| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.1276324186343331, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0025526483726866626, | |
| "grad_norm": 0.38045117259025574, | |
| "learning_rate": 5e-06, | |
| "loss": 4.6611, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0025526483726866626, | |
| "eval_loss": 1.226615071296692, | |
| "eval_runtime": 18.849, | |
| "eval_samples_per_second": 8.754, | |
| "eval_steps_per_second": 4.403, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005105296745373325, | |
| "grad_norm": 0.506873369216919, | |
| "learning_rate": 1e-05, | |
| "loss": 5.3237, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.007657945118059987, | |
| "grad_norm": 0.4798293113708496, | |
| "learning_rate": 1.5e-05, | |
| "loss": 4.8629, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01021059349074665, | |
| "grad_norm": 0.47835850715637207, | |
| "learning_rate": 2e-05, | |
| "loss": 5.1539, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.012763241863433313, | |
| "grad_norm": 0.5494751334190369, | |
| "learning_rate": 2.5e-05, | |
| "loss": 5.3459, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.015315890236119975, | |
| "grad_norm": 0.441150426864624, | |
| "learning_rate": 3e-05, | |
| "loss": 5.294, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.017868538608806637, | |
| "grad_norm": 0.4212581217288971, | |
| "learning_rate": 3.5e-05, | |
| "loss": 4.4479, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0204211869814933, | |
| "grad_norm": 0.4588180184364319, | |
| "learning_rate": 4e-05, | |
| "loss": 5.1455, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.02297383535417996, | |
| "grad_norm": 0.45286649465560913, | |
| "learning_rate": 4.5e-05, | |
| "loss": 4.7709, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.025526483726866625, | |
| "grad_norm": 0.4099409580230713, | |
| "learning_rate": 5e-05, | |
| "loss": 5.016, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.028079132099553285, | |
| "grad_norm": 0.5061376094818115, | |
| "learning_rate": 4.99229333433282e-05, | |
| "loss": 4.831, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.03063178047223995, | |
| "grad_norm": 0.5543558597564697, | |
| "learning_rate": 4.9692208514878444e-05, | |
| "loss": 5.1186, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03318442884492661, | |
| "grad_norm": 0.5037564039230347, | |
| "learning_rate": 4.9309248009941914e-05, | |
| "loss": 4.8973, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.03318442884492661, | |
| "eval_loss": 1.221695065498352, | |
| "eval_runtime": 18.8578, | |
| "eval_samples_per_second": 8.75, | |
| "eval_steps_per_second": 4.401, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.035737077217613274, | |
| "grad_norm": 0.45682233572006226, | |
| "learning_rate": 4.877641290737884e-05, | |
| "loss": 4.858, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.03828972559029994, | |
| "grad_norm": 0.5056037306785583, | |
| "learning_rate": 4.8096988312782174e-05, | |
| "loss": 5.0241, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0408423739629866, | |
| "grad_norm": 0.5200461745262146, | |
| "learning_rate": 4.72751631047092e-05, | |
| "loss": 4.9421, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04339502233567326, | |
| "grad_norm": 0.5702153444290161, | |
| "learning_rate": 4.6316004108852305e-05, | |
| "loss": 5.3207, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.04594767070835992, | |
| "grad_norm": 0.5843238234519958, | |
| "learning_rate": 4.522542485937369e-05, | |
| "loss": 4.5834, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.048500319081046586, | |
| "grad_norm": 0.6220148801803589, | |
| "learning_rate": 4.401014914000078e-05, | |
| "loss": 5.3389, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.05105296745373325, | |
| "grad_norm": 0.5212782621383667, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 4.9851, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.053605615826419914, | |
| "grad_norm": 0.6411041617393494, | |
| "learning_rate": 4.123620120825459e-05, | |
| "loss": 5.4835, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.05615826419910657, | |
| "grad_norm": 0.509672224521637, | |
| "learning_rate": 3.969463130731183e-05, | |
| "loss": 4.7967, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.058710912571793235, | |
| "grad_norm": 6.348986625671387, | |
| "learning_rate": 3.8062464117898724e-05, | |
| "loss": 4.2609, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0612635609444799, | |
| "grad_norm": 0.5605723261833191, | |
| "learning_rate": 3.634976249348867e-05, | |
| "loss": 4.7142, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06381620931716656, | |
| "grad_norm": 0.5585960745811462, | |
| "learning_rate": 3.456708580912725e-05, | |
| "loss": 4.9871, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06636885768985322, | |
| "grad_norm": 0.5804082155227661, | |
| "learning_rate": 3.272542485937369e-05, | |
| "loss": 4.8049, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.06636885768985322, | |
| "eval_loss": 1.200016975402832, | |
| "eval_runtime": 18.8842, | |
| "eval_samples_per_second": 8.737, | |
| "eval_steps_per_second": 4.395, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.06892150606253988, | |
| "grad_norm": 0.5582651495933533, | |
| "learning_rate": 3.083613409639764e-05, | |
| "loss": 4.7835, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.07147415443522655, | |
| "grad_norm": 0.6535090208053589, | |
| "learning_rate": 2.8910861626005776e-05, | |
| "loss": 5.1362, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.07402680280791321, | |
| "grad_norm": 0.649197518825531, | |
| "learning_rate": 2.6961477393196126e-05, | |
| "loss": 5.1688, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.07657945118059988, | |
| "grad_norm": 0.6213428378105164, | |
| "learning_rate": 2.5e-05, | |
| "loss": 4.7665, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07913209955328654, | |
| "grad_norm": 0.5954240560531616, | |
| "learning_rate": 2.303852260680388e-05, | |
| "loss": 4.6579, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0816847479259732, | |
| "grad_norm": 0.6680808067321777, | |
| "learning_rate": 2.1089138373994223e-05, | |
| "loss": 4.7746, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08423739629865985, | |
| "grad_norm": 0.8420075178146362, | |
| "learning_rate": 1.9163865903602374e-05, | |
| "loss": 4.3421, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.08679004467134652, | |
| "grad_norm": 0.6272104382514954, | |
| "learning_rate": 1.7274575140626318e-05, | |
| "loss": 5.0857, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.08934269304403318, | |
| "grad_norm": 0.6511965990066528, | |
| "learning_rate": 1.5432914190872757e-05, | |
| "loss": 4.2033, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.09189534141671984, | |
| "grad_norm": 0.6219887733459473, | |
| "learning_rate": 1.3650237506511331e-05, | |
| "loss": 5.0471, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.09444798978940651, | |
| "grad_norm": 0.6609371304512024, | |
| "learning_rate": 1.1937535882101281e-05, | |
| "loss": 5.1857, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.09700063816209317, | |
| "grad_norm": 0.6283613443374634, | |
| "learning_rate": 1.0305368692688174e-05, | |
| "loss": 4.5755, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.09955328653477984, | |
| "grad_norm": 0.6079770922660828, | |
| "learning_rate": 8.763798791745411e-06, | |
| "loss": 4.6142, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.09955328653477984, | |
| "eval_loss": 1.1924939155578613, | |
| "eval_runtime": 18.8948, | |
| "eval_samples_per_second": 8.733, | |
| "eval_steps_per_second": 4.393, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.1021059349074665, | |
| "grad_norm": 0.742786705493927, | |
| "learning_rate": 7.3223304703363135e-06, | |
| "loss": 5.2024, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10465858328015316, | |
| "grad_norm": 0.637015163898468, | |
| "learning_rate": 5.989850859999227e-06, | |
| "loss": 4.9553, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.10721123165283983, | |
| "grad_norm": 0.6206537485122681, | |
| "learning_rate": 4.7745751406263165e-06, | |
| "loss": 4.4764, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.10976388002552648, | |
| "grad_norm": 0.7534269690513611, | |
| "learning_rate": 3.6839958911476957e-06, | |
| "loss": 5.0251, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.11231652839821314, | |
| "grad_norm": 0.642451286315918, | |
| "learning_rate": 2.7248368952908053e-06, | |
| "loss": 4.9982, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.1148691767708998, | |
| "grad_norm": 0.6466948986053467, | |
| "learning_rate": 1.9030116872178316e-06, | |
| "loss": 5.0004, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.11742182514358647, | |
| "grad_norm": 0.6484655141830444, | |
| "learning_rate": 1.2235870926211619e-06, | |
| "loss": 4.6644, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.11997447351627313, | |
| "grad_norm": 0.6601917147636414, | |
| "learning_rate": 6.907519900580861e-07, | |
| "loss": 4.7506, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.1225271218889598, | |
| "grad_norm": 0.6565570831298828, | |
| "learning_rate": 3.077914851215585e-07, | |
| "loss": 5.1511, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.12507977026164646, | |
| "grad_norm": 0.6291937232017517, | |
| "learning_rate": 7.706665667180091e-08, | |
| "loss": 4.7335, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.1276324186343331, | |
| "grad_norm": 0.6736301183700562, | |
| "learning_rate": 0.0, | |
| "loss": 5.0993, | |
| "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": 3.37188711038976e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |