| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.946273830155979, | |
| "eval_steps": 25, | |
| "global_step": 425, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1733102253032929, | |
| "grad_norm": 1.3994637727737427, | |
| "learning_rate": 2.43993993993994e-05, | |
| "loss": 1.2043, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.1733102253032929, | |
| "eval_loss": 1.1417787075042725, | |
| "eval_runtime": 158.1666, | |
| "eval_samples_per_second": 1.081, | |
| "eval_steps_per_second": 0.139, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.3466204506065858, | |
| "grad_norm": 1.4478472471237183, | |
| "learning_rate": 2.3773773773773775e-05, | |
| "loss": 1.0312, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3466204506065858, | |
| "eval_loss": 0.9970389008522034, | |
| "eval_runtime": 158.0459, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5199306759098787, | |
| "grad_norm": 2.2421517372131348, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.8606, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.5199306759098787, | |
| "eval_loss": 0.8508256077766418, | |
| "eval_runtime": 158.0428, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.6932409012131716, | |
| "grad_norm": 2.5336129665374756, | |
| "learning_rate": 2.2522522522522523e-05, | |
| "loss": 0.7317, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6932409012131716, | |
| "eval_loss": 0.7086797952651978, | |
| "eval_runtime": 158.0434, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8665511265164645, | |
| "grad_norm": 2.8436946868896484, | |
| "learning_rate": 2.18968968968969e-05, | |
| "loss": 0.5988, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.8665511265164645, | |
| "eval_loss": 0.5647777318954468, | |
| "eval_runtime": 158.0443, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.0398613518197575, | |
| "grad_norm": 2.979930877685547, | |
| "learning_rate": 2.1271271271271275e-05, | |
| "loss": 0.4824, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0398613518197575, | |
| "eval_loss": 0.48206475377082825, | |
| "eval_runtime": 158.0456, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2131715771230502, | |
| "grad_norm": 3.3822202682495117, | |
| "learning_rate": 2.0645645645645647e-05, | |
| "loss": 0.3579, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.2131715771230502, | |
| "eval_loss": 0.4603586792945862, | |
| "eval_runtime": 158.0395, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.3864818024263432, | |
| "grad_norm": 2.0061120986938477, | |
| "learning_rate": 2.0020020020020023e-05, | |
| "loss": 0.319, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.3864818024263432, | |
| "eval_loss": 0.4302046597003937, | |
| "eval_runtime": 158.0413, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.5597920277296362, | |
| "grad_norm": 3.4909021854400635, | |
| "learning_rate": 1.9394394394394395e-05, | |
| "loss": 0.2902, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.5597920277296362, | |
| "eval_loss": 0.42577695846557617, | |
| "eval_runtime": 158.0406, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.733102253032929, | |
| "grad_norm": 3.723992347717285, | |
| "learning_rate": 1.8768768768768768e-05, | |
| "loss": 0.2714, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.733102253032929, | |
| "eval_loss": 0.412419855594635, | |
| "eval_runtime": 158.0404, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.9064124783362217, | |
| "grad_norm": 1.7258232831954956, | |
| "learning_rate": 1.8143143143143144e-05, | |
| "loss": 0.2338, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.9064124783362217, | |
| "eval_loss": 0.4015367329120636, | |
| "eval_runtime": 158.032, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.079722703639515, | |
| "grad_norm": 2.5275797843933105, | |
| "learning_rate": 1.7517517517517516e-05, | |
| "loss": 0.1714, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.079722703639515, | |
| "eval_loss": 0.4162571430206299, | |
| "eval_runtime": 158.0293, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.2530329289428077, | |
| "grad_norm": 3.2901697158813477, | |
| "learning_rate": 1.6891891891891892e-05, | |
| "loss": 0.1779, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.2530329289428077, | |
| "eval_loss": 0.41930997371673584, | |
| "eval_runtime": 158.0285, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.4263431542461005, | |
| "grad_norm": 1.777195930480957, | |
| "learning_rate": 1.6266266266266268e-05, | |
| "loss": 0.1357, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.4263431542461005, | |
| "eval_loss": 0.4277823865413666, | |
| "eval_runtime": 158.0295, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.5996533795493937, | |
| "grad_norm": 2.714945077896118, | |
| "learning_rate": 1.564064064064064e-05, | |
| "loss": 0.1451, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.5996533795493937, | |
| "eval_loss": 0.4059053063392639, | |
| "eval_runtime": 158.0297, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.7729636048526864, | |
| "grad_norm": 3.0253992080688477, | |
| "learning_rate": 1.5015015015015016e-05, | |
| "loss": 0.1273, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.7729636048526864, | |
| "eval_loss": 0.4206814169883728, | |
| "eval_runtime": 158.0304, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.946273830155979, | |
| "grad_norm": 4.61653995513916, | |
| "learning_rate": 1.438938938938939e-05, | |
| "loss": 0.1339, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.946273830155979, | |
| "eval_loss": 0.3988127112388611, | |
| "eval_runtime": 158.0299, | |
| "eval_samples_per_second": 1.082, | |
| "eval_steps_per_second": 0.139, | |
| "step": 425 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 25, | |
| "total_flos": 2.494001269751808e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |