| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5319148936170213, | |
| "eval_steps": 500, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010638297872340425, | |
| "grad_norm": 0.2887957990169525, | |
| "learning_rate": 0.0002, | |
| "loss": 0.9327, | |
| "mean_token_accuracy": 0.7623532712459564, | |
| "num_tokens": 1982.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02127659574468085, | |
| "grad_norm": 0.6030351519584656, | |
| "learning_rate": 0.000196, | |
| "loss": 1.8105, | |
| "mean_token_accuracy": 0.6288281381130219, | |
| "num_tokens": 2943.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.031914893617021274, | |
| "grad_norm": 0.3874857723712921, | |
| "learning_rate": 0.000192, | |
| "loss": 1.0589, | |
| "mean_token_accuracy": 0.7225219160318375, | |
| "num_tokens": 4484.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0425531914893617, | |
| "grad_norm": 0.24475964903831482, | |
| "learning_rate": 0.000188, | |
| "loss": 1.0091, | |
| "mean_token_accuracy": 0.7289294451475143, | |
| "num_tokens": 6727.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.05319148936170213, | |
| "grad_norm": 0.27278101444244385, | |
| "learning_rate": 0.00018400000000000003, | |
| "loss": 0.7239, | |
| "mean_token_accuracy": 0.7673206329345703, | |
| "num_tokens": 8935.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.06382978723404255, | |
| "grad_norm": 0.687332272529602, | |
| "learning_rate": 0.00018, | |
| "loss": 1.3729, | |
| "mean_token_accuracy": 0.6930361837148666, | |
| "num_tokens": 9786.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.07446808510638298, | |
| "grad_norm": 1.9471864700317383, | |
| "learning_rate": 0.00017600000000000002, | |
| "loss": 1.4784, | |
| "mean_token_accuracy": 0.6858641654253006, | |
| "num_tokens": 10553.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0851063829787234, | |
| "grad_norm": 0.5966729521751404, | |
| "learning_rate": 0.000172, | |
| "loss": 0.9692, | |
| "mean_token_accuracy": 0.7569561004638672, | |
| "num_tokens": 11778.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.09574468085106383, | |
| "grad_norm": 0.8323692679405212, | |
| "learning_rate": 0.000168, | |
| "loss": 1.1015, | |
| "mean_token_accuracy": 0.7499070465564728, | |
| "num_tokens": 12686.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.10638297872340426, | |
| "grad_norm": 1.2347114086151123, | |
| "learning_rate": 0.000164, | |
| "loss": 1.1746, | |
| "mean_token_accuracy": 0.740473136305809, | |
| "num_tokens": 13444.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11702127659574468, | |
| "grad_norm": 0.528573751449585, | |
| "learning_rate": 0.00016, | |
| "loss": 0.5423, | |
| "mean_token_accuracy": 0.8274708986282349, | |
| "num_tokens": 15704.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.1276595744680851, | |
| "grad_norm": 2.641803026199341, | |
| "learning_rate": 0.00015600000000000002, | |
| "loss": 0.863, | |
| "mean_token_accuracy": 0.7851854711771011, | |
| "num_tokens": 16706.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.13829787234042554, | |
| "grad_norm": 1.9446417093276978, | |
| "learning_rate": 0.000152, | |
| "loss": 0.949, | |
| "mean_token_accuracy": 0.7901278287172318, | |
| "num_tokens": 17562.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.14893617021276595, | |
| "grad_norm": 1.0860435962677002, | |
| "learning_rate": 0.000148, | |
| "loss": 0.917, | |
| "mean_token_accuracy": 0.7842272222042084, | |
| "num_tokens": 18558.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.1595744680851064, | |
| "grad_norm": 1.0175093412399292, | |
| "learning_rate": 0.000144, | |
| "loss": 0.6745, | |
| "mean_token_accuracy": 0.8425705283880234, | |
| "num_tokens": 19700.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1702127659574468, | |
| "grad_norm": 1.4986056089401245, | |
| "learning_rate": 0.00014, | |
| "loss": 0.6603, | |
| "mean_token_accuracy": 0.8449233621358871, | |
| "num_tokens": 20420.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.18085106382978725, | |
| "grad_norm": 0.8947276473045349, | |
| "learning_rate": 0.00013600000000000003, | |
| "loss": 0.539, | |
| "mean_token_accuracy": 0.8811111599206924, | |
| "num_tokens": 21419.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.19148936170212766, | |
| "grad_norm": 0.42721500992774963, | |
| "learning_rate": 0.000132, | |
| "loss": 0.4788, | |
| "mean_token_accuracy": 0.8939242213964462, | |
| "num_tokens": 23846.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.20212765957446807, | |
| "grad_norm": 0.48865628242492676, | |
| "learning_rate": 0.00012800000000000002, | |
| "loss": 0.6326, | |
| "mean_token_accuracy": 0.8746728301048279, | |
| "num_tokens": 25747.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.2127659574468085, | |
| "grad_norm": 0.8984824419021606, | |
| "learning_rate": 0.000124, | |
| "loss": 0.4745, | |
| "mean_token_accuracy": 0.8928907662630081, | |
| "num_tokens": 26627.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.22340425531914893, | |
| "grad_norm": 0.9669850468635559, | |
| "learning_rate": 0.00012, | |
| "loss": 0.4962, | |
| "mean_token_accuracy": 0.8934834748506546, | |
| "num_tokens": 27436.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.23404255319148937, | |
| "grad_norm": 0.36605075001716614, | |
| "learning_rate": 0.000116, | |
| "loss": 0.3258, | |
| "mean_token_accuracy": 0.9293452799320221, | |
| "num_tokens": 29420.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.24468085106382978, | |
| "grad_norm": 0.894280731678009, | |
| "learning_rate": 0.00011200000000000001, | |
| "loss": 0.4877, | |
| "mean_token_accuracy": 0.8980903029441833, | |
| "num_tokens": 30216.0, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.2553191489361702, | |
| "grad_norm": 0.7681854963302612, | |
| "learning_rate": 0.00010800000000000001, | |
| "loss": 0.541, | |
| "mean_token_accuracy": 0.8775683045387268, | |
| "num_tokens": 31099.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.26595744680851063, | |
| "grad_norm": 0.7193688750267029, | |
| "learning_rate": 0.00010400000000000001, | |
| "loss": 0.5943, | |
| "mean_token_accuracy": 0.8750757277011871, | |
| "num_tokens": 31956.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2765957446808511, | |
| "grad_norm": 0.6575931310653687, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4527, | |
| "mean_token_accuracy": 0.8902412056922913, | |
| "num_tokens": 32786.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.2872340425531915, | |
| "grad_norm": 0.6464940309524536, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.4924, | |
| "mean_token_accuracy": 0.8929924517869949, | |
| "num_tokens": 33631.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.2978723404255319, | |
| "grad_norm": 0.7342627644538879, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 0.4438, | |
| "mean_token_accuracy": 0.8995166420936584, | |
| "num_tokens": 34486.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.30851063829787234, | |
| "grad_norm": 0.41542309522628784, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.3749, | |
| "mean_token_accuracy": 0.9000015705823898, | |
| "num_tokens": 36295.0, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.3191489361702128, | |
| "grad_norm": 0.7227439880371094, | |
| "learning_rate": 8.4e-05, | |
| "loss": 0.6178, | |
| "mean_token_accuracy": 0.8595587909221649, | |
| "num_tokens": 37104.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.32978723404255317, | |
| "grad_norm": 0.5684676766395569, | |
| "learning_rate": 8e-05, | |
| "loss": 0.4551, | |
| "mean_token_accuracy": 0.8971092998981476, | |
| "num_tokens": 38549.0, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.3404255319148936, | |
| "grad_norm": 0.550665020942688, | |
| "learning_rate": 7.6e-05, | |
| "loss": 0.3505, | |
| "mean_token_accuracy": 0.9074584692716599, | |
| "num_tokens": 39872.0, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.35106382978723405, | |
| "grad_norm": 0.6100918650627136, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.3216, | |
| "mean_token_accuracy": 0.9257150143384933, | |
| "num_tokens": 40735.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.3617021276595745, | |
| "grad_norm": 0.4749443233013153, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 0.5295, | |
| "mean_token_accuracy": 0.8798354268074036, | |
| "num_tokens": 42345.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.3723404255319149, | |
| "grad_norm": 0.6380234360694885, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.5781, | |
| "mean_token_accuracy": 0.8827451318502426, | |
| "num_tokens": 43313.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.3829787234042553, | |
| "grad_norm": 0.8250994086265564, | |
| "learning_rate": 6e-05, | |
| "loss": 0.6299, | |
| "mean_token_accuracy": 0.8456272482872009, | |
| "num_tokens": 44173.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.39361702127659576, | |
| "grad_norm": 0.5874150395393372, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.4274, | |
| "mean_token_accuracy": 0.8966661095619202, | |
| "num_tokens": 45785.0, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.40425531914893614, | |
| "grad_norm": 0.42270627617836, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 0.3216, | |
| "mean_token_accuracy": 0.9102791994810104, | |
| "num_tokens": 47201.0, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.4148936170212766, | |
| "grad_norm": 0.5454854965209961, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.4902, | |
| "mean_token_accuracy": 0.9055827260017395, | |
| "num_tokens": 48182.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.425531914893617, | |
| "grad_norm": 0.6554898023605347, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.3607, | |
| "mean_token_accuracy": 0.9176944196224213, | |
| "num_tokens": 48959.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.43617021276595747, | |
| "grad_norm": 0.35927918553352356, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2911, | |
| "mean_token_accuracy": 0.9149797260761261, | |
| "num_tokens": 50485.0, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.44680851063829785, | |
| "grad_norm": 0.3777298331260681, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.4113, | |
| "mean_token_accuracy": 0.9079957902431488, | |
| "num_tokens": 52011.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.4574468085106383, | |
| "grad_norm": 0.5052103996276855, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.4112, | |
| "mean_token_accuracy": 0.8977669477462769, | |
| "num_tokens": 53670.0, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.46808510638297873, | |
| "grad_norm": 0.7540407180786133, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.6973, | |
| "mean_token_accuracy": 0.8849319517612457, | |
| "num_tokens": 55106.0, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.4787234042553192, | |
| "grad_norm": 0.5548056960105896, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.4593, | |
| "mean_token_accuracy": 0.8819023072719574, | |
| "num_tokens": 56178.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.48936170212765956, | |
| "grad_norm": 0.8761438131332397, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3054, | |
| "mean_token_accuracy": 0.9220689237117767, | |
| "num_tokens": 57070.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.917113184928894, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.619, | |
| "mean_token_accuracy": 0.8847331702709198, | |
| "num_tokens": 58389.0, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.5106382978723404, | |
| "grad_norm": 0.35633689165115356, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.2867, | |
| "mean_token_accuracy": 0.9246316254138947, | |
| "num_tokens": 60319.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.5212765957446809, | |
| "grad_norm": 0.5241644978523254, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.3784, | |
| "mean_token_accuracy": 0.8954263031482697, | |
| "num_tokens": 61355.0, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.5319148936170213, | |
| "grad_norm": 0.39193060994148254, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.399, | |
| "mean_token_accuracy": 0.9024893939495087, | |
| "num_tokens": 63093.0, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 485900913143808.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |