| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9671179883945842, | |
| "eval_steps": 50, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09671179883945841, | |
| "grad_norm": 0.035103704780340195, | |
| "learning_rate": 0.00016025641025641028, | |
| "loss": 1.0483, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09671179883945841, | |
| "eval_loss": 0.9260319471359253, | |
| "eval_runtime": 60.9012, | |
| "eval_samples_per_second": 1.642, | |
| "eval_steps_per_second": 0.213, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.19342359767891681, | |
| "grad_norm": 0.030684856697916985, | |
| "learning_rate": 0.00032051282051282057, | |
| "loss": 0.8577, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.19342359767891681, | |
| "eval_loss": 0.820220410823822, | |
| "eval_runtime": 56.5241, | |
| "eval_samples_per_second": 1.769, | |
| "eval_steps_per_second": 0.23, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2901353965183752, | |
| "grad_norm": 0.02802114002406597, | |
| "learning_rate": 0.0004807692307692308, | |
| "loss": 0.7996, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2901353965183752, | |
| "eval_loss": 0.7894760370254517, | |
| "eval_runtime": 56.6018, | |
| "eval_samples_per_second": 1.767, | |
| "eval_steps_per_second": 0.23, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.38684719535783363, | |
| "grad_norm": 0.03189089894294739, | |
| "learning_rate": 0.0004842293906810036, | |
| "loss": 0.7802, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.38684719535783363, | |
| "eval_loss": 0.7783958315849304, | |
| "eval_runtime": 56.5799, | |
| "eval_samples_per_second": 1.767, | |
| "eval_steps_per_second": 0.23, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4835589941972921, | |
| "grad_norm": 0.03580320626497269, | |
| "learning_rate": 0.00046630824372759856, | |
| "loss": 0.7671, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.4835589941972921, | |
| "eval_loss": 0.7721498012542725, | |
| "eval_runtime": 57.9466, | |
| "eval_samples_per_second": 1.726, | |
| "eval_steps_per_second": 0.224, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5802707930367504, | |
| "grad_norm": 0.034203894436359406, | |
| "learning_rate": 0.00044838709677419355, | |
| "loss": 0.761, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5802707930367504, | |
| "eval_loss": 0.7687731385231018, | |
| "eval_runtime": 60.6867, | |
| "eval_samples_per_second": 1.648, | |
| "eval_steps_per_second": 0.214, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6769825918762089, | |
| "grad_norm": 0.03307868540287018, | |
| "learning_rate": 0.00043046594982078853, | |
| "loss": 0.7587, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6769825918762089, | |
| "eval_loss": 0.766252338886261, | |
| "eval_runtime": 56.5991, | |
| "eval_samples_per_second": 1.767, | |
| "eval_steps_per_second": 0.23, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7736943907156673, | |
| "grad_norm": 0.03772694244980812, | |
| "learning_rate": 0.0004125448028673835, | |
| "loss": 0.7529, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7736943907156673, | |
| "eval_loss": 0.7637045383453369, | |
| "eval_runtime": 56.5561, | |
| "eval_samples_per_second": 1.768, | |
| "eval_steps_per_second": 0.23, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8704061895551257, | |
| "grad_norm": 0.04153395816683769, | |
| "learning_rate": 0.00039462365591397855, | |
| "loss": 0.7562, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.8704061895551257, | |
| "eval_loss": 0.7616338133811951, | |
| "eval_runtime": 56.6718, | |
| "eval_samples_per_second": 1.765, | |
| "eval_steps_per_second": 0.229, | |
| "step": 450 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 1551, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.844805825019904e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |