| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5024812976816532, | |
| "eval_steps": 11, | |
| "global_step": 53, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.009480779201540626, | |
| "grad_norm": 0.4609375, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3706, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.009480779201540626, | |
| "eval_loss": 0.35137510299682617, | |
| "eval_runtime": 37.2418, | |
| "eval_samples_per_second": 19.091, | |
| "eval_steps_per_second": 19.091, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.018961558403081252, | |
| "grad_norm": 0.419921875, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3413, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.02844233760462188, | |
| "grad_norm": 0.4375, | |
| "learning_rate": 6e-05, | |
| "loss": 0.3573, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.037923116806162505, | |
| "grad_norm": 0.37890625, | |
| "learning_rate": 8e-05, | |
| "loss": 0.3077, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.04740389600770313, | |
| "grad_norm": 0.33984375, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2305, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05688467520924376, | |
| "grad_norm": 0.337890625, | |
| "learning_rate": 0.00012, | |
| "loss": 0.1989, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.06636545441078438, | |
| "grad_norm": 0.2275390625, | |
| "learning_rate": 0.00014, | |
| "loss": 0.1523, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.07584623361232501, | |
| "grad_norm": 0.1767578125, | |
| "learning_rate": 0.00016, | |
| "loss": 0.1402, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.08532701281386564, | |
| "grad_norm": 0.1494140625, | |
| "learning_rate": 0.00018, | |
| "loss": 0.1156, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.09480779201540626, | |
| "grad_norm": 0.140625, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0993, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10428857121694689, | |
| "grad_norm": 0.1357421875, | |
| "learning_rate": 0.00019999469523400122, | |
| "loss": 0.1022, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.10428857121694689, | |
| "eval_loss": 0.09353966265916824, | |
| "eval_runtime": 37.0224, | |
| "eval_samples_per_second": 19.205, | |
| "eval_steps_per_second": 19.205, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.11376935041848751, | |
| "grad_norm": 0.08740234375, | |
| "learning_rate": 0.00019997878149881574, | |
| "loss": 0.0939, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.12325012962002814, | |
| "grad_norm": 0.08837890625, | |
| "learning_rate": 0.0001999522604828164, | |
| "loss": 0.0876, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.13273090882156877, | |
| "grad_norm": 0.08984375, | |
| "learning_rate": 0.00019991513499975882, | |
| "loss": 0.0984, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.1422116880231094, | |
| "grad_norm": 0.091796875, | |
| "learning_rate": 0.00019986740898848306, | |
| "loss": 0.0825, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.15169246722465002, | |
| "grad_norm": 0.06640625, | |
| "learning_rate": 0.00019980908751249555, | |
| "loss": 0.0801, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.16117324642619066, | |
| "grad_norm": 0.0703125, | |
| "learning_rate": 0.00019974017675943192, | |
| "loss": 0.0801, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.17065402562773127, | |
| "grad_norm": 0.0634765625, | |
| "learning_rate": 0.0001996606840404006, | |
| "loss": 0.0845, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1801348048292719, | |
| "grad_norm": 0.060546875, | |
| "learning_rate": 0.00019957061778920701, | |
| "loss": 0.0652, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.18961558403081252, | |
| "grad_norm": 0.06689453125, | |
| "learning_rate": 0.0001994699875614589, | |
| "loss": 0.0768, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.19909636323235316, | |
| "grad_norm": 0.059326171875, | |
| "learning_rate": 0.00019935880403355253, | |
| "loss": 0.0689, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.20857714243389378, | |
| "grad_norm": 0.0693359375, | |
| "learning_rate": 0.00019923707900153982, | |
| "loss": 0.0636, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.20857714243389378, | |
| "eval_loss": 0.07002714276313782, | |
| "eval_runtime": 37.2313, | |
| "eval_samples_per_second": 19.097, | |
| "eval_steps_per_second": 19.097, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.21805792163543442, | |
| "grad_norm": 0.060302734375, | |
| "learning_rate": 0.00019910482537987702, | |
| "loss": 0.0717, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.22753870083697503, | |
| "grad_norm": 0.06396484375, | |
| "learning_rate": 0.0001989620572000544, | |
| "loss": 0.0703, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.23701948003851567, | |
| "grad_norm": 0.08544921875, | |
| "learning_rate": 0.00019880878960910772, | |
| "loss": 0.078, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.24650025924005628, | |
| "grad_norm": 0.0654296875, | |
| "learning_rate": 0.00019864503886801106, | |
| "loss": 0.0762, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.2559810384415969, | |
| "grad_norm": 0.06689453125, | |
| "learning_rate": 0.00019847082234995171, | |
| "loss": 0.0634, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.26546181764313753, | |
| "grad_norm": 0.0693359375, | |
| "learning_rate": 0.00019828615853848688, | |
| "loss": 0.062, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.27494259684467814, | |
| "grad_norm": 0.056640625, | |
| "learning_rate": 0.00019809106702558277, | |
| "loss": 0.0594, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.2844233760462188, | |
| "grad_norm": 0.05859375, | |
| "learning_rate": 0.0001978855685095358, | |
| "loss": 0.0592, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2939041552477594, | |
| "grad_norm": 0.06787109375, | |
| "learning_rate": 0.00019766968479277683, | |
| "loss": 0.0721, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.30338493444930004, | |
| "grad_norm": 0.0595703125, | |
| "learning_rate": 0.00019744343877955788, | |
| "loss": 0.0625, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.3128657136508407, | |
| "grad_norm": 0.0576171875, | |
| "learning_rate": 0.00019720685447352209, | |
| "loss": 0.0624, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.3128657136508407, | |
| "eval_loss": 0.06330914050340652, | |
| "eval_runtime": 37.0111, | |
| "eval_samples_per_second": 19.21, | |
| "eval_steps_per_second": 19.21, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.3223464928523813, | |
| "grad_norm": 0.0615234375, | |
| "learning_rate": 0.0001969599569751571, | |
| "loss": 0.0631, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.33182727205392193, | |
| "grad_norm": 0.056884765625, | |
| "learning_rate": 0.00019670277247913205, | |
| "loss": 0.0651, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.34130805125546254, | |
| "grad_norm": 0.061279296875, | |
| "learning_rate": 0.0001964353282715183, | |
| "loss": 0.0667, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.3507888304570032, | |
| "grad_norm": 0.060791015625, | |
| "learning_rate": 0.00019615765272689461, | |
| "loss": 0.064, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.3602696096585438, | |
| "grad_norm": 0.064453125, | |
| "learning_rate": 0.00019586977530533677, | |
| "loss": 0.0737, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.36975038886008443, | |
| "grad_norm": 0.059326171875, | |
| "learning_rate": 0.00019557172654929196, | |
| "loss": 0.0652, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.37923116806162505, | |
| "grad_norm": 0.05908203125, | |
| "learning_rate": 0.00019526353808033825, | |
| "loss": 0.0592, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3887119472631657, | |
| "grad_norm": 0.053955078125, | |
| "learning_rate": 0.00019494524259582992, | |
| "loss": 0.0599, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.3981927264647063, | |
| "grad_norm": 0.052978515625, | |
| "learning_rate": 0.00019461687386542826, | |
| "loss": 0.0574, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.40767350566624694, | |
| "grad_norm": 0.0537109375, | |
| "learning_rate": 0.00019427846672751873, | |
| "loss": 0.0528, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.41715428486778755, | |
| "grad_norm": 0.0576171875, | |
| "learning_rate": 0.00019393005708551498, | |
| "loss": 0.0638, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.41715428486778755, | |
| "eval_loss": 0.05790511518716812, | |
| "eval_runtime": 37.3066, | |
| "eval_samples_per_second": 19.058, | |
| "eval_steps_per_second": 19.058, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.4266350640693282, | |
| "grad_norm": 0.05224609375, | |
| "learning_rate": 0.00019357168190404936, | |
| "loss": 0.0585, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.43611584327086883, | |
| "grad_norm": 0.0625, | |
| "learning_rate": 0.00019320337920505153, | |
| "loss": 0.0707, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.44559662247240944, | |
| "grad_norm": 0.06005859375, | |
| "learning_rate": 0.00019282518806371414, | |
| "loss": 0.0647, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.45507740167395005, | |
| "grad_norm": 0.056884765625, | |
| "learning_rate": 0.0001924371486043473, | |
| "loss": 0.0651, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.4645581808754907, | |
| "grad_norm": 0.052978515625, | |
| "learning_rate": 0.0001920393019961217, | |
| "loss": 0.053, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.47403896007703133, | |
| "grad_norm": 0.057861328125, | |
| "learning_rate": 0.0001916316904487005, | |
| "loss": 0.0567, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.48351973927857195, | |
| "grad_norm": 0.053955078125, | |
| "learning_rate": 0.00019121435720776122, | |
| "loss": 0.054, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.49300051848011256, | |
| "grad_norm": 0.05615234375, | |
| "learning_rate": 0.0001907873465504076, | |
| "loss": 0.0536, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.5024812976816532, | |
| "grad_norm": 0.05615234375, | |
| "learning_rate": 0.00019035070378047204, | |
| "loss": 0.0491, | |
| "step": 53 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 315, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 53, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.911562387377357e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |