| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.12, |
| "eval_steps": 500, |
| "global_step": 600, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.004, |
| "grad_norm": 2.131298542022705, |
| "learning_rate": 0.00019962000000000002, |
| "loss": 2.0121, |
| "mean_token_accuracy": 0.6703190118074417, |
| "num_tokens": 2348.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.008, |
| "grad_norm": 1.2890418767929077, |
| "learning_rate": 0.00019922, |
| "loss": 0.2751, |
| "mean_token_accuracy": 0.9120438575744629, |
| "num_tokens": 4697.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.012, |
| "grad_norm": 0.8126867413520813, |
| "learning_rate": 0.00019882, |
| "loss": 0.1966, |
| "mean_token_accuracy": 0.9203487157821655, |
| "num_tokens": 7014.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.016, |
| "grad_norm": 0.6051881313323975, |
| "learning_rate": 0.00019842000000000001, |
| "loss": 0.1851, |
| "mean_token_accuracy": 0.9291799515485764, |
| "num_tokens": 9327.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.02, |
| "grad_norm": 0.6378348469734192, |
| "learning_rate": 0.00019802, |
| "loss": 0.1766, |
| "mean_token_accuracy": 0.9286984890699387, |
| "num_tokens": 11670.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.024, |
| "grad_norm": 0.624138593673706, |
| "learning_rate": 0.00019762, |
| "loss": 0.1784, |
| "mean_token_accuracy": 0.9264282643795013, |
| "num_tokens": 14000.0, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.028, |
| "grad_norm": 0.2111046463251114, |
| "learning_rate": 0.00019722, |
| "loss": 0.1702, |
| "mean_token_accuracy": 0.9321970880031586, |
| "num_tokens": 16329.0, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.032, |
| "grad_norm": 0.5350440740585327, |
| "learning_rate": 0.00019682, |
| "loss": 0.171, |
| "mean_token_accuracy": 0.9311463803052902, |
| "num_tokens": 18667.0, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.036, |
| "grad_norm": 0.19237647950649261, |
| "learning_rate": 0.00019642, |
| "loss": 0.167, |
| "mean_token_accuracy": 0.9345656305551528, |
| "num_tokens": 20985.0, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.04, |
| "grad_norm": 0.28153756260871887, |
| "learning_rate": 0.00019602, |
| "loss": 0.1674, |
| "mean_token_accuracy": 0.9329667061567306, |
| "num_tokens": 23325.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.044, |
| "grad_norm": 0.8545331954956055, |
| "learning_rate": 0.00019562, |
| "loss": 0.166, |
| "mean_token_accuracy": 0.9344212204217911, |
| "num_tokens": 25670.0, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.048, |
| "grad_norm": 0.24941129982471466, |
| "learning_rate": 0.00019522, |
| "loss": 0.1661, |
| "mean_token_accuracy": 0.934859549999237, |
| "num_tokens": 28016.0, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.052, |
| "grad_norm": 0.29549548029899597, |
| "learning_rate": 0.00019482, |
| "loss": 0.1707, |
| "mean_token_accuracy": 0.9345517784357071, |
| "num_tokens": 30345.0, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.056, |
| "grad_norm": 0.20388178527355194, |
| "learning_rate": 0.00019442, |
| "loss": 0.1673, |
| "mean_token_accuracy": 0.9353183209896088, |
| "num_tokens": 32691.0, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.06, |
| "grad_norm": 0.10762794315814972, |
| "learning_rate": 0.00019402, |
| "loss": 0.1642, |
| "mean_token_accuracy": 0.9325390756130219, |
| "num_tokens": 35030.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.064, |
| "grad_norm": 0.07676753401756287, |
| "learning_rate": 0.00019362, |
| "loss": 0.1633, |
| "mean_token_accuracy": 0.9348760217428207, |
| "num_tokens": 37359.0, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.068, |
| "grad_norm": 0.06781225651502609, |
| "learning_rate": 0.00019322, |
| "loss": 0.1589, |
| "mean_token_accuracy": 0.936154904961586, |
| "num_tokens": 39707.0, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.072, |
| "grad_norm": 0.10010460019111633, |
| "learning_rate": 0.00019282000000000001, |
| "loss": 0.1583, |
| "mean_token_accuracy": 0.9410124599933625, |
| "num_tokens": 42071.0, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.076, |
| "grad_norm": 0.07932794839143753, |
| "learning_rate": 0.00019242, |
| "loss": 0.1608, |
| "mean_token_accuracy": 0.9380002528429031, |
| "num_tokens": 44404.0, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.08, |
| "grad_norm": 0.06678586453199387, |
| "learning_rate": 0.00019202, |
| "loss": 0.1633, |
| "mean_token_accuracy": 0.9347729980945587, |
| "num_tokens": 46715.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.084, |
| "grad_norm": 0.05118393525481224, |
| "learning_rate": 0.00019162, |
| "loss": 0.1621, |
| "mean_token_accuracy": 0.9356200367212295, |
| "num_tokens": 49030.0, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.088, |
| "grad_norm": 0.07563836127519608, |
| "learning_rate": 0.00019122, |
| "loss": 0.1603, |
| "mean_token_accuracy": 0.9362942427396774, |
| "num_tokens": 51366.0, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.092, |
| "grad_norm": 0.053388580679893494, |
| "learning_rate": 0.00019082, |
| "loss": 0.1585, |
| "mean_token_accuracy": 0.9377258807420731, |
| "num_tokens": 53706.0, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.096, |
| "grad_norm": 0.05659119412302971, |
| "learning_rate": 0.00019042, |
| "loss": 0.1575, |
| "mean_token_accuracy": 0.937972965836525, |
| "num_tokens": 56052.0, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.1, |
| "grad_norm": 0.04934714362025261, |
| "learning_rate": 0.00019002, |
| "loss": 0.1606, |
| "mean_token_accuracy": 0.9356311202049256, |
| "num_tokens": 58374.0, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.104, |
| "grad_norm": 0.05647804215550423, |
| "learning_rate": 0.00018962000000000002, |
| "loss": 0.1587, |
| "mean_token_accuracy": 0.9353525519371033, |
| "num_tokens": 60706.0, |
| "step": 520 |
| }, |
| { |
| "epoch": 0.108, |
| "grad_norm": 0.058523017913103104, |
| "learning_rate": 0.00018922, |
| "loss": 0.1595, |
| "mean_token_accuracy": 0.9371987581253052, |
| "num_tokens": 63021.0, |
| "step": 540 |
| }, |
| { |
| "epoch": 0.112, |
| "grad_norm": 0.03793497756123543, |
| "learning_rate": 0.00018882000000000003, |
| "loss": 0.1584, |
| "mean_token_accuracy": 0.9341553807258606, |
| "num_tokens": 65369.0, |
| "step": 560 |
| }, |
| { |
| "epoch": 0.116, |
| "grad_norm": 0.04743633046746254, |
| "learning_rate": 0.00018842000000000002, |
| "loss": 0.1586, |
| "mean_token_accuracy": 0.936157900094986, |
| "num_tokens": 67700.0, |
| "step": 580 |
| }, |
| { |
| "epoch": 0.12, |
| "grad_norm": 0.05463261529803276, |
| "learning_rate": 0.00018802, |
| "loss": 0.1578, |
| "mean_token_accuracy": 0.937694975733757, |
| "num_tokens": 70038.0, |
| "step": 600 |
| } |
| ], |
| "logging_steps": 20, |
| "max_steps": 10000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 200, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 847458860064768.0, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
