| { | |
| "best_metric": 0.9301473498344421, | |
| "best_model_checkpoint": "data/Mistral-7B_task-1_120-samples_config-2_full/checkpoint-38", | |
| "epoch": 14.0, | |
| "eval_steps": 500, | |
| "global_step": 77, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 2.584413766860962, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.1072, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 2.6920111179351807, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 2.1328, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 2.3574070930480957, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 2.0764, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "eval_loss": 1.9517040252685547, | |
| "eval_runtime": 9.6589, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 2.351407051086426, | |
| "learning_rate": 2.4e-05, | |
| "loss": 2.0112, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 1.702775239944458, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.8747, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 1.2940586805343628, | |
| "learning_rate": 4e-05, | |
| "loss": 1.7328, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.5257502794265747, | |
| "eval_runtime": 9.6591, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 1.1890767812728882, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.5111, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 1.6276880502700806, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.4218, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 1.2439768314361572, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.2472, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "eval_loss": 1.1677581071853638, | |
| "eval_runtime": 9.6591, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 1.1022446155548096, | |
| "learning_rate": 7.2e-05, | |
| "loss": 1.1306, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.8354489803314209, | |
| "learning_rate": 8e-05, | |
| "loss": 1.0059, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.6951958537101746, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.979, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.9953374862670898, | |
| "eval_runtime": 9.6427, | |
| "eval_samples_per_second": 2.489, | |
| "eval_steps_per_second": 2.489, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 0.7623617053031921, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.9318, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.5785163044929504, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.909, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "eval_loss": 0.9436513781547546, | |
| "eval_runtime": 9.6649, | |
| "eval_samples_per_second": 2.483, | |
| "eval_steps_per_second": 2.483, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 0.6469224095344543, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.8443, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 0.6425813436508179, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.7984, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 0.600364625453949, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.7945, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.9338257908821106, | |
| "eval_runtime": 9.6486, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 0.6566327810287476, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.7414, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 0.7668809294700623, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.7177, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 0.7392465472221375, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.6672, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "eval_loss": 0.9301473498344421, | |
| "eval_runtime": 9.6559, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 0.8255151510238647, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.5998, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 0.968456506729126, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.5428, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.1418453454971313, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.5517, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.9630338549613953, | |
| "eval_runtime": 9.6529, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 1.042122721672058, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.4632, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 1.2745144367218018, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.411, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "eval_loss": 1.0644561052322388, | |
| "eval_runtime": 9.6572, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 1.2450038194656372, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.3976, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 1.1493616104125977, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.3276, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 1.4712613821029663, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.296, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.156125545501709, | |
| "eval_runtime": 9.653, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 10.181818181818182, | |
| "grad_norm": 1.0641038417816162, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.2698, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 10.545454545454545, | |
| "grad_norm": 1.0415575504302979, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.2237, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 1.4111512899398804, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.2302, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "eval_loss": 1.3205782175064087, | |
| "eval_runtime": 9.6497, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 11.272727272727273, | |
| "grad_norm": 1.0967178344726562, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.182, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 11.636363636363637, | |
| "grad_norm": 0.911393940448761, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.1651, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.382616639137268, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.1752, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.3892408609390259, | |
| "eval_runtime": 9.6473, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.363636363636363, | |
| "grad_norm": 1.0780975818634033, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.1329, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.727272727272727, | |
| "grad_norm": 1.1387547254562378, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.1498, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.909090909090908, | |
| "eval_loss": 1.3748698234558105, | |
| "eval_runtime": 9.6457, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 13.090909090909092, | |
| "grad_norm": 0.933902382850647, | |
| "learning_rate": 8.961448216775954e-05, | |
| "loss": 0.1342, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 13.454545454545455, | |
| "grad_norm": 0.8743265867233276, | |
| "learning_rate": 8.874722443520899e-05, | |
| "loss": 0.119, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 13.818181818181818, | |
| "grad_norm": 1.562985897064209, | |
| "learning_rate": 8.784975278258783e-05, | |
| "loss": 0.1138, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 1.4864697456359863, | |
| "eval_runtime": 9.6505, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "step": 77, | |
| "total_flos": 1.855843474394317e+16, | |
| "train_loss": 0.7548286336195933, | |
| "train_runtime": 1448.4727, | |
| "train_samples_per_second": 3.038, | |
| "train_steps_per_second": 0.173 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 7, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.855843474394317e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |