| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.96, | |
| "eval_steps": 500, | |
| "global_step": 60, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.016, | |
| "grad_norm": 0.6046149134635925, | |
| "learning_rate": 0.0, | |
| "loss": 2.0633, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.032, | |
| "grad_norm": 0.5796938538551331, | |
| "learning_rate": 2e-05, | |
| "loss": 2.034, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.048, | |
| "grad_norm": 0.6046639084815979, | |
| "learning_rate": 4e-05, | |
| "loss": 2.0676, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.064, | |
| "grad_norm": 0.6537912487983704, | |
| "learning_rate": 6e-05, | |
| "loss": 2.0168, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.7197220325469971, | |
| "learning_rate": 8e-05, | |
| "loss": 1.9651, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 0.7753214240074158, | |
| "learning_rate": 0.0001, | |
| "loss": 1.9283, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.112, | |
| "grad_norm": 0.7212746143341064, | |
| "learning_rate": 0.00012, | |
| "loss": 1.7437, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.128, | |
| "grad_norm": 0.7059862017631531, | |
| "learning_rate": 0.00014, | |
| "loss": 1.7018, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.144, | |
| "grad_norm": 0.7816401720046997, | |
| "learning_rate": 0.00016, | |
| "loss": 1.5029, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.8176925778388977, | |
| "learning_rate": 0.00018, | |
| "loss": 1.3778, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.176, | |
| "grad_norm": 0.8414409160614014, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3103, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 0.8573537468910217, | |
| "learning_rate": 0.000196, | |
| "loss": 1.1423, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.208, | |
| "grad_norm": 0.7942666411399841, | |
| "learning_rate": 0.000192, | |
| "loss": 0.9792, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.224, | |
| "grad_norm": 1.1660468578338623, | |
| "learning_rate": 0.000188, | |
| "loss": 0.8941, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.6082952618598938, | |
| "learning_rate": 0.00018400000000000003, | |
| "loss": 0.8822, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.256, | |
| "grad_norm": 0.6617238521575928, | |
| "learning_rate": 0.00018, | |
| "loss": 0.9272, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.272, | |
| "grad_norm": 2.0397157669067383, | |
| "learning_rate": 0.00017600000000000002, | |
| "loss": 0.8929, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 0.5073072910308838, | |
| "learning_rate": 0.000172, | |
| "loss": 0.8766, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.304, | |
| "grad_norm": 0.536578357219696, | |
| "learning_rate": 0.000168, | |
| "loss": 0.8599, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.4412708282470703, | |
| "learning_rate": 0.000164, | |
| "loss": 0.737, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.336, | |
| "grad_norm": 0.41358307003974915, | |
| "learning_rate": 0.00016, | |
| "loss": 0.8145, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.352, | |
| "grad_norm": 0.38526788353919983, | |
| "learning_rate": 0.00015600000000000002, | |
| "loss": 0.802, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.368, | |
| "grad_norm": 0.34247690439224243, | |
| "learning_rate": 0.000152, | |
| "loss": 0.7904, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 0.38441890478134155, | |
| "learning_rate": 0.000148, | |
| "loss": 0.7924, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.3929769992828369, | |
| "learning_rate": 0.000144, | |
| "loss": 0.8502, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.416, | |
| "grad_norm": 0.3987599313259125, | |
| "learning_rate": 0.00014, | |
| "loss": 0.7837, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.432, | |
| "grad_norm": 0.36407092213630676, | |
| "learning_rate": 0.00013600000000000003, | |
| "loss": 0.7029, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.448, | |
| "grad_norm": 0.47272396087646484, | |
| "learning_rate": 0.000132, | |
| "loss": 0.7621, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.464, | |
| "grad_norm": 0.3687835931777954, | |
| "learning_rate": 0.00012800000000000002, | |
| "loss": 0.767, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.38991373777389526, | |
| "learning_rate": 0.000124, | |
| "loss": 0.7381, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.496, | |
| "grad_norm": 0.31758126616477966, | |
| "learning_rate": 0.00012, | |
| "loss": 0.7326, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.512, | |
| "grad_norm": 0.3498470187187195, | |
| "learning_rate": 0.000116, | |
| "loss": 0.7858, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.528, | |
| "grad_norm": 0.339616984128952, | |
| "learning_rate": 0.00011200000000000001, | |
| "loss": 0.7636, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.544, | |
| "grad_norm": 0.3355540335178375, | |
| "learning_rate": 0.00010800000000000001, | |
| "loss": 0.7011, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.38637426495552063, | |
| "learning_rate": 0.00010400000000000001, | |
| "loss": 0.7206, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 0.39599090814590454, | |
| "learning_rate": 0.0001, | |
| "loss": 0.7352, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.592, | |
| "grad_norm": 0.34623557329177856, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.6788, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.608, | |
| "grad_norm": 0.3236479163169861, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 0.6575, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.624, | |
| "grad_norm": 0.3357284367084503, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.7061, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.3321351408958435, | |
| "learning_rate": 8.4e-05, | |
| "loss": 0.6878, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.656, | |
| "grad_norm": 0.30003687739372253, | |
| "learning_rate": 8e-05, | |
| "loss": 0.6769, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 0.3306218385696411, | |
| "learning_rate": 7.6e-05, | |
| "loss": 0.7436, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.688, | |
| "grad_norm": 0.3028322756290436, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.678, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.704, | |
| "grad_norm": 0.31864282488822937, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 0.6989, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.3383725881576538, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.7458, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.736, | |
| "grad_norm": 0.35683438181877136, | |
| "learning_rate": 6e-05, | |
| "loss": 0.7546, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.752, | |
| "grad_norm": 0.3352407217025757, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.7256, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 0.34041017293930054, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 0.7068, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.784, | |
| "grad_norm": 0.339663565158844, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6839, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.29457777738571167, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.6694, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.816, | |
| "grad_norm": 0.3308852016925812, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7289, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.832, | |
| "grad_norm": 0.36052626371383667, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.6802, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.848, | |
| "grad_norm": 0.3117854595184326, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.6572, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 0.3308817446231842, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.6628, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.3344241976737976, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.7066, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.896, | |
| "grad_norm": 0.3327839970588684, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7066, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.912, | |
| "grad_norm": 0.36788785457611084, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.6946, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.928, | |
| "grad_norm": 0.38319170475006104, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.7339, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.944, | |
| "grad_norm": 0.31625112891197205, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.7204, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.323129802942276, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.6615, | |
| "step": 60 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 60, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3241268608206720.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |