokens_per_second": 32754.156 + }, + { + "epoch": 1.218112, + "grad_norm": 0.8125, + "learning_rate": 4.726877717434773e-05, + "loss": 0.6644347190856934, + "num_input_tokens_seen": 232273792, + "step": 1190, + "train_runtime": 7091.4876, + "train_tokens_per_second": 32753.888 + }, + { + "epoch": 1.228352, + "grad_norm": 0.828125, + "learning_rate": 4.724766844415013e-05, + "loss": 0.662445068359375, + "num_input_tokens_seen": 234234368, + "step": 1200, + "train_runtime": 7151.1612, + "train_tokens_per_second": 32754.732 + }, + { + "epoch": 1.238592, + "grad_norm": 0.7734375, + "learning_rate": 4.722658796817813e-05, + "loss": 0.6602650165557862, + "num_input_tokens_seen": 236184128, + "step": 1210, + "train_runtime": 7211.6529, + "train_tokens_per_second": 32750.346 + }, + { + "epoch": 1.248832, + "grad_norm": 0.8046875, + "learning_rate": 4.7205535683457044e-05, + "loss": 0.6661070823669434, + "num_input_tokens_seen": 238148992, + "step": 1220, + "train_runtime": 7272.0994, + "train_tokens_per_second": 32748.314 + }, + { + "epoch": 1.259072, + "grad_norm": 0.78125, + "learning_rate": 4.7184511527208484e-05, + "loss": 0.6634117603302002, + "num_input_tokens_seen": 240113216, + "step": 1230, + "train_runtime": 7332.2674, + "train_tokens_per_second": 32747.471 + }, + { + "epoch": 1.269312, + "grad_norm": 0.859375, + "learning_rate": 4.7163515436849644e-05, + "loss": 0.659183406829834, + "num_input_tokens_seen": 242056192, + "step": 1240, + "train_runtime": 7391.9286, + "train_tokens_per_second": 32746.013 + }, + { + "epoch": 1.279552, + "grad_norm": 0.8046875, + "learning_rate": 4.714254734999245e-05, + "loss": 0.6668022632598877, + "num_input_tokens_seen": 244001472, + "step": 1250, + "train_runtime": 7450.765, + "train_tokens_per_second": 32748.512 + }, + { + "epoch": 1.289792, + "grad_norm": 0.8125, + "learning_rate": 4.712160720444284e-05, + "loss": 0.6643959999084472, + "num_input_tokens_seen": 245958592, + "step": 1260, + "train_runtime": 7510.6075, + "train_tokens_per_second": 32748.162 + }, + { + "epoch": 1.300032, + "grad_norm": 0.79296875, + "learning_rate": 4.710069493819992e-05, + "loss": 0.6648179054260254, + "num_input_tokens_seen": 247917248, + "step": 1270, + "train_runtime": 7570.2595, + "train_tokens_per_second": 32748.844 + }, + { + "epoch": 1.3102719999999999, + "grad_norm": 0.75390625, + "learning_rate": 4.70798104894553e-05, + "loss": 0.6611180305480957, + "num_input_tokens_seen": 249863104, + "step": 1280, + "train_runtime": 7629.8975, + "train_tokens_per_second": 32747.898 + }, + { + "epoch": 1.320512, + "grad_norm": 0.765625, + "learning_rate": 4.705895379659219e-05, + "loss": 0.663090705871582, + "num_input_tokens_seen": 251816832, + "step": 1290, + "train_runtime": 7690.1349, + "train_tokens_per_second": 32745.438 + }, + { + "epoch": 1.330752, + "grad_norm": 0.80859375, + "learning_rate": 4.7038124798184766e-05, + "loss": 0.6652997016906739, + "num_input_tokens_seen": 253758592, + "step": 1300, + "train_runtime": 7750.7327, + "train_tokens_per_second": 32739.949 + }, + { + "epoch": 1.340992, + "grad_norm": 0.83984375, + "learning_rate": 4.7017323432997304e-05, + "loss": 0.6616836547851562, + "num_input_tokens_seen": 255687360, + "step": 1310, + "train_runtime": 7809.8266, + "train_tokens_per_second": 32739.185 + }, + { + "epoch": 1.351232, + "grad_norm": 0.796875, + "learning_rate": 4.6996549639983506e-05, + "loss": 0.6657767295837402, + "num_input_tokens_seen": 257640064, + "step": 1320, + "train_runtime": 7867.9946, + "train_tokens_per_second": 32745.328 + }, + { + "epoch": 1.361472, + "grad_norm": 0.78515625, + "learning_rate": 4.697580335828569e-05, + "loss": 0.6621747970581054, + "num_input_tokens_seen": 259595136, + "step": 1330, + "train_runtime": 7927.9177, + "train_tokens_per_second": 32744.429 + }, + { + "epoch": 1.371712, + "grad_norm": 0.84375, + "learning_rate": 4.6955084527234076e-05, + "loss": 0.6663068771362305, + "num_input_tokens_seen": 261559488, + "step": 1340, + "train_runtime": 7988.0773, + "train_tokens_per_second": 32743.735 + }, + { + "epoch": 1.381952, + "grad_norm": 0.8515625, + "learning_rate": 4.6934393086346034e-05, + "loss": 0.6628549098968506, + "num_input_tokens_seen": 263504960, + "step": 1350, + "train_runtime": 8047.4348, + "train_tokens_per_second": 32743.97 + }, + { + "epoch": 1.392192, + "grad_norm": 0.77734375, + "learning_rate": 4.6913728975325324e-05, + "loss": 0.6543691158294678, + "num_input_tokens_seen": 265475904, + "step": 1360, + "train_runtime": 8108.7462, + "train_tokens_per_second": 32739.451 + }, + { + "epoch": 1.4024320000000001, + "grad_norm": 0.7734375, + "learning_rate": 4.6893092134061393e-05, + "loss": 0.6563886165618896, + "num_input_tokens_seen": 267417152, + "step": 1370, + "train_runtime": 8167.6607, + "train_tokens_per_second": 32740.972 + }, + { + "epoch": 1.412672, + "grad_norm": 0.8203125, + "learning_rate": 4.687248250262859e-05, + "loss": 0.6647259712219238, + "num_input_tokens_seen": 269353152, + "step": 1380, + "train_runtime": 8226.112, + "train_tokens_per_second": 32743.677 + }, + { + "epoch": 1.422912, + "grad_norm": 0.77734375, + "learning_rate": 4.685190002128548e-05, + "loss": 0.6596537113189698, + "num_input_tokens_seen": 271311872, + "step": 1390, + "train_runtime": 8285.6877, + "train_tokens_per_second": 32744.641 + }, + { + "epoch": 1.433152, + "grad_norm": 0.796875, + "learning_rate": 4.6831344630474114e-05, + "loss": 0.6623442649841309, + "num_input_tokens_seen": 273284288, + "step": 1400, + "train_runtime": 8345.7692, + "train_tokens_per_second": 32745.249 + }, + { + "epoch": 1.443392, + "grad_norm": 0.8359375, + "learning_rate": 4.6810816270819276e-05, + "loss": 0.6568885803222656, + "num_input_tokens_seen": 275251328, + "step": 1410, + "train_runtime": 8404.8193, + "train_tokens_per_second": 32749.226 + }, + { + "epoch": 1.453632, + "grad_norm": 0.828125, + "learning_rate": 4.679031488312777e-05, + "loss": 0.6608157157897949, + "num_input_tokens_seen": 277202496, + "step": 1420, + "train_runtime": 8465.0573, + "train_tokens_per_second": 32746.677 + }, + { + "epoch": 1.463872, + "grad_norm": 0.7265625, + "learning_rate": 4.6769840408387717e-05, + "loss": 0.6612616539001465, + "num_input_tokens_seen": 279162944, + "step": 1430, + "train_runtime": 8526.0479, + "train_tokens_per_second": 32742.362 + }, + { + "epoch": 1.4741119999999999, + "grad_norm": 0.8125, + "learning_rate": 4.674939278776787e-05, + "loss": 0.6639165878295898, + "num_input_tokens_seen": 281109696, + "step": 1440, + "train_runtime": 8585.1435, + "train_tokens_per_second": 32743.739 + }, + { + "epoch": 1.484352, + "grad_norm": 0.8046875, + "learning_rate": 4.672897196261683e-05, + "loss": 0.6613688945770264, + "num_input_tokens_seen": 283056064, + "step": 1450, + "train_runtime": 8643.8923, + "train_tokens_per_second": 32746.366 + }, + { + "epoch": 1.494592, + "grad_norm": 0.78515625, + "learning_rate": 4.670857787446238e-05, + "loss": 0.6640623569488525, + "num_input_tokens_seen": 285002496, + "step": 1460, + "train_runtime": 8703.5414, + "train_tokens_per_second": 32745.578 + }, + { + "epoch": 1.504832, + "grad_norm": 0.8359375, + "learning_rate": 4.668821046501082e-05, + "loss": 0.6640773773193359, + "num_input_tokens_seen": 286959616, + "step": 1470, + "train_runtime": 8763.9486, + "train_tokens_per_second": 32743.188 + }, + { + "epoch": 1.515072, + "grad_norm": 0.79296875, + "learning_rate": 4.6667869676146194e-05, + "loss": 0.6624667167663574, + "num_input_tokens_seen": 288905280, + "step": 1480, + "train_runtime": 8822.4748, + "train_tokens_per_second": 32746.512 + }, + { + "epoch": 1.525312, + "grad_norm": 0.77734375, + "learning_rate": 4.6647555449929645e-05, + "loss": 0.6601789474487305, + "num_input_tokens_seen": 290846400, + "step": 1490, + "train_runtime": 8881.7856, + "train_tokens_per_second": 32746.388 + }, + { + "epoch": 1.535552, + "grad_norm": 0.83984375, + "learning_rate": 4.662726772859869e-05, + "loss": 0.6581646919250488, + "num_input_tokens_seen": 292768064, + "step": 1500, + "train_runtime": 8939.6828, + "train_tokens_per_second": 32749.268 + }, + { + "epoch": 1.545792, + "grad_norm": 0.765625, + "learning_rate": 4.660700645456655e-05, + "loss": 0.6602959632873535, + "num_input_tokens_seen": 294723392, + "step": 1510, + "train_runtime": 8999.2269, + "train_tokens_per_second": 32749.857 + }, + { + "epoch": 1.556032, + "grad_norm": 0.80859375, + "learning_rate": 4.658677157042149e-05, + "loss": 0.6587230682373046, + "num_input_tokens_seen": 296664896, + "step": 1520, + "train_runtime": 9057.6637, + "train_tokens_per_second": 32752.916 + }, + { + "epoch": 1.566272, + "grad_norm": 0.796875, + "learning_rate": 4.656656301892605e-05, + "loss": 0.6597561836242676, + "num_input_tokens_seen": 298606592, + "step": 1530, + "train_runtime": 9116.7544, + "train_tokens_per_second": 32753.607 + }, + { + "epoch": 1.5765120000000001, + "grad_norm": 0.8203125, + "learning_rate": 4.6546380743016465e-05, + "loss": 0.6614864349365235, + "num_input_tokens_seen": 300562496, + "step": 1540, + "train_runtime": 9177.0472, + "train_tokens_per_second": 32751.547 + }, + { + "epoch": 1.5867520000000002, + "grad_norm": 0.8046875, + "learning_rate": 4.652622468580193e-05, + "loss": 0.6568780422210694, + "num_input_tokens_seen": 302519104, + "step": 1550, + "train_runtime": 9236.6099, + "train_tokens_per_second": 32752.179 + }, + { + "epoch": 1.596992, + "grad_norm": 0.8046875, + "learning_rate": 4.650609479056392e-05, + "loss": 0.6616767883300781, + "num_input_tokens_seen": 304449152, + "step": 1560, + "train_runtime": 9296.2427, + "train_tokens_per_second": 32749.699 + }, + { + "epoch": 1.607232, + "grad_norm": 0.7734375, + "learning_rate": 4.648599100075556e-05, + "loss": 0.6637643814086914, + "num_input_tokens_seen": 306388224, + "step": 1570, + "train_runtime": 9355.7549, + "train_tokens_per_second": 32748.637 + }, + { + "epoch": 1.617472, + "grad_norm": 0.74609375, + "learning_rate": 4.6465913260000945e-05, + "loss": 0.6638627052307129, + "num_input_tokens_seen": 308336512, + "step": 1580, + "train_runtime": 9414.6843, + "train_tokens_per_second": 32750.595 + }, + { + "epoch": 1.627712, + "grad_norm": 0.7734375, + "learning_rate": 4.644586151209444e-05, + "loss": 0.661181640625, + "num_input_tokens_seen": 310314752, + "step": 1590, + "train_runtime": 9476.2434, + "train_tokens_per_second": 32746.6 + }, + { + "epoch": 1.6379519999999999, + "grad_norm": 0.8203125, + "learning_rate": 4.6425835701000084e-05, + "loss": 0.6550232887268066, + "num_input_tokens_seen": 312281088, + "step": 1600, + "train_runtime": 9537.1306, + "train_tokens_per_second": 32743.715 + }, + { + "epoch": 1.6481919999999999, + "grad_norm": 0.74609375, + "learning_rate": 4.640583577085084e-05, + "loss": 0.6569297790527344, + "num_input_tokens_seen": 314245824, + "step": 1610, + "train_runtime": 9597.5734, + "train_tokens_per_second": 32742.216 + }, + { + "epoch": 1.658432, + "grad_norm": 0.796875, + "learning_rate": 4.638586166594806e-05, + "loss": 0.652769947052002, + "num_input_tokens_seen": 316216768, + "step": 1620, + "train_runtime": 9658.3636, + "train_tokens_per_second": 32740.201 + }, + { + "epoch": 1.668672, + "grad_norm": 0.8125, + "learning_rate": 4.6365913330760726e-05, + "loss": 0.6508107662200928, + "num_input_tokens_seen": 318168448, + "step": 1630, + "train_runtime": 9716.9225, + "train_tokens_per_second": 32743.747 + }, + { + "epoch": 1.678912, + "grad_norm": 0.80859375, + "learning_rate": 4.6345990709924855e-05, + "loss": 0.6558364868164063, + "num_input_tokens_seen": 320136128, + "step": 1640, + "train_runtime": 9776.5171, + "train_tokens_per_second": 32745.417 + }, + { + "epoch": 1.689152, + "grad_norm": 0.8203125, + "learning_rate": 4.632609374824284e-05, + "loss": 0.6574318885803223, + "num_input_tokens_seen": 322099456, + "step": 1650, + "train_runtime": 9836.4836, + "train_tokens_per_second": 32745.387 + }, + { + "epoch": 1.699392, + "grad_norm": 0.7734375, + "learning_rate": 4.630622239068285e-05, + "loss": 0.6548179626464844, + "num_input_tokens_seen": 324049408, + "step": 1660, + "train_runtime": 9895.3464, + "train_tokens_per_second": 32747.657 + }, + { + "epoch": 1.709632, + "grad_norm": 0.828125, + "learning_rate": 4.628637658237808e-05, + "loss": 0.6559741497039795, + "num_input_tokens_seen": 325990016, + "step": 1670, + "train_runtime": 9953.6672, + "train_tokens_per_second": 32750.745 + }, + { + "epoch": 1.719872, + "grad_norm": 0.82421875, + "learning_rate": 4.626655626862625e-05, + "loss": 0.6587584495544434, + "num_input_tokens_seen": 327937472, + "step": 1680, + "train_runtime": 10012.7266, + "train_tokens_per_second": 32752.065 + }, + { + "epoch": 1.730112, + "grad_norm": 0.79296875, + "learning_rate": 4.624676139488888e-05, + "loss": 0.656014347076416, + "num_input_tokens_seen": 329908352, + "step": 1690, + "train_runtime": 10072.9189, + "train_tokens_per_second": 32752.011 + }, + { + "epoch": 1.7403520000000001, + "grad_norm": 0.80078125, + "learning_rate": 4.6226991906790686e-05, + "loss": 0.658781623840332, + "num_input_tokens_seen": 331853696, + "step": 1700, + "train_runtime": 10131.69, + "train_tokens_per_second": 32754.032 + }, + { + "epoch": 1.7505920000000001, + "grad_norm": 0.78125, + "learning_rate": 4.620724775011897e-05, + "loss": 0.6597333908081054, + "num_input_tokens_seen": 333782656, + "step": 1710, + "train_runtime": 10190.1862, + "train_tokens_per_second": 32755.305 + }, + { + "epoch": 1.760832, + "grad_norm": 0.76953125, + "learning_rate": 4.618752887082297e-05, + "loss": 0.6641686439514161, + "num_input_tokens_seen": 335741568, + "step": 1720, + "train_runtime": 10248.8549, + "train_tokens_per_second": 32758.935 + }, + { + "epoch": 1.771072, + "grad_norm": 0.81640625, + "learning_rate": 4.616783521501325e-05, + "loss": 0.6582834720611572, + "num_input_tokens_seen": 337673856, + "step": 1730, + "train_runtime": 10308.2562, + "train_tokens_per_second": 32757.612 + }, + { + "epoch": 1.781312, + "grad_norm": 0.8125, + "learning_rate": 4.614816672896108e-05, + "loss": 0.6613027572631835, + "num_input_tokens_seen": 339633088, + "step": 1740, + "train_runtime": 10368.3354, + "train_tokens_per_second": 32756.761 + }, + { + "epoch": 1.791552, + "grad_norm": 0.80078125, + "learning_rate": 4.612852335909782e-05, + "loss": 0.6542655467987061, + "num_input_tokens_seen": 341581696, + "step": 1750, + "train_runtime": 10428.6115, + "train_tokens_per_second": 32754.283 + }, + { + "epoch": 1.8017919999999998, + "grad_norm": 0.7734375, + "learning_rate": 4.6108905052014323e-05, + "loss": 0.6533420562744141, + "num_input_tokens_seen": 343534720, + "step": 1760, + "train_runtime": 10488.0364, + "train_tokens_per_second": 32754.913 + }, + { + "epoch": 1.8120319999999999, + "grad_norm": 0.76171875, + "learning_rate": 4.608931175446027e-05, + "loss": 0.6449230194091797, + "num_input_tokens_seen": 345495296, + "step": 1770, + "train_runtime": 10547.857, + "train_tokens_per_second": 32755.023 + }, + { + "epoch": 1.822272, + "grad_norm": 0.796875, + "learning_rate": 4.606974341334367e-05, + "loss": 0.6538689613342286, + "num_input_tokens_seen": 347431552, + "step": 1780, + "train_runtime": 10606.573, + "train_tokens_per_second": 32756.25 + }, + { + "epoch": 1.832512, + "grad_norm": 0.7734375, + "learning_rate": 4.605019997573011e-05, + "loss": 0.6490047454833985, + "num_input_tokens_seen": 349382976, + "step": 1790, + "train_runtime": 10666.3471, + "train_tokens_per_second": 32755.635 + }, + { + "epoch": 1.842752, + "grad_norm": 0.79296875, + "learning_rate": 4.603068138884229e-05, + "loss": 0.6522298812866211, + "num_input_tokens_seen": 351343680, + "step": 1800, + "train_runtime": 10726.0131, + "train_tokens_per_second": 32756.223 + }, + { + "epoch": 1.852992, + "grad_norm": 0.765625, + "learning_rate": 4.6011187600059345e-05, + "loss": 0.6543211936950684, + "num_input_tokens_seen": 353284096, + "step": 1810, + "train_runtime": 10785.9291, + "train_tokens_per_second": 32754.165 + }, + { + "epoch": 1.863232, + "grad_norm": 0.7734375, + "learning_rate": 4.599171855691629e-05, + "loss": 0.6490991115570068, + "num_input_tokens_seen": 355207296, + "step": 1820, + "train_runtime": 10845.1553, + "train_tokens_per_second": 32752.624 + }, + { + "epoch": 1.873472, + "grad_norm": 0.80078125, + "learning_rate": 4.597227420710335e-05, + "loss": 0.6518874168395996, + "num_input_tokens_seen": 357148288, + "step": 1830, + "train_runtime": 10904.8516, + "train_tokens_per_second": 32751.32 + }, + { + "epoch": 1.883712, + "grad_norm": 0.84765625, + "learning_rate": 4.595285449846551e-05, + "loss": 0.6608992099761963, + "num_input_tokens_seen": 359101824, + "step": 1840, + "train_runtime": 10964.5719, + "train_tokens_per_second": 32751.103 + }, + { + "epoch": 1.893952, + "grad_norm": 0.76171875, + "learning_rate": 4.593345937900178e-05, + "loss": 0.6517756462097168, + "num_input_tokens_seen": 361063168, + "step": 1850, + "train_runtime": 11023.8575, + "train_tokens_per_second": 32752.888 + }, + { + "epoch": 1.904192, + "grad_norm": 0.79296875, + "learning_rate": 4.591408879686472e-05, + "loss": 0.6512045860290527, + "num_input_tokens_seen": 362991872, + "step": 1860, + "train_runtime": 11081.5644, + "train_tokens_per_second": 32756.374 + }, + { + "epoch": 1.9144320000000001, + "grad_norm": 0.7578125, + "learning_rate": 4.5894742700359775e-05, + "loss": 0.6591498374938964, + "num_input_tokens_seen": 364946688, + "step": 1870, + "train_runtime": 11140.0845, + "train_tokens_per_second": 32759.777 + }, + { + "epoch": 1.9246720000000002, + "grad_norm": 0.7578125, + "learning_rate": 4.587542103794477e-05, + "loss": 0.6532726287841797, + "num_input_tokens_seen": 366891072, + "step": 1880, + "train_runtime": 11198.9489, + "train_tokens_per_second": 32761.206 + }, + { + "epoch": 1.934912, + "grad_norm": 0.7578125, + "learning_rate": 4.5856123758229247e-05, + "loss": 0.6575124740600586, + "num_input_tokens_seen": 368845632, + "step": 1890, + "train_runtime": 11258.7243, + "train_tokens_per_second": 32760.873 + }, + { + "epoch": 1.945152, + "grad_norm": 0.8046875, + "learning_rate": 4.5836850809973993e-05, + "loss": 0.6555071830749511, + "num_input_tokens_seen": 370797568, + "step": 1900, + "train_runtime": 11318.3367, + "train_tokens_per_second": 32760.783 + }, + { + "epoch": 1.955392, + "grad_norm": 0.796875, + "learning_rate": 4.5817602142090385e-05, + "loss": 0.6574622631072998, + "num_input_tokens_seen": 372745728, + "step": 1910, + "train_runtime": 11376.6941, + "train_tokens_per_second": 32763.976 + }, + { + "epoch": 1.965632, + "grad_norm": 0.76953125, + "learning_rate": 4.579837770363989e-05, + "loss": 0.6541840553283691, + "num_input_tokens_seen": 374707904, + "step": 1920, + "train_runtime": 11435.9496, + "train_tokens_per_second": 32765.788 + }, + { + "epoch": 1.9758719999999999, + "grad_norm": 0.8125, + "learning_rate": 4.57791774438334e-05, + "loss": 0.6511831283569336, + "num_input_tokens_seen": 376672064, + "step": 1930, + "train_runtime": 11495.5055, + "train_tokens_per_second": 32766.899 + }, + { + "epoch": 1.9861119999999999, + "grad_norm": 0.75390625, + "learning_rate": 4.576000131203078e-05, + "loss": 0.653932523727417, + "num_input_tokens_seen": 378613568, + "step": 1940, + "train_runtime": 11555.2052, + "train_tokens_per_second": 32765.629 + }, + { + "epoch": 1.996352, + "grad_norm": 0.78515625, + "learning_rate": 4.574084925774023e-05, + "loss": 0.6487459659576416, + "num_input_tokens_seen": 380541952, + "step": 1950, + "train_runtime": 11613.8196, + "train_tokens_per_second": 32766.305 + }, + { + "epoch": 2.006144, + "grad_norm": 0.859375, + "learning_rate": 4.5721721230617795e-05, + "loss": 0.6155517578125, + "num_input_tokens_seen": 382413824, + "step": 1960, + "train_runtime": 11670.483, + "train_tokens_per_second": 32767.609 + }, + { + "epoch": 2.016384, + "grad_norm": 0.80859375, + "learning_rate": 4.57026171804667e-05, + "loss": 0.5954423427581788, + "num_input_tokens_seen": 384361152, + "step": 1970, + "train_runtime": 11729.6669, + "train_tokens_per_second": 32768.292 + }, + { + "epoch": 2.026624, + "grad_norm": 0.859375, + "learning_rate": 4.568353705723692e-05, + "loss": 0.5828520774841308, + "num_input_tokens_seen": 386292032, + "step": 1980, + "train_runtime": 11788.2977, + "train_tokens_per_second": 32769.111 + }, + { + "epoch": 2.036864, + "grad_norm": 0.828125, + "learning_rate": 4.566448081102455e-05, + "loss": 0.5893785953521729, + "num_input_tokens_seen": 388221376, + "step": 1990, + "train_runtime": 11845.8075, + "train_tokens_per_second": 32772.893 + }, + { + "epoch": 2.047104, + "grad_norm": 0.80859375, + "learning_rate": 4.564544839207128e-05, + "loss": 0.5838380813598633, + "num_input_tokens_seen": 390196224, + "step": 2000, + "train_runtime": 11907.5584, + "train_tokens_per_second": 32768.785 + }, + { + "epoch": 2.057344, + "grad_norm": 0.78125, + "learning_rate": 4.562643975076387e-05, + "loss": 0.5901473999023438, + "num_input_tokens_seen": 392126976, + "step": 2010, + "train_runtime": 11965.9242, + "train_tokens_per_second": 32770.304 + }, + { + "epoch": 2.067584, + "grad_norm": 0.80859375, + "learning_rate": 4.560745483763357e-05, + "loss": 0.5894758224487304, + "num_input_tokens_seen": 394095104, + "step": 2020, + "train_runtime": 12026.1967, + "train_tokens_per_second": 32769.72 + }, + { + "epoch": 2.077824, + "grad_norm": 0.84765625, + "learning_rate": 4.5588493603355595e-05, + "loss": 0.5862543106079101, + "num_input_tokens_seen": 396042368, + "step": 2030, + "train_runtime": 12085.3433, + "train_tokens_per_second": 32770.469 + }, + { + "epoch": 2.088064, + "grad_norm": 0.80078125, + "learning_rate": 4.556955599874859e-05, + "loss": 0.5956791400909424, + "num_input_tokens_seen": 397981952, + "step": 2040, + "train_runtime": 12144.0426, + "train_tokens_per_second": 32771.785 + }, + { + "epoch": 2.098304, + "grad_norm": 0.8125, + "learning_rate": 4.555064197477409e-05, + "loss": 0.5897563934326172, + "num_input_tokens_seen": 399907072, + "step": 2050, + "train_runtime": 12201.9828, + "train_tokens_per_second": 32773.942 + }, + { + "epoch": 2.108544, + "grad_norm": 0.78515625, + "learning_rate": 4.5531751482536e-05, + "loss": 0.5853563785552979, + "num_input_tokens_seen": 401862848, + "step": 2060, + "train_runtime": 12262.5862, + "train_tokens_per_second": 32771.46 + }, + { + "epoch": 2.118784, + "grad_norm": 0.83203125, + "learning_rate": 4.5512884473280024e-05, + "loss": 0.5935741424560547, + "num_input_tokens_seen": 403810240, + "step": 2070, + "train_runtime": 12322.1702, + "train_tokens_per_second": 32771.033 + }, + { + "epoch": 2.129024, + "grad_norm": 0.87890625, + "learning_rate": 4.549404089839322e-05, + "loss": 0.59158935546875, + "num_input_tokens_seen": 405776256, + "step": 2080, + "train_runtime": 12383.1293, + "train_tokens_per_second": 32768.475 + }, + { + "epoch": 2.139264, + "grad_norm": 0.8125, + "learning_rate": 4.547522070940335e-05, + "loss": 0.5877013206481934, + "num_input_tokens_seen": 407735744, + "step": 2090, + "train_runtime": 12443.2739, + "train_tokens_per_second": 32767.562 + }, + { + "epoch": 2.149504, + "grad_norm": 0.99609375, + "learning_rate": 4.545642385797848e-05, + "loss": 0.5920863628387452, + "num_input_tokens_seen": 409688064, + "step": 2100, + "train_runtime": 12503.2878, + "train_tokens_per_second": 32766.427 + }, + { + "epoch": 2.159744, + "grad_norm": 0.84375, + "learning_rate": 4.543765029592637e-05, + "loss": 0.5942326545715332, + "num_input_tokens_seen": 411607680, + "step": 2110, + "train_runtime": 12561.0265, + "train_tokens_per_second": 32768.634 + }, + { + "epoch": 2.169984, + "grad_norm": 0.84765625, + "learning_rate": 4.541889997519403e-05, + "loss": 0.5903902053833008, + "num_input_tokens_seen": 413559232, + "step": 2120, + "train_runtime": 12620.5344, + "train_tokens_per_second": 32768.758 + }, + { + "epoch": 2.180224, + "grad_norm": 1.0546875, + "learning_rate": 4.5400172847867095e-05, + "loss": 0.5925062656402588, + "num_input_tokens_seen": 415502272, + "step": 2130, + "train_runtime": 12679.2897, + "train_tokens_per_second": 32770.154 + }, + { + "epoch": 2.190464, + "grad_norm": 0.84375, + "learning_rate": 4.5381468866169466e-05, + "loss": 0.5988312244415284, + "num_input_tokens_seen": 417454208, + "step": 2140, + "train_runtime": 12738.0225, + "train_tokens_per_second": 32772.293 + }, + { + "epoch": 2.200704, + "grad_norm": 0.8125, + "learning_rate": 4.5362787982462616e-05, + "loss": 0.5885556221008301, + "num_input_tokens_seen": 419408384, + "step": 2150, + "train_runtime": 12798.0808, + "train_tokens_per_second": 32771.194 + }, + { + "epoch": 2.210944, + "grad_norm": 0.8125, + "learning_rate": 4.5344130149245275e-05, + "loss": 0.5948306560516358, + "num_input_tokens_seen": 421340992, + "step": 2160, + "train_runtime": 12857.292, + "train_tokens_per_second": 32770.586 + }, + { + "epoch": 2.221184, + "grad_norm": 0.82421875, + "learning_rate": 4.5325495319152715e-05, + "loss": 0.595841646194458, + "num_input_tokens_seen": 423294080, + "step": 2170, + "train_runtime": 12915.8392, + "train_tokens_per_second": 32773.254 + }, + { + "epoch": 2.231424, + "grad_norm": 0.828125, + "learning_rate": 4.530688344495644e-05, + "loss": 0.5914995193481445, + "num_input_tokens_seen": 425247872, + "step": 2180, + "train_runtime": 12976.1802, + "train_tokens_per_second": 32771.422 + }, + { + "epoch": 2.241664, + "grad_norm": 0.82421875, + "learning_rate": 4.528829447956357e-05, + "loss": 0.5963719367980957, + "num_input_tokens_seen": 427203328, + "step": 2190, + "train_runtime": 13034.7435, + "train_tokens_per_second": 32774.203 + }, + { + "epoch": 2.251904, + "grad_norm": 0.81640625, + "learning_rate": 4.526972837601633e-05, + "loss": 0.5922994613647461, + "num_input_tokens_seen": 429173888, + "step": 2200, + "train_runtime": 13094.3153, + "train_tokens_per_second": 32775.588 + }, + { + "epoch": 2.262144, + "grad_norm": 0.9609375, + "learning_rate": 4.525118508749165e-05, + "loss": 0.5888791561126709, + "num_input_tokens_seen": 431118720, + "step": 2210, + "train_runtime": 13154.6565, + "train_tokens_per_second": 32773.088 + }, + { + "epoch": 2.272384, + "grad_norm": 0.84765625, + "learning_rate": 4.5232664567300546e-05, + "loss": 0.5937693119049072, + "num_input_tokens_seen": 433071488, + "step": 2220, + "train_runtime": 13213.6282, + "train_tokens_per_second": 32774.608 + }, + { + "epoch": 2.282624, + "grad_norm": 0.83203125, + "learning_rate": 4.521416676888773e-05, + "loss": 0.5958268165588378, + "num_input_tokens_seen": 435030464, + "step": 2230, + "train_runtime": 13273.5292, + "train_tokens_per_second": 32774.288 + }, + { + "epoch": 2.292864, + "grad_norm": 0.91796875, + "learning_rate": 4.519569164583107e-05, + "loss": 0.5926674842834473, + "num_input_tokens_seen": 436981504, + "step": 2240, + "train_runtime": 13333.8604, + "train_tokens_per_second": 32772.317 + }, + { + "epoch": 2.303104, + "grad_norm": 0.83984375, + "learning_rate": 4.517723915184109e-05, + "loss": 0.5918882846832275, + "num_input_tokens_seen": 438963648, + "step": 2250, + "train_runtime": 13394.8028, + "train_tokens_per_second": 32771.191 + }, + { + "epoch": 2.313344, + "grad_norm": 0.8203125, + "learning_rate": 4.5158809240760506e-05, + "loss": 0.5919432163238525, + "num_input_tokens_seen": 440906176, + "step": 2260, + "train_runtime": 13454.508, + "train_tokens_per_second": 32770.145 + }, + { + "epoch": 2.323584, + "grad_norm": 0.8515625, + "learning_rate": 4.514040186656375e-05, + "loss": 0.5900529384613037, + "num_input_tokens_seen": 442885504, + "step": 2270, + "train_runtime": 13515.0748, + "train_tokens_per_second": 32769.741 + }, + { + "epoch": 2.333824, + "grad_norm": 0.81640625, + "learning_rate": 4.512201698335644e-05, + "loss": 0.5980159759521484, + "num_input_tokens_seen": 444823488, + "step": 2280, + "train_runtime": 13575.1652, + "train_tokens_per_second": 32767.446 + }, + { + "epoch": 2.344064, + "grad_norm": 0.80078125, + "learning_rate": 4.510365454537496e-05, + "loss": 0.5917848587036133, + "num_input_tokens_seen": 446776320, + "step": 2290, + "train_runtime": 13634.2022, + "train_tokens_per_second": 32768.791 + }, + { + "epoch": 2.354304, + "grad_norm": 0.78125, + "learning_rate": 4.5085314506985945e-05, + "loss": 0.6014077186584472, + "num_input_tokens_seen": 448734784, + "step": 2300, + "train_runtime": 13693.3577, + "train_tokens_per_second": 32770.252 + }, + { + "epoch": 2.364544, + "grad_norm": 0.83203125, + "learning_rate": 4.50669968226858e-05, + "loss": 0.5941033363342285, + "num_input_tokens_seen": 450674688, + "step": 2310, + "train_runtime": 13751.6852, + "train_tokens_per_second": 32772.324 + }, + { + "epoch": 2.374784, + "grad_norm": 0.86328125, + "learning_rate": 4.504870144710027e-05, + "loss": 0.5903738021850586, + "num_input_tokens_seen": 452620544, + "step": 2320, + "train_runtime": 13810.0811, + "train_tokens_per_second": 32774.648 + }, + { + "epoch": 2.385024, + "grad_norm": 0.828125, + "learning_rate": 4.5030428334983884e-05, + "loss": 0.59536771774292, + "num_input_tokens_seen": 454586752, + "step": 2330, + "train_runtime": 13870.4679, + "train_tokens_per_second": 32773.714 + }, + { + "epoch": 2.395264, + "grad_norm": 0.80078125, + "learning_rate": 4.501217744121959e-05, + "loss": 0.5988545417785645, + "num_input_tokens_seen": 456539968, + "step": 2340, + "train_runtime": 13931.2089, + "train_tokens_per_second": 32771.023 + }, + { + "epoch": 2.405504, + "grad_norm": 0.796875, + "learning_rate": 4.499394872081821e-05, + "loss": 0.5976475715637207, + "num_input_tokens_seen": 458487936, + "step": 2350, + "train_runtime": 13991.1659, + "train_tokens_per_second": 32769.816 + }, + { + "epoch": 2.415744, + "grad_norm": 0.83203125, + "learning_rate": 4.4975742128918e-05, + "loss": 0.5921101570129395, + "num_input_tokens_seen": 460446336, + "step": 2360, + "train_runtime": 14051.3115, + "train_tokens_per_second": 32768.922 + }, + { + "epoch": 2.425984, + "grad_norm": 0.85546875, + "learning_rate": 4.495755762078418e-05, + "loss": 0.5958994388580322, + "num_input_tokens_seen": 462432640, + "step": 2370, + "train_runtime": 14112.7434, + "train_tokens_per_second": 32767.027 + }, + { + "epoch": 2.436224, + "grad_norm": 0.8359375, + "learning_rate": 4.49393951518085e-05, + "loss": 0.5889082908630371, + "num_input_tokens_seen": 464370368, + "step": 2380, + "train_runtime": 14172.1972, + "train_tokens_per_second": 32766.293 + }, + { + "epoch": 2.446464, + "grad_norm": 0.79296875, + "learning_rate": 4.4921254677508716e-05, + "loss": 0.5928831100463867, + "num_input_tokens_seen": 466318912, + "step": 2390, + "train_runtime": 14232.8255, + "train_tokens_per_second": 32763.622 + }, + { + "epoch": 2.456704, + "grad_norm": 0.8125, + "learning_rate": 4.490313615352821e-05, + "loss": 0.5945481777191162, + "num_input_tokens_seen": 468270848, + "step": 2400, + "train_runtime": 14292.55, + "train_tokens_per_second": 32763.282 + }, + { + "epoch": 2.466944, + "grad_norm": 0.81640625, + "learning_rate": 4.48850395356355e-05, + "loss": 0.5897410392761231, + "num_input_tokens_seen": 470260224, + "step": 2410, + "train_runtime": 14353.7153, + "train_tokens_per_second": 32762.265 + }, + { + "epoch": 2.477184, + "grad_norm": 0.8515625, + "learning_rate": 4.486696477972375e-05, + "loss": 0.5936931133270263, + "num_input_tokens_seen": 472188416, + "step": 2420, + "train_runtime": 14411.0634, + "train_tokens_per_second": 32765.689 + }, + { + "epoch": 2.487424, + "grad_norm": 0.7890625, + "learning_rate": 4.484891184181041e-05, + "loss": 0.5995767116546631, + "num_input_tokens_seen": 474117824, + "step": 2430, + "train_runtime": 14469.5287, + "train_tokens_per_second": 32766.639 + }, + { + "epoch": 2.497664, + "grad_norm": 0.8125, + "learning_rate": 4.483088067803662e-05, + "loss": 0.6011977195739746, + "num_input_tokens_seen": 476069504, + "step": 2440, + "train_runtime": 14528.4206, + "train_tokens_per_second": 32768.153 + }, + { + "epoch": 2.507904, + "grad_norm": 0.77734375, + "learning_rate": 4.481287124466697e-05, + "loss": 0.5965440273284912, + "num_input_tokens_seen": 478029056, + "step": 2450, + "train_runtime": 14588.4906, + "train_tokens_per_second": 32767.547 + }, + { + "epoch": 2.518144, + "grad_norm": 0.84765625, + "learning_rate": 4.479488349808885e-05, + "loss": 0.5970401763916016, + "num_input_tokens_seen": 479992832, + "step": 2460, + "train_runtime": 14648.8001, + "train_tokens_per_second": 32766.7 + }, + { + "epoch": 2.528384, + "grad_norm": 0.7890625, + "learning_rate": 4.4776917394812114e-05, + "loss": 0.5875445365905761, + "num_input_tokens_seen": 481965248, + "step": 2470, + "train_runtime": 14709.6035, + "train_tokens_per_second": 32765.346 + }, + { + "epoch": 2.538624, + "grad_norm": 0.83203125, + "learning_rate": 4.475897289146862e-05, + "loss": 0.5980213165283204, + "num_input_tokens_seen": 483911872, + "step": 2480, + "train_runtime": 14768.9505, + "train_tokens_per_second": 32765.488 + }, + { + "epoch": 2.548864, + "grad_norm": 0.83203125, + "learning_rate": 4.4741049944811806e-05, + "loss": 0.5927760601043701, + "num_input_tokens_seen": 485858560, + "step": 2490, + "train_runtime": 14828.1435, + "train_tokens_per_second": 32765.974 + }, + { + "epoch": 2.559104, + "grad_norm": 0.82421875, + "learning_rate": 4.472314851171621e-05, + "loss": 0.5933934211730957, + "num_input_tokens_seen": 487808832, + "step": 2500, + "train_runtime": 14886.7134, + "train_tokens_per_second": 32768.068 + }, + { + "epoch": 2.569344, + "grad_norm": 0.83984375, + "learning_rate": 4.4705268549177084e-05, + "loss": 0.5965798854827881, + "num_input_tokens_seen": 489736832, + "step": 2510, + "train_runtime": 14944.9788, + "train_tokens_per_second": 32769.323 + }, + { + "epoch": 2.579584, + "grad_norm": 0.8359375, + "learning_rate": 4.468741001430989e-05, + "loss": 0.5946637153625488, + "num_input_tokens_seen": 491707968, + "step": 2520, + "train_runtime": 15005.5446, + "train_tokens_per_second": 32768.419 + }, + { + "epoch": 2.589824, + "grad_norm": 0.90625, + "learning_rate": 4.466957286434997e-05, + "loss": 0.5971454620361328, + "num_input_tokens_seen": 493661504, + "step": 2530, + "train_runtime": 15065.0579, + "train_tokens_per_second": 32768.643 + }, + { + "epoch": 2.600064, + "grad_norm": 0.83203125, + "learning_rate": 4.4651757056652e-05, + "loss": 0.6013389587402344, + "num_input_tokens_seen": 495603392, + "step": 2540, + "train_runtime": 15124.1333, + "train_tokens_per_second": 32769.044 + }, + { + "epoch": 2.610304, + "grad_norm": 0.78515625, + "learning_rate": 4.463396254868968e-05, + "loss": 0.5951766014099121, + "num_input_tokens_seen": 497553920, + "step": 2550, + "train_runtime": 15183.8819, + "train_tokens_per_second": 32768.558 + }, + { + "epoch": 2.6205439999999998, + "grad_norm": 0.828125, + "learning_rate": 4.461618929805519e-05, + "loss": 0.5954748153686523, + "num_input_tokens_seen": 499495872, + "step": 2560, + "train_runtime": 15243.1326, + "train_tokens_per_second": 32768.584 + }, + { + "epoch": 2.6307840000000002, + "grad_norm": 0.859375, + "learning_rate": 4.459843726245888e-05, + "loss": 0.586335563659668, + "num_input_tokens_seen": 501455808, + "step": 2570, + "train_runtime": 15302.5352, + "train_tokens_per_second": 32769.46 + }, + { + "epoch": 2.641024, + "grad_norm": 0.80859375, + "learning_rate": 4.458070639972875e-05, + "loss": 0.5931864738464355, + "num_input_tokens_seen": 503391232, + "step": 2580, + "train_runtime": 15361.7014, + "train_tokens_per_second": 32769.237 + }, + { + "epoch": 2.651264, + "grad_norm": 0.87109375, + "learning_rate": 4.456299666781007e-05, + "loss": 0.5972167015075683, + "num_input_tokens_seen": 505354624, + "step": 2590, + "train_runtime": 15422.5101, + "train_tokens_per_second": 32767.34 + }, + { + "epoch": 2.661504, + "grad_norm": 0.84375, + "learning_rate": 4.4545308024764984e-05, + "loss": 0.597212266921997, + "num_input_tokens_seen": 507291008, + "step": 2600, + "train_runtime": 15482.1176, + "train_tokens_per_second": 32766.255 + }, + { + "epoch": 2.671744, + "grad_norm": 0.80078125, + "learning_rate": 4.452764042877207e-05, + "loss": 0.59915771484375, + "num_input_tokens_seen": 509231104, + "step": 2610, + "train_runtime": 15541.0581, + "train_tokens_per_second": 32766.823 + }, + { + "epoch": 2.681984, + "grad_norm": 0.84765625, + "learning_rate": 4.45099938381259e-05, + "loss": 0.5966706275939941, + "num_input_tokens_seen": 511185152, + "step": 2620, + "train_runtime": 15601.7043, + "train_tokens_per_second": 32764.699 + }, + { + "epoch": 2.692224, + "grad_norm": 0.84375, + "learning_rate": 4.449236821123667e-05, + "loss": 0.5944993019104003, + "num_input_tokens_seen": 513121152, + "step": 2630, + "train_runtime": 15660.022, + "train_tokens_per_second": 32766.311 + }, + { + "epoch": 2.702464, + "grad_norm": 0.83984375, + "learning_rate": 4.447476350662976e-05, + "loss": 0.5976039409637451, + "num_input_tokens_seen": 515076928, + "step": 2640, + "train_runtime": 15719.1649, + "train_tokens_per_second": 32767.449 + }, + { + "epoch": 2.712704, + "grad_norm": 0.87109375, + "learning_rate": 4.4457179682945346e-05, + "loss": 0.5955704689025879, + "num_input_tokens_seen": 517031488, + "step": 2650, + "train_runtime": 15778.7122, + "train_tokens_per_second": 32767.661 + }, + { + "epoch": 2.722944, + "grad_norm": 0.8125, + "learning_rate": 4.443961669893798e-05, + "loss": 0.5922414779663085, + "num_input_tokens_seen": 518978880, + "step": 2660, + "train_runtime": 15838.7635, + "train_tokens_per_second": 32766.376 + }, + { + "epoch": 2.733184, + "grad_norm": 0.80859375, + "learning_rate": 4.4422074513476155e-05, + "loss": 0.5948915481567383, + "num_input_tokens_seen": 520932416, + "step": 2670, + "train_runtime": 15898.2387, + "train_tokens_per_second": 32766.675 + }, + { + "epoch": 2.743424, + "grad_norm": 0.90234375, + "learning_rate": 4.4404553085541955e-05, + "loss": 0.603402853012085, + "num_input_tokens_seen": 522866176, + "step": 2680, + "train_runtime": 15956.9495, + "train_tokens_per_second": 32767.302 + }, + { + "epoch": 2.753664, + "grad_norm": 0.8203125, + "learning_rate": 4.438705237423063e-05, + "loss": 0.5967090606689454, + "num_input_tokens_seen": 524830528, + "step": 2690, + "train_runtime": 16016.5448, + "train_tokens_per_second": 32768.024 + }, + { + "epoch": 2.763904, + "grad_norm": 0.82421875, + "learning_rate": 4.436957233875017e-05, + "loss": 0.5950541496276855, + "num_input_tokens_seen": 526763648, + "step": 2700, + "train_runtime": 16074.8455, + "train_tokens_per_second": 32769.438 + }, + { + "epoch": 2.774144, + "grad_norm": 0.8359375, + "learning_rate": 4.4352112938420956e-05, + "loss": 0.5917387962341308, + "num_input_tokens_seen": 528714368, + "step": 2710, + "train_runtime": 16134.1716, + "train_tokens_per_second": 32769.849 + }, + { + "epoch": 2.784384, + "grad_norm": 0.77734375, + "learning_rate": 4.433467413267529e-05, + "loss": 0.592765474319458, + "num_input_tokens_seen": 530652416, + "step": 2720, + "train_runtime": 16192.9592, + "train_tokens_per_second": 32770.565 + }, + { + "epoch": 2.7946239999999998, + "grad_norm": 0.87109375, + "learning_rate": 4.431725588105708e-05, + "loss": 0.5953330993652344, + "num_input_tokens_seen": 532596928, + "step": 2730, + "train_runtime": 16251.8008, + "train_tokens_per_second": 32771.564 + }, + { + "epoch": 2.8048640000000002, + "grad_norm": 0.8046875, + "learning_rate": 4.4299858143221377e-05, + "loss": 0.5934776306152344, + "num_input_tokens_seen": 534571776, + "step": 2740, + "train_runtime": 16311.457, + "train_tokens_per_second": 32772.779 + }, + { + "epoch": 2.815104, + "grad_norm": 0.8125, + "learning_rate": 4.4282480878934065e-05, + "loss": 0.5929346561431885, + "num_input_tokens_seen": 536513472, + "step": 2750, + "train_runtime": 16370.5136, + "train_tokens_per_second": 32773.161 + }, + { + "epoch": 2.825344, + "grad_norm": 0.8203125, + "learning_rate": 4.4265124048071346e-05, + "loss": 0.5918947219848633, + "num_input_tokens_seen": 538483968, + "step": 2760, + "train_runtime": 16430.5232, + "train_tokens_per_second": 32773.391 + }, + { + "epoch": 2.835584, + "grad_norm": 0.82421875, + "learning_rate": 4.4247787610619477e-05, + "loss": 0.594475269317627, + "num_input_tokens_seen": 540438592, + "step": 2770, + "train_runtime": 16489.7455, + "train_tokens_per_second": 32774.223 + }, + { + "epoch": 2.845824, + "grad_norm": 0.84375, + "learning_rate": 4.42304715266743e-05, + "loss": 0.5964062690734864, + "num_input_tokens_seen": 542392128, + "step": 2780, + "train_runtime": 16549.9267, + "train_tokens_per_second": 32773.083 + }, + { + "epoch": 2.856064, + "grad_norm": 0.83203125, + "learning_rate": 4.421317575644092e-05, + "loss": 0.5967954635620117, + "num_input_tokens_seen": 544377664, + "step": 2790, + "train_runtime": 16610.0177, + "train_tokens_per_second": 32774.057 + }, + { + "epoch": 2.866304, + "grad_norm": 0.8046875, + "learning_rate": 4.419590026023325e-05, + "loss": 0.5921245098114014, + "num_input_tokens_seen": 546344448, + "step": 2800, + "train_runtime": 16668.9664, + "train_tokens_per_second": 32776.144 + }, + { + "epoch": 2.876544, + "grad_norm": 0.78515625, + "learning_rate": 4.417864499847368e-05, + "loss": 0.5939229965209961, + "num_input_tokens_seen": 548293568, + "step": 2810, + "train_runtime": 16728.7214, + "train_tokens_per_second": 32775.581 + }, + { + "epoch": 2.886784, + "grad_norm": 0.8828125, + "learning_rate": 4.4161409931692676e-05, + "loss": 0.6008837699890137, + "num_input_tokens_seen": 550250688, + "step": 2820, + "train_runtime": 16788.0142, + "train_tokens_per_second": 32776.401 + }, + { + "epoch": 2.897024, + "grad_norm": 0.875, + "learning_rate": 4.414419502052841e-05, + "loss": 0.5984635353088379, + "num_input_tokens_seen": 552207104, + "step": 2830, + "train_runtime": 16846.6414, + "train_tokens_per_second": 32778.469 + }, + { + "epoch": 2.907264, + "grad_norm": 0.81640625, + "learning_rate": 4.412700022572637e-05, + "loss": 0.5894341945648194, + "num_input_tokens_seen": 554150528, + "step": 2840, + "train_runtime": 16906.9041, + "train_tokens_per_second": 32776.582 + }, + { + "epoch": 2.917504, + "grad_norm": 0.77734375, + "learning_rate": 4.410982550813902e-05, + "loss": 0.5969189643859864, + "num_input_tokens_seen": 556108544, + "step": 2850, + "train_runtime": 16966.3284, + "train_tokens_per_second": 32777.189 + }, + { + "epoch": 2.927744, + "grad_norm": 0.84765625, + "learning_rate": 4.409267082872535e-05, + "loss": 0.5878054618835449, + "num_input_tokens_seen": 558057216, + "step": 2860, + "train_runtime": 17025.2622, + "train_tokens_per_second": 32778.186 + }, + { + "epoch": 2.937984, + "grad_norm": 0.796875, + "learning_rate": 4.407553614855059e-05, + "loss": 0.5897252082824707, + "num_input_tokens_seen": 560011008, + "step": 2870, + "train_runtime": 17084.3194, + "train_tokens_per_second": 32779.24 + }, + { + "epoch": 2.9482239999999997, + "grad_norm": 0.80859375, + "learning_rate": 4.405842142878579e-05, + "loss": 0.5927898406982421, + "num_input_tokens_seen": 561967104, + "step": 2880, + "train_runtime": 17144.0476, + "train_tokens_per_second": 32779.138 + }, + { + "epoch": 2.958464, + "grad_norm": 0.80859375, + "learning_rate": 4.404132663070745e-05, + "loss": 0.5948840141296386, + "num_input_tokens_seen": 563924032, + "step": 2890, + "train_runtime": 17203.5627, + "train_tokens_per_second": 32779.491 + }, + { + "epoch": 2.968704, + "grad_norm": 0.80859375, + "learning_rate": 4.402425171569716e-05, + "loss": 0.5974791526794434, + "num_input_tokens_seen": 565866240, + "step": 2900, + "train_runtime": 17264.9688, + "train_tokens_per_second": 32775.399 + }, + { + "epoch": 2.9789440000000003, + "grad_norm": 0.83984375, + "learning_rate": 4.400719664524127e-05, + "loss": 0.5980208396911622, + "num_input_tokens_seen": 567839616, + "step": 2910, + "train_runtime": 17324.5301, + "train_tokens_per_second": 32776.624 + }, + { + "epoch": 2.989184, + "grad_norm": 0.80859375, + "learning_rate": 4.399016138093044e-05, + "loss": 0.5934501647949219, + "num_input_tokens_seen": 569804928, + "step": 2920, + "train_runtime": 17384.4555, + "train_tokens_per_second": 32776.691 + }, + { + "epoch": 2.999424, + "grad_norm": 0.8046875, + "learning_rate": 4.397314588445937e-05, + "loss": 0.5971685409545898, + "num_input_tokens_seen": 571794176, + "step": 2930, + "train_runtime": 17444.6674, + "train_tokens_per_second": 32777.591 + }, + { + "epoch": 3.0, + "eval_loss": 0.7842721939086914, + "eval_runtime": 1.0376, + "eval_samples_per_second": 960.831, + "eval_steps_per_second": 7.71, + "num_input_tokens_seen": 571898240, + "step": 2931 + }, + { + "epoch": 3.0, + "num_input_tokens_seen": 571898240, + "step": 2931, + "total_flos": 9.521329376839336e+18, + "train_loss": 0.6782078807039174, + "train_runtime": 17469.688, + "train_samples_per_second": 171.726, + "train_steps_per_second": 0.168 + } + ], + "logging_steps": 10, + "max_steps": 2931, + "num_input_tokens_seen": 571898240, + "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": 9.521329376839336e+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..0d5e7e5 --- /dev/null +++ b/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:64efc489757ec84ad026e0ea7c47c79b079f6b17f5ef57f722feaedf062e2a6a +size 5905 diff --git a/training_eval_loss.png b/training_eval_loss.png new file mode 100644 index 0000000..d65ecce 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..087103d Binary files /dev/null and b/training_loss.png differ