| { | |
| "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": 1.421875, | |
| "learning_rate": 2e-05, | |
| "loss": 0.4503, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.009480779201540626, | |
| "eval_loss": 0.4263582229614258, | |
| "eval_runtime": 33.6286, | |
| "eval_samples_per_second": 21.143, | |
| "eval_steps_per_second": 21.143, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.018961558403081252, | |
| "grad_norm": 1.359375, | |
| "learning_rate": 4e-05, | |
| "loss": 0.412, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.02844233760462188, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 6e-05, | |
| "loss": 0.4214, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.037923116806162505, | |
| "grad_norm": 0.80859375, | |
| "learning_rate": 8e-05, | |
| "loss": 0.2924, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.04740389600770313, | |
| "grad_norm": 0.51953125, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1896, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05688467520924376, | |
| "grad_norm": 0.4453125, | |
| "learning_rate": 0.00012, | |
| "loss": 0.1531, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.06636545441078438, | |
| "grad_norm": 0.27734375, | |
| "learning_rate": 0.00014, | |
| "loss": 0.1181, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.07584623361232501, | |
| "grad_norm": 0.263671875, | |
| "learning_rate": 0.00016, | |
| "loss": 0.1143, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.08532701281386564, | |
| "grad_norm": 0.2197265625, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0952, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.09480779201540626, | |
| "grad_norm": 0.1708984375, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0767, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10428857121694689, | |
| "grad_norm": 0.1484375, | |
| "learning_rate": 0.00019999469523400122, | |
| "loss": 0.0836, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.10428857121694689, | |
| "eval_loss": 0.07918477058410645, | |
| "eval_runtime": 33.4231, | |
| "eval_samples_per_second": 21.273, | |
| "eval_steps_per_second": 21.273, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.11376935041848751, | |
| "grad_norm": 0.154296875, | |
| "learning_rate": 0.00019997878149881574, | |
| "loss": 0.0757, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.12325012962002814, | |
| "grad_norm": 0.1513671875, | |
| "learning_rate": 0.0001999522604828164, | |
| "loss": 0.0767, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.13273090882156877, | |
| "grad_norm": 0.1416015625, | |
| "learning_rate": 0.00019991513499975882, | |
| "loss": 0.0809, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.1422116880231094, | |
| "grad_norm": 0.09765625, | |
| "learning_rate": 0.00019986740898848306, | |
| "loss": 0.0634, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.15169246722465002, | |
| "grad_norm": 0.099609375, | |
| "learning_rate": 0.00019980908751249555, | |
| "loss": 0.0674, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.16117324642619066, | |
| "grad_norm": 0.119140625, | |
| "learning_rate": 0.00019974017675943192, | |
| "loss": 0.0667, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.17065402562773127, | |
| "grad_norm": 0.09619140625, | |
| "learning_rate": 0.0001996606840404006, | |
| "loss": 0.0632, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1801348048292719, | |
| "grad_norm": 0.09130859375, | |
| "learning_rate": 0.00019957061778920701, | |
| "loss": 0.0488, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.18961558403081252, | |
| "grad_norm": 0.0947265625, | |
| "learning_rate": 0.0001994699875614589, | |
| "loss": 0.0627, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.19909636323235316, | |
| "grad_norm": 0.08203125, | |
| "learning_rate": 0.00019935880403355253, | |
| "loss": 0.0528, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.20857714243389378, | |
| "grad_norm": 0.1123046875, | |
| "learning_rate": 0.00019923707900153982, | |
| "loss": 0.0532, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.20857714243389378, | |
| "eval_loss": 0.056647635996341705, | |
| "eval_runtime": 33.3595, | |
| "eval_samples_per_second": 21.313, | |
| "eval_steps_per_second": 21.313, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.21805792163543442, | |
| "grad_norm": 0.10693359375, | |
| "learning_rate": 0.00019910482537987702, | |
| "loss": 0.0583, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.22753870083697503, | |
| "grad_norm": 0.0791015625, | |
| "learning_rate": 0.0001989620572000544, | |
| "loss": 0.0554, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.23701948003851567, | |
| "grad_norm": 0.123046875, | |
| "learning_rate": 0.00019880878960910772, | |
| "loss": 0.0688, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.24650025924005628, | |
| "grad_norm": 0.10498046875, | |
| "learning_rate": 0.00019864503886801106, | |
| "loss": 0.0655, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.2559810384415969, | |
| "grad_norm": 0.07421875, | |
| "learning_rate": 0.00019847082234995171, | |
| "loss": 0.0471, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.26546181764313753, | |
| "grad_norm": 0.080078125, | |
| "learning_rate": 0.00019828615853848688, | |
| "loss": 0.0518, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.27494259684467814, | |
| "grad_norm": 0.07421875, | |
| "learning_rate": 0.00019809106702558277, | |
| "loss": 0.0481, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.2844233760462188, | |
| "grad_norm": 0.0703125, | |
| "learning_rate": 0.0001978855685095358, | |
| "loss": 0.0464, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2939041552477594, | |
| "grad_norm": 0.0849609375, | |
| "learning_rate": 0.00019766968479277683, | |
| "loss": 0.0566, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.30338493444930004, | |
| "grad_norm": 0.0859375, | |
| "learning_rate": 0.00019744343877955788, | |
| "loss": 0.0517, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.3128657136508407, | |
| "grad_norm": 0.07177734375, | |
| "learning_rate": 0.00019720685447352209, | |
| "loss": 0.0511, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.3128657136508407, | |
| "eval_loss": 0.04964344948530197, | |
| "eval_runtime": 33.3741, | |
| "eval_samples_per_second": 21.304, | |
| "eval_steps_per_second": 21.304, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.3223464928523813, | |
| "grad_norm": 0.076171875, | |
| "learning_rate": 0.0001969599569751571, | |
| "loss": 0.045, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.33182727205392193, | |
| "grad_norm": 0.07568359375, | |
| "learning_rate": 0.00019670277247913205, | |
| "loss": 0.0543, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.34130805125546254, | |
| "grad_norm": 0.0791015625, | |
| "learning_rate": 0.0001964353282715183, | |
| "loss": 0.0444, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.3507888304570032, | |
| "grad_norm": 0.0771484375, | |
| "learning_rate": 0.00019615765272689461, | |
| "loss": 0.0506, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.3602696096585438, | |
| "grad_norm": 0.080078125, | |
| "learning_rate": 0.00019586977530533677, | |
| "loss": 0.0558, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.36975038886008443, | |
| "grad_norm": 0.078125, | |
| "learning_rate": 0.00019557172654929196, | |
| "loss": 0.0507, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.37923116806162505, | |
| "grad_norm": 0.06396484375, | |
| "learning_rate": 0.00019526353808033825, | |
| "loss": 0.0452, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3887119472631657, | |
| "grad_norm": 0.07275390625, | |
| "learning_rate": 0.00019494524259582992, | |
| "loss": 0.0481, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.3981927264647063, | |
| "grad_norm": 0.06591796875, | |
| "learning_rate": 0.00019461687386542826, | |
| "loss": 0.0464, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.40767350566624694, | |
| "grad_norm": 0.06591796875, | |
| "learning_rate": 0.00019427846672751873, | |
| "loss": 0.0431, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.41715428486778755, | |
| "grad_norm": 0.06982421875, | |
| "learning_rate": 0.00019393005708551498, | |
| "loss": 0.0511, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.41715428486778755, | |
| "eval_loss": 0.04574437811970711, | |
| "eval_runtime": 33.3514, | |
| "eval_samples_per_second": 21.318, | |
| "eval_steps_per_second": 21.318, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.4266350640693282, | |
| "grad_norm": 0.06787109375, | |
| "learning_rate": 0.00019357168190404936, | |
| "loss": 0.0443, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.43611584327086883, | |
| "grad_norm": 0.0751953125, | |
| "learning_rate": 0.00019320337920505153, | |
| "loss": 0.0545, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.44559662247240944, | |
| "grad_norm": 0.07666015625, | |
| "learning_rate": 0.00019282518806371414, | |
| "loss": 0.0542, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.45507740167395005, | |
| "grad_norm": 0.07666015625, | |
| "learning_rate": 0.0001924371486043473, | |
| "loss": 0.0589, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.4645581808754907, | |
| "grad_norm": 0.059326171875, | |
| "learning_rate": 0.0001920393019961217, | |
| "loss": 0.0444, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.47403896007703133, | |
| "grad_norm": 0.0732421875, | |
| "learning_rate": 0.0001916316904487005, | |
| "loss": 0.0485, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.48351973927857195, | |
| "grad_norm": 0.060546875, | |
| "learning_rate": 0.00019121435720776122, | |
| "loss": 0.0408, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.49300051848011256, | |
| "grad_norm": 0.068359375, | |
| "learning_rate": 0.0001907873465504076, | |
| "loss": 0.0466, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.5024812976816532, | |
| "grad_norm": 0.07763671875, | |
| "learning_rate": 0.00019035070378047204, | |
| "loss": 0.0426, | |
| "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": 9.521933077020672e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |