| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.0, |
| "eval_steps": 500, |
| "global_step": 200, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.05, |
| "grad_norm": 2.859375, |
| "learning_rate": 8.330360163265307e-06, |
| "loss": 1.1812, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.1, |
| "grad_norm": 2.09375, |
| "learning_rate": 1.874331036734694e-05, |
| "loss": 1.0718, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.15, |
| "grad_norm": 1.421875, |
| "learning_rate": 2.9156260571428573e-05, |
| "loss": 0.9789, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.2, |
| "grad_norm": 1.1796875, |
| "learning_rate": 3.9569210775510205e-05, |
| "loss": 0.9218, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.25, |
| "grad_norm": 1.171875, |
| "learning_rate": 4.998216097959184e-05, |
| "loss": 0.8882, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.3, |
| "grad_norm": 1.1875, |
| "learning_rate": 6.0395111183673474e-05, |
| "loss": 0.899, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.35, |
| "grad_norm": 1.125, |
| "learning_rate": 7.080806138775511e-05, |
| "loss": 0.8827, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.4, |
| "grad_norm": 1.140625, |
| "learning_rate": 7.285992446116292e-05, |
| "loss": 0.8456, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.45, |
| "grad_norm": 1.046875, |
| "learning_rate": 7.27352145443898e-05, |
| "loss": 0.8275, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 1.0234375, |
| "learning_rate": 7.25150389535049e-05, |
| "loss": 0.8417, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.55, |
| "grad_norm": 1.078125, |
| "learning_rate": 7.220017106330133e-05, |
| "loss": 0.8078, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.6, |
| "grad_norm": 0.89453125, |
| "learning_rate": 7.17917168587271e-05, |
| "loss": 0.8155, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.65, |
| "grad_norm": 1.0078125, |
| "learning_rate": 7.129111105007264e-05, |
| "loss": 0.7951, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.7, |
| "grad_norm": 0.9765625, |
| "learning_rate": 7.070011203349862e-05, |
| "loss": 0.8144, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.75, |
| "grad_norm": 1.0078125, |
| "learning_rate": 7.00207957146053e-05, |
| "loss": 0.8017, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.8, |
| "grad_norm": 1.0078125, |
| "learning_rate": 6.925554821673839e-05, |
| "loss": 0.8014, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.85, |
| "grad_norm": 0.92578125, |
| "learning_rate": 6.840705749964374e-05, |
| "loss": 0.7924, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.9, |
| "grad_norm": 0.91015625, |
| "learning_rate": 6.747830391791077e-05, |
| "loss": 0.786, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.95, |
| "grad_norm": 0.9453125, |
| "learning_rate": 6.64725497523682e-05, |
| "loss": 0.7999, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.90234375, |
| "learning_rate": 6.539332775120374e-05, |
| "loss": 0.7758, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.6963938474655151, |
| "eval_runtime": 2.3054, |
| "eval_samples_per_second": 16.049, |
| "eval_steps_per_second": 16.049, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.05, |
| "grad_norm": 1.0, |
| "learning_rate": 6.424442872105717e-05, |
| "loss": 0.5848, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.1, |
| "grad_norm": 0.95703125, |
| "learning_rate": 6.302988821167378e-05, |
| "loss": 0.569, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.15, |
| "grad_norm": 0.97265625, |
| "learning_rate": 6.175397234088872e-05, |
| "loss": 0.5524, |
| "step": 115 |
| }, |
| { |
| "epoch": 1.2, |
| "grad_norm": 0.90234375, |
| "learning_rate": 6.04211628097325e-05, |
| "loss": 0.5769, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.25, |
| "grad_norm": 0.9765625, |
| "learning_rate": 5.9036141160292685e-05, |
| "loss": 0.5548, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.3, |
| "grad_norm": 0.92578125, |
| "learning_rate": 5.7603772331626554e-05, |
| "loss": 0.5558, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.35, |
| "grad_norm": 1.0234375, |
| "learning_rate": 5.612908757148523e-05, |
| "loss": 0.5536, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.4, |
| "grad_norm": 0.93359375, |
| "learning_rate": 5.461726676387246e-05, |
| "loss": 0.5525, |
| "step": 140 |
| }, |
| { |
| "epoch": 1.45, |
| "grad_norm": 0.90234375, |
| "learning_rate": 5.3073620234513154e-05, |
| "loss": 0.5547, |
| "step": 145 |
| }, |
| { |
| "epoch": 1.5, |
| "grad_norm": 1.0, |
| "learning_rate": 5.150357009814056e-05, |
| "loss": 0.5492, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.55, |
| "grad_norm": 1.0078125, |
| "learning_rate": 4.9912631213120535e-05, |
| "loss": 0.554, |
| "step": 155 |
| }, |
| { |
| "epoch": 1.6, |
| "grad_norm": 0.9296875, |
| "learning_rate": 4.830639181031043e-05, |
| "loss": 0.5588, |
| "step": 160 |
| }, |
| { |
| "epoch": 1.65, |
| "grad_norm": 1.0, |
| "learning_rate": 4.669049386419433e-05, |
| "loss": 0.54, |
| "step": 165 |
| }, |
| { |
| "epoch": 1.7, |
| "grad_norm": 1.0546875, |
| "learning_rate": 4.507061327524199e-05, |
| "loss": 0.5272, |
| "step": 170 |
| }, |
| { |
| "epoch": 1.75, |
| "grad_norm": 0.92578125, |
| "learning_rate": 4.345243993310132e-05, |
| "loss": 0.5414, |
| "step": 175 |
| }, |
| { |
| "epoch": 1.8, |
| "grad_norm": 0.90234375, |
| "learning_rate": 4.1841657730653567e-05, |
| "loss": 0.5376, |
| "step": 180 |
| }, |
| { |
| "epoch": 1.85, |
| "grad_norm": 0.89453125, |
| "learning_rate": 4.0243924599132264e-05, |
| "loss": 0.5308, |
| "step": 185 |
| }, |
| { |
| "epoch": 1.9, |
| "grad_norm": 0.8984375, |
| "learning_rate": 3.866485263443334e-05, |
| "loss": 0.5084, |
| "step": 190 |
| }, |
| { |
| "epoch": 1.95, |
| "grad_norm": 0.9765625, |
| "learning_rate": 3.710998838442374e-05, |
| "loss": 0.54, |
| "step": 195 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.9296875, |
| "learning_rate": 3.5584793366489594e-05, |
| "loss": 0.5312, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.6541135907173157, |
| "eval_runtime": 2.1847, |
| "eval_samples_per_second": 16.936, |
| "eval_steps_per_second": 16.936, |
| "step": 200 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 300, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.9212367740928e+17, |
| "train_batch_size": 48, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
