| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03537944454272068, | |
| "eval_steps": 5, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0014151777817088271, | |
| "grad_norm": 0.20514807105064392, | |
| "learning_rate": 2e-05, | |
| "loss": 11.9175, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0014151777817088271, | |
| "eval_loss": 11.916790962219238, | |
| "eval_runtime": 8.5875, | |
| "eval_samples_per_second": 69.403, | |
| "eval_steps_per_second": 34.702, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0028303555634176542, | |
| "grad_norm": 0.23273983597755432, | |
| "learning_rate": 4e-05, | |
| "loss": 11.9373, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.004245533345126481, | |
| "grad_norm": 0.2802961468696594, | |
| "learning_rate": 6e-05, | |
| "loss": 11.9104, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0056607111268353084, | |
| "grad_norm": 0.23625600337982178, | |
| "learning_rate": 8e-05, | |
| "loss": 11.9218, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0070758889085441356, | |
| "grad_norm": 0.22699324786663055, | |
| "learning_rate": 0.0001, | |
| "loss": 11.915, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0070758889085441356, | |
| "eval_loss": 11.916035652160645, | |
| "eval_runtime": 8.7182, | |
| "eval_samples_per_second": 68.362, | |
| "eval_steps_per_second": 34.181, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.008491066690252963, | |
| "grad_norm": 0.21260926127433777, | |
| "learning_rate": 0.00012, | |
| "loss": 11.9222, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00990624447196179, | |
| "grad_norm": 0.2168704867362976, | |
| "learning_rate": 0.00014, | |
| "loss": 11.9087, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.011321422253670617, | |
| "grad_norm": 0.2431064248085022, | |
| "learning_rate": 0.00016, | |
| "loss": 11.9209, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.012736600035379445, | |
| "grad_norm": 0.19750989973545074, | |
| "learning_rate": 0.00018, | |
| "loss": 11.9141, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.014151777817088271, | |
| "grad_norm": 0.21232645213603973, | |
| "learning_rate": 0.0002, | |
| "loss": 11.9101, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.014151777817088271, | |
| "eval_loss": 11.913423538208008, | |
| "eval_runtime": 8.7577, | |
| "eval_samples_per_second": 68.054, | |
| "eval_steps_per_second": 34.027, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.015566955598797099, | |
| "grad_norm": 0.2188003957271576, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 11.9215, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.016982133380505925, | |
| "grad_norm": 0.19326922297477722, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 11.9316, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.01839731116221475, | |
| "grad_norm": 0.22555257380008698, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 11.9136, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.01981248894392358, | |
| "grad_norm": 0.2110656350851059, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 11.8945, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.021227666725632408, | |
| "grad_norm": 0.20976051688194275, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 11.9, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.021227666725632408, | |
| "eval_loss": 11.909821510314941, | |
| "eval_runtime": 8.6935, | |
| "eval_samples_per_second": 68.557, | |
| "eval_steps_per_second": 34.279, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.022642844507341234, | |
| "grad_norm": 0.24981027841567993, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 11.8935, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.024058022289050063, | |
| "grad_norm": 0.2591237723827362, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 11.905, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.02547320007075889, | |
| "grad_norm": 0.22017155587673187, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 11.9214, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.026888377852467716, | |
| "grad_norm": 0.25297868251800537, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 11.9002, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.028303555634176542, | |
| "grad_norm": 0.25911155343055725, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 11.9091, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.028303555634176542, | |
| "eval_loss": 11.90591049194336, | |
| "eval_runtime": 8.6221, | |
| "eval_samples_per_second": 69.125, | |
| "eval_steps_per_second": 34.562, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.029718733415885372, | |
| "grad_norm": 0.24711765348911285, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 11.9044, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.031133911197594198, | |
| "grad_norm": 0.2823926508426666, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 11.8933, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.032549088979303024, | |
| "grad_norm": 0.2736799418926239, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 11.9053, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.03396426676101185, | |
| "grad_norm": 0.31680405139923096, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 11.9123, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03537944454272068, | |
| "grad_norm": 0.31065115332603455, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 11.9045, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03537944454272068, | |
| "eval_loss": 11.901829719543457, | |
| "eval_runtime": 8.5918, | |
| "eval_samples_per_second": 69.369, | |
| "eval_steps_per_second": 34.684, | |
| "step": 25 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 37473484800.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |