| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.002473563787026158, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.947127574052316e-05, | |
| "grad_norm": 9.065704345703125, | |
| "learning_rate": 1e-05, | |
| "loss": 9.6599, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 4.947127574052316e-05, | |
| "eval_loss": 9.460058212280273, | |
| "eval_runtime": 95.1531, | |
| "eval_samples_per_second": 89.445, | |
| "eval_steps_per_second": 44.728, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 9.894255148104632e-05, | |
| "grad_norm": 9.010805130004883, | |
| "learning_rate": 2e-05, | |
| "loss": 9.2264, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00014841382722156947, | |
| "grad_norm": 10.453004837036133, | |
| "learning_rate": 3e-05, | |
| "loss": 9.5811, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00019788510296209265, | |
| "grad_norm": 8.284311294555664, | |
| "learning_rate": 4e-05, | |
| "loss": 9.5017, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00024735637870261577, | |
| "grad_norm": 8.38084602355957, | |
| "learning_rate": 5e-05, | |
| "loss": 9.5888, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00024735637870261577, | |
| "eval_loss": 9.321613311767578, | |
| "eval_runtime": 95.4249, | |
| "eval_samples_per_second": 89.191, | |
| "eval_steps_per_second": 44.601, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00029682765444313895, | |
| "grad_norm": 8.923850059509277, | |
| "learning_rate": 6e-05, | |
| "loss": 9.4131, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0003462989301836621, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 9.0528, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0003957702059241853, | |
| "grad_norm": 10.302660942077637, | |
| "learning_rate": 7e-05, | |
| "loss": 9.3181, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0004452414816647084, | |
| "grad_norm": 10.739596366882324, | |
| "learning_rate": 8e-05, | |
| "loss": 9.1572, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0004947127574052315, | |
| "grad_norm": 9.674470901489258, | |
| "learning_rate": 9e-05, | |
| "loss": 9.2498, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0004947127574052315, | |
| "eval_loss": 8.945061683654785, | |
| "eval_runtime": 94.8571, | |
| "eval_samples_per_second": 89.724, | |
| "eval_steps_per_second": 44.867, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0005441840331457548, | |
| "grad_norm": 8.420089721679688, | |
| "learning_rate": 0.0001, | |
| "loss": 9.144, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0005936553088862779, | |
| "grad_norm": 8.640538215637207, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 8.9871, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.000643126584626801, | |
| "grad_norm": 9.995044708251953, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 8.3837, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0006925978603673242, | |
| "grad_norm": 9.764483451843262, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 8.608, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0007420691361078474, | |
| "grad_norm": 9.012595176696777, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 8.3834, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0007420691361078474, | |
| "eval_loss": 8.214789390563965, | |
| "eval_runtime": 94.8878, | |
| "eval_samples_per_second": 89.695, | |
| "eval_steps_per_second": 44.853, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0007915404118483706, | |
| "grad_norm": 8.971148490905762, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 8.3348, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0008410116875888937, | |
| "grad_norm": 9.869667053222656, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 8.1376, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0008904829633294168, | |
| "grad_norm": 10.327800750732422, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 7.8432, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0009399542390699401, | |
| "grad_norm": 9.640523910522461, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 7.9228, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.000989425514810463, | |
| "grad_norm": 8.59736156463623, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 7.5875, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.000989425514810463, | |
| "eval_loss": 7.525243282318115, | |
| "eval_runtime": 94.8698, | |
| "eval_samples_per_second": 89.712, | |
| "eval_steps_per_second": 44.862, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0010388967905509863, | |
| "grad_norm": 7.832772731781006, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 7.5962, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0010883680662915095, | |
| "grad_norm": 8.253615379333496, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 7.2084, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0011378393420320326, | |
| "grad_norm": 7.1926188468933105, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 7.2504, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0011873106177725558, | |
| "grad_norm": 8.011531829833984, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 7.3374, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.001236781893513079, | |
| "grad_norm": 6.147706985473633, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 7.134, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.001236781893513079, | |
| "eval_loss": 7.0482401847839355, | |
| "eval_runtime": 94.9771, | |
| "eval_samples_per_second": 89.611, | |
| "eval_steps_per_second": 44.811, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.001286253169253602, | |
| "grad_norm": 5.88217830657959, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 7.0338, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0013357244449941253, | |
| "grad_norm": 5.846419811248779, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 7.1329, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0013851957207346485, | |
| "grad_norm": 5.849457740783691, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 6.7678, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0014346669964751715, | |
| "grad_norm": 5.783066749572754, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 6.8641, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0014841382722156947, | |
| "grad_norm": 5.95844030380249, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 6.7875, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0014841382722156947, | |
| "eval_loss": 6.735217094421387, | |
| "eval_runtime": 95.4239, | |
| "eval_samples_per_second": 89.192, | |
| "eval_steps_per_second": 44.601, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.001533609547956218, | |
| "grad_norm": 5.942857265472412, | |
| "learning_rate": 5e-05, | |
| "loss": 6.7855, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0015830808236967412, | |
| "grad_norm": 5.439502716064453, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 6.6782, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0016325520994372642, | |
| "grad_norm": 5.478299617767334, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 6.6382, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0016820233751777874, | |
| "grad_norm": 6.273443698883057, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 6.5828, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0017314946509183107, | |
| "grad_norm": 6.156881332397461, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 6.6771, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0017314946509183107, | |
| "eval_loss": 6.5112690925598145, | |
| "eval_runtime": 95.7427, | |
| "eval_samples_per_second": 88.895, | |
| "eval_steps_per_second": 44.452, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0017809659266588337, | |
| "grad_norm": 5.457279205322266, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 6.563, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.001830437202399357, | |
| "grad_norm": 5.992366790771484, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 6.4556, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0018799084781398801, | |
| "grad_norm": 6.0998029708862305, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 6.4219, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0019293797538804031, | |
| "grad_norm": 5.670993328094482, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 6.4434, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.001978851029620926, | |
| "grad_norm": 5.718242168426514, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 6.4194, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.001978851029620926, | |
| "eval_loss": 6.370116233825684, | |
| "eval_runtime": 95.6, | |
| "eval_samples_per_second": 89.027, | |
| "eval_steps_per_second": 44.519, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0020283223053614496, | |
| "grad_norm": 5.832897663116455, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 6.417, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0020777935811019726, | |
| "grad_norm": 5.707322120666504, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 6.3907, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0021272648568424956, | |
| "grad_norm": 6.1590800285339355, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 6.3305, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.002176736132583019, | |
| "grad_norm": 5.986573696136475, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 6.4059, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.002226207408323542, | |
| "grad_norm": 5.721052169799805, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 6.3249, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.002226207408323542, | |
| "eval_loss": 6.304693698883057, | |
| "eval_runtime": 94.88, | |
| "eval_samples_per_second": 89.703, | |
| "eval_steps_per_second": 44.857, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.002275678684064065, | |
| "grad_norm": 6.428943634033203, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 6.2422, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.0023251499598045886, | |
| "grad_norm": 5.835705280303955, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 6.2458, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0023746212355451116, | |
| "grad_norm": 6.010042667388916, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 6.4163, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0024240925112856346, | |
| "grad_norm": 5.472652912139893, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 6.2867, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.002473563787026158, | |
| "grad_norm": 5.85940408706665, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 6.1921, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.002473563787026158, | |
| "eval_loss": 6.28983211517334, | |
| "eval_runtime": 94.9476, | |
| "eval_samples_per_second": 89.639, | |
| "eval_steps_per_second": 44.825, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6787380019200.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |