| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00040705304197946504, | |
| "eval_steps": 3, | |
| "global_step": 48, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.48027170790552e-06, | |
| "eval_loss": 8.904634475708008, | |
| "eval_runtime": 395.5428, | |
| "eval_samples_per_second": 125.526, | |
| "eval_steps_per_second": 62.764, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.5440815123716565e-05, | |
| "grad_norm": 3.2092926502227783, | |
| "learning_rate": 6e-05, | |
| "loss": 9.4435, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.5440815123716565e-05, | |
| "eval_loss": 8.903043746948242, | |
| "eval_runtime": 394.4845, | |
| "eval_samples_per_second": 125.863, | |
| "eval_steps_per_second": 62.933, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 5.088163024743313e-05, | |
| "grad_norm": 2.6875100135803223, | |
| "learning_rate": 0.00012, | |
| "loss": 8.8735, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 5.088163024743313e-05, | |
| "eval_loss": 8.875833511352539, | |
| "eval_runtime": 393.8802, | |
| "eval_samples_per_second": 126.056, | |
| "eval_steps_per_second": 63.029, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 7.632244537114969e-05, | |
| "grad_norm": 2.1503734588623047, | |
| "learning_rate": 0.00018, | |
| "loss": 8.243, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 7.632244537114969e-05, | |
| "eval_loss": 8.804418563842773, | |
| "eval_runtime": 393.1137, | |
| "eval_samples_per_second": 126.302, | |
| "eval_steps_per_second": 63.152, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00010176326049486626, | |
| "grad_norm": 2.0084388256073, | |
| "learning_rate": 0.00019863613034027224, | |
| "loss": 9.4, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.00010176326049486626, | |
| "eval_loss": 8.705506324768066, | |
| "eval_runtime": 393.3705, | |
| "eval_samples_per_second": 126.219, | |
| "eval_steps_per_second": 63.111, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.00012720407561858283, | |
| "grad_norm": 2.128262758255005, | |
| "learning_rate": 0.00019157733266550575, | |
| "loss": 8.7753, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00012720407561858283, | |
| "eval_loss": 8.609003067016602, | |
| "eval_runtime": 393.4243, | |
| "eval_samples_per_second": 126.202, | |
| "eval_steps_per_second": 63.102, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00015264489074229938, | |
| "grad_norm": 2.6665070056915283, | |
| "learning_rate": 0.00017891405093963938, | |
| "loss": 8.7284, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00015264489074229938, | |
| "eval_loss": 8.514460563659668, | |
| "eval_runtime": 393.1462, | |
| "eval_samples_per_second": 126.291, | |
| "eval_steps_per_second": 63.147, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00017808570586601595, | |
| "grad_norm": 1.9979753494262695, | |
| "learning_rate": 0.0001614212712689668, | |
| "loss": 8.4017, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.00017808570586601595, | |
| "eval_loss": 8.427648544311523, | |
| "eval_runtime": 393.6383, | |
| "eval_samples_per_second": 126.134, | |
| "eval_steps_per_second": 63.068, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.00020352652098973252, | |
| "grad_norm": 1.9800701141357422, | |
| "learning_rate": 0.00014016954246529696, | |
| "loss": 8.6508, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.00020352652098973252, | |
| "eval_loss": 8.347582817077637, | |
| "eval_runtime": 393.648, | |
| "eval_samples_per_second": 126.13, | |
| "eval_steps_per_second": 63.066, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.00022896733611344906, | |
| "grad_norm": 1.8114262819290161, | |
| "learning_rate": 0.00011645945902807341, | |
| "loss": 7.9676, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.00022896733611344906, | |
| "eval_loss": 8.279464721679688, | |
| "eval_runtime": 394.6493, | |
| "eval_samples_per_second": 125.81, | |
| "eval_steps_per_second": 62.906, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.00025440815123716566, | |
| "grad_norm": 3.259230852127075, | |
| "learning_rate": 9.174206545276677e-05, | |
| "loss": 7.7184, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.00025440815123716566, | |
| "eval_loss": 8.224762916564941, | |
| "eval_runtime": 393.2672, | |
| "eval_samples_per_second": 126.253, | |
| "eval_steps_per_second": 63.128, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0002798489663608822, | |
| "grad_norm": 2.8306350708007812, | |
| "learning_rate": 6.753005307953167e-05, | |
| "loss": 9.016, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0002798489663608822, | |
| "eval_loss": 8.18464469909668, | |
| "eval_runtime": 394.6805, | |
| "eval_samples_per_second": 125.8, | |
| "eval_steps_per_second": 62.902, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.00030528978148459875, | |
| "grad_norm": 2.145679235458374, | |
| "learning_rate": 4.530518418775733e-05, | |
| "loss": 8.1484, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.00030528978148459875, | |
| "eval_loss": 8.153741836547852, | |
| "eval_runtime": 395.8325, | |
| "eval_samples_per_second": 125.434, | |
| "eval_steps_per_second": 62.718, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0003307305966083153, | |
| "grad_norm": 1.9868985414505005, | |
| "learning_rate": 2.6427608932686843e-05, | |
| "loss": 8.136, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0003307305966083153, | |
| "eval_loss": 8.135424613952637, | |
| "eval_runtime": 396.6057, | |
| "eval_samples_per_second": 125.19, | |
| "eval_steps_per_second": 62.596, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0003561714117320319, | |
| "grad_norm": 1.6023919582366943, | |
| "learning_rate": 1.2052624879351104e-05, | |
| "loss": 7.4801, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0003561714117320319, | |
| "eval_loss": 8.125348091125488, | |
| "eval_runtime": 393.4071, | |
| "eval_samples_per_second": 126.208, | |
| "eval_steps_per_second": 63.105, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.00038161222685574847, | |
| "grad_norm": 2.529416561126709, | |
| "learning_rate": 3.059973406066963e-06, | |
| "loss": 8.0106, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.00038161222685574847, | |
| "eval_loss": 8.12152099609375, | |
| "eval_runtime": 393.2608, | |
| "eval_samples_per_second": 126.255, | |
| "eval_steps_per_second": 63.129, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.00040705304197946504, | |
| "grad_norm": 1.6257084608078003, | |
| "learning_rate": 0.0, | |
| "loss": 7.7403, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.00040705304197946504, | |
| "eval_loss": 8.12084674835205, | |
| "eval_runtime": 392.7186, | |
| "eval_samples_per_second": 126.429, | |
| "eval_steps_per_second": 63.216, | |
| "step": 48 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 48, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 70, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6359178805248.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |