|
{ |
|
"best_metric": 1.9845335483551025, |
|
"best_model_checkpoint": "temp_debug/axolotl_outputs/model/checkpoint-35", |
|
"epoch": 8.88888888888889, |
|
"eval_steps": 5, |
|
"global_step": 50, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.18, |
|
"grad_norm": 0.09891793876886368, |
|
"learning_rate": 2e-05, |
|
"loss": 2.1801, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"eval_loss": 2.0421414375305176, |
|
"eval_runtime": 5.3893, |
|
"eval_samples_per_second": 14.844, |
|
"eval_steps_per_second": 1.856, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.36, |
|
"grad_norm": 0.09653851389884949, |
|
"learning_rate": 4e-05, |
|
"loss": 2.1696, |
|
"step": 2 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"grad_norm": 0.09466348588466644, |
|
"learning_rate": 6e-05, |
|
"loss": 2.1678, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"grad_norm": 0.0848536491394043, |
|
"learning_rate": 8e-05, |
|
"loss": 2.1566, |
|
"step": 4 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"grad_norm": 0.07991650700569153, |
|
"learning_rate": 0.0001, |
|
"loss": 2.1988, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"eval_loss": 2.0347237586975098, |
|
"eval_runtime": 5.3587, |
|
"eval_samples_per_second": 14.929, |
|
"eval_steps_per_second": 1.866, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 1.07, |
|
"grad_norm": 0.06940381973981857, |
|
"learning_rate": 0.00012, |
|
"loss": 2.1301, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 1.24, |
|
"grad_norm": 0.0641147643327713, |
|
"learning_rate": 0.00014, |
|
"loss": 2.1693, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 1.42, |
|
"grad_norm": 0.06445202231407166, |
|
"learning_rate": 0.00016, |
|
"loss": 2.1262, |
|
"step": 8 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"grad_norm": 0.07305111736059189, |
|
"learning_rate": 0.00018, |
|
"loss": 2.1775, |
|
"step": 9 |
|
}, |
|
{ |
|
"epoch": 1.78, |
|
"grad_norm": 0.06370216608047485, |
|
"learning_rate": 0.0002, |
|
"loss": 2.1806, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 1.78, |
|
"eval_loss": 2.0147595405578613, |
|
"eval_runtime": 5.3657, |
|
"eval_samples_per_second": 14.91, |
|
"eval_steps_per_second": 1.864, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 1.96, |
|
"grad_norm": 0.0525302030146122, |
|
"learning_rate": 0.0001996917333733128, |
|
"loss": 2.036, |
|
"step": 11 |
|
}, |
|
{ |
|
"epoch": 2.13, |
|
"grad_norm": 0.06005712226033211, |
|
"learning_rate": 0.00019876883405951377, |
|
"loss": 2.063, |
|
"step": 12 |
|
}, |
|
{ |
|
"epoch": 2.31, |
|
"grad_norm": 0.06996417045593262, |
|
"learning_rate": 0.00019723699203976766, |
|
"loss": 2.0135, |
|
"step": 13 |
|
}, |
|
{ |
|
"epoch": 2.49, |
|
"grad_norm": 0.07619830965995789, |
|
"learning_rate": 0.00019510565162951537, |
|
"loss": 2.0959, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 2.67, |
|
"grad_norm": 0.07416332513093948, |
|
"learning_rate": 0.0001923879532511287, |
|
"loss": 2.1273, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 2.67, |
|
"eval_loss": 2.0056827068328857, |
|
"eval_runtime": 5.442, |
|
"eval_samples_per_second": 14.7, |
|
"eval_steps_per_second": 1.838, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 2.84, |
|
"grad_norm": 0.06484115868806839, |
|
"learning_rate": 0.0001891006524188368, |
|
"loss": 2.1218, |
|
"step": 16 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"grad_norm": 0.06188470497727394, |
|
"learning_rate": 0.00018526401643540922, |
|
"loss": 1.992, |
|
"step": 17 |
|
}, |
|
{ |
|
"epoch": 3.2, |
|
"grad_norm": 0.06161986663937569, |
|
"learning_rate": 0.00018090169943749476, |
|
"loss": 2.1295, |
|
"step": 18 |
|
}, |
|
{ |
|
"epoch": 3.38, |
|
"grad_norm": 0.06910485774278641, |
|
"learning_rate": 0.0001760405965600031, |
|
"loss": 1.9729, |
|
"step": 19 |
|
}, |
|
{ |
|
"epoch": 3.56, |
|
"grad_norm": 0.054629288613796234, |
|
"learning_rate": 0.00017071067811865476, |
|
"loss": 2.0685, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 3.56, |
|
"eval_loss": 1.9929816722869873, |
|
"eval_runtime": 5.361, |
|
"eval_samples_per_second": 14.923, |
|
"eval_steps_per_second": 1.865, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 3.73, |
|
"grad_norm": 0.05759178102016449, |
|
"learning_rate": 0.00016494480483301836, |
|
"loss": 2.0524, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 3.91, |
|
"grad_norm": 0.06020751968026161, |
|
"learning_rate": 0.00015877852522924732, |
|
"loss": 2.0323, |
|
"step": 22 |
|
}, |
|
{ |
|
"epoch": 4.09, |
|
"grad_norm": 0.06324882060289383, |
|
"learning_rate": 0.0001522498564715949, |
|
"loss": 2.017, |
|
"step": 23 |
|
}, |
|
{ |
|
"epoch": 4.27, |
|
"grad_norm": 0.06709667295217514, |
|
"learning_rate": 0.00014539904997395468, |
|
"loss": 2.0264, |
|
"step": 24 |
|
}, |
|
{ |
|
"epoch": 4.44, |
|
"grad_norm": 0.057694002985954285, |
|
"learning_rate": 0.000138268343236509, |
|
"loss": 1.9596, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 4.44, |
|
"eval_loss": 1.986427664756775, |
|
"eval_runtime": 5.3574, |
|
"eval_samples_per_second": 14.933, |
|
"eval_steps_per_second": 1.867, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 4.62, |
|
"grad_norm": 0.06380322575569153, |
|
"learning_rate": 0.00013090169943749476, |
|
"loss": 2.0995, |
|
"step": 26 |
|
}, |
|
{ |
|
"epoch": 4.8, |
|
"grad_norm": 0.059418462216854095, |
|
"learning_rate": 0.00012334453638559057, |
|
"loss": 2.0501, |
|
"step": 27 |
|
}, |
|
{ |
|
"epoch": 4.98, |
|
"grad_norm": 0.06417760252952576, |
|
"learning_rate": 0.0001156434465040231, |
|
"loss": 1.9904, |
|
"step": 28 |
|
}, |
|
{ |
|
"epoch": 5.16, |
|
"grad_norm": 0.06266569346189499, |
|
"learning_rate": 0.0001078459095727845, |
|
"loss": 2.0275, |
|
"step": 29 |
|
}, |
|
{ |
|
"epoch": 5.33, |
|
"grad_norm": 0.06987171620130539, |
|
"learning_rate": 0.0001, |
|
"loss": 2.043, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 5.33, |
|
"eval_loss": 1.9850692749023438, |
|
"eval_runtime": 5.5829, |
|
"eval_samples_per_second": 14.329, |
|
"eval_steps_per_second": 1.791, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 5.51, |
|
"grad_norm": 0.06864842027425766, |
|
"learning_rate": 9.215409042721552e-05, |
|
"loss": 1.9196, |
|
"step": 31 |
|
}, |
|
{ |
|
"epoch": 5.69, |
|
"grad_norm": 0.06972988694906235, |
|
"learning_rate": 8.435655349597689e-05, |
|
"loss": 1.9468, |
|
"step": 32 |
|
}, |
|
{ |
|
"epoch": 5.87, |
|
"grad_norm": 0.0746963694691658, |
|
"learning_rate": 7.66554636144095e-05, |
|
"loss": 1.9266, |
|
"step": 33 |
|
}, |
|
{ |
|
"epoch": 6.04, |
|
"grad_norm": 0.07660232484340668, |
|
"learning_rate": 6.909830056250527e-05, |
|
"loss": 1.9507, |
|
"step": 34 |
|
}, |
|
{ |
|
"epoch": 6.22, |
|
"grad_norm": 0.06880834698677063, |
|
"learning_rate": 6.173165676349103e-05, |
|
"loss": 1.9464, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 6.22, |
|
"eval_loss": 1.9845335483551025, |
|
"eval_runtime": 5.4837, |
|
"eval_samples_per_second": 14.589, |
|
"eval_steps_per_second": 1.824, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 6.4, |
|
"grad_norm": 0.07336819916963577, |
|
"learning_rate": 5.4600950026045326e-05, |
|
"loss": 1.9353, |
|
"step": 36 |
|
}, |
|
{ |
|
"epoch": 6.58, |
|
"grad_norm": 0.07693742960691452, |
|
"learning_rate": 4.7750143528405126e-05, |
|
"loss": 1.9385, |
|
"step": 37 |
|
}, |
|
{ |
|
"epoch": 6.76, |
|
"grad_norm": 0.07842959463596344, |
|
"learning_rate": 4.12214747707527e-05, |
|
"loss": 1.9153, |
|
"step": 38 |
|
}, |
|
{ |
|
"epoch": 6.93, |
|
"grad_norm": 0.08660242706537247, |
|
"learning_rate": 3.5055195166981645e-05, |
|
"loss": 1.934, |
|
"step": 39 |
|
}, |
|
{ |
|
"epoch": 7.11, |
|
"grad_norm": 0.0903269574046135, |
|
"learning_rate": 2.9289321881345254e-05, |
|
"loss": 2.0036, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 7.11, |
|
"eval_loss": 1.985894799232483, |
|
"eval_runtime": 5.3759, |
|
"eval_samples_per_second": 14.881, |
|
"eval_steps_per_second": 1.86, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 7.29, |
|
"grad_norm": 0.08571472018957138, |
|
"learning_rate": 2.3959403439996907e-05, |
|
"loss": 1.9909, |
|
"step": 41 |
|
}, |
|
{ |
|
"epoch": 7.47, |
|
"grad_norm": 0.07964807748794556, |
|
"learning_rate": 1.9098300562505266e-05, |
|
"loss": 1.8645, |
|
"step": 42 |
|
}, |
|
{ |
|
"epoch": 7.64, |
|
"grad_norm": 0.07694341242313385, |
|
"learning_rate": 1.4735983564590783e-05, |
|
"loss": 1.8933, |
|
"step": 43 |
|
}, |
|
{ |
|
"epoch": 7.82, |
|
"grad_norm": 0.08349137008190155, |
|
"learning_rate": 1.0899347581163221e-05, |
|
"loss": 2.0064, |
|
"step": 44 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"grad_norm": 0.08065728098154068, |
|
"learning_rate": 7.612046748871327e-06, |
|
"loss": 1.9168, |
|
"step": 45 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_loss": 1.986413598060608, |
|
"eval_runtime": 5.5847, |
|
"eval_samples_per_second": 14.325, |
|
"eval_steps_per_second": 1.791, |
|
"step": 45 |
|
}, |
|
{ |
|
"epoch": 8.18, |
|
"grad_norm": 0.08433607220649719, |
|
"learning_rate": 4.8943483704846475e-06, |
|
"loss": 1.9723, |
|
"step": 46 |
|
}, |
|
{ |
|
"epoch": 8.36, |
|
"grad_norm": 0.08004725724458694, |
|
"learning_rate": 2.7630079602323442e-06, |
|
"loss": 1.8944, |
|
"step": 47 |
|
}, |
|
{ |
|
"epoch": 8.53, |
|
"grad_norm": 0.07930443435907364, |
|
"learning_rate": 1.231165940486234e-06, |
|
"loss": 1.925, |
|
"step": 48 |
|
}, |
|
{ |
|
"epoch": 8.71, |
|
"grad_norm": 0.08285905420780182, |
|
"learning_rate": 3.0826662668720364e-07, |
|
"loss": 1.9438, |
|
"step": 49 |
|
}, |
|
{ |
|
"epoch": 8.89, |
|
"grad_norm": 0.08028247207403183, |
|
"learning_rate": 0.0, |
|
"loss": 1.9258, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 8.89, |
|
"eval_loss": 1.9865543842315674, |
|
"eval_runtime": 5.3735, |
|
"eval_samples_per_second": 14.888, |
|
"eval_steps_per_second": 1.861, |
|
"step": 50 |
|
} |
|
], |
|
"logging_steps": 1, |
|
"max_steps": 50, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 10, |
|
"save_steps": 5, |
|
"total_flos": 9.328263092502528e+17, |
|
"train_batch_size": 2, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|