| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.07243752263672583, | |
| "eval_steps": 25, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0014487504527345165, | |
| "grad_norm": 5.941494464874268, | |
| "learning_rate": 5e-05, | |
| "loss": 6.016, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0014487504527345165, | |
| "eval_loss": 6.004227161407471, | |
| "eval_runtime": 1.1418, | |
| "eval_samples_per_second": 43.789, | |
| "eval_steps_per_second": 11.385, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002897500905469033, | |
| "grad_norm": 5.850800037384033, | |
| "learning_rate": 0.0001, | |
| "loss": 6.0326, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.004346251358203549, | |
| "grad_norm": 5.415925025939941, | |
| "learning_rate": 9.990365154573717e-05, | |
| "loss": 5.7164, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.005795001810938066, | |
| "grad_norm": 4.989433765411377, | |
| "learning_rate": 9.961501876182148e-05, | |
| "loss": 4.9589, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.007243752263672582, | |
| "grad_norm": 6.119456768035889, | |
| "learning_rate": 9.913533761814537e-05, | |
| "loss": 4.4917, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.008692502716407098, | |
| "grad_norm": 6.897642135620117, | |
| "learning_rate": 9.846666218300807e-05, | |
| "loss": 3.9613, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.010141253169141615, | |
| "grad_norm": 6.291426658630371, | |
| "learning_rate": 9.761185582727977e-05, | |
| "loss": 3.484, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.011590003621876132, | |
| "grad_norm": 6.523845672607422, | |
| "learning_rate": 9.657457896300791e-05, | |
| "loss": 3.0865, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.013038754074610649, | |
| "grad_norm": 6.794440269470215, | |
| "learning_rate": 9.535927336897098e-05, | |
| "loss": 2.7457, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.014487504527345164, | |
| "grad_norm": 8.122562408447266, | |
| "learning_rate": 9.397114317029975e-05, | |
| "loss": 2.4849, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01593625498007968, | |
| "grad_norm": 9.729490280151367, | |
| "learning_rate": 9.241613255361455e-05, | |
| "loss": 2.1615, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.017385005432814196, | |
| "grad_norm": 7.469081401824951, | |
| "learning_rate": 9.070090031310558e-05, | |
| "loss": 1.8752, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.018833755885548715, | |
| "grad_norm": 4.167898178100586, | |
| "learning_rate": 8.883279133655399e-05, | |
| "loss": 1.7962, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.02028250633828323, | |
| "grad_norm": 3.0158345699310303, | |
| "learning_rate": 8.681980515339464e-05, | |
| "loss": 1.652, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.02173125679101775, | |
| "grad_norm": 1.1410199403762817, | |
| "learning_rate": 8.467056167950311e-05, | |
| "loss": 1.6153, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.023180007243752264, | |
| "grad_norm": 1.4439971446990967, | |
| "learning_rate": 8.239426430539243e-05, | |
| "loss": 1.6087, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.02462875769648678, | |
| "grad_norm": 1.1010881662368774, | |
| "learning_rate": 8.000066048588211e-05, | |
| "loss": 1.5891, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.026077508149221298, | |
| "grad_norm": 3.1430435180664062, | |
| "learning_rate": 7.75e-05, | |
| "loss": 1.624, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.027526258601955813, | |
| "grad_norm": 1.2153921127319336, | |
| "learning_rate": 7.490299105985507e-05, | |
| "loss": 1.619, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.028975009054690328, | |
| "grad_norm": 0.8797547817230225, | |
| "learning_rate": 7.222075445642904e-05, | |
| "loss": 1.5787, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.030423759507424847, | |
| "grad_norm": 1.1267567873001099, | |
| "learning_rate": 6.946477593864228e-05, | |
| "loss": 1.5738, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.03187250996015936, | |
| "grad_norm": 0.7399479746818542, | |
| "learning_rate": 6.664685702961344e-05, | |
| "loss": 1.5835, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.03332126041289388, | |
| "grad_norm": 0.8881580233573914, | |
| "learning_rate": 6.377906449072578e-05, | |
| "loss": 1.5823, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.03477001086562839, | |
| "grad_norm": 2.5095949172973633, | |
| "learning_rate": 6.087367864990233e-05, | |
| "loss": 1.5911, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.036218761318362915, | |
| "grad_norm": 0.7536084055900574, | |
| "learning_rate": 5.794314081535644e-05, | |
| "loss": 1.5854, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.036218761318362915, | |
| "eval_loss": 1.5820493698120117, | |
| "eval_runtime": 1.1609, | |
| "eval_samples_per_second": 43.069, | |
| "eval_steps_per_second": 11.198, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03766751177109743, | |
| "grad_norm": 1.519792914390564, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 1.6345, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.039116262223831945, | |
| "grad_norm": 0.8911117315292358, | |
| "learning_rate": 5.205685918464356e-05, | |
| "loss": 1.5823, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.04056501267656646, | |
| "grad_norm": 0.8245032429695129, | |
| "learning_rate": 4.912632135009769e-05, | |
| "loss": 1.5606, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.042013763129300975, | |
| "grad_norm": 0.8425853848457336, | |
| "learning_rate": 4.6220935509274235e-05, | |
| "loss": 1.5497, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0434625135820355, | |
| "grad_norm": 1.2193143367767334, | |
| "learning_rate": 4.3353142970386564e-05, | |
| "loss": 1.5908, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04491126403477001, | |
| "grad_norm": 1.1511098146438599, | |
| "learning_rate": 4.053522406135775e-05, | |
| "loss": 1.5944, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.04636001448750453, | |
| "grad_norm": 0.8153553605079651, | |
| "learning_rate": 3.777924554357096e-05, | |
| "loss": 1.5599, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.04780876494023904, | |
| "grad_norm": 0.9374620914459229, | |
| "learning_rate": 3.509700894014496e-05, | |
| "loss": 1.5604, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.04925751539297356, | |
| "grad_norm": 0.5474502444267273, | |
| "learning_rate": 3.250000000000001e-05, | |
| "loss": 1.5657, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.05070626584570807, | |
| "grad_norm": 0.7612038254737854, | |
| "learning_rate": 2.9999339514117912e-05, | |
| "loss": 1.5785, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.052155016298442596, | |
| "grad_norm": 0.6650935411453247, | |
| "learning_rate": 2.760573569460757e-05, | |
| "loss": 1.5666, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.05360376675117711, | |
| "grad_norm": 0.7727776169776917, | |
| "learning_rate": 2.53294383204969e-05, | |
| "loss": 1.5634, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.055052517203911626, | |
| "grad_norm": 0.5608248114585876, | |
| "learning_rate": 2.3180194846605367e-05, | |
| "loss": 1.5904, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.05650126765664614, | |
| "grad_norm": 0.879086434841156, | |
| "learning_rate": 2.1167208663446025e-05, | |
| "loss": 1.5817, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.057950018109380656, | |
| "grad_norm": 0.8060774207115173, | |
| "learning_rate": 1.9299099686894423e-05, | |
| "loss": 1.5599, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05939876856211518, | |
| "grad_norm": 0.7871993184089661, | |
| "learning_rate": 1.758386744638546e-05, | |
| "loss": 1.5396, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.060847519014849694, | |
| "grad_norm": 0.3978239893913269, | |
| "learning_rate": 1.602885682970026e-05, | |
| "loss": 1.5451, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.06229626946758421, | |
| "grad_norm": 0.7827884554862976, | |
| "learning_rate": 1.464072663102903e-05, | |
| "loss": 1.5745, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.06374501992031872, | |
| "grad_norm": 1.1979758739471436, | |
| "learning_rate": 1.3425421036992098e-05, | |
| "loss": 1.5875, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.06519377037305324, | |
| "grad_norm": 1.3340702056884766, | |
| "learning_rate": 1.2388144172720251e-05, | |
| "loss": 1.5744, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06664252082578775, | |
| "grad_norm": 1.1701022386550903, | |
| "learning_rate": 1.1533337816991932e-05, | |
| "loss": 1.5734, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.06809127127852227, | |
| "grad_norm": 1.051703691482544, | |
| "learning_rate": 1.0864662381854632e-05, | |
| "loss": 1.5751, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.06954002173125678, | |
| "grad_norm": 0.8547968864440918, | |
| "learning_rate": 1.0384981238178534e-05, | |
| "loss": 1.5754, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.07098877218399131, | |
| "grad_norm": 0.6540709137916565, | |
| "learning_rate": 1.0096348454262845e-05, | |
| "loss": 1.5709, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.07243752263672583, | |
| "grad_norm": 0.7267070412635803, | |
| "learning_rate": 1e-05, | |
| "loss": 1.5731, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07243752263672583, | |
| "eval_loss": 1.5619583129882812, | |
| "eval_runtime": 1.1496, | |
| "eval_samples_per_second": 43.492, | |
| "eval_steps_per_second": 11.308, | |
| "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": 1.0606952192776602e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |