| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.7199999999999998, | |
| "eval_steps": 3, | |
| "global_step": 48, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.1656520962715149, | |
| "learning_rate": 2e-05, | |
| "loss": 1.4615, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 1.4899382591247559, | |
| "eval_runtime": 29.186, | |
| "eval_samples_per_second": 3.426, | |
| "eval_steps_per_second": 1.713, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.1882542371749878, | |
| "learning_rate": 4e-05, | |
| "loss": 1.4241, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.15945176780223846, | |
| "learning_rate": 6e-05, | |
| "loss": 1.3849, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 1.485183835029602, | |
| "eval_runtime": 29.3484, | |
| "eval_samples_per_second": 3.407, | |
| "eval_steps_per_second": 1.704, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.13675835728645325, | |
| "learning_rate": 8e-05, | |
| "loss": 1.2212, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.1532098948955536, | |
| "learning_rate": 0.0001, | |
| "loss": 1.3626, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.16159594058990479, | |
| "learning_rate": 0.00012, | |
| "loss": 1.3665, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 1.441084623336792, | |
| "eval_runtime": 29.4785, | |
| "eval_samples_per_second": 3.392, | |
| "eval_steps_per_second": 1.696, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.1462002545595169, | |
| "learning_rate": 0.00014, | |
| "loss": 1.3003, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.13418763875961304, | |
| "learning_rate": 0.00016, | |
| "loss": 1.3331, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.10984567552804947, | |
| "learning_rate": 0.00018, | |
| "loss": 1.2689, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_loss": 1.3380621671676636, | |
| "eval_runtime": 29.5305, | |
| "eval_samples_per_second": 3.386, | |
| "eval_steps_per_second": 1.693, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.10075916349887848, | |
| "learning_rate": 0.0002, | |
| "loss": 1.2942, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.11774784326553345, | |
| "learning_rate": 0.000199658449300667, | |
| "loss": 1.2936, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.10691336542367935, | |
| "learning_rate": 0.00019863613034027224, | |
| "loss": 1.2258, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 1.296021580696106, | |
| "eval_runtime": 29.6291, | |
| "eval_samples_per_second": 3.375, | |
| "eval_steps_per_second": 1.688, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 0.10832437872886658, | |
| "learning_rate": 0.00019694002659393305, | |
| "loss": 1.2647, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 0.11240752041339874, | |
| "learning_rate": 0.00019458172417006347, | |
| "loss": 1.2595, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.10769112408161163, | |
| "learning_rate": 0.00019157733266550575, | |
| "loss": 1.2518, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_loss": 1.279650092124939, | |
| "eval_runtime": 29.6168, | |
| "eval_samples_per_second": 3.376, | |
| "eval_steps_per_second": 1.688, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.09908384829759598, | |
| "learning_rate": 0.0001879473751206489, | |
| "loss": 1.1644, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 0.09107685834169388, | |
| "learning_rate": 0.00018371664782625287, | |
| "loss": 1.1601, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.09361294656991959, | |
| "learning_rate": 0.00017891405093963938, | |
| "loss": 1.2263, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_loss": 1.2533847093582153, | |
| "eval_runtime": 29.5676, | |
| "eval_samples_per_second": 3.382, | |
| "eval_steps_per_second": 1.691, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 0.0980026125907898, | |
| "learning_rate": 0.00017357239106731317, | |
| "loss": 1.2272, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.07957063615322113, | |
| "learning_rate": 0.00016772815716257412, | |
| "loss": 1.1913, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 0.07226990163326263, | |
| "learning_rate": 0.0001614212712689668, | |
| "loss": 1.1343, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "eval_loss": 1.2354038953781128, | |
| "eval_runtime": 29.5859, | |
| "eval_samples_per_second": 3.38, | |
| "eval_steps_per_second": 1.69, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 0.0797078013420105, | |
| "learning_rate": 0.00015469481581224272, | |
| "loss": 1.202, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.0746772438287735, | |
| "learning_rate": 0.00014759473930370736, | |
| "loss": 1.2479, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.07190073281526566, | |
| "learning_rate": 0.00014016954246529696, | |
| "loss": 1.2699, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "eval_loss": 1.2254745960235596, | |
| "eval_runtime": 29.6412, | |
| "eval_samples_per_second": 3.374, | |
| "eval_steps_per_second": 1.687, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 0.06926661729812622, | |
| "learning_rate": 0.00013246994692046836, | |
| "loss": 1.2042, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 0.07788683474063873, | |
| "learning_rate": 0.00012454854871407994, | |
| "loss": 1.1925, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.06513918191194534, | |
| "learning_rate": 0.00011645945902807341, | |
| "loss": 1.1493, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "eval_loss": 1.2227890491485596, | |
| "eval_runtime": 29.6808, | |
| "eval_samples_per_second": 3.369, | |
| "eval_steps_per_second": 1.685, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 0.07514671981334686, | |
| "learning_rate": 0.00010825793454723325, | |
| "loss": 1.1685, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.06782150268554688, | |
| "learning_rate": 0.0001, | |
| "loss": 1.2049, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 0.06837104260921478, | |
| "learning_rate": 9.174206545276677e-05, | |
| "loss": 1.153, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "eval_loss": 1.2187583446502686, | |
| "eval_runtime": 29.5599, | |
| "eval_samples_per_second": 3.383, | |
| "eval_steps_per_second": 1.691, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.0675550326704979, | |
| "learning_rate": 8.35405409719266e-05, | |
| "loss": 1.1826, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 0.06812074780464172, | |
| "learning_rate": 7.54514512859201e-05, | |
| "loss": 1.2106, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.06854978948831558, | |
| "learning_rate": 6.753005307953167e-05, | |
| "loss": 1.1947, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_loss": 1.2183395624160767, | |
| "eval_runtime": 29.5611, | |
| "eval_samples_per_second": 3.383, | |
| "eval_steps_per_second": 1.691, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 0.06954147666692734, | |
| "learning_rate": 5.983045753470308e-05, | |
| "loss": 1.1887, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 0.07219364494085312, | |
| "learning_rate": 5.240526069629265e-05, | |
| "loss": 1.1784, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.07003732025623322, | |
| "learning_rate": 4.530518418775733e-05, | |
| "loss": 1.1125, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_loss": 1.2157045602798462, | |
| "eval_runtime": 29.5711, | |
| "eval_samples_per_second": 3.382, | |
| "eval_steps_per_second": 1.691, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.06867608428001404, | |
| "learning_rate": 3.857872873103322e-05, | |
| "loss": 1.1847, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 0.06787115335464478, | |
| "learning_rate": 3.227184283742591e-05, | |
| "loss": 1.1651, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 0.06784378737211227, | |
| "learning_rate": 2.6427608932686843e-05, | |
| "loss": 1.1512, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "eval_loss": 1.2122877836227417, | |
| "eval_runtime": 29.5849, | |
| "eval_samples_per_second": 3.38, | |
| "eval_steps_per_second": 1.69, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 0.064244844019413, | |
| "learning_rate": 2.1085949060360654e-05, | |
| "loss": 1.154, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 0.0654948428273201, | |
| "learning_rate": 1.6283352173747145e-05, | |
| "loss": 1.1454, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "grad_norm": 0.06797634065151215, | |
| "learning_rate": 1.2052624879351104e-05, | |
| "loss": 1.1883, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "eval_loss": 1.209986686706543, | |
| "eval_runtime": 29.54, | |
| "eval_samples_per_second": 3.385, | |
| "eval_steps_per_second": 1.693, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "grad_norm": 0.06889563798904419, | |
| "learning_rate": 8.422667334494249e-06, | |
| "loss": 1.174, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 0.06667731702327728, | |
| "learning_rate": 5.418275829936537e-06, | |
| "loss": 1.0968, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "grad_norm": 0.0668441653251648, | |
| "learning_rate": 3.059973406066963e-06, | |
| "loss": 1.1012, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "eval_loss": 1.2118505239486694, | |
| "eval_runtime": 29.5358, | |
| "eval_samples_per_second": 3.386, | |
| "eval_steps_per_second": 1.693, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "grad_norm": 0.06772288680076599, | |
| "learning_rate": 1.3638696597277679e-06, | |
| "loss": 1.1264, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "grad_norm": 0.06901554763317108, | |
| "learning_rate": 3.415506993330153e-07, | |
| "loss": 1.1457, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 3.7199999999999998, | |
| "grad_norm": 0.06979726254940033, | |
| "learning_rate": 0.0, | |
| "loss": 1.1891, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 3.7199999999999998, | |
| "eval_loss": 1.2122316360473633, | |
| "eval_runtime": 29.5331, | |
| "eval_samples_per_second": 3.386, | |
| "eval_steps_per_second": 1.693, | |
| "step": 48 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 48, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 12, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9974952442724352.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |