| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.0, |
| "eval_steps": 500, |
| "global_step": 396, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.050505050505050504, |
| "grad_norm": 2.394207715988159, |
| "learning_rate": 1.9545454545454546e-05, |
| "loss": 5.1582, |
| "mean_token_accuracy": 0.2001233294606209, |
| "num_tokens": 870609.0, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.10101010101010101, |
| "grad_norm": 1.912199854850769, |
| "learning_rate": 1.904040404040404e-05, |
| "loss": 4.9452, |
| "mean_token_accuracy": 0.2211832121014595, |
| "num_tokens": 1740025.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.15151515151515152, |
| "grad_norm": 1.7625455856323242, |
| "learning_rate": 1.8535353535353537e-05, |
| "loss": 4.793, |
| "mean_token_accuracy": 0.23438227660953997, |
| "num_tokens": 2612799.0, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.20202020202020202, |
| "grad_norm": 1.8361155986785889, |
| "learning_rate": 1.803030303030303e-05, |
| "loss": 4.6586, |
| "mean_token_accuracy": 0.2492609653621912, |
| "num_tokens": 3481685.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.25252525252525254, |
| "grad_norm": 1.7993879318237305, |
| "learning_rate": 1.752525252525253e-05, |
| "loss": 4.5569, |
| "mean_token_accuracy": 0.2619336560368538, |
| "num_tokens": 4345672.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.30303030303030304, |
| "grad_norm": 1.399275541305542, |
| "learning_rate": 1.7020202020202023e-05, |
| "loss": 4.4388, |
| "mean_token_accuracy": 0.27335850447416304, |
| "num_tokens": 5211312.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.35353535353535354, |
| "grad_norm": 1.0864430665969849, |
| "learning_rate": 1.6515151515151517e-05, |
| "loss": 4.3731, |
| "mean_token_accuracy": 0.2805166557431221, |
| "num_tokens": 6083451.0, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.40404040404040403, |
| "grad_norm": 1.1086381673812866, |
| "learning_rate": 1.601010101010101e-05, |
| "loss": 4.2975, |
| "mean_token_accuracy": 0.28774467632174494, |
| "num_tokens": 6945294.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.45454545454545453, |
| "grad_norm": 3.123107671737671, |
| "learning_rate": 1.5505050505050508e-05, |
| "loss": 4.2324, |
| "mean_token_accuracy": 0.2966753713786602, |
| "num_tokens": 7817939.0, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.5050505050505051, |
| "grad_norm": 0.9251999855041504, |
| "learning_rate": 1.5000000000000002e-05, |
| "loss": 4.1625, |
| "mean_token_accuracy": 0.3053535230457783, |
| "num_tokens": 8691417.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.5555555555555556, |
| "grad_norm": 0.9401581287384033, |
| "learning_rate": 1.4494949494949494e-05, |
| "loss": 4.1047, |
| "mean_token_accuracy": 0.31215064153075217, |
| "num_tokens": 9562161.0, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.6060606060606061, |
| "grad_norm": 1.956444263458252, |
| "learning_rate": 1.3989898989898992e-05, |
| "loss": 4.0634, |
| "mean_token_accuracy": 0.31729411631822585, |
| "num_tokens": 10428287.0, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.6565656565656566, |
| "grad_norm": 1.4297130107879639, |
| "learning_rate": 1.3484848484848486e-05, |
| "loss": 4.0134, |
| "mean_token_accuracy": 0.321382661908865, |
| "num_tokens": 11297263.0, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.7070707070707071, |
| "grad_norm": 0.9279031753540039, |
| "learning_rate": 1.2979797979797981e-05, |
| "loss": 3.9757, |
| "mean_token_accuracy": 0.3249970510601997, |
| "num_tokens": 12166798.0, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.7575757575757576, |
| "grad_norm": 0.7721190452575684, |
| "learning_rate": 1.2474747474747475e-05, |
| "loss": 3.9472, |
| "mean_token_accuracy": 0.32724879309535027, |
| "num_tokens": 13039570.0, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.8080808080808081, |
| "grad_norm": 0.8442203402519226, |
| "learning_rate": 1.1969696969696971e-05, |
| "loss": 3.9148, |
| "mean_token_accuracy": 0.3321701854467392, |
| "num_tokens": 13914152.0, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.8585858585858586, |
| "grad_norm": 2.19596266746521, |
| "learning_rate": 1.1464646464646465e-05, |
| "loss": 3.8928, |
| "mean_token_accuracy": 0.3358906783163548, |
| "num_tokens": 14784924.0, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.9090909090909091, |
| "grad_norm": 0.7308172583580017, |
| "learning_rate": 1.0959595959595961e-05, |
| "loss": 3.8529, |
| "mean_token_accuracy": 0.33917868658900263, |
| "num_tokens": 15653435.0, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.9595959595959596, |
| "grad_norm": 0.6991971135139465, |
| "learning_rate": 1.0454545454545455e-05, |
| "loss": 3.847, |
| "mean_token_accuracy": 0.3425107732415199, |
| "num_tokens": 16520037.0, |
| "step": 190 |
| }, |
| { |
| "epoch": 1.0101010101010102, |
| "grad_norm": 0.6827145218849182, |
| "learning_rate": 9.94949494949495e-06, |
| "loss": 3.832, |
| "mean_token_accuracy": 0.34413780048489573, |
| "num_tokens": 17380162.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.0606060606060606, |
| "grad_norm": 0.7619866132736206, |
| "learning_rate": 9.444444444444445e-06, |
| "loss": 3.7904, |
| "mean_token_accuracy": 0.3476664200425148, |
| "num_tokens": 18252280.0, |
| "step": 210 |
| }, |
| { |
| "epoch": 1.1111111111111112, |
| "grad_norm": 0.6569390892982483, |
| "learning_rate": 8.93939393939394e-06, |
| "loss": 3.7948, |
| "mean_token_accuracy": 0.3474963568150997, |
| "num_tokens": 19118807.0, |
| "step": 220 |
| }, |
| { |
| "epoch": 1.1616161616161615, |
| "grad_norm": 1.1422395706176758, |
| "learning_rate": 8.434343434343434e-06, |
| "loss": 3.7545, |
| "mean_token_accuracy": 0.350558465719223, |
| "num_tokens": 19987183.0, |
| "step": 230 |
| }, |
| { |
| "epoch": 1.2121212121212122, |
| "grad_norm": 0.9118938446044922, |
| "learning_rate": 7.92929292929293e-06, |
| "loss": 3.7356, |
| "mean_token_accuracy": 0.3532609537243843, |
| "num_tokens": 20864146.0, |
| "step": 240 |
| }, |
| { |
| "epoch": 1.2626262626262625, |
| "grad_norm": 0.830457329750061, |
| "learning_rate": 7.424242424242425e-06, |
| "loss": 3.7153, |
| "mean_token_accuracy": 0.3551630362868309, |
| "num_tokens": 21719950.0, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.3131313131313131, |
| "grad_norm": 0.735416054725647, |
| "learning_rate": 6.91919191919192e-06, |
| "loss": 3.719, |
| "mean_token_accuracy": 0.35566510185599326, |
| "num_tokens": 22598888.0, |
| "step": 260 |
| }, |
| { |
| "epoch": 1.3636363636363638, |
| "grad_norm": 0.7242897152900696, |
| "learning_rate": 6.4141414141414145e-06, |
| "loss": 3.699, |
| "mean_token_accuracy": 0.35784361362457273, |
| "num_tokens": 23475457.0, |
| "step": 270 |
| }, |
| { |
| "epoch": 1.4141414141414141, |
| "grad_norm": 0.876636803150177, |
| "learning_rate": 5.90909090909091e-06, |
| "loss": 3.6957, |
| "mean_token_accuracy": 0.35799761712551115, |
| "num_tokens": 24343304.0, |
| "step": 280 |
| }, |
| { |
| "epoch": 1.4646464646464645, |
| "grad_norm": 0.7917715907096863, |
| "learning_rate": 5.404040404040405e-06, |
| "loss": 3.6889, |
| "mean_token_accuracy": 0.3586253359913826, |
| "num_tokens": 25207489.0, |
| "step": 290 |
| }, |
| { |
| "epoch": 1.5151515151515151, |
| "grad_norm": 0.7085571885108948, |
| "learning_rate": 4.898989898989899e-06, |
| "loss": 3.668, |
| "mean_token_accuracy": 0.36179669201374054, |
| "num_tokens": 26072804.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.5656565656565657, |
| "grad_norm": 0.5749636292457581, |
| "learning_rate": 4.393939393939394e-06, |
| "loss": 3.657, |
| "mean_token_accuracy": 0.3627041429281235, |
| "num_tokens": 26939192.0, |
| "step": 310 |
| }, |
| { |
| "epoch": 1.6161616161616161, |
| "grad_norm": 0.8387623429298401, |
| "learning_rate": 3.88888888888889e-06, |
| "loss": 3.668, |
| "mean_token_accuracy": 0.3618996299803257, |
| "num_tokens": 27805084.0, |
| "step": 320 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 0.6559562683105469, |
| "learning_rate": 3.3838383838383844e-06, |
| "loss": 3.6492, |
| "mean_token_accuracy": 0.36368392035365105, |
| "num_tokens": 28676488.0, |
| "step": 330 |
| }, |
| { |
| "epoch": 1.7171717171717171, |
| "grad_norm": 0.6847353577613831, |
| "learning_rate": 2.8787878787878793e-06, |
| "loss": 3.6622, |
| "mean_token_accuracy": 0.3614747479557991, |
| "num_tokens": 29545994.0, |
| "step": 340 |
| }, |
| { |
| "epoch": 1.7676767676767677, |
| "grad_norm": 0.8682920336723328, |
| "learning_rate": 2.373737373737374e-06, |
| "loss": 3.648, |
| "mean_token_accuracy": 0.36393540278077124, |
| "num_tokens": 30421682.0, |
| "step": 350 |
| }, |
| { |
| "epoch": 1.8181818181818183, |
| "grad_norm": 0.7822595834732056, |
| "learning_rate": 1.868686868686869e-06, |
| "loss": 3.6468, |
| "mean_token_accuracy": 0.36377949416637423, |
| "num_tokens": 31290360.0, |
| "step": 360 |
| }, |
| { |
| "epoch": 1.8686868686868687, |
| "grad_norm": 0.6070267558097839, |
| "learning_rate": 1.3636363636363636e-06, |
| "loss": 3.6441, |
| "mean_token_accuracy": 0.3641804397106171, |
| "num_tokens": 32154120.0, |
| "step": 370 |
| }, |
| { |
| "epoch": 1.9191919191919191, |
| "grad_norm": 0.7802349328994751, |
| "learning_rate": 8.585858585858587e-07, |
| "loss": 3.6413, |
| "mean_token_accuracy": 0.3650804080069065, |
| "num_tokens": 33021167.0, |
| "step": 380 |
| }, |
| { |
| "epoch": 1.9696969696969697, |
| "grad_norm": 0.5713841319084167, |
| "learning_rate": 3.535353535353536e-07, |
| "loss": 3.6402, |
| "mean_token_accuracy": 0.3654905065894127, |
| "num_tokens": 33894053.0, |
| "step": 390 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 396, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.6464949686042624e+16, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
