| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.043835616438356165, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0008767123287671233, | |
| "grad_norm": 1992.504150390625, | |
| "learning_rate": 5e-05, | |
| "loss": 103.2688, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008767123287671233, | |
| "eval_loss": 3.141178607940674, | |
| "eval_runtime": 695.8138, | |
| "eval_samples_per_second": 11.045, | |
| "eval_steps_per_second": 2.762, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0017534246575342466, | |
| "grad_norm": 1830.1170654296875, | |
| "learning_rate": 0.0001, | |
| "loss": 99.0502, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00263013698630137, | |
| "grad_norm": 717.9740600585938, | |
| "learning_rate": 9.989294616193017e-05, | |
| "loss": 37.293, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0035068493150684932, | |
| "grad_norm": 266.412109375, | |
| "learning_rate": 9.957224306869053e-05, | |
| "loss": 8.7787, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.004383561643835616, | |
| "grad_norm": 248.5948028564453, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 3.353, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00526027397260274, | |
| "grad_norm": 24.0151424407959, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 0.5836, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.006136986301369863, | |
| "grad_norm": 5.856022357940674, | |
| "learning_rate": 9.73465064747553e-05, | |
| "loss": 0.0592, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0070136986301369865, | |
| "grad_norm": 51.24010467529297, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.3567, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00789041095890411, | |
| "grad_norm": 0.2786119282245636, | |
| "learning_rate": 9.484363707663442e-05, | |
| "loss": 0.0026, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.008767123287671232, | |
| "grad_norm": 0.6561084389686584, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.0021, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.009643835616438357, | |
| "grad_norm": 0.08350276947021484, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.0004, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.01052054794520548, | |
| "grad_norm": 4.876893997192383, | |
| "learning_rate": 8.966766701456177e-05, | |
| "loss": 0.0055, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.011397260273972603, | |
| "grad_norm": 0.1872442364692688, | |
| "learning_rate": 8.759199037394887e-05, | |
| "loss": 0.0005, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.011397260273972603, | |
| "eval_loss": 0.008148307912051678, | |
| "eval_runtime": 696.3466, | |
| "eval_samples_per_second": 11.036, | |
| "eval_steps_per_second": 2.76, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.012273972602739725, | |
| "grad_norm": 0.7512439489364624, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0017, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.01315068493150685, | |
| "grad_norm": 22.033903121948242, | |
| "learning_rate": 8.296729075500344e-05, | |
| "loss": 0.7813, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.014027397260273973, | |
| "grad_norm": 56.58907699584961, | |
| "learning_rate": 8.043807145043604e-05, | |
| "loss": 1.5295, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.014904109589041096, | |
| "grad_norm": 2.5578157901763916, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.0037, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.01578082191780822, | |
| "grad_norm": 0.013324792496860027, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0001, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.01665753424657534, | |
| "grad_norm": 0.09056846052408218, | |
| "learning_rate": 7.211443451095007e-05, | |
| "loss": 0.0005, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.017534246575342465, | |
| "grad_norm": 0.35177427530288696, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.0007, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.018410958904109587, | |
| "grad_norm": 39.4155387878418, | |
| "learning_rate": 6.607197326515808e-05, | |
| "loss": 1.0277, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.019287671232876714, | |
| "grad_norm": 0.11208546161651611, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 0.0005, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.020164383561643837, | |
| "grad_norm": 23.243501663208008, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 1.4354, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.02104109589041096, | |
| "grad_norm": 0.19026409089565277, | |
| "learning_rate": 5.6526309611002594e-05, | |
| "loss": 0.0009, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.021917808219178082, | |
| "grad_norm": 45.536373138427734, | |
| "learning_rate": 5.327015646150716e-05, | |
| "loss": 1.5384, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.022794520547945205, | |
| "grad_norm": 0.026117563247680664, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0001, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.022794520547945205, | |
| "eval_loss": 0.002191019244492054, | |
| "eval_runtime": 696.2694, | |
| "eval_samples_per_second": 11.037, | |
| "eval_steps_per_second": 2.76, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.023671232876712328, | |
| "grad_norm": 0.03182300180196762, | |
| "learning_rate": 4.6729843538492847e-05, | |
| "loss": 0.0002, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.02454794520547945, | |
| "grad_norm": 0.10340346395969391, | |
| "learning_rate": 4.347369038899744e-05, | |
| "loss": 0.0004, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.025424657534246574, | |
| "grad_norm": 0.028125710785388947, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.0002, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0263013698630137, | |
| "grad_norm": 0.09296861290931702, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 0.0007, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.027178082191780823, | |
| "grad_norm": 0.008260289207100868, | |
| "learning_rate": 3.392802673484193e-05, | |
| "loss": 0.0, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.028054794520547946, | |
| "grad_norm": 0.05867616832256317, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0004, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.02893150684931507, | |
| "grad_norm": 0.01258253213018179, | |
| "learning_rate": 2.7885565489049946e-05, | |
| "loss": 0.0001, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.029808219178082192, | |
| "grad_norm": 0.006502221804112196, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.0001, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.030684931506849315, | |
| "grad_norm": 0.005238674581050873, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.0001, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.03156164383561644, | |
| "grad_norm": 0.05581780895590782, | |
| "learning_rate": 1.9561928549563968e-05, | |
| "loss": 0.0003, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.032438356164383564, | |
| "grad_norm": 20.235116958618164, | |
| "learning_rate": 1.703270924499656e-05, | |
| "loss": 0.3343, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.03331506849315068, | |
| "grad_norm": 40.50947570800781, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.3323, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.03419178082191781, | |
| "grad_norm": 0.002002500696107745, | |
| "learning_rate": 1.2408009626051137e-05, | |
| "loss": 0.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.03419178082191781, | |
| "eval_loss": 0.0014001160161569715, | |
| "eval_runtime": 696.3004, | |
| "eval_samples_per_second": 11.037, | |
| "eval_steps_per_second": 2.76, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.03506849315068493, | |
| "grad_norm": 0.008319096639752388, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 0.0001, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.035945205479452055, | |
| "grad_norm": 0.028254352509975433, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.0002, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.036821917808219175, | |
| "grad_norm": 0.005224841181188822, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0376986301369863, | |
| "grad_norm": 0.0015793335624039173, | |
| "learning_rate": 5.156362923365588e-06, | |
| "loss": 0.0, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.03857534246575343, | |
| "grad_norm": 0.012530317530035973, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0002, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.03945205479452055, | |
| "grad_norm": 0.004576420411467552, | |
| "learning_rate": 2.653493525244721e-06, | |
| "loss": 0.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.04032876712328767, | |
| "grad_norm": 0.009576981887221336, | |
| "learning_rate": 1.70370868554659e-06, | |
| "loss": 0.0001, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.04120547945205479, | |
| "grad_norm": 0.002640854800119996, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.0, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.04208219178082192, | |
| "grad_norm": 0.0029024560935795307, | |
| "learning_rate": 4.277569313094809e-07, | |
| "loss": 0.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.04295890410958904, | |
| "grad_norm": 0.011004402302205563, | |
| "learning_rate": 1.0705383806982606e-07, | |
| "loss": 0.0001, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.043835616438356165, | |
| "grad_norm": 0.0039349207654595375, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1255019781226496e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |