| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9591836734693877, | |
| "eval_steps": 2, | |
| "global_step": 120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0, | |
| "eval_loss": 1.2313029766082764, | |
| "eval_runtime": 18.2352, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 0 | |
| }, | |
| { | |
| "epoch": 0.0163265306122449, | |
| "grad_norm": 0.37113556265830994, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.4085, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0326530612244898, | |
| "grad_norm": 0.35803329944610596, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3876, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0326530612244898, | |
| "eval_loss": 1.2231345176696777, | |
| "eval_runtime": 18.2243, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.04897959183673469, | |
| "grad_norm": 0.3112759590148926, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.3946, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0653061224489796, | |
| "grad_norm": 0.2448713332414627, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4363, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0653061224489796, | |
| "eval_loss": 1.1564743518829346, | |
| "eval_runtime": 18.253, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.08163265306122448, | |
| "grad_norm": 0.2955642342567444, | |
| "learning_rate": 0.000125, | |
| "loss": 0.4394, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.09795918367346938, | |
| "grad_norm": 0.41399946808815, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.4902, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.09795918367346938, | |
| "eval_loss": 1.1353044509887695, | |
| "eval_runtime": 18.2664, | |
| "eval_samples_per_second": 0.876, | |
| "eval_steps_per_second": 0.876, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.11428571428571428, | |
| "grad_norm": 0.2643347978591919, | |
| "learning_rate": 0.000175, | |
| "loss": 0.3528, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.1306122448979592, | |
| "grad_norm": 0.21472330391407013, | |
| "learning_rate": 0.0002, | |
| "loss": 0.357, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1306122448979592, | |
| "eval_loss": 1.1218546628952026, | |
| "eval_runtime": 18.2139, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1469387755102041, | |
| "grad_norm": 0.23261462152004242, | |
| "learning_rate": 0.00022500000000000002, | |
| "loss": 0.3924, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.16326530612244897, | |
| "grad_norm": 0.18365171551704407, | |
| "learning_rate": 0.00025, | |
| "loss": 0.283, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16326530612244897, | |
| "eval_loss": 1.1094393730163574, | |
| "eval_runtime": 18.249, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17959183673469387, | |
| "grad_norm": 0.20431634783744812, | |
| "learning_rate": 0.000275, | |
| "loss": 0.3178, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.19591836734693877, | |
| "grad_norm": 0.2033773809671402, | |
| "learning_rate": 0.00030000000000000003, | |
| "loss": 0.3366, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.19591836734693877, | |
| "eval_loss": 1.1021316051483154, | |
| "eval_runtime": 18.2562, | |
| "eval_samples_per_second": 0.876, | |
| "eval_steps_per_second": 0.876, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.21224489795918366, | |
| "grad_norm": 0.27416911721229553, | |
| "learning_rate": 0.00032500000000000004, | |
| "loss": 0.3367, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.22857142857142856, | |
| "grad_norm": 0.32283130288124084, | |
| "learning_rate": 0.00035, | |
| "loss": 0.3364, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.22857142857142856, | |
| "eval_loss": 1.0523244142532349, | |
| "eval_runtime": 18.2347, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.24489795918367346, | |
| "grad_norm": 0.21469368040561676, | |
| "learning_rate": 0.000375, | |
| "loss": 0.3496, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2612244897959184, | |
| "grad_norm": 0.19361759722232819, | |
| "learning_rate": 0.0004, | |
| "loss": 0.2961, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.2612244897959184, | |
| "eval_loss": 1.0401124954223633, | |
| "eval_runtime": 18.2288, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.27755102040816326, | |
| "grad_norm": 0.1797463297843933, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3016, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.2938775510204082, | |
| "grad_norm": 0.28122591972351074, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3656, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2938775510204082, | |
| "eval_loss": 1.0228931903839111, | |
| "eval_runtime": 18.2429, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.31020408163265306, | |
| "grad_norm": 0.227556511759758, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3246, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.32653061224489793, | |
| "grad_norm": 0.211012601852417, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3453, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.32653061224489793, | |
| "eval_loss": 1.017521858215332, | |
| "eval_runtime": 18.2213, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.34285714285714286, | |
| "grad_norm": 0.20422972738742828, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3258, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.35918367346938773, | |
| "grad_norm": 0.206649512052536, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3248, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.35918367346938773, | |
| "eval_loss": 0.9992413520812988, | |
| "eval_runtime": 18.249, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.37551020408163266, | |
| "grad_norm": 0.19837401807308197, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3181, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.39183673469387753, | |
| "grad_norm": 0.20325997471809387, | |
| "learning_rate": 0.0004, | |
| "loss": 0.2826, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.39183673469387753, | |
| "eval_loss": 0.9943413734436035, | |
| "eval_runtime": 18.2482, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.40816326530612246, | |
| "grad_norm": 0.3371317386627197, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3197, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.42448979591836733, | |
| "grad_norm": 0.21709182858467102, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3272, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.42448979591836733, | |
| "eval_loss": 0.9733779430389404, | |
| "eval_runtime": 18.2232, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.44081632653061226, | |
| "grad_norm": 0.2420857548713684, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3293, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.45714285714285713, | |
| "grad_norm": 0.30486994981765747, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3958, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.45714285714285713, | |
| "eval_loss": 0.9588731527328491, | |
| "eval_runtime": 18.2458, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.47346938775510206, | |
| "grad_norm": 0.24714964628219604, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3305, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.4897959183673469, | |
| "grad_norm": 0.21984225511550903, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3395, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4897959183673469, | |
| "eval_loss": 0.9427903890609741, | |
| "eval_runtime": 18.2414, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5061224489795918, | |
| "grad_norm": 0.19778016209602356, | |
| "learning_rate": 0.0004, | |
| "loss": 0.2918, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.5224489795918368, | |
| "grad_norm": 0.21754617989063263, | |
| "learning_rate": 0.0004, | |
| "loss": 0.2855, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.5224489795918368, | |
| "eval_loss": 0.9335330128669739, | |
| "eval_runtime": 18.275, | |
| "eval_samples_per_second": 0.876, | |
| "eval_steps_per_second": 0.876, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.5387755102040817, | |
| "grad_norm": 0.2221430391073227, | |
| "learning_rate": 0.0004, | |
| "loss": 0.2946, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.5551020408163265, | |
| "grad_norm": 0.2112974375486374, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3149, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.5551020408163265, | |
| "eval_loss": 0.9311869144439697, | |
| "eval_runtime": 18.2431, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 0.19651219248771667, | |
| "learning_rate": 0.0004, | |
| "loss": 0.2606, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.5877551020408164, | |
| "grad_norm": 0.22742077708244324, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3245, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5877551020408164, | |
| "eval_loss": 0.9281033873558044, | |
| "eval_runtime": 18.2248, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.6040816326530613, | |
| "grad_norm": 0.2320890575647354, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3532, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.6204081632653061, | |
| "grad_norm": 0.21191200613975525, | |
| "learning_rate": 0.0004, | |
| "loss": 0.2973, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.6204081632653061, | |
| "eval_loss": 0.9274996519088745, | |
| "eval_runtime": 18.2309, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.636734693877551, | |
| "grad_norm": 0.24098484218120575, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3194, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.6530612244897959, | |
| "grad_norm": 0.3358725607395172, | |
| "learning_rate": 0.0004, | |
| "loss": 0.2833, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6530612244897959, | |
| "eval_loss": 0.9254183769226074, | |
| "eval_runtime": 18.2685, | |
| "eval_samples_per_second": 0.876, | |
| "eval_steps_per_second": 0.876, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6693877551020408, | |
| "grad_norm": 0.2399401068687439, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3381, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.6857142857142857, | |
| "grad_norm": 0.23229075968265533, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3501, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.6857142857142857, | |
| "eval_loss": 0.9213573932647705, | |
| "eval_runtime": 18.2788, | |
| "eval_samples_per_second": 0.875, | |
| "eval_steps_per_second": 0.875, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.7020408163265306, | |
| "grad_norm": 0.2341497391462326, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3033, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.7183673469387755, | |
| "grad_norm": 0.22997914254665375, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3675, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.7183673469387755, | |
| "eval_loss": 0.9130539298057556, | |
| "eval_runtime": 18.2601, | |
| "eval_samples_per_second": 0.876, | |
| "eval_steps_per_second": 0.876, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.7346938775510204, | |
| "grad_norm": 0.23445634543895721, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3113, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.7510204081632653, | |
| "grad_norm": 0.22852188348770142, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3593, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.7510204081632653, | |
| "eval_loss": 0.9100953936576843, | |
| "eval_runtime": 18.2446, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.7673469387755102, | |
| "grad_norm": 0.2316325306892395, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3121, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.7836734693877551, | |
| "grad_norm": 0.2397606372833252, | |
| "learning_rate": 0.0004, | |
| "loss": 0.2979, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.7836734693877551, | |
| "eval_loss": 0.9087210297584534, | |
| "eval_runtime": 18.2833, | |
| "eval_samples_per_second": 0.875, | |
| "eval_steps_per_second": 0.875, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.22637659311294556, | |
| "learning_rate": 0.0004, | |
| "loss": 0.2919, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.8163265306122449, | |
| "grad_norm": 0.255599707365036, | |
| "learning_rate": 0.0004, | |
| "loss": 0.3741, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8163265306122449, | |
| "eval_loss": 0.9020435810089111, | |
| "eval_runtime": 18.2931, | |
| "eval_samples_per_second": 0.875, | |
| "eval_steps_per_second": 0.875, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8163265306122449, | |
| "eval_loss": 0.9020435810089111, | |
| "eval_runtime": 18.1276, | |
| "eval_samples_per_second": 0.883, | |
| "eval_steps_per_second": 0.883, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8326530612244898, | |
| "grad_norm": 0.2062978744506836, | |
| "learning_rate": 0.0003877467715307749, | |
| "loss": 0.3303, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.8489795918367347, | |
| "grad_norm": 0.20149795711040497, | |
| "learning_rate": 0.0003870443502801494, | |
| "loss": 0.3212, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.8489795918367347, | |
| "eval_loss": 0.8994156718254089, | |
| "eval_runtime": 18.1853, | |
| "eval_samples_per_second": 0.88, | |
| "eval_steps_per_second": 0.88, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.8653061224489796, | |
| "grad_norm": 0.17913399636745453, | |
| "learning_rate": 0.0003863230255984052, | |
| "loss": 0.2637, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.8816326530612245, | |
| "grad_norm": 0.19331133365631104, | |
| "learning_rate": 0.00038558287038542615, | |
| "loss": 0.3564, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.8816326530612245, | |
| "eval_loss": 0.8963940143585205, | |
| "eval_runtime": 18.223, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.8979591836734694, | |
| "grad_norm": 0.20378044247627258, | |
| "learning_rate": 0.00038482395944418313, | |
| "loss": 0.3253, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.9142857142857143, | |
| "grad_norm": 0.19742076098918915, | |
| "learning_rate": 0.0003840463694731741, | |
| "loss": 0.3715, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.9142857142857143, | |
| "eval_loss": 0.8902382850646973, | |
| "eval_runtime": 18.2093, | |
| "eval_samples_per_second": 0.879, | |
| "eval_steps_per_second": 0.879, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.9306122448979591, | |
| "grad_norm": 0.1908055692911148, | |
| "learning_rate": 0.0003832501790586724, | |
| "loss": 0.3305, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.9469387755102041, | |
| "grad_norm": 0.21527761220932007, | |
| "learning_rate": 0.0003824354686667848, | |
| "loss": 0.3523, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.9469387755102041, | |
| "eval_loss": 0.8786917328834534, | |
| "eval_runtime": 18.2124, | |
| "eval_samples_per_second": 0.879, | |
| "eval_steps_per_second": 0.879, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.963265306122449, | |
| "grad_norm": 0.1817687749862671, | |
| "learning_rate": 0.00038160232063531917, | |
| "loss": 0.2616, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.9795918367346939, | |
| "grad_norm": 0.17659063637256622, | |
| "learning_rate": 0.0003807508191654631, | |
| "loss": 0.251, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.9795918367346939, | |
| "eval_loss": 0.8799586296081543, | |
| "eval_runtime": 18.2543, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.9959183673469387, | |
| "grad_norm": 0.22647982835769653, | |
| "learning_rate": 0.0003798810503132742, | |
| "loss": 0.2881, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.0122448979591836, | |
| "grad_norm": 0.45377442240715027, | |
| "learning_rate": 0.00037899310198098295, | |
| "loss": 0.5515, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.0122448979591836, | |
| "eval_loss": 0.8895297050476074, | |
| "eval_runtime": 18.2212, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.0285714285714285, | |
| "grad_norm": 0.3527546226978302, | |
| "learning_rate": 0.000378087063908109, | |
| "loss": 0.2427, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.0448979591836736, | |
| "grad_norm": 0.2781178951263428, | |
| "learning_rate": 0.0003771630276623915, | |
| "loss": 0.2215, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.0448979591836736, | |
| "eval_loss": 0.9294220209121704, | |
| "eval_runtime": 18.226, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.0612244897959184, | |
| "grad_norm": 0.2543565034866333, | |
| "learning_rate": 0.00037622108663053536, | |
| "loss": 0.2761, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.0775510204081633, | |
| "grad_norm": 0.26259011030197144, | |
| "learning_rate": 0.0003752613360087727, | |
| "loss": 0.2425, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.0775510204081633, | |
| "eval_loss": 0.9212721586227417, | |
| "eval_runtime": 18.2309, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.0938775510204082, | |
| "grad_norm": 0.36850446462631226, | |
| "learning_rate": 0.00037428387279324257, | |
| "loss": 0.225, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.110204081632653, | |
| "grad_norm": 0.23748713731765747, | |
| "learning_rate": 0.0003732887957701874, | |
| "loss": 0.2363, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.110204081632653, | |
| "eval_loss": 0.9431418776512146, | |
| "eval_runtime": 18.2471, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.126530612244898, | |
| "grad_norm": 0.23217403888702393, | |
| "learning_rate": 0.00037227620550597, | |
| "loss": 0.2364, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 0.28124868869781494, | |
| "learning_rate": 0.0003712462043369093, | |
| "loss": 0.2197, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "eval_loss": 0.9205393195152283, | |
| "eval_runtime": 18.2662, | |
| "eval_samples_per_second": 0.876, | |
| "eval_steps_per_second": 0.876, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.1591836734693877, | |
| "grad_norm": 0.2142041176557541, | |
| "learning_rate": 0.00037019889635893843, | |
| "loss": 0.1958, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 1.1755102040816325, | |
| "grad_norm": 0.34761127829551697, | |
| "learning_rate": 0.0003691343874170838, | |
| "loss": 0.2183, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.1755102040816325, | |
| "eval_loss": 0.9254322648048401, | |
| "eval_runtime": 18.2303, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.1918367346938776, | |
| "grad_norm": 0.2560065686702728, | |
| "learning_rate": 0.00036805278509476844, | |
| "loss": 0.248, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.2081632653061225, | |
| "grad_norm": 0.7435296773910522, | |
| "learning_rate": 0.00036695419870293915, | |
| "loss": 0.2356, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.2081632653061225, | |
| "eval_loss": 0.9271378517150879, | |
| "eval_runtime": 18.2503, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.2244897959183674, | |
| "grad_norm": 0.8685758113861084, | |
| "learning_rate": 0.00036583873926901867, | |
| "loss": 0.2129, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.2408163265306122, | |
| "grad_norm": 1.2740998268127441, | |
| "learning_rate": 0.0003647065195256855, | |
| "loss": 0.2469, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.2408163265306122, | |
| "eval_loss": 0.9236885905265808, | |
| "eval_runtime": 18.2226, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.2571428571428571, | |
| "grad_norm": 0.28524693846702576, | |
| "learning_rate": 0.0003635576538994801, | |
| "loss": 0.2278, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 1.273469387755102, | |
| "grad_norm": 1.0116885900497437, | |
| "learning_rate": 0.0003623922584992409, | |
| "loss": 0.2488, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.273469387755102, | |
| "eval_loss": 0.923646867275238, | |
| "eval_runtime": 18.2519, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.2897959183673469, | |
| "grad_norm": 0.9336573481559753, | |
| "learning_rate": 0.0003612104511043694, | |
| "loss": 0.3095, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 1.306122448979592, | |
| "grad_norm": 0.43459734320640564, | |
| "learning_rate": 0.0003600123511529278, | |
| "loss": 0.2215, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.306122448979592, | |
| "eval_loss": 0.941472589969635, | |
| "eval_runtime": 18.251, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.3224489795918366, | |
| "grad_norm": 0.39941656589508057, | |
| "learning_rate": 0.0003587980797295671, | |
| "loss": 0.251, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.3387755102040817, | |
| "grad_norm": 0.2876632511615753, | |
| "learning_rate": 0.0003575677595532904, | |
| "loss": 0.208, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.3387755102040817, | |
| "eval_loss": 0.932316780090332, | |
| "eval_runtime": 18.2326, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.3551020408163266, | |
| "grad_norm": 0.30556565523147583, | |
| "learning_rate": 0.0003563215149650505, | |
| "loss": 0.2821, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 1.3714285714285714, | |
| "grad_norm": 0.3039940595626831, | |
| "learning_rate": 0.00035505947191518316, | |
| "loss": 0.2177, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.3714285714285714, | |
| "eval_loss": 0.9171479344367981, | |
| "eval_runtime": 18.2397, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.3877551020408163, | |
| "grad_norm": 0.32876476645469666, | |
| "learning_rate": 0.0003537817579506783, | |
| "loss": 0.2142, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.4040816326530612, | |
| "grad_norm": 0.26468542218208313, | |
| "learning_rate": 0.0003524885022022896, | |
| "loss": 0.2293, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.4040816326530612, | |
| "eval_loss": 0.9176874756813049, | |
| "eval_runtime": 18.2553, | |
| "eval_samples_per_second": 0.876, | |
| "eval_steps_per_second": 0.876, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.420408163265306, | |
| "grad_norm": 0.9715031981468201, | |
| "learning_rate": 0.000351179835371484, | |
| "loss": 0.235, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.436734693877551, | |
| "grad_norm": 0.8392952680587769, | |
| "learning_rate": 0.00034985588971723233, | |
| "loss": 0.2307, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.436734693877551, | |
| "eval_loss": 0.9272938370704651, | |
| "eval_runtime": 18.2194, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.453061224489796, | |
| "grad_norm": 0.28850337862968445, | |
| "learning_rate": 0.00034851679904264314, | |
| "loss": 0.2403, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 1.469387755102041, | |
| "grad_norm": 0.46389341354370117, | |
| "learning_rate": 0.00034716269868143956, | |
| "loss": 0.2464, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.469387755102041, | |
| "eval_loss": 0.9267984628677368, | |
| "eval_runtime": 18.2514, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.4857142857142858, | |
| "grad_norm": 0.29597121477127075, | |
| "learning_rate": 0.00034579372548428235, | |
| "loss": 0.2307, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 1.5020408163265306, | |
| "grad_norm": 0.3392711877822876, | |
| "learning_rate": 0.00034441001780493886, | |
| "loss": 0.231, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.5020408163265306, | |
| "eval_loss": 0.9267009496688843, | |
| "eval_runtime": 18.2179, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.5183673469387755, | |
| "grad_norm": 0.2309531569480896, | |
| "learning_rate": 0.00034301171548630063, | |
| "loss": 0.2307, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.5346938775510204, | |
| "grad_norm": 0.28941744565963745, | |
| "learning_rate": 0.0003415989598462506, | |
| "loss": 0.2384, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 1.5346938775510204, | |
| "eval_loss": 0.9336121082305908, | |
| "eval_runtime": 18.2184, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 1.5510204081632653, | |
| "grad_norm": 0.22579389810562134, | |
| "learning_rate": 0.00034017189366338034, | |
| "loss": 0.1949, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.5673469387755103, | |
| "grad_norm": 0.2894729971885681, | |
| "learning_rate": 0.000338730661162561, | |
| "loss": 0.2348, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.5673469387755103, | |
| "eval_loss": 0.9432627558708191, | |
| "eval_runtime": 18.2158, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.583673469387755, | |
| "grad_norm": 0.24218647181987762, | |
| "learning_rate": 0.00033727540800036683, | |
| "loss": 0.2113, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.2695946991443634, | |
| "learning_rate": 0.0003358062812503548, | |
| "loss": 0.2209, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.9510765075683594, | |
| "eval_runtime": 18.2181, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 1.6163265306122447, | |
| "grad_norm": 0.27196425199508667, | |
| "learning_rate": 0.00033432342938820086, | |
| "loss": 0.2366, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.6326530612244898, | |
| "grad_norm": 0.2302471101284027, | |
| "learning_rate": 0.0003328270022766941, | |
| "loss": 0.2099, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.6326530612244898, | |
| "eval_loss": 0.9363111853599548, | |
| "eval_runtime": 18.2175, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.6489795918367347, | |
| "grad_norm": 0.24810455739498138, | |
| "learning_rate": 0.0003313171511505913, | |
| "loss": 0.2196, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 1.6653061224489796, | |
| "grad_norm": 0.25561368465423584, | |
| "learning_rate": 0.0003297940286013326, | |
| "loss": 0.253, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 1.6653061224489796, | |
| "eval_loss": 0.942488431930542, | |
| "eval_runtime": 18.2068, | |
| "eval_samples_per_second": 0.879, | |
| "eval_steps_per_second": 0.879, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 1.6816326530612244, | |
| "grad_norm": 0.22677375376224518, | |
| "learning_rate": 0.0003282577885616198, | |
| "loss": 0.1789, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 1.6979591836734693, | |
| "grad_norm": 0.26009589433670044, | |
| "learning_rate": 0.0003267085862898594, | |
| "loss": 0.2541, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 1.6979591836734693, | |
| "eval_loss": 0.9335595369338989, | |
| "eval_runtime": 18.2507, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.20782141387462616, | |
| "learning_rate": 0.0003251465783544716, | |
| "loss": 0.2308, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.730612244897959, | |
| "grad_norm": 0.22768278419971466, | |
| "learning_rate": 0.0003235719226180669, | |
| "loss": 0.2354, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 1.730612244897959, | |
| "eval_loss": 0.9376662373542786, | |
| "eval_runtime": 18.2284, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 1.7469387755102042, | |
| "grad_norm": 0.23742002248764038, | |
| "learning_rate": 0.00032198477822149185, | |
| "loss": 0.2306, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 1.763265306122449, | |
| "grad_norm": 0.25688832998275757, | |
| "learning_rate": 0.0003203853055677457, | |
| "loss": 0.2816, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 1.763265306122449, | |
| "eval_loss": 0.9434496164321899, | |
| "eval_runtime": 18.2263, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 1.779591836734694, | |
| "grad_norm": 0.2690117359161377, | |
| "learning_rate": 0.00031877366630576946, | |
| "loss": 0.2013, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 1.7959183673469388, | |
| "grad_norm": 0.2317412942647934, | |
| "learning_rate": 0.0003171500233141089, | |
| "loss": 0.2398, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.7959183673469388, | |
| "eval_loss": 0.9329774975776672, | |
| "eval_runtime": 18.2112, | |
| "eval_samples_per_second": 0.879, | |
| "eval_steps_per_second": 0.879, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.8122448979591836, | |
| "grad_norm": 0.2262117713689804, | |
| "learning_rate": 0.0003155145406844535, | |
| "loss": 0.207, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 1.8285714285714287, | |
| "grad_norm": 0.22458741068840027, | |
| "learning_rate": 0.00031386738370505293, | |
| "loss": 0.2093, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 1.8285714285714287, | |
| "eval_loss": 0.9222925901412964, | |
| "eval_runtime": 18.2333, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 1.8448979591836734, | |
| "grad_norm": 0.24273625016212463, | |
| "learning_rate": 0.0003122087188440118, | |
| "loss": 0.2434, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 1.8612244897959185, | |
| "grad_norm": 0.2206314653158188, | |
| "learning_rate": 0.0003105387137324663, | |
| "loss": 0.2056, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 1.8612244897959185, | |
| "eval_loss": 0.9319095015525818, | |
| "eval_runtime": 18.2327, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 1.8775510204081631, | |
| "grad_norm": 0.2606954574584961, | |
| "learning_rate": 0.0003088575371476426, | |
| "loss": 0.2198, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.8938775510204082, | |
| "grad_norm": 0.23156337440013885, | |
| "learning_rate": 0.0003071653589957993, | |
| "loss": 0.1957, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 1.8938775510204082, | |
| "eval_loss": 0.9314719438552856, | |
| "eval_runtime": 18.2411, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 1.910204081632653, | |
| "grad_norm": 0.2253030389547348, | |
| "learning_rate": 0.0003054623502950565, | |
| "loss": 0.266, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 1.926530612244898, | |
| "grad_norm": 0.24510890245437622, | |
| "learning_rate": 0.0003037486831581115, | |
| "loss": 0.2493, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 1.926530612244898, | |
| "eval_loss": 0.926245927810669, | |
| "eval_runtime": 18.2428, | |
| "eval_samples_per_second": 0.877, | |
| "eval_steps_per_second": 0.877, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 1.9428571428571428, | |
| "grad_norm": 0.22151368856430054, | |
| "learning_rate": 0.00030202453077484494, | |
| "loss": 0.2666, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 1.9591836734693877, | |
| "grad_norm": 0.21714863181114197, | |
| "learning_rate": 0.0003002900673948173, | |
| "loss": 0.253, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.9591836734693877, | |
| "eval_loss": 0.9263309240341187, | |
| "eval_runtime": 18.2196, | |
| "eval_samples_per_second": 0.878, | |
| "eval_steps_per_second": 0.878, | |
| "step": 120 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 305, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 30, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4518993974018048e+18, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |