| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5524861878453039, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0055248618784530384, | |
| "eval_loss": 1.9245201349258423, | |
| "eval_runtime": 5.5755, | |
| "eval_samples_per_second": 54.703, | |
| "eval_steps_per_second": 6.995, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.016574585635359115, | |
| "grad_norm": 1.1449600458145142, | |
| "learning_rate": 3e-05, | |
| "loss": 1.9242, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.03314917127071823, | |
| "grad_norm": 1.1835942268371582, | |
| "learning_rate": 6e-05, | |
| "loss": 1.8927, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.049723756906077346, | |
| "grad_norm": 1.033794641494751, | |
| "learning_rate": 9e-05, | |
| "loss": 1.8456, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.049723756906077346, | |
| "eval_loss": 1.7091096639633179, | |
| "eval_runtime": 5.5811, | |
| "eval_samples_per_second": 54.649, | |
| "eval_steps_per_second": 6.988, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.06629834254143646, | |
| "grad_norm": 0.9931322336196899, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.5976, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.08287292817679558, | |
| "grad_norm": 0.8187804222106934, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.481, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.09944751381215469, | |
| "grad_norm": 0.7247737646102905, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 1.2972, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.09944751381215469, | |
| "eval_loss": 1.2055176496505737, | |
| "eval_runtime": 5.5902, | |
| "eval_samples_per_second": 54.56, | |
| "eval_steps_per_second": 6.976, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.11602209944751381, | |
| "grad_norm": 0.6320592761039734, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 1.2298, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.13259668508287292, | |
| "grad_norm": 0.7799755930900574, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.0673, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.14917127071823205, | |
| "grad_norm": 0.7171224355697632, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.9548, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.14917127071823205, | |
| "eval_loss": 0.962252140045166, | |
| "eval_runtime": 5.6052, | |
| "eval_samples_per_second": 54.414, | |
| "eval_steps_per_second": 6.958, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.16574585635359115, | |
| "grad_norm": 0.5219327211380005, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.9444, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.18232044198895028, | |
| "grad_norm": 0.45712268352508545, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.9104, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.19889502762430938, | |
| "grad_norm": 0.452898234128952, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.8865, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.19889502762430938, | |
| "eval_loss": 0.8845100998878479, | |
| "eval_runtime": 5.6201, | |
| "eval_samples_per_second": 54.269, | |
| "eval_steps_per_second": 6.939, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.2154696132596685, | |
| "grad_norm": 0.4160885512828827, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.9234, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.23204419889502761, | |
| "grad_norm": 0.38359537720680237, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.8605, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.24861878453038674, | |
| "grad_norm": 0.42660069465637207, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.8852, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.24861878453038674, | |
| "eval_loss": 0.8424320220947266, | |
| "eval_runtime": 5.6306, | |
| "eval_samples_per_second": 54.168, | |
| "eval_steps_per_second": 6.926, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.26519337016574585, | |
| "grad_norm": 0.41090133786201477, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.8215, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.281767955801105, | |
| "grad_norm": 0.3953685164451599, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.8451, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.2983425414364641, | |
| "grad_norm": 0.38556817173957825, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.8105, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.2983425414364641, | |
| "eval_loss": 0.8192790150642395, | |
| "eval_runtime": 5.6342, | |
| "eval_samples_per_second": 54.134, | |
| "eval_steps_per_second": 6.922, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.3149171270718232, | |
| "grad_norm": 0.3790028691291809, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.8161, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.3314917127071823, | |
| "grad_norm": 0.4070817530155182, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.8064, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.34806629834254144, | |
| "grad_norm": 0.3971315920352936, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.8088, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.34806629834254144, | |
| "eval_loss": 0.8077210187911987, | |
| "eval_runtime": 5.6475, | |
| "eval_samples_per_second": 54.006, | |
| "eval_steps_per_second": 6.906, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.36464088397790057, | |
| "grad_norm": 0.3867045044898987, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.7798, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.3812154696132597, | |
| "grad_norm": 0.40124425292015076, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.7491, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.39779005524861877, | |
| "grad_norm": 0.38526755571365356, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.8083, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.39779005524861877, | |
| "eval_loss": 0.7986339926719666, | |
| "eval_runtime": 5.639, | |
| "eval_samples_per_second": 54.088, | |
| "eval_steps_per_second": 6.916, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.4143646408839779, | |
| "grad_norm": 0.3931930959224701, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.7764, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.430939226519337, | |
| "grad_norm": 0.4095585346221924, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.8311, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.44751381215469616, | |
| "grad_norm": 0.42162761092185974, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.7614, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.44751381215469616, | |
| "eval_loss": 0.7943903803825378, | |
| "eval_runtime": 5.6404, | |
| "eval_samples_per_second": 54.074, | |
| "eval_steps_per_second": 6.914, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.46408839779005523, | |
| "grad_norm": 0.4192807376384735, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.8023, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.48066298342541436, | |
| "grad_norm": 0.42631134390830994, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.752, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.4972375690607735, | |
| "grad_norm": 0.39116641879081726, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.7912, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4972375690607735, | |
| "eval_loss": 0.7925387620925903, | |
| "eval_runtime": 5.6625, | |
| "eval_samples_per_second": 53.863, | |
| "eval_steps_per_second": 6.887, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5138121546961326, | |
| "grad_norm": 0.377729207277298, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.7606, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.5303867403314917, | |
| "grad_norm": 0.3727019131183624, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.7929, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.5469613259668509, | |
| "grad_norm": 0.42312800884246826, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.7776, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.5469613259668509, | |
| "eval_loss": 0.7921723127365112, | |
| "eval_runtime": 5.6531, | |
| "eval_samples_per_second": 53.953, | |
| "eval_steps_per_second": 6.899, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.04633582600192e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |