| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00177210703526493, | |
| "eval_steps": 3, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 7.08842814105972e-05, | |
| "grad_norm": 0.07005972415208817, | |
| "learning_rate": 2e-05, | |
| "loss": 10.3888, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 7.08842814105972e-05, | |
| "eval_loss": 10.386326789855957, | |
| "eval_runtime": 21.6585, | |
| "eval_samples_per_second": 137.129, | |
| "eval_steps_per_second": 137.129, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001417685628211944, | |
| "grad_norm": 0.07696182280778885, | |
| "learning_rate": 4e-05, | |
| "loss": 10.3974, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0002126528442317916, | |
| "grad_norm": 0.08747204393148422, | |
| "learning_rate": 6e-05, | |
| "loss": 10.3903, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0002126528442317916, | |
| "eval_loss": 10.386242866516113, | |
| "eval_runtime": 21.8717, | |
| "eval_samples_per_second": 135.792, | |
| "eval_steps_per_second": 135.792, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0002835371256423888, | |
| "grad_norm": 0.07742352038621902, | |
| "learning_rate": 8e-05, | |
| "loss": 10.383, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.000354421407052986, | |
| "grad_norm": 0.08146929740905762, | |
| "learning_rate": 0.0001, | |
| "loss": 10.3837, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0004253056884635832, | |
| "grad_norm": 0.10105450451374054, | |
| "learning_rate": 0.00012, | |
| "loss": 10.3935, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0004253056884635832, | |
| "eval_loss": 10.385862350463867, | |
| "eval_runtime": 21.8973, | |
| "eval_samples_per_second": 135.633, | |
| "eval_steps_per_second": 135.633, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0004961899698741804, | |
| "grad_norm": 0.07793600857257843, | |
| "learning_rate": 0.00014, | |
| "loss": 10.392, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0005670742512847776, | |
| "grad_norm": 0.1045127809047699, | |
| "learning_rate": 0.00016, | |
| "loss": 10.3973, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0006379585326953748, | |
| "grad_norm": 0.08697844296693802, | |
| "learning_rate": 0.00018, | |
| "loss": 10.3811, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0006379585326953748, | |
| "eval_loss": 10.385165214538574, | |
| "eval_runtime": 21.8138, | |
| "eval_samples_per_second": 136.152, | |
| "eval_steps_per_second": 136.152, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.000708842814105972, | |
| "grad_norm": 0.08561215549707413, | |
| "learning_rate": 0.0002, | |
| "loss": 10.382, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0007797270955165692, | |
| "grad_norm": 0.09970103204250336, | |
| "learning_rate": 0.00019781476007338058, | |
| "loss": 10.389, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0008506113769271664, | |
| "grad_norm": 0.09796269983053207, | |
| "learning_rate": 0.0001913545457642601, | |
| "loss": 10.385, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0008506113769271664, | |
| "eval_loss": 10.384222984313965, | |
| "eval_runtime": 21.8059, | |
| "eval_samples_per_second": 136.202, | |
| "eval_steps_per_second": 136.202, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0009214956583377636, | |
| "grad_norm": 0.08635301142930984, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 10.3813, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0009923799397483608, | |
| "grad_norm": 0.07206328958272934, | |
| "learning_rate": 0.00016691306063588583, | |
| "loss": 10.3874, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.001063264221158958, | |
| "grad_norm": 0.09540396928787231, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 10.3825, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.001063264221158958, | |
| "eval_loss": 10.38329792022705, | |
| "eval_runtime": 23.4703, | |
| "eval_samples_per_second": 126.543, | |
| "eval_steps_per_second": 126.543, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0011341485025695552, | |
| "grad_norm": 0.09925121068954468, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 10.3858, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0012050327839801525, | |
| "grad_norm": 0.12503716349601746, | |
| "learning_rate": 0.00011045284632676536, | |
| "loss": 10.3875, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0012759170653907496, | |
| "grad_norm": 0.08620873838663101, | |
| "learning_rate": 8.954715367323468e-05, | |
| "loss": 10.385, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0012759170653907496, | |
| "eval_loss": 10.382616996765137, | |
| "eval_runtime": 21.933, | |
| "eval_samples_per_second": 135.412, | |
| "eval_steps_per_second": 135.412, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0013468013468013469, | |
| "grad_norm": 0.0887003093957901, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 10.3812, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.001417685628211944, | |
| "grad_norm": 0.10247205197811127, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 10.3803, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0014885699096225413, | |
| "grad_norm": 0.10386703908443451, | |
| "learning_rate": 3.308693936411421e-05, | |
| "loss": 10.3872, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0014885699096225413, | |
| "eval_loss": 10.382221221923828, | |
| "eval_runtime": 21.9648, | |
| "eval_samples_per_second": 135.217, | |
| "eval_steps_per_second": 135.217, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0015594541910331384, | |
| "grad_norm": 0.11468103528022766, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 10.3818, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0016303384724437357, | |
| "grad_norm": 0.10616473853588104, | |
| "learning_rate": 8.645454235739903e-06, | |
| "loss": 10.3837, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0017012227538543328, | |
| "grad_norm": 0.10667341947555542, | |
| "learning_rate": 2.1852399266194314e-06, | |
| "loss": 10.3874, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0017012227538543328, | |
| "eval_loss": 10.382101058959961, | |
| "eval_runtime": 23.3378, | |
| "eval_samples_per_second": 127.262, | |
| "eval_steps_per_second": 127.262, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.00177210703526493, | |
| "grad_norm": 0.07770507037639618, | |
| "learning_rate": 0.0, | |
| "loss": 10.3784, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 646545408000.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |