| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.192, | |
| "eval_steps": 500, | |
| "global_step": 60, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0032, | |
| "grad_norm": 0.7807257175445557, | |
| "learning_rate": 4e-05, | |
| "loss": 2.509, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0064, | |
| "grad_norm": 0.7869868874549866, | |
| "learning_rate": 8e-05, | |
| "loss": 2.5141, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0096, | |
| "grad_norm": 0.7782315611839294, | |
| "learning_rate": 0.00012, | |
| "loss": 2.4717, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0128, | |
| "grad_norm": 0.8121834993362427, | |
| "learning_rate": 0.00016, | |
| "loss": 2.3125, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.016, | |
| "grad_norm": 0.8421923518180847, | |
| "learning_rate": 0.0002, | |
| "loss": 2.087, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0192, | |
| "grad_norm": 1.1386044025421143, | |
| "learning_rate": 0.00019636363636363636, | |
| "loss": 1.8111, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0224, | |
| "grad_norm": 1.2734416723251343, | |
| "learning_rate": 0.00019272727272727274, | |
| "loss": 1.497, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0256, | |
| "grad_norm": 1.3242508172988892, | |
| "learning_rate": 0.0001890909090909091, | |
| "loss": 1.1827, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0288, | |
| "grad_norm": 1.5871871709823608, | |
| "learning_rate": 0.00018545454545454545, | |
| "loss": 0.9281, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.032, | |
| "grad_norm": 1.4778683185577393, | |
| "learning_rate": 0.00018181818181818183, | |
| "loss": 0.7139, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0352, | |
| "grad_norm": 1.833769679069519, | |
| "learning_rate": 0.0001781818181818182, | |
| "loss": 0.572, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0384, | |
| "grad_norm": 1.3144052028656006, | |
| "learning_rate": 0.00017454545454545454, | |
| "loss": 0.4447, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0416, | |
| "grad_norm": 3.582252264022827, | |
| "learning_rate": 0.0001709090909090909, | |
| "loss": 0.38, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0448, | |
| "grad_norm": 2.804332971572876, | |
| "learning_rate": 0.00016727272727272728, | |
| "loss": 0.3184, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.048, | |
| "grad_norm": 1.5338329076766968, | |
| "learning_rate": 0.00016363636363636366, | |
| "loss": 0.2658, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0512, | |
| "grad_norm": 0.9025959968566895, | |
| "learning_rate": 0.00016, | |
| "loss": 0.2266, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0544, | |
| "grad_norm": 0.9067522883415222, | |
| "learning_rate": 0.00015636363636363637, | |
| "loss": 0.1958, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0576, | |
| "grad_norm": 0.5175400376319885, | |
| "learning_rate": 0.00015272727272727275, | |
| "loss": 0.1932, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0608, | |
| "grad_norm": 0.30395233631134033, | |
| "learning_rate": 0.0001490909090909091, | |
| "loss": 0.1829, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.064, | |
| "grad_norm": 0.28186506032943726, | |
| "learning_rate": 0.00014545454545454546, | |
| "loss": 0.1788, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0672, | |
| "grad_norm": 0.21213509142398834, | |
| "learning_rate": 0.00014181818181818184, | |
| "loss": 0.1711, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0704, | |
| "grad_norm": 0.2244960516691208, | |
| "learning_rate": 0.0001381818181818182, | |
| "loss": 0.1712, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0736, | |
| "grad_norm": 0.22173970937728882, | |
| "learning_rate": 0.00013454545454545455, | |
| "loss": 0.1652, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0768, | |
| "grad_norm": 0.16787873208522797, | |
| "learning_rate": 0.00013090909090909093, | |
| "loss": 0.166, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.20377585291862488, | |
| "learning_rate": 0.00012727272727272728, | |
| "loss": 0.1593, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0832, | |
| "grad_norm": 0.17485077679157257, | |
| "learning_rate": 0.00012363636363636364, | |
| "loss": 0.1591, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0864, | |
| "grad_norm": 0.1908087134361267, | |
| "learning_rate": 0.00012, | |
| "loss": 0.1585, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0896, | |
| "grad_norm": 0.21306680142879486, | |
| "learning_rate": 0.00011636363636363636, | |
| "loss": 0.1612, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0928, | |
| "grad_norm": 0.1579781174659729, | |
| "learning_rate": 0.00011272727272727272, | |
| "loss": 0.1532, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 0.21352051198482513, | |
| "learning_rate": 0.00010909090909090909, | |
| "loss": 0.1464, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0992, | |
| "grad_norm": 0.2137880176305771, | |
| "learning_rate": 0.00010545454545454545, | |
| "loss": 0.1527, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.1024, | |
| "grad_norm": 0.21048834919929504, | |
| "learning_rate": 0.00010181818181818181, | |
| "loss": 0.1499, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.1056, | |
| "grad_norm": 0.206934854388237, | |
| "learning_rate": 9.818181818181818e-05, | |
| "loss": 0.1457, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.1088, | |
| "grad_norm": 0.1933310180902481, | |
| "learning_rate": 9.454545454545455e-05, | |
| "loss": 0.1487, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.112, | |
| "grad_norm": 0.19178827106952667, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 0.1474, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.1152, | |
| "grad_norm": 0.24533748626708984, | |
| "learning_rate": 8.727272727272727e-05, | |
| "loss": 0.1495, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.1184, | |
| "grad_norm": 0.29513201117515564, | |
| "learning_rate": 8.363636363636364e-05, | |
| "loss": 0.1469, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.1216, | |
| "grad_norm": 0.19339747726917267, | |
| "learning_rate": 8e-05, | |
| "loss": 0.1409, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.1248, | |
| "grad_norm": 0.21743986010551453, | |
| "learning_rate": 7.636363636363637e-05, | |
| "loss": 0.1418, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.128, | |
| "grad_norm": 0.23053643107414246, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.1418, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1312, | |
| "grad_norm": 0.22498559951782227, | |
| "learning_rate": 6.90909090909091e-05, | |
| "loss": 0.1403, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.1344, | |
| "grad_norm": 0.2134191244840622, | |
| "learning_rate": 6.545454545454546e-05, | |
| "loss": 0.1416, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.1376, | |
| "grad_norm": 0.20188340544700623, | |
| "learning_rate": 6.181818181818182e-05, | |
| "loss": 0.1388, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.1408, | |
| "grad_norm": 0.22138598561286926, | |
| "learning_rate": 5.818181818181818e-05, | |
| "loss": 0.137, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.144, | |
| "grad_norm": 0.2358243465423584, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 0.1386, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.1472, | |
| "grad_norm": 0.2473626732826233, | |
| "learning_rate": 5.090909090909091e-05, | |
| "loss": 0.1367, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.1504, | |
| "grad_norm": 0.23792380094528198, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 0.1335, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.1536, | |
| "grad_norm": 0.2288408726453781, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 0.1345, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1568, | |
| "grad_norm": 0.2344062626361847, | |
| "learning_rate": 4e-05, | |
| "loss": 0.1319, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.23560936748981476, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.131, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1632, | |
| "grad_norm": 0.24920542538166046, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 0.1351, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.1664, | |
| "grad_norm": 0.2517445683479309, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 0.1344, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.1696, | |
| "grad_norm": 0.24988840520381927, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 0.1302, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.1728, | |
| "grad_norm": 0.2461981624364853, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 0.1284, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.176, | |
| "grad_norm": 0.25435972213745117, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.1302, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.1792, | |
| "grad_norm": 0.2487025409936905, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 0.1309, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.1824, | |
| "grad_norm": 0.24456636607646942, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 0.1275, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.1856, | |
| "grad_norm": 0.2547883689403534, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 0.1254, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.1888, | |
| "grad_norm": 0.2569814622402191, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 0.1296, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 0.2542433440685272, | |
| "learning_rate": 0.0, | |
| "loss": 0.126, | |
| "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": 5.55328713719808e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |