| { | |
| "best_metric": 0.8465077877044678, | |
| "best_model_checkpoint": "outputs/checkpoint-92", | |
| "epoch": 1.0, | |
| "eval_steps": 23, | |
| "global_step": 92, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010869565217391304, | |
| "grad_norm": 0.23393695056438446, | |
| "learning_rate": 0.001, | |
| "loss": 2.6154, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.021739130434782608, | |
| "grad_norm": 0.4464223086833954, | |
| "learning_rate": 0.001, | |
| "loss": 2.3194, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.03260869565217391, | |
| "grad_norm": 1.6278090476989746, | |
| "learning_rate": 0.001, | |
| "loss": 2.1998, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.043478260869565216, | |
| "grad_norm": 0.32962527871131897, | |
| "learning_rate": 0.001, | |
| "loss": 1.9167, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.05434782608695652, | |
| "grad_norm": 0.4716239273548126, | |
| "learning_rate": 0.001, | |
| "loss": 1.7546, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.06521739130434782, | |
| "grad_norm": 0.38779595494270325, | |
| "learning_rate": 0.001, | |
| "loss": 1.6069, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.07608695652173914, | |
| "grad_norm": 0.3187870681285858, | |
| "learning_rate": 0.001, | |
| "loss": 1.5629, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.08695652173913043, | |
| "grad_norm": 0.27891236543655396, | |
| "learning_rate": 0.001, | |
| "loss": 1.4564, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.09782608695652174, | |
| "grad_norm": 0.2857877314090729, | |
| "learning_rate": 0.001, | |
| "loss": 1.4014, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.10869565217391304, | |
| "grad_norm": 0.2176678329706192, | |
| "learning_rate": 0.001, | |
| "loss": 1.3423, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11956521739130435, | |
| "grad_norm": 0.27025488018989563, | |
| "learning_rate": 0.001, | |
| "loss": 1.3101, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.13043478260869565, | |
| "grad_norm": 0.2788834571838379, | |
| "learning_rate": 0.001, | |
| "loss": 1.2676, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.14130434782608695, | |
| "grad_norm": 0.24594657123088837, | |
| "learning_rate": 0.001, | |
| "loss": 1.3237, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.15217391304347827, | |
| "grad_norm": 0.2215934842824936, | |
| "learning_rate": 0.001, | |
| "loss": 1.2517, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.16304347826086957, | |
| "grad_norm": 0.1776692122220993, | |
| "learning_rate": 0.001, | |
| "loss": 1.2289, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 0.18647059798240662, | |
| "learning_rate": 0.001, | |
| "loss": 1.2342, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.18478260869565216, | |
| "grad_norm": 0.2081366926431656, | |
| "learning_rate": 0.001, | |
| "loss": 1.2139, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.1956521739130435, | |
| "grad_norm": 0.18122343719005585, | |
| "learning_rate": 0.001, | |
| "loss": 1.1598, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.20652173913043478, | |
| "grad_norm": 0.19013595581054688, | |
| "learning_rate": 0.001, | |
| "loss": 1.2007, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.21739130434782608, | |
| "grad_norm": 0.17870813608169556, | |
| "learning_rate": 0.001, | |
| "loss": 1.1708, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.22826086956521738, | |
| "grad_norm": 0.19605986773967743, | |
| "learning_rate": 0.001, | |
| "loss": 1.0883, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.2391304347826087, | |
| "grad_norm": 0.1825353056192398, | |
| "learning_rate": 0.001, | |
| "loss": 1.1113, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.20768754184246063, | |
| "learning_rate": 0.001, | |
| "loss": 1.1057, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 1.0819175243377686, | |
| "eval_runtime": 39.6272, | |
| "eval_samples_per_second": 29.55, | |
| "eval_steps_per_second": 3.71, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.2608695652173913, | |
| "grad_norm": 0.24947208166122437, | |
| "learning_rate": 0.001, | |
| "loss": 1.0427, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2717391304347826, | |
| "grad_norm": 0.41416075825691223, | |
| "learning_rate": 0.001, | |
| "loss": 1.0768, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2826086956521739, | |
| "grad_norm": 3.9600319862365723, | |
| "learning_rate": 0.001, | |
| "loss": 1.069, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.29347826086956524, | |
| "grad_norm": 0.8276275992393494, | |
| "learning_rate": 0.001, | |
| "loss": 1.1223, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.30434782608695654, | |
| "grad_norm": 0.23127484321594238, | |
| "learning_rate": 0.001, | |
| "loss": 1.0725, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.31521739130434784, | |
| "grad_norm": 0.22777508199214935, | |
| "learning_rate": 0.001, | |
| "loss": 1.1025, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.32608695652173914, | |
| "grad_norm": 0.23111239075660706, | |
| "learning_rate": 0.001, | |
| "loss": 1.0722, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.33695652173913043, | |
| "grad_norm": 0.24532677233219147, | |
| "learning_rate": 0.001, | |
| "loss": 1.1259, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 0.7043253183364868, | |
| "learning_rate": 0.001, | |
| "loss": 1.0502, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.358695652173913, | |
| "grad_norm": 0.22311310470104218, | |
| "learning_rate": 0.001, | |
| "loss": 0.9918, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.3695652173913043, | |
| "grad_norm": 0.2729714810848236, | |
| "learning_rate": 0.001, | |
| "loss": 1.0666, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.3804347826086957, | |
| "grad_norm": 0.1948792189359665, | |
| "learning_rate": 0.001, | |
| "loss": 0.9834, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.391304347826087, | |
| "grad_norm": 0.17322953045368195, | |
| "learning_rate": 0.001, | |
| "loss": 0.9522, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.40217391304347827, | |
| "grad_norm": 0.15274113416671753, | |
| "learning_rate": 0.001, | |
| "loss": 0.9644, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.41304347826086957, | |
| "grad_norm": 0.18090985715389252, | |
| "learning_rate": 0.001, | |
| "loss": 0.991, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.42391304347826086, | |
| "grad_norm": 0.1719120740890503, | |
| "learning_rate": 0.001, | |
| "loss": 0.9659, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.43478260869565216, | |
| "grad_norm": 0.17336900532245636, | |
| "learning_rate": 0.001, | |
| "loss": 1.0173, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.44565217391304346, | |
| "grad_norm": 0.16952969133853912, | |
| "learning_rate": 0.001, | |
| "loss": 0.9946, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.45652173913043476, | |
| "grad_norm": 0.1436886191368103, | |
| "learning_rate": 0.001, | |
| "loss": 0.9536, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.4673913043478261, | |
| "grad_norm": 0.16588632762432098, | |
| "learning_rate": 0.001, | |
| "loss": 0.9723, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.4782608695652174, | |
| "grad_norm": 0.1483495533466339, | |
| "learning_rate": 0.001, | |
| "loss": 0.9297, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.4891304347826087, | |
| "grad_norm": 0.1516825258731842, | |
| "learning_rate": 0.001, | |
| "loss": 0.9911, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.16181206703186035, | |
| "learning_rate": 0.001, | |
| "loss": 0.9688, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.9558140635490417, | |
| "eval_runtime": 39.5819, | |
| "eval_samples_per_second": 29.584, | |
| "eval_steps_per_second": 3.714, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.5108695652173914, | |
| "grad_norm": 0.14223167300224304, | |
| "learning_rate": 0.001, | |
| "loss": 0.8418, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.5217391304347826, | |
| "grad_norm": 0.14156702160835266, | |
| "learning_rate": 0.001, | |
| "loss": 0.868, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.532608695652174, | |
| "grad_norm": 0.1668156236410141, | |
| "learning_rate": 0.001, | |
| "loss": 0.9643, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.5434782608695652, | |
| "grad_norm": 0.14512571692466736, | |
| "learning_rate": 0.001, | |
| "loss": 0.9395, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5543478260869565, | |
| "grad_norm": 0.1591370701789856, | |
| "learning_rate": 0.001, | |
| "loss": 0.9467, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.5652173913043478, | |
| "grad_norm": 0.15379436314105988, | |
| "learning_rate": 0.001, | |
| "loss": 0.984, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.5760869565217391, | |
| "grad_norm": 0.15107646584510803, | |
| "learning_rate": 0.001, | |
| "loss": 0.924, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.5869565217391305, | |
| "grad_norm": 0.14596650004386902, | |
| "learning_rate": 0.001, | |
| "loss": 0.933, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.5978260869565217, | |
| "grad_norm": 0.13844196498394012, | |
| "learning_rate": 0.001, | |
| "loss": 0.8861, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.6086956521739131, | |
| "grad_norm": 0.1443144679069519, | |
| "learning_rate": 0.001, | |
| "loss": 0.9239, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.6195652173913043, | |
| "grad_norm": 0.1431627869606018, | |
| "learning_rate": 0.001, | |
| "loss": 0.9399, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.6304347826086957, | |
| "grad_norm": 0.13777939975261688, | |
| "learning_rate": 0.001, | |
| "loss": 0.8712, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.6413043478260869, | |
| "grad_norm": 0.14304274320602417, | |
| "learning_rate": 0.001, | |
| "loss": 0.977, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.6521739130434783, | |
| "grad_norm": 0.13079452514648438, | |
| "learning_rate": 0.001, | |
| "loss": 0.8102, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.6630434782608695, | |
| "grad_norm": 0.13848628103733063, | |
| "learning_rate": 0.001, | |
| "loss": 0.8854, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.6739130434782609, | |
| "grad_norm": 0.13032200932502747, | |
| "learning_rate": 0.001, | |
| "loss": 0.9307, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.6847826086956522, | |
| "grad_norm": 0.13883714377880096, | |
| "learning_rate": 0.001, | |
| "loss": 0.9464, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.12983089685440063, | |
| "learning_rate": 0.001, | |
| "loss": 0.9065, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.7065217391304348, | |
| "grad_norm": 0.12657804787158966, | |
| "learning_rate": 0.001, | |
| "loss": 0.8866, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.717391304347826, | |
| "grad_norm": 0.14905163645744324, | |
| "learning_rate": 0.001, | |
| "loss": 0.9583, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.7282608695652174, | |
| "grad_norm": 0.1388324350118637, | |
| "learning_rate": 0.001, | |
| "loss": 0.7862, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.7391304347826086, | |
| "grad_norm": 0.13726989924907684, | |
| "learning_rate": 0.001, | |
| "loss": 0.8841, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.14547140896320343, | |
| "learning_rate": 0.001, | |
| "loss": 0.9234, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.8918075561523438, | |
| "eval_runtime": 39.653, | |
| "eval_samples_per_second": 29.531, | |
| "eval_steps_per_second": 3.707, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.7608695652173914, | |
| "grad_norm": 0.12735989689826965, | |
| "learning_rate": 0.001, | |
| "loss": 0.8529, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.7717391304347826, | |
| "grad_norm": 0.12790235877037048, | |
| "learning_rate": 0.001, | |
| "loss": 0.8211, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.782608695652174, | |
| "grad_norm": 0.14512644708156586, | |
| "learning_rate": 0.001, | |
| "loss": 0.9544, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.7934782608695652, | |
| "grad_norm": 0.1390400230884552, | |
| "learning_rate": 0.001, | |
| "loss": 0.8758, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.8043478260869565, | |
| "grad_norm": 0.12742573022842407, | |
| "learning_rate": 0.001, | |
| "loss": 0.8937, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.8152173913043478, | |
| "grad_norm": 0.1397847682237625, | |
| "learning_rate": 0.001, | |
| "loss": 0.9319, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.8260869565217391, | |
| "grad_norm": 0.13563187420368195, | |
| "learning_rate": 0.001, | |
| "loss": 0.9594, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.8369565217391305, | |
| "grad_norm": 0.13903594017028809, | |
| "learning_rate": 0.001, | |
| "loss": 0.8688, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.8478260869565217, | |
| "grad_norm": 0.12712402641773224, | |
| "learning_rate": 0.001, | |
| "loss": 0.8964, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.8586956521739131, | |
| "grad_norm": 0.131251260638237, | |
| "learning_rate": 0.001, | |
| "loss": 0.858, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "grad_norm": 0.1342822164297104, | |
| "learning_rate": 0.001, | |
| "loss": 0.8941, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.8804347826086957, | |
| "grad_norm": 0.1401877999305725, | |
| "learning_rate": 0.001, | |
| "loss": 0.8575, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.8913043478260869, | |
| "grad_norm": 0.13211235404014587, | |
| "learning_rate": 0.001, | |
| "loss": 0.8475, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.9021739130434783, | |
| "grad_norm": 0.13683001697063446, | |
| "learning_rate": 0.001, | |
| "loss": 0.8534, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.9130434782608695, | |
| "grad_norm": 0.13229840993881226, | |
| "learning_rate": 0.001, | |
| "loss": 0.8567, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.9239130434782609, | |
| "grad_norm": 0.12229685485363007, | |
| "learning_rate": 0.001, | |
| "loss": 0.7974, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.9347826086956522, | |
| "grad_norm": 0.14769427478313446, | |
| "learning_rate": 0.001, | |
| "loss": 0.8962, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.9456521739130435, | |
| "grad_norm": 0.12334488332271576, | |
| "learning_rate": 0.001, | |
| "loss": 0.817, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.9565217391304348, | |
| "grad_norm": 0.11790772527456284, | |
| "learning_rate": 0.001, | |
| "loss": 0.8305, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.967391304347826, | |
| "grad_norm": 0.15138116478919983, | |
| "learning_rate": 0.001, | |
| "loss": 0.8873, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.9782608695652174, | |
| "grad_norm": 0.14076858758926392, | |
| "learning_rate": 0.001, | |
| "loss": 0.841, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.9891304347826086, | |
| "grad_norm": 0.15599556267261505, | |
| "learning_rate": 0.001, | |
| "loss": 0.9173, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.19630737602710724, | |
| "learning_rate": 0.001, | |
| "loss": 0.8713, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.8465077877044678, | |
| "eval_runtime": 39.5565, | |
| "eval_samples_per_second": 29.603, | |
| "eval_steps_per_second": 3.716, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 92, | |
| "total_flos": 2.3790959179333632e+17, | |
| "train_loss": 1.0686528734538867, | |
| "train_runtime": 1550.4974, | |
| "train_samples_per_second": 7.55, | |
| "train_steps_per_second": 0.059 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 92, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 23, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.3790959179333632e+17, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |