| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.948453608247423, | |
| "eval_steps": 500, | |
| "global_step": 240, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.020618556701030927, | |
| "grad_norm": 1.546875, | |
| "learning_rate": 8.333333333333333e-08, | |
| "loss": 2.5196, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.10309278350515463, | |
| "grad_norm": 1.5390625, | |
| "learning_rate": 4.1666666666666667e-07, | |
| "loss": 2.458, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.20618556701030927, | |
| "grad_norm": 1.4375, | |
| "learning_rate": 8.333333333333333e-07, | |
| "loss": 2.4466, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.30927835051546393, | |
| "grad_norm": 1.515625, | |
| "learning_rate": 1.2499999999999999e-06, | |
| "loss": 2.4248, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.41237113402061853, | |
| "grad_norm": 1.6171875, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 2.4622, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5154639175257731, | |
| "grad_norm": 1.4921875, | |
| "learning_rate": 1.9998942319271077e-06, | |
| "loss": 2.4147, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.6185567010309279, | |
| "grad_norm": 1.46875, | |
| "learning_rate": 1.9961946980917456e-06, | |
| "loss": 2.3747, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.7216494845360825, | |
| "grad_norm": 1.453125, | |
| "learning_rate": 1.987229113117374e-06, | |
| "loss": 2.4189, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.8247422680412371, | |
| "grad_norm": 1.578125, | |
| "learning_rate": 1.9730448705798236e-06, | |
| "loss": 2.4126, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.9278350515463918, | |
| "grad_norm": 1.4765625, | |
| "learning_rate": 1.953716950748227e-06, | |
| "loss": 2.3912, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.9896907216494846, | |
| "eval_loss": 2.246396064758301, | |
| "eval_runtime": 1335.3788, | |
| "eval_samples_per_second": 4.423, | |
| "eval_steps_per_second": 0.553, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.0309278350515463, | |
| "grad_norm": 1.4453125, | |
| "learning_rate": 1.929347524226822e-06, | |
| "loss": 2.3653, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.134020618556701, | |
| "grad_norm": 1.546875, | |
| "learning_rate": 1.900065411864121e-06, | |
| "loss": 2.3567, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.2371134020618557, | |
| "grad_norm": 1.5, | |
| "learning_rate": 1.8660254037844386e-06, | |
| "loss": 2.3555, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.3402061855670104, | |
| "grad_norm": 1.6015625, | |
| "learning_rate": 1.8274074411415103e-06, | |
| "loss": 2.2988, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.443298969072165, | |
| "grad_norm": 1.53125, | |
| "learning_rate": 1.7844156649195757e-06, | |
| "loss": 2.288, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.5463917525773194, | |
| "grad_norm": 1.3671875, | |
| "learning_rate": 1.737277336810124e-06, | |
| "loss": 2.2871, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.6494845360824741, | |
| "grad_norm": 1.59375, | |
| "learning_rate": 1.6862416378687337e-06, | |
| "loss": 2.2903, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.7525773195876289, | |
| "grad_norm": 1.484375, | |
| "learning_rate": 1.6315783513024974e-06, | |
| "loss": 2.2465, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.8556701030927836, | |
| "grad_norm": 1.484375, | |
| "learning_rate": 1.573576436351046e-06, | |
| "loss": 2.2434, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.9587628865979383, | |
| "grad_norm": 1.421875, | |
| "learning_rate": 1.5125425007998652e-06, | |
| "loss": 2.2442, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.1167430877685547, | |
| "eval_runtime": 1327.822, | |
| "eval_samples_per_second": 4.449, | |
| "eval_steps_per_second": 0.557, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 2.0618556701030926, | |
| "grad_norm": 1.4765625, | |
| "learning_rate": 1.4487991802004622e-06, | |
| "loss": 2.2931, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.1649484536082473, | |
| "grad_norm": 1.390625, | |
| "learning_rate": 1.3826834323650898e-06, | |
| "loss": 2.2062, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.268041237113402, | |
| "grad_norm": 1.4921875, | |
| "learning_rate": 1.3145447561516136e-06, | |
| "loss": 2.1501, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.3711340206185567, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 1.2447433439543238e-06, | |
| "loss": 2.1248, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.4742268041237114, | |
| "grad_norm": 1.3984375, | |
| "learning_rate": 1.1736481776669305e-06, | |
| "loss": 2.1624, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.5773195876288657, | |
| "grad_norm": 1.2265625, | |
| "learning_rate": 1.101635078182802e-06, | |
| "loss": 2.1479, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.680412371134021, | |
| "grad_norm": 1.3359375, | |
| "learning_rate": 1.0290847187431114e-06, | |
| "loss": 2.1655, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.783505154639175, | |
| "grad_norm": 1.3203125, | |
| "learning_rate": 9.56380612634664e-07, | |
| "loss": 2.1369, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.88659793814433, | |
| "grad_norm": 1.28125, | |
| "learning_rate": 8.839070858747696e-07, | |
| "loss": 2.1003, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.9896907216494846, | |
| "grad_norm": 1.2109375, | |
| "learning_rate": 8.120472455998881e-07, | |
| "loss": 2.1047, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 2.9896907216494846, | |
| "eval_loss": 2.0316832065582275, | |
| "eval_runtime": 1327.2975, | |
| "eval_samples_per_second": 4.45, | |
| "eval_steps_per_second": 0.557, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 3.0927835051546393, | |
| "grad_norm": 1.25, | |
| "learning_rate": 7.411809548974791e-07, | |
| "loss": 2.0675, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.195876288659794, | |
| "grad_norm": 1.3125, | |
| "learning_rate": 6.71682824786439e-07, | |
| "loss": 2.0934, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 3.2989690721649483, | |
| "grad_norm": 1.0234375, | |
| "learning_rate": 6.039202339608431e-07, | |
| "loss": 2.081, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.402061855670103, | |
| "grad_norm": 1.0625, | |
| "learning_rate": 5.382513867649663e-07, | |
| "loss": 2.0219, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.5051546391752577, | |
| "grad_norm": 1.3203125, | |
| "learning_rate": 4.750234196654399e-07, | |
| "loss": 2.0877, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.6082474226804124, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 4.1457056623005947e-07, | |
| "loss": 2.1019, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 3.711340206185567, | |
| "grad_norm": 1.1328125, | |
| "learning_rate": 3.5721239031346063e-07, | |
| "loss": 2.0828, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.8144329896907214, | |
| "grad_norm": 1.2421875, | |
| "learning_rate": 3.032520967893453e-07, | |
| "loss": 2.09, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 3.917525773195876, | |
| "grad_norm": 1.3125, | |
| "learning_rate": 2.5297492875900415e-07, | |
| "loss": 2.05, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.0067081451416016, | |
| "eval_runtime": 1327.277, | |
| "eval_samples_per_second": 4.45, | |
| "eval_steps_per_second": 0.557, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 4.020618556701031, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 2.0664665970876495e-07, | |
| "loss": 2.1246, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 4.123711340206185, | |
| "grad_norm": 1.1796875, | |
| "learning_rate": 1.6451218858706372e-07, | |
| "loss": 2.0933, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.22680412371134, | |
| "grad_norm": 1.1328125, | |
| "learning_rate": 1.2679424522780425e-07, | |
| "loss": 2.0561, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 4.329896907216495, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 9.369221296335006e-08, | |
| "loss": 2.0946, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.43298969072165, | |
| "grad_norm": 1.125, | |
| "learning_rate": 6.538107465101162e-08, | |
| "loss": 2.0797, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 4.536082474226804, | |
| "grad_norm": 1.125, | |
| "learning_rate": 4.20104876845111e-08, | |
| "loss": 2.0907, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.639175257731958, | |
| "grad_norm": 1.171875, | |
| "learning_rate": 2.3703992880066636e-08, | |
| "loss": 2.1295, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 4.742268041237113, | |
| "grad_norm": 1.1640625, | |
| "learning_rate": 1.0558361419055529e-08, | |
| "loss": 2.0247, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.845360824742268, | |
| "grad_norm": 1.046875, | |
| "learning_rate": 2.643083299427751e-09, | |
| "loss": 2.0658, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 4.948453608247423, | |
| "grad_norm": 1.046875, | |
| "learning_rate": 0.0, | |
| "loss": 2.0626, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.948453608247423, | |
| "eval_loss": 2.0068044662475586, | |
| "eval_runtime": 1327.4294, | |
| "eval_samples_per_second": 4.45, | |
| "eval_steps_per_second": 0.557, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.948453608247423, | |
| "step": 240, | |
| "total_flos": 1.0715672433026662e+17, | |
| "train_loss": 2.205865615606308, | |
| "train_runtime": 7479.3037, | |
| "train_samples_per_second": 0.257, | |
| "train_steps_per_second": 0.032 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 240, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 1, | |
| "total_flos": 1.0715672433026662e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |