| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0645879732739423, | |
| "eval_steps": 25, | |
| "global_step": 43, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07126948775055679, | |
| "grad_norm": 10.804896354675293, | |
| "learning_rate": 5e-05, | |
| "loss": 4.5978, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07126948775055679, | |
| "eval_loss": 4.046659469604492, | |
| "eval_runtime": 2.4557, | |
| "eval_samples_per_second": 20.361, | |
| "eval_steps_per_second": 5.294, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.14253897550111358, | |
| "grad_norm": 10.301349639892578, | |
| "learning_rate": 0.0001, | |
| "loss": 4.8501, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.21380846325167038, | |
| "grad_norm": 12.16850471496582, | |
| "learning_rate": 9.986796105326832e-05, | |
| "loss": 2.3801, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.28507795100222716, | |
| "grad_norm": 7.317905902862549, | |
| "learning_rate": 9.947261907011569e-05, | |
| "loss": 2.669, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.35634743875278396, | |
| "grad_norm": 7.833179950714111, | |
| "learning_rate": 9.881629407450006e-05, | |
| "loss": 2.5579, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.42761692650334077, | |
| "grad_norm": 5.9061079025268555, | |
| "learning_rate": 9.790283764247188e-05, | |
| "loss": 1.8147, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.49888641425389757, | |
| "grad_norm": 4.455478191375732, | |
| "learning_rate": 9.673761029959427e-05, | |
| "loss": 0.7854, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.5701559020044543, | |
| "grad_norm": 4.549908638000488, | |
| "learning_rate": 9.53274500632475e-05, | |
| "loss": 1.3198, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.6414253897550112, | |
| "grad_norm": 4.589344024658203, | |
| "learning_rate": 9.368063231442406e-05, | |
| "loss": 1.093, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.7126948775055679, | |
| "grad_norm": 3.6987669467926025, | |
| "learning_rate": 9.18068212345023e-05, | |
| "loss": 0.3916, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7839643652561247, | |
| "grad_norm": 4.129584312438965, | |
| "learning_rate": 8.971701309198744e-05, | |
| "loss": 0.5841, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.8552338530066815, | |
| "grad_norm": 3.923628330230713, | |
| "learning_rate": 8.742347171203542e-05, | |
| "loss": 0.6434, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.9265033407572383, | |
| "grad_norm": 2.7002556324005127, | |
| "learning_rate": 8.493965650745043e-05, | |
| "loss": 0.5206, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.9977728285077951, | |
| "grad_norm": 1.9785927534103394, | |
| "learning_rate": 8.228014349349872e-05, | |
| "loss": 0.3233, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.069042316258352, | |
| "grad_norm": 3.3591909408569336, | |
| "learning_rate": 7.946053975005496e-05, | |
| "loss": 0.5516, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.1403118040089086, | |
| "grad_norm": 1.316374659538269, | |
| "learning_rate": 7.649739183305184e-05, | |
| "loss": 0.3752, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.2115812917594655, | |
| "grad_norm": 1.1888165473937988, | |
| "learning_rate": 7.34080886727103e-05, | |
| "loss": 0.2189, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.2828507795100224, | |
| "grad_norm": 1.096291184425354, | |
| "learning_rate": 7.021075952838262e-05, | |
| "loss": 0.2563, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.354120267260579, | |
| "grad_norm": 1.2140592336654663, | |
| "learning_rate": 6.692416759884978e-05, | |
| "loss": 0.211, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.4253897550111359, | |
| "grad_norm": 1.2319027185440063, | |
| "learning_rate": 6.356759991241008e-05, | |
| "loss": 0.2163, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.4966592427616927, | |
| "grad_norm": 1.153494954109192, | |
| "learning_rate": 6.016075414292803e-05, | |
| "loss": 0.2663, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.5679287305122496, | |
| "grad_norm": 0.7327424883842468, | |
| "learning_rate": 5.6723623016051596e-05, | |
| "loss": 0.174, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.6391982182628062, | |
| "grad_norm": 1.0020651817321777, | |
| "learning_rate": 5.3276376983948415e-05, | |
| "loss": 0.2659, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.710467706013363, | |
| "grad_norm": 1.2167654037475586, | |
| "learning_rate": 4.9839245857071994e-05, | |
| "loss": 0.1188, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.7817371937639197, | |
| "grad_norm": 0.9578669667243958, | |
| "learning_rate": 4.643240008758992e-05, | |
| "loss": 0.3337, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.7817371937639197, | |
| "eval_loss": 0.1293194591999054, | |
| "eval_runtime": 1.4921, | |
| "eval_samples_per_second": 33.509, | |
| "eval_steps_per_second": 8.712, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.8530066815144766, | |
| "grad_norm": 0.8329769372940063, | |
| "learning_rate": 4.307583240115024e-05, | |
| "loss": 0.1588, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.9242761692650334, | |
| "grad_norm": 1.176580548286438, | |
| "learning_rate": 3.9789240471617385e-05, | |
| "loss": 0.1374, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.9955456570155903, | |
| "grad_norm": 0.6185368895530701, | |
| "learning_rate": 3.659191132728971e-05, | |
| "loss": 0.2032, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.066815144766147, | |
| "grad_norm": 3.2315216064453125, | |
| "learning_rate": 3.350260816694816e-05, | |
| "loss": 0.514, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.138084632516704, | |
| "grad_norm": 0.7700513601303101, | |
| "learning_rate": 3.0539460249945056e-05, | |
| "loss": 0.1931, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.2093541202672604, | |
| "grad_norm": 0.5963460206985474, | |
| "learning_rate": 2.7719856506501306e-05, | |
| "loss": 0.1288, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.2806236080178173, | |
| "grad_norm": 0.5300996899604797, | |
| "learning_rate": 2.5060343492549566e-05, | |
| "loss": 0.1928, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.351893095768374, | |
| "grad_norm": 0.9159259796142578, | |
| "learning_rate": 2.257652828796459e-05, | |
| "loss": 0.1928, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.423162583518931, | |
| "grad_norm": 0.8254880905151367, | |
| "learning_rate": 2.0282986908012567e-05, | |
| "loss": 0.1319, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.494432071269488, | |
| "grad_norm": 0.6946773529052734, | |
| "learning_rate": 1.819317876549771e-05, | |
| "loss": 0.2188, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.5657015590200447, | |
| "grad_norm": 0.7235588431358337, | |
| "learning_rate": 1.631936768557596e-05, | |
| "loss": 0.2069, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.6369710467706016, | |
| "grad_norm": 0.5640904903411865, | |
| "learning_rate": 1.4672549936752506e-05, | |
| "loss": 0.1033, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.708240534521158, | |
| "grad_norm": 0.5563797354698181, | |
| "learning_rate": 1.3262389700405744e-05, | |
| "loss": 0.1157, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.779510022271715, | |
| "grad_norm": 0.6494319438934326, | |
| "learning_rate": 1.2097162357528127e-05, | |
| "loss": 0.2194, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.8507795100222717, | |
| "grad_norm": 0.6779353022575378, | |
| "learning_rate": 1.1183705925499948e-05, | |
| "loss": 0.1103, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.9220489977728286, | |
| "grad_norm": 0.7211727499961853, | |
| "learning_rate": 1.0527380929884324e-05, | |
| "loss": 0.1513, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 2.9933184855233854, | |
| "grad_norm": 0.5112365484237671, | |
| "learning_rate": 1.013203894673168e-05, | |
| "loss": 0.1698, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 3.0645879732739423, | |
| "grad_norm": 5.871841907501221, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5043, | |
| "step": 43 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 43, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 4.049313816222106e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |