| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.05812263876780006, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0011624527753560012, | |
| "grad_norm": 0.9013734459877014, | |
| "learning_rate": 5e-06, | |
| "loss": 3.2382, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0011624527753560012, | |
| "eval_loss": 0.6915596127510071, | |
| "eval_runtime": 18.725, | |
| "eval_samples_per_second": 19.386, | |
| "eval_steps_per_second": 9.72, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0023249055507120024, | |
| "grad_norm": 0.7998698949813843, | |
| "learning_rate": 1e-05, | |
| "loss": 2.4713, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0034873583260680036, | |
| "grad_norm": 0.691608190536499, | |
| "learning_rate": 1.5e-05, | |
| "loss": 2.6607, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004649811101424005, | |
| "grad_norm": 0.9019598960876465, | |
| "learning_rate": 2e-05, | |
| "loss": 3.0399, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.005812263876780006, | |
| "grad_norm": 0.6667659878730774, | |
| "learning_rate": 2.5e-05, | |
| "loss": 2.9629, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.006974716652136007, | |
| "grad_norm": 0.7605751156806946, | |
| "learning_rate": 3e-05, | |
| "loss": 2.5889, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.008137169427492008, | |
| "grad_norm": 0.8346095085144043, | |
| "learning_rate": 3.5e-05, | |
| "loss": 2.8887, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00929962220284801, | |
| "grad_norm": 0.6577608585357666, | |
| "learning_rate": 4e-05, | |
| "loss": 2.8019, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.010462074978204011, | |
| "grad_norm": 0.8766710758209229, | |
| "learning_rate": 4.5e-05, | |
| "loss": 3.0156, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.011624527753560012, | |
| "grad_norm": 0.8462783098220825, | |
| "learning_rate": 5e-05, | |
| "loss": 2.9731, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.012786980528916013, | |
| "grad_norm": 0.8758143782615662, | |
| "learning_rate": 4.99229333433282e-05, | |
| "loss": 2.7932, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.013949433304272014, | |
| "grad_norm": 0.9164470434188843, | |
| "learning_rate": 4.9692208514878444e-05, | |
| "loss": 2.5973, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.015111886079628016, | |
| "grad_norm": 0.9943875670433044, | |
| "learning_rate": 4.9309248009941914e-05, | |
| "loss": 2.8627, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.015111886079628016, | |
| "eval_loss": 0.6720152497291565, | |
| "eval_runtime": 18.6397, | |
| "eval_samples_per_second": 19.475, | |
| "eval_steps_per_second": 9.764, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.016274338854984017, | |
| "grad_norm": 1.0360639095306396, | |
| "learning_rate": 4.877641290737884e-05, | |
| "loss": 2.784, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.017436791630340016, | |
| "grad_norm": 0.9277502298355103, | |
| "learning_rate": 4.8096988312782174e-05, | |
| "loss": 2.1715, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.01859924440569602, | |
| "grad_norm": 1.2977659702301025, | |
| "learning_rate": 4.72751631047092e-05, | |
| "loss": 2.3805, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.01976169718105202, | |
| "grad_norm": 1.1046712398529053, | |
| "learning_rate": 4.6316004108852305e-05, | |
| "loss": 2.6283, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.020924149956408022, | |
| "grad_norm": 1.2673217058181763, | |
| "learning_rate": 4.522542485937369e-05, | |
| "loss": 3.021, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02208660273176402, | |
| "grad_norm": 0.9734105467796326, | |
| "learning_rate": 4.401014914000078e-05, | |
| "loss": 2.597, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.023249055507120024, | |
| "grad_norm": 1.0393081903457642, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 2.5482, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.024411508282476024, | |
| "grad_norm": 1.247772455215454, | |
| "learning_rate": 4.123620120825459e-05, | |
| "loss": 2.6122, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.025573961057832027, | |
| "grad_norm": 1.2255369424819946, | |
| "learning_rate": 3.969463130731183e-05, | |
| "loss": 2.8942, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.026736413833188026, | |
| "grad_norm": 1.286831259727478, | |
| "learning_rate": 3.8062464117898724e-05, | |
| "loss": 2.2942, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.02789886660854403, | |
| "grad_norm": 1.4110021591186523, | |
| "learning_rate": 3.634976249348867e-05, | |
| "loss": 2.113, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.02906131938390003, | |
| "grad_norm": 1.2104796171188354, | |
| "learning_rate": 3.456708580912725e-05, | |
| "loss": 2.1373, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03022377215925603, | |
| "grad_norm": 1.2794564962387085, | |
| "learning_rate": 3.272542485937369e-05, | |
| "loss": 2.465, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.03022377215925603, | |
| "eval_loss": 0.6091086268424988, | |
| "eval_runtime": 18.6804, | |
| "eval_samples_per_second": 19.432, | |
| "eval_steps_per_second": 9.743, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.03138622493461203, | |
| "grad_norm": 1.445319414138794, | |
| "learning_rate": 3.083613409639764e-05, | |
| "loss": 2.0072, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.032548677709968034, | |
| "grad_norm": 1.4100600481033325, | |
| "learning_rate": 2.8910861626005776e-05, | |
| "loss": 2.9271, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.03371113048532404, | |
| "grad_norm": 1.2174726724624634, | |
| "learning_rate": 2.6961477393196126e-05, | |
| "loss": 2.325, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.03487358326068003, | |
| "grad_norm": 1.3934134244918823, | |
| "learning_rate": 2.5e-05, | |
| "loss": 2.4305, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.036036036036036036, | |
| "grad_norm": 1.353877067565918, | |
| "learning_rate": 2.303852260680388e-05, | |
| "loss": 1.9587, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.03719848881139204, | |
| "grad_norm": 1.3453905582427979, | |
| "learning_rate": 2.1089138373994223e-05, | |
| "loss": 2.4014, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.03836094158674804, | |
| "grad_norm": 1.4014428853988647, | |
| "learning_rate": 1.9163865903602374e-05, | |
| "loss": 2.4628, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.03952339436210404, | |
| "grad_norm": 1.7237356901168823, | |
| "learning_rate": 1.7274575140626318e-05, | |
| "loss": 2.5534, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.04068584713746004, | |
| "grad_norm": 1.6201252937316895, | |
| "learning_rate": 1.5432914190872757e-05, | |
| "loss": 2.6219, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.041848299912816043, | |
| "grad_norm": 1.7005544900894165, | |
| "learning_rate": 1.3650237506511331e-05, | |
| "loss": 2.2468, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.043010752688172046, | |
| "grad_norm": 1.5233935117721558, | |
| "learning_rate": 1.1937535882101281e-05, | |
| "loss": 2.4295, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.04417320546352804, | |
| "grad_norm": 1.4735093116760254, | |
| "learning_rate": 1.0305368692688174e-05, | |
| "loss": 2.3018, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.045335658238884045, | |
| "grad_norm": 1.803377389907837, | |
| "learning_rate": 8.763798791745411e-06, | |
| "loss": 2.6965, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.045335658238884045, | |
| "eval_loss": 0.5883385539054871, | |
| "eval_runtime": 18.5863, | |
| "eval_samples_per_second": 19.531, | |
| "eval_steps_per_second": 9.792, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.04649811101424005, | |
| "grad_norm": 1.3496054410934448, | |
| "learning_rate": 7.3223304703363135e-06, | |
| "loss": 2.3177, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.047660563789596044, | |
| "grad_norm": 1.4967939853668213, | |
| "learning_rate": 5.989850859999227e-06, | |
| "loss": 2.5048, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.04882301656495205, | |
| "grad_norm": 1.4733471870422363, | |
| "learning_rate": 4.7745751406263165e-06, | |
| "loss": 2.2494, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.04998546934030805, | |
| "grad_norm": 1.2914958000183105, | |
| "learning_rate": 3.6839958911476957e-06, | |
| "loss": 2.0449, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.05114792211566405, | |
| "grad_norm": 1.4443607330322266, | |
| "learning_rate": 2.7248368952908053e-06, | |
| "loss": 2.2681, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.05231037489102005, | |
| "grad_norm": 1.3686310052871704, | |
| "learning_rate": 1.9030116872178316e-06, | |
| "loss": 2.5226, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05347282766637605, | |
| "grad_norm": 1.6216455698013306, | |
| "learning_rate": 1.2235870926211619e-06, | |
| "loss": 2.3361, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.054635280441732055, | |
| "grad_norm": 2.076577663421631, | |
| "learning_rate": 6.907519900580861e-07, | |
| "loss": 2.9312, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.05579773321708806, | |
| "grad_norm": 1.4256387948989868, | |
| "learning_rate": 3.077914851215585e-07, | |
| "loss": 2.2066, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.056960185992444054, | |
| "grad_norm": 1.4510419368743896, | |
| "learning_rate": 7.706665667180091e-08, | |
| "loss": 2.1987, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.05812263876780006, | |
| "grad_norm": 1.2870044708251953, | |
| "learning_rate": 0.0, | |
| "loss": 2.4237, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3803513326862336e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |