| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.009321401938851604, | |
| "eval_steps": 9, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0003728560775540641, | |
| "grad_norm": 3.2273905277252197, | |
| "learning_rate": 1e-05, | |
| "loss": 3.4355, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0003728560775540641, | |
| "eval_loss": 1.8008577823638916, | |
| "eval_runtime": 353.1503, | |
| "eval_samples_per_second": 6.397, | |
| "eval_steps_per_second": 0.801, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0007457121551081282, | |
| "grad_norm": 4.027805805206299, | |
| "learning_rate": 2e-05, | |
| "loss": 4.0185, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0011185682326621924, | |
| "grad_norm": 3.179356575012207, | |
| "learning_rate": 3e-05, | |
| "loss": 3.541, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0014914243102162564, | |
| "grad_norm": 3.339418888092041, | |
| "learning_rate": 4e-05, | |
| "loss": 3.4222, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0018642803877703207, | |
| "grad_norm": 4.543241500854492, | |
| "learning_rate": 5e-05, | |
| "loss": 3.6497, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0022371364653243847, | |
| "grad_norm": 3.6381893157958984, | |
| "learning_rate": 6e-05, | |
| "loss": 3.4862, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.002609992542878449, | |
| "grad_norm": 11.36868953704834, | |
| "learning_rate": 7e-05, | |
| "loss": 3.2618, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.002982848620432513, | |
| "grad_norm": 4.990373134613037, | |
| "learning_rate": 8e-05, | |
| "loss": 3.0643, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.003355704697986577, | |
| "grad_norm": 11.262433052062988, | |
| "learning_rate": 9e-05, | |
| "loss": 2.5547, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.003355704697986577, | |
| "eval_loss": 1.2176220417022705, | |
| "eval_runtime": 353.0167, | |
| "eval_samples_per_second": 6.399, | |
| "eval_steps_per_second": 0.802, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0037285607755406414, | |
| "grad_norm": 4.229122161865234, | |
| "learning_rate": 0.0001, | |
| "loss": 2.5665, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004101416853094705, | |
| "grad_norm": 3.833820104598999, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 2.2817, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0044742729306487695, | |
| "grad_norm": 4.755269527435303, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.9641, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.004847129008202834, | |
| "grad_norm": 4.510311603546143, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 1.7572, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.005219985085756898, | |
| "grad_norm": 4.087654113769531, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 1.4029, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.005592841163310962, | |
| "grad_norm": 4.565130710601807, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.1817, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.005965697240865026, | |
| "grad_norm": 3.981370687484741, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 1.2368, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.00633855331841909, | |
| "grad_norm": 4.3100972175598145, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 1.1073, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.006711409395973154, | |
| "grad_norm": 9.070611953735352, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 1.239, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.006711409395973154, | |
| "eval_loss": 0.49113982915878296, | |
| "eval_runtime": 352.9587, | |
| "eval_samples_per_second": 6.4, | |
| "eval_steps_per_second": 0.802, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0070842654735272185, | |
| "grad_norm": 3.0325357913970947, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.7823, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.007457121551081283, | |
| "grad_norm": 3.4878969192504883, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.7623, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.007829977628635347, | |
| "grad_norm": 2.9287140369415283, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.8463, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.00820283370618941, | |
| "grad_norm": 3.594860553741455, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.6903, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.008575689783743476, | |
| "grad_norm": 4.352069854736328, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.927, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.008948545861297539, | |
| "grad_norm": 2.5088510513305664, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.6223, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.009321401938851604, | |
| "grad_norm": 2.696772813796997, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.7409, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.173894848512e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |