| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.04731488052992666, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0009462976105985332, | |
| "eval_loss": 3.693946361541748, | |
| "eval_runtime": 31.6412, | |
| "eval_samples_per_second": 14.064, | |
| "eval_steps_per_second": 7.048, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0047314880529926665, | |
| "grad_norm": 3.518923044204712, | |
| "learning_rate": 5e-05, | |
| "loss": 3.6618, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.009462976105985333, | |
| "grad_norm": 5.546475887298584, | |
| "learning_rate": 0.0001, | |
| "loss": 3.3702, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.009462976105985333, | |
| "eval_loss": 2.7480592727661133, | |
| "eval_runtime": 32.018, | |
| "eval_samples_per_second": 13.898, | |
| "eval_steps_per_second": 6.965, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.014194464158977998, | |
| "grad_norm": 6.178415775299072, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.1327, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.018925952211970666, | |
| "grad_norm": 5.005501747131348, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.2552, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.018925952211970666, | |
| "eval_loss": 0.77772057056427, | |
| "eval_runtime": 32.1434, | |
| "eval_samples_per_second": 13.844, | |
| "eval_steps_per_second": 6.938, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02365744026496333, | |
| "grad_norm": 3.465031862258911, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.8897, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.028388928317955996, | |
| "grad_norm": 4.336249828338623, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6237, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.028388928317955996, | |
| "eval_loss": 0.6297822594642639, | |
| "eval_runtime": 32.0394, | |
| "eval_samples_per_second": 13.889, | |
| "eval_steps_per_second": 6.96, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03312041637094866, | |
| "grad_norm": 2.2877960205078125, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.5804, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.03785190442394133, | |
| "grad_norm": 2.645491361618042, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.5657, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03785190442394133, | |
| "eval_loss": 0.5742623209953308, | |
| "eval_runtime": 32.2961, | |
| "eval_samples_per_second": 13.779, | |
| "eval_steps_per_second": 6.905, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.042583392476934, | |
| "grad_norm": 2.466954469680786, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.5091, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.04731488052992666, | |
| "grad_norm": 2.9833056926727295, | |
| "learning_rate": 0.0, | |
| "loss": 0.6479, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04731488052992666, | |
| "eval_loss": 0.568959653377533, | |
| "eval_runtime": 32.0433, | |
| "eval_samples_per_second": 13.887, | |
| "eval_steps_per_second": 6.959, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9479016771747840.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |