| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.02831657935721365, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.000566331587144273, | |
| "eval_loss": 7.698707103729248, | |
| "eval_runtime": 235.8933, | |
| "eval_samples_per_second": 12.607, | |
| "eval_steps_per_second": 3.154, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016989947614328188, | |
| "grad_norm": 10.413496971130371, | |
| "learning_rate": 3e-05, | |
| "loss": 7.7694, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.002831657935721365, | |
| "eval_loss": 6.380946636199951, | |
| "eval_runtime": 237.9505, | |
| "eval_samples_per_second": 12.498, | |
| "eval_steps_per_second": 3.127, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0033979895228656376, | |
| "grad_norm": 7.785809516906738, | |
| "learning_rate": 6e-05, | |
| "loss": 6.8429, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005096984284298457, | |
| "grad_norm": 6.8363213539123535, | |
| "learning_rate": 9e-05, | |
| "loss": 5.5567, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00566331587144273, | |
| "eval_loss": 5.027488708496094, | |
| "eval_runtime": 237.9356, | |
| "eval_samples_per_second": 12.499, | |
| "eval_steps_per_second": 3.127, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.006795979045731275, | |
| "grad_norm": 5.078495979309082, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 5.1733, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.008494973807164095, | |
| "grad_norm": 3.602994441986084, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 4.8478, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.008494973807164095, | |
| "eval_loss": 4.539829730987549, | |
| "eval_runtime": 237.9504, | |
| "eval_samples_per_second": 12.498, | |
| "eval_steps_per_second": 3.127, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.010193968568596914, | |
| "grad_norm": 4.215411186218262, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 4.5758, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.01132663174288546, | |
| "eval_loss": 4.316648006439209, | |
| "eval_runtime": 237.9369, | |
| "eval_samples_per_second": 12.499, | |
| "eval_steps_per_second": 3.127, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.011892963330029732, | |
| "grad_norm": 3.9418630599975586, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 4.3036, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.01359195809146255, | |
| "grad_norm": 3.5897490978240967, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 4.2336, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.014158289678606824, | |
| "eval_loss": 4.144935131072998, | |
| "eval_runtime": 237.9531, | |
| "eval_samples_per_second": 12.498, | |
| "eval_steps_per_second": 3.127, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01529095285289537, | |
| "grad_norm": 4.833578109741211, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 4.1645, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.01698994761432819, | |
| "grad_norm": 3.4795303344726562, | |
| "learning_rate": 5e-05, | |
| "loss": 4.1821, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.01698994761432819, | |
| "eval_loss": 4.064857006072998, | |
| "eval_runtime": 237.9521, | |
| "eval_samples_per_second": 12.498, | |
| "eval_steps_per_second": 3.127, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.01868894237576101, | |
| "grad_norm": 3.493994951248169, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 4.0664, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.019821605550049554, | |
| "eval_loss": 4.008715629577637, | |
| "eval_runtime": 237.9336, | |
| "eval_samples_per_second": 12.499, | |
| "eval_steps_per_second": 3.127, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.020387937137193828, | |
| "grad_norm": 3.5762040615081787, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 3.9659, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.022086931898626646, | |
| "grad_norm": 3.549154043197632, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 3.8698, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.02265326348577092, | |
| "eval_loss": 3.9499382972717285, | |
| "eval_runtime": 237.9362, | |
| "eval_samples_per_second": 12.499, | |
| "eval_steps_per_second": 3.127, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.023785926660059464, | |
| "grad_norm": 3.726226329803467, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 3.8956, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.025484921421492283, | |
| "grad_norm": 3.9398977756500244, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 3.8685, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.025484921421492283, | |
| "eval_loss": 3.9185895919799805, | |
| "eval_runtime": 237.9499, | |
| "eval_samples_per_second": 12.498, | |
| "eval_steps_per_second": 3.127, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0271839161829251, | |
| "grad_norm": 3.495335578918457, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 4.0048, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.02831657935721365, | |
| "eval_loss": 3.912412166595459, | |
| "eval_runtime": 237.9471, | |
| "eval_samples_per_second": 12.499, | |
| "eval_steps_per_second": 3.127, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.54256789372928e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |