| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 500, |
| "global_step": 400, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.5130800098180771, |
| "epoch": 0.025, |
| "grad_norm": 0.130859375, |
| "learning_rate": 0.00015000000000000001, |
| "loss": 0.8656, |
| "mean_token_accuracy": 0.802799665927887, |
| "num_tokens": 1952251.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.5367302268743515, |
| "epoch": 0.05, |
| "grad_norm": 0.06494140625, |
| "learning_rate": 0.00019639175257731958, |
| "loss": 0.511, |
| "mean_token_accuracy": 0.8456412851810455, |
| "num_tokens": 3913906.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.45613302290439606, |
| "epoch": 0.075, |
| "grad_norm": 0.0281982421875, |
| "learning_rate": 0.00019123711340206188, |
| "loss": 0.4604, |
| "mean_token_accuracy": 0.8565289735794067, |
| "num_tokens": 5864430.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.4395095020532608, |
| "epoch": 0.1, |
| "grad_norm": 0.0255126953125, |
| "learning_rate": 0.00018608247422680413, |
| "loss": 0.4418, |
| "mean_token_accuracy": 0.8613654911518097, |
| "num_tokens": 7824557.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.4373431891202927, |
| "epoch": 0.125, |
| "grad_norm": 0.021728515625, |
| "learning_rate": 0.0001809278350515464, |
| "loss": 0.4392, |
| "mean_token_accuracy": 0.8622795283794403, |
| "num_tokens": 9780531.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.44023202657699584, |
| "epoch": 0.15, |
| "grad_norm": 0.02294921875, |
| "learning_rate": 0.00017577319587628866, |
| "loss": 0.4401, |
| "mean_token_accuracy": 0.8610924303531646, |
| "num_tokens": 11737624.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.428512442111969, |
| "epoch": 0.175, |
| "grad_norm": 0.022705078125, |
| "learning_rate": 0.00017061855670103093, |
| "loss": 0.4313, |
| "mean_token_accuracy": 0.8635753691196442, |
| "num_tokens": 13697322.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.43105508685112, |
| "epoch": 0.2, |
| "grad_norm": 0.0228271484375, |
| "learning_rate": 0.00016546391752577318, |
| "loss": 0.4318, |
| "mean_token_accuracy": 0.863618916273117, |
| "num_tokens": 15649945.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.424772709608078, |
| "epoch": 0.225, |
| "grad_norm": 0.0255126953125, |
| "learning_rate": 0.0001603092783505155, |
| "loss": 0.4253, |
| "mean_token_accuracy": 0.8657742798328399, |
| "num_tokens": 17593762.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.4235113948583603, |
| "epoch": 0.25, |
| "grad_norm": 0.024658203125, |
| "learning_rate": 0.00015515463917525774, |
| "loss": 0.4243, |
| "mean_token_accuracy": 0.8653883039951324, |
| "num_tokens": 19542045.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.41819455325603483, |
| "epoch": 0.275, |
| "grad_norm": 0.0244140625, |
| "learning_rate": 0.00015000000000000001, |
| "loss": 0.4193, |
| "mean_token_accuracy": 0.8669234216213226, |
| "num_tokens": 21495444.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.41760264337062836, |
| "epoch": 0.3, |
| "grad_norm": 0.024169921875, |
| "learning_rate": 0.00014484536082474226, |
| "loss": 0.4192, |
| "mean_token_accuracy": 0.8668062031269074, |
| "num_tokens": 23454845.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.4235453844070435, |
| "epoch": 0.325, |
| "grad_norm": 0.0247802734375, |
| "learning_rate": 0.00013969072164948454, |
| "loss": 0.4255, |
| "mean_token_accuracy": 0.8658895254135132, |
| "num_tokens": 25407964.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.41925162076950073, |
| "epoch": 0.35, |
| "grad_norm": 0.0252685546875, |
| "learning_rate": 0.00013453608247422682, |
| "loss": 0.4194, |
| "mean_token_accuracy": 0.8667669773101807, |
| "num_tokens": 27366353.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.4193441987037659, |
| "epoch": 0.375, |
| "grad_norm": 0.024658203125, |
| "learning_rate": 0.0001293814432989691, |
| "loss": 0.4219, |
| "mean_token_accuracy": 0.8661471247673035, |
| "num_tokens": 29324217.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.41715689599514005, |
| "epoch": 0.4, |
| "grad_norm": 0.0255126953125, |
| "learning_rate": 0.00012422680412371134, |
| "loss": 0.417, |
| "mean_token_accuracy": 0.8673597633838653, |
| "num_tokens": 31274352.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.41513965725898744, |
| "epoch": 0.425, |
| "grad_norm": 0.02490234375, |
| "learning_rate": 0.00011907216494845362, |
| "loss": 0.4144, |
| "mean_token_accuracy": 0.8679460406303405, |
| "num_tokens": 33232032.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.41377590596675873, |
| "epoch": 0.45, |
| "grad_norm": 0.02587890625, |
| "learning_rate": 0.00011391752577319588, |
| "loss": 0.4151, |
| "mean_token_accuracy": 0.8679844081401825, |
| "num_tokens": 35185807.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.41528587639331815, |
| "epoch": 0.475, |
| "grad_norm": 0.0262451171875, |
| "learning_rate": 0.00010876288659793815, |
| "loss": 0.4163, |
| "mean_token_accuracy": 0.8680092096328735, |
| "num_tokens": 37142040.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.4145059704780579, |
| "epoch": 0.5, |
| "grad_norm": 0.025390625, |
| "learning_rate": 0.00010360824742268042, |
| "loss": 0.4155, |
| "mean_token_accuracy": 0.868000590801239, |
| "num_tokens": 39089347.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.4087012320756912, |
| "epoch": 0.525, |
| "grad_norm": 0.0260009765625, |
| "learning_rate": 9.845360824742269e-05, |
| "loss": 0.4089, |
| "mean_token_accuracy": 0.8693469941616059, |
| "num_tokens": 41036290.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.40894885659217833, |
| "epoch": 0.55, |
| "grad_norm": 0.0257568359375, |
| "learning_rate": 9.329896907216495e-05, |
| "loss": 0.411, |
| "mean_token_accuracy": 0.8692088723182678, |
| "num_tokens": 42977921.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.4120857834815979, |
| "epoch": 0.575, |
| "grad_norm": 0.0257568359375, |
| "learning_rate": 8.814432989690722e-05, |
| "loss": 0.4121, |
| "mean_token_accuracy": 0.868464720249176, |
| "num_tokens": 44926629.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.4110714465379715, |
| "epoch": 0.6, |
| "grad_norm": 0.025634765625, |
| "learning_rate": 8.298969072164949e-05, |
| "loss": 0.4118, |
| "mean_token_accuracy": 0.8687853872776031, |
| "num_tokens": 46881619.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.4152237057685852, |
| "epoch": 0.625, |
| "grad_norm": 0.026123046875, |
| "learning_rate": 7.783505154639175e-05, |
| "loss": 0.4164, |
| "mean_token_accuracy": 0.867627602815628, |
| "num_tokens": 48844100.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.40879290401935575, |
| "epoch": 0.65, |
| "grad_norm": 0.0267333984375, |
| "learning_rate": 7.268041237113403e-05, |
| "loss": 0.4087, |
| "mean_token_accuracy": 0.869741427898407, |
| "num_tokens": 50790693.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.4091353178024292, |
| "epoch": 0.675, |
| "grad_norm": 0.0257568359375, |
| "learning_rate": 6.752577319587629e-05, |
| "loss": 0.4103, |
| "mean_token_accuracy": 0.8694662570953369, |
| "num_tokens": 52742161.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.4107410669326782, |
| "epoch": 0.7, |
| "grad_norm": 0.0269775390625, |
| "learning_rate": 6.237113402061855e-05, |
| "loss": 0.4111, |
| "mean_token_accuracy": 0.8692059934139251, |
| "num_tokens": 54692664.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.4057494968175888, |
| "epoch": 0.725, |
| "grad_norm": 0.0263671875, |
| "learning_rate": 5.721649484536082e-05, |
| "loss": 0.407, |
| "mean_token_accuracy": 0.8700494408607483, |
| "num_tokens": 56654881.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.40529186129570005, |
| "epoch": 0.75, |
| "grad_norm": 0.0263671875, |
| "learning_rate": 5.20618556701031e-05, |
| "loss": 0.4057, |
| "mean_token_accuracy": 0.8709144175052643, |
| "num_tokens": 58612358.0, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.4049283146858215, |
| "epoch": 0.775, |
| "grad_norm": 0.0264892578125, |
| "learning_rate": 4.690721649484536e-05, |
| "loss": 0.4059, |
| "mean_token_accuracy": 0.8705782651901245, |
| "num_tokens": 60570532.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.41278861463069916, |
| "epoch": 0.8, |
| "grad_norm": 0.0264892578125, |
| "learning_rate": 4.175257731958763e-05, |
| "loss": 0.4129, |
| "mean_token_accuracy": 0.8685104608535766, |
| "num_tokens": 62523745.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.4059197723865509, |
| "epoch": 0.825, |
| "grad_norm": 0.0262451171875, |
| "learning_rate": 3.6597938144329896e-05, |
| "loss": 0.4061, |
| "mean_token_accuracy": 0.8701611340045929, |
| "num_tokens": 64468966.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.4075415819883347, |
| "epoch": 0.85, |
| "grad_norm": 0.026611328125, |
| "learning_rate": 3.1443298969072166e-05, |
| "loss": 0.4084, |
| "mean_token_accuracy": 0.8696109414100647, |
| "num_tokens": 66421242.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.4091123640537262, |
| "epoch": 0.875, |
| "grad_norm": 0.0269775390625, |
| "learning_rate": 2.6288659793814435e-05, |
| "loss": 0.4095, |
| "mean_token_accuracy": 0.8693838775157928, |
| "num_tokens": 68383604.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.4036696970462799, |
| "epoch": 0.9, |
| "grad_norm": 0.026611328125, |
| "learning_rate": 2.1134020618556702e-05, |
| "loss": 0.4036, |
| "mean_token_accuracy": 0.8708941996097564, |
| "num_tokens": 70340165.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.4065678924322128, |
| "epoch": 0.925, |
| "grad_norm": 0.0260009765625, |
| "learning_rate": 1.5979381443298968e-05, |
| "loss": 0.4081, |
| "mean_token_accuracy": 0.8699499130249023, |
| "num_tokens": 72299574.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.40871464014053344, |
| "epoch": 0.95, |
| "grad_norm": 0.0263671875, |
| "learning_rate": 1.0824742268041238e-05, |
| "loss": 0.4097, |
| "mean_token_accuracy": 0.8694815635681152, |
| "num_tokens": 74258475.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.4087704449892044, |
| "epoch": 0.975, |
| "grad_norm": 0.026123046875, |
| "learning_rate": 5.670103092783505e-06, |
| "loss": 0.409, |
| "mean_token_accuracy": 0.8698881447315217, |
| "num_tokens": 76213204.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.4095977067947388, |
| "epoch": 1.0, |
| "grad_norm": 0.02587890625, |
| "learning_rate": 5.154639175257732e-07, |
| "loss": 0.4106, |
| "mean_token_accuracy": 0.8693839371204376, |
| "num_tokens": 78069199.0, |
| "step": 400 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 400, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 50, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.5378917481854796e+19, |
| "train_batch_size": 6, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
