| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9747899159663865, | |
| "eval_steps": 8, | |
| "global_step": 29, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03361344537815126, | |
| "grad_norm": 1.921875, | |
| "learning_rate": 2e-05, | |
| "loss": 3.1589, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03361344537815126, | |
| "eval_loss": 3.214388847351074, | |
| "eval_runtime": 9.5403, | |
| "eval_samples_per_second": 5.241, | |
| "eval_steps_per_second": 0.734, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06722689075630252, | |
| "grad_norm": 1.8984375, | |
| "learning_rate": 4e-05, | |
| "loss": 3.2696, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.10084033613445378, | |
| "grad_norm": 1.7109375, | |
| "learning_rate": 6e-05, | |
| "loss": 3.4298, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.13445378151260504, | |
| "grad_norm": 1.84375, | |
| "learning_rate": 8e-05, | |
| "loss": 3.1625, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.16806722689075632, | |
| "grad_norm": 1.875, | |
| "learning_rate": 0.0001, | |
| "loss": 3.2503, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.20168067226890757, | |
| "grad_norm": 1.7734375, | |
| "learning_rate": 0.00012, | |
| "loss": 3.1762, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 1.3828125, | |
| "learning_rate": 0.00014, | |
| "loss": 2.8709, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.2689075630252101, | |
| "grad_norm": 1.171875, | |
| "learning_rate": 0.00016, | |
| "loss": 2.8091, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.2689075630252101, | |
| "eval_loss": 2.6286215782165527, | |
| "eval_runtime": 10.6222, | |
| "eval_samples_per_second": 4.707, | |
| "eval_steps_per_second": 0.659, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.3025210084033613, | |
| "grad_norm": 1.0078125, | |
| "learning_rate": 0.00018, | |
| "loss": 2.6691, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.33613445378151263, | |
| "grad_norm": 0.82421875, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6851, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3697478991596639, | |
| "grad_norm": 0.828125, | |
| "learning_rate": 0.00019995608365087946, | |
| "loss": 2.5119, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.40336134453781514, | |
| "grad_norm": 0.65234375, | |
| "learning_rate": 0.00019982437317643217, | |
| "loss": 2.3433, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.4369747899159664, | |
| "grad_norm": 0.73046875, | |
| "learning_rate": 0.0001996049842615217, | |
| "loss": 2.3176, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 0.7109375, | |
| "learning_rate": 0.00019929810960135172, | |
| "loss": 2.2936, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5042016806722689, | |
| "grad_norm": 0.7890625, | |
| "learning_rate": 0.0001989040187322164, | |
| "loss": 2.3125, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5378151260504201, | |
| "grad_norm": 0.69921875, | |
| "learning_rate": 0.00019842305779475968, | |
| "loss": 2.312, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5378151260504201, | |
| "eval_loss": 2.242363691329956, | |
| "eval_runtime": 10.6302, | |
| "eval_samples_per_second": 4.704, | |
| "eval_steps_per_second": 0.659, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 0.70703125, | |
| "learning_rate": 0.0001978556492299504, | |
| "loss": 2.2591, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.6050420168067226, | |
| "grad_norm": 0.70703125, | |
| "learning_rate": 0.0001972022914080411, | |
| "loss": 2.3639, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.6386554621848739, | |
| "grad_norm": 0.7734375, | |
| "learning_rate": 0.00019646355819083589, | |
| "loss": 2.2533, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.6722689075630253, | |
| "grad_norm": 0.58984375, | |
| "learning_rate": 0.00019564009842765225, | |
| "loss": 2.2628, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 0.7265625, | |
| "learning_rate": 0.00019473263538541914, | |
| "loss": 2.2931, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.7394957983193278, | |
| "grad_norm": 0.69140625, | |
| "learning_rate": 0.0001937419661134121, | |
| "loss": 2.2773, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.773109243697479, | |
| "grad_norm": 0.7109375, | |
| "learning_rate": 0.00019266896074318334, | |
| "loss": 2.3788, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.8067226890756303, | |
| "grad_norm": 0.59375, | |
| "learning_rate": 0.00019151456172430183, | |
| "loss": 2.0133, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8067226890756303, | |
| "eval_loss": 2.153235673904419, | |
| "eval_runtime": 10.6243, | |
| "eval_samples_per_second": 4.706, | |
| "eval_steps_per_second": 0.659, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8403361344537815, | |
| "grad_norm": 0.53515625, | |
| "learning_rate": 0.00019027978299657436, | |
| "loss": 2.086, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8739495798319328, | |
| "grad_norm": 0.70703125, | |
| "learning_rate": 0.00018896570909947475, | |
| "loss": 2.2269, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.907563025210084, | |
| "grad_norm": 0.65625, | |
| "learning_rate": 0.0001875734942195637, | |
| "loss": 2.3534, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 0.55078125, | |
| "learning_rate": 0.00018610436117673555, | |
| "loss": 2.1109, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.9747899159663865, | |
| "grad_norm": 0.52734375, | |
| "learning_rate": 0.0001845596003501826, | |
| "loss": 2.2316, | |
| "step": 29 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 116, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 29, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1832864956022784.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |