| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 59.67567567567568, | |
| "eval_steps": 276, | |
| "global_step": 2760, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.967567567567568, | |
| "grad_norm": 1.919080138206482, | |
| "learning_rate": 9.007246376811595e-06, | |
| "loss": 1.2383, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 5.967567567567568, | |
| "eval_accuracy": 0.24171382376717865, | |
| "eval_loss": 2.6388285160064697, | |
| "eval_runtime": 4.0175, | |
| "eval_samples_per_second": 307.904, | |
| "eval_steps_per_second": 15.433, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 11.935135135135136, | |
| "grad_norm": 4.005405426025391, | |
| "learning_rate": 8.007246376811595e-06, | |
| "loss": 0.9769, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 11.935135135135136, | |
| "eval_accuracy": 0.22473726758286175, | |
| "eval_loss": 2.817387580871582, | |
| "eval_runtime": 3.8543, | |
| "eval_samples_per_second": 320.941, | |
| "eval_steps_per_second": 16.086, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 17.9027027027027, | |
| "grad_norm": 5.196224689483643, | |
| "learning_rate": 7.007246376811595e-06, | |
| "loss": 0.8157, | |
| "step": 828 | |
| }, | |
| { | |
| "epoch": 17.9027027027027, | |
| "eval_accuracy": 0.21476690918889788, | |
| "eval_loss": 3.148566961288452, | |
| "eval_runtime": 3.8369, | |
| "eval_samples_per_second": 322.399, | |
| "eval_steps_per_second": 16.159, | |
| "step": 828 | |
| }, | |
| { | |
| "epoch": 23.87027027027027, | |
| "grad_norm": 5.394821643829346, | |
| "learning_rate": 6.007246376811595e-06, | |
| "loss": 0.7322, | |
| "step": 1104 | |
| }, | |
| { | |
| "epoch": 23.87027027027027, | |
| "eval_accuracy": 0.20796281325788196, | |
| "eval_loss": 3.3019931316375732, | |
| "eval_runtime": 3.8785, | |
| "eval_samples_per_second": 318.936, | |
| "eval_steps_per_second": 15.985, | |
| "step": 1104 | |
| }, | |
| { | |
| "epoch": 29.83783783783784, | |
| "grad_norm": 4.061520576477051, | |
| "learning_rate": 5.0072463768115946e-06, | |
| "loss": 0.6777, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 29.83783783783784, | |
| "eval_accuracy": 0.20258690379951497, | |
| "eval_loss": 3.393256187438965, | |
| "eval_runtime": 3.8192, | |
| "eval_samples_per_second": 323.891, | |
| "eval_steps_per_second": 16.234, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 35.8054054054054, | |
| "grad_norm": 4.986082077026367, | |
| "learning_rate": 4.007246376811594e-06, | |
| "loss": 0.6466, | |
| "step": 1656 | |
| }, | |
| { | |
| "epoch": 35.8054054054054, | |
| "eval_accuracy": 0.19954190245216924, | |
| "eval_loss": 3.4180140495300293, | |
| "eval_runtime": 3.841, | |
| "eval_samples_per_second": 322.055, | |
| "eval_steps_per_second": 16.142, | |
| "step": 1656 | |
| }, | |
| { | |
| "epoch": 41.77297297297297, | |
| "grad_norm": 4.486128807067871, | |
| "learning_rate": 3.0072463768115946e-06, | |
| "loss": 0.6273, | |
| "step": 1932 | |
| }, | |
| { | |
| "epoch": 41.77297297297297, | |
| "eval_accuracy": 0.1971359279362513, | |
| "eval_loss": 3.438514471054077, | |
| "eval_runtime": 3.7802, | |
| "eval_samples_per_second": 327.233, | |
| "eval_steps_per_second": 16.401, | |
| "step": 1932 | |
| }, | |
| { | |
| "epoch": 47.74054054054054, | |
| "grad_norm": 3.447795867919922, | |
| "learning_rate": 2.0108695652173916e-06, | |
| "loss": 0.6188, | |
| "step": 2208 | |
| }, | |
| { | |
| "epoch": 47.74054054054054, | |
| "eval_accuracy": 0.19533144704931285, | |
| "eval_loss": 3.467123031616211, | |
| "eval_runtime": 3.8671, | |
| "eval_samples_per_second": 319.876, | |
| "eval_steps_per_second": 16.033, | |
| "step": 2208 | |
| }, | |
| { | |
| "epoch": 53.70810810810811, | |
| "grad_norm": 5.680208683013916, | |
| "learning_rate": 1.0108695652173914e-06, | |
| "loss": 0.6051, | |
| "step": 2484 | |
| }, | |
| { | |
| "epoch": 53.70810810810811, | |
| "eval_accuracy": 0.19437707715799873, | |
| "eval_loss": 3.4630978107452393, | |
| "eval_runtime": 3.837, | |
| "eval_samples_per_second": 322.388, | |
| "eval_steps_per_second": 16.158, | |
| "step": 2484 | |
| }, | |
| { | |
| "epoch": 59.67567567567568, | |
| "grad_norm": 3.523763656616211, | |
| "learning_rate": 1.0869565217391305e-08, | |
| "loss": 0.6049, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 59.67567567567568, | |
| "eval_accuracy": 0.19345189975747776, | |
| "eval_loss": 3.4764957427978516, | |
| "eval_runtime": 3.842, | |
| "eval_samples_per_second": 321.968, | |
| "eval_steps_per_second": 16.137, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 59.67567567567568, | |
| "step": 2760, | |
| "total_flos": 1.4462122119191654e+17, | |
| "train_loss": 0.7543496670930282, | |
| "train_runtime": 3692.6545, | |
| "train_samples_per_second": 150.006, | |
| "train_steps_per_second": 0.747 | |
| } | |
| ], | |
| "logging_steps": 276, | |
| "max_steps": 2760, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 60, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4462122119191654e+17, | |
| "train_batch_size": 25, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |