| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9814814814814814, | |
| "eval_steps": 14, | |
| "global_step": 54, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.037037037037037035, | |
| "grad_norm": 13.693925857543945, | |
| "learning_rate": 2.0000000000000002e-07, | |
| "loss": 1.4366, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.037037037037037035, | |
| "eval_loss": 1.4367035627365112, | |
| "eval_runtime": 19.7904, | |
| "eval_samples_per_second": 35.27, | |
| "eval_steps_per_second": 4.447, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07407407407407407, | |
| "grad_norm": 13.327592849731445, | |
| "learning_rate": 4.0000000000000003e-07, | |
| "loss": 1.4017, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 13.196478843688965, | |
| "learning_rate": 6.000000000000001e-07, | |
| "loss": 1.4017, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.14814814814814814, | |
| "grad_norm": 13.321550369262695, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 1.4058, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.18518518518518517, | |
| "grad_norm": 12.840996742248535, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.4149, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 11.752096176147461, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "loss": 1.3709, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.25925925925925924, | |
| "grad_norm": 11.358497619628906, | |
| "learning_rate": 1.4000000000000001e-06, | |
| "loss": 1.3723, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.2962962962962963, | |
| "grad_norm": 7.285672187805176, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "loss": 1.3024, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 6.211453914642334, | |
| "learning_rate": 1.8000000000000001e-06, | |
| "loss": 1.3428, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.37037037037037035, | |
| "grad_norm": 4.729733467102051, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.3021, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4074074074074074, | |
| "grad_norm": 5.45022439956665, | |
| "learning_rate": 2.2e-06, | |
| "loss": 1.2561, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 6.256317138671875, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "loss": 1.3131, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.48148148148148145, | |
| "grad_norm": 5.992193698883057, | |
| "learning_rate": 2.6e-06, | |
| "loss": 1.2764, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5185185185185185, | |
| "grad_norm": 5.3906707763671875, | |
| "learning_rate": 2.8000000000000003e-06, | |
| "loss": 1.2883, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5185185185185185, | |
| "eval_loss": 1.2674976587295532, | |
| "eval_runtime": 18.8899, | |
| "eval_samples_per_second": 36.951, | |
| "eval_steps_per_second": 4.659, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 4.330776691436768, | |
| "learning_rate": 3e-06, | |
| "loss": 1.2553, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 3.870635509490967, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "loss": 1.2092, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.6296296296296297, | |
| "grad_norm": 3.076308012008667, | |
| "learning_rate": 3.4000000000000005e-06, | |
| "loss": 1.2735, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 2.6835415363311768, | |
| "learning_rate": 3.6000000000000003e-06, | |
| "loss": 1.2449, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7037037037037037, | |
| "grad_norm": 2.1219379901885986, | |
| "learning_rate": 3.8000000000000005e-06, | |
| "loss": 1.2051, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7407407407407407, | |
| "grad_norm": 1.8215879201889038, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.171, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7777777777777778, | |
| "grad_norm": 2.0634374618530273, | |
| "learning_rate": 4.2000000000000004e-06, | |
| "loss": 1.2243, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8148148148148148, | |
| "grad_norm": 1.9009621143341064, | |
| "learning_rate": 4.4e-06, | |
| "loss": 1.1914, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.8518518518518519, | |
| "grad_norm": 1.8763676881790161, | |
| "learning_rate": 4.600000000000001e-06, | |
| "loss": 1.1752, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 1.8934900760650635, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 1.1793, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9259259259259259, | |
| "grad_norm": 1.7864941358566284, | |
| "learning_rate": 5e-06, | |
| "loss": 1.1839, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.9629629629629629, | |
| "grad_norm": 1.810880184173584, | |
| "learning_rate": 5.2e-06, | |
| "loss": 1.1728, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.7052356004714966, | |
| "learning_rate": 5.400000000000001e-06, | |
| "loss": 1.1623, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.0185185185185186, | |
| "grad_norm": 1.6250964403152466, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 1.1254, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.0185185185185186, | |
| "eval_loss": 1.160744071006775, | |
| "eval_runtime": 18.8648, | |
| "eval_samples_per_second": 37.0, | |
| "eval_steps_per_second": 4.665, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.0555555555555556, | |
| "grad_norm": 1.8527966737747192, | |
| "learning_rate": 5.8e-06, | |
| "loss": 1.0638, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.0925925925925926, | |
| "grad_norm": 1.7427172660827637, | |
| "learning_rate": 6e-06, | |
| "loss": 1.0382, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1296296296296295, | |
| "grad_norm": 1.7577691078186035, | |
| "learning_rate": 6.200000000000001e-06, | |
| "loss": 1.0523, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.1666666666666667, | |
| "grad_norm": 1.928122639656067, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 1.0711, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.2037037037037037, | |
| "grad_norm": 1.7540444135665894, | |
| "learning_rate": 6.600000000000001e-06, | |
| "loss": 1.0296, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.2407407407407407, | |
| "grad_norm": 1.704374074935913, | |
| "learning_rate": 6.800000000000001e-06, | |
| "loss": 1.0126, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.2777777777777777, | |
| "grad_norm": 1.7199629545211792, | |
| "learning_rate": 7e-06, | |
| "loss": 1.0091, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.3148148148148149, | |
| "grad_norm": 1.6979806423187256, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 1.0189, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.3518518518518519, | |
| "grad_norm": 1.7349421977996826, | |
| "learning_rate": 7.4e-06, | |
| "loss": 0.9761, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.3888888888888888, | |
| "grad_norm": 1.5777740478515625, | |
| "learning_rate": 7.600000000000001e-06, | |
| "loss": 0.9847, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.425925925925926, | |
| "grad_norm": 1.9043402671813965, | |
| "learning_rate": 7.800000000000002e-06, | |
| "loss": 0.9688, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.462962962962963, | |
| "grad_norm": 1.5200198888778687, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.9511, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 1.7094305753707886, | |
| "learning_rate": 8.2e-06, | |
| "loss": 0.9597, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.5370370370370372, | |
| "grad_norm": 1.7840018272399902, | |
| "learning_rate": 8.400000000000001e-06, | |
| "loss": 0.9361, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.5370370370370372, | |
| "eval_loss": 1.13677179813385, | |
| "eval_runtime": 18.8462, | |
| "eval_samples_per_second": 37.037, | |
| "eval_steps_per_second": 4.669, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.574074074074074, | |
| "grad_norm": 1.6459747552871704, | |
| "learning_rate": 8.6e-06, | |
| "loss": 0.9506, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.6111111111111112, | |
| "grad_norm": 1.922658085823059, | |
| "learning_rate": 8.8e-06, | |
| "loss": 0.9846, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.6481481481481481, | |
| "grad_norm": 1.8302316665649414, | |
| "learning_rate": 9e-06, | |
| "loss": 0.9371, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.6851851851851851, | |
| "grad_norm": 1.6393502950668335, | |
| "learning_rate": 9.200000000000002e-06, | |
| "loss": 0.8898, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.7222222222222223, | |
| "grad_norm": 1.9181392192840576, | |
| "learning_rate": 9.4e-06, | |
| "loss": 0.9555, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.7592592592592593, | |
| "grad_norm": 1.7563830614089966, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 0.943, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.7962962962962963, | |
| "grad_norm": 1.8117369413375854, | |
| "learning_rate": 9.800000000000001e-06, | |
| "loss": 0.9278, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.8333333333333335, | |
| "grad_norm": 1.6542695760726929, | |
| "learning_rate": 1e-05, | |
| "loss": 0.973, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.8703703703703702, | |
| "grad_norm": 1.7787063121795654, | |
| "learning_rate": 8.535533905932739e-06, | |
| "loss": 0.951, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.9074074074074074, | |
| "grad_norm": 1.6953744888305664, | |
| "learning_rate": 5e-06, | |
| "loss": 0.951, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.9444444444444444, | |
| "grad_norm": 1.9485039710998535, | |
| "learning_rate": 1.4644660940672628e-06, | |
| "loss": 0.9492, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.9814814814814814, | |
| "grad_norm": 1.531893253326416, | |
| "learning_rate": 0.0, | |
| "loss": 0.9236, | |
| "step": 54 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 54, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 14, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.9665250622279516e+18, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |