| { | |
| "best_metric": 0.7166422009468079, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-2_60-samples_config-4/checkpoint-143", | |
| "epoch": 56.69565217391305, | |
| "eval_steps": 500, | |
| "global_step": 163, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 0.6984618902206421, | |
| "learning_rate": 3.3333333333333335e-07, | |
| "loss": 1.1182, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.6601165533065796, | |
| "learning_rate": 6.666666666666667e-07, | |
| "loss": 1.0749, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "eval_loss": 1.0965907573699951, | |
| "eval_runtime": 12.9785, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 0.7221220135688782, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 1.0739, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "eval_loss": 1.0941747426986694, | |
| "eval_runtime": 12.9817, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 0.6844056844711304, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.0805, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 0.7226310968399048, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 1.0883, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "eval_loss": 1.0905098915100098, | |
| "eval_runtime": 12.9814, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 0.6965072751045227, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.0572, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "eval_loss": 1.084372878074646, | |
| "eval_runtime": 12.9774, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.6885063648223877, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.0867, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.7193288803100586, | |
| "learning_rate": 4.666666666666667e-06, | |
| "loss": 1.0814, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 1.0741270780563354, | |
| "eval_runtime": 12.9811, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.6680200099945068, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 1.0423, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 1.0622462034225464, | |
| "eval_runtime": 12.9802, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.669226348400116, | |
| "learning_rate": 6e-06, | |
| "loss": 1.0248, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 0.6525329947471619, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.0626, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 1.0461969375610352, | |
| "eval_runtime": 12.9785, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.6599993705749512, | |
| "learning_rate": 7.333333333333333e-06, | |
| "loss": 1.0118, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.0248109102249146, | |
| "eval_runtime": 12.9803, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.6169376969337463, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.0176, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "eval_loss": 1.0099031925201416, | |
| "eval_runtime": 12.9784, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.6581985950469971, | |
| "learning_rate": 8.666666666666668e-06, | |
| "loss": 1.0046, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.5997404456138611, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 0.9728, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "eval_loss": 0.9821717143058777, | |
| "eval_runtime": 12.9796, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.5669417977333069, | |
| "learning_rate": 1e-05, | |
| "loss": 0.9567, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "eval_loss": 0.9526691436767578, | |
| "eval_runtime": 12.98, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.5114875435829163, | |
| "learning_rate": 9.99864620589731e-06, | |
| "loss": 0.9557, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 0.4725043773651123, | |
| "learning_rate": 9.994585556692624e-06, | |
| "loss": 0.9202, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "eval_loss": 0.9259089827537537, | |
| "eval_runtime": 12.9791, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 0.4576855003833771, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 0.9099, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 0.9014627933502197, | |
| "eval_runtime": 12.9828, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 13.217391304347826, | |
| "grad_norm": 0.3902890384197235, | |
| "learning_rate": 9.978353953249023e-06, | |
| "loss": 0.87, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 0.3489598333835602, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 0.8806, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_loss": 0.8827962279319763, | |
| "eval_runtime": 12.9816, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 14.608695652173914, | |
| "grad_norm": 0.32044491171836853, | |
| "learning_rate": 9.951340343707852e-06, | |
| "loss": 0.7975, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 14.956521739130435, | |
| "eval_loss": 0.8661110997200012, | |
| "eval_runtime": 12.9821, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 15.304347826086957, | |
| "grad_norm": 0.2890640199184418, | |
| "learning_rate": 9.933807660562898e-06, | |
| "loss": 0.8663, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.26249009370803833, | |
| "learning_rate": 9.913603233532067e-06, | |
| "loss": 0.8572, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.8533071875572205, | |
| "eval_runtime": 12.9792, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 16.695652173913043, | |
| "grad_norm": 0.3019982874393463, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.8342, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 16.695652173913043, | |
| "eval_loss": 0.8447268009185791, | |
| "eval_runtime": 12.9818, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 17.391304347826086, | |
| "grad_norm": 0.3173466920852661, | |
| "learning_rate": 9.86522435289912e-06, | |
| "loss": 0.8242, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 17.73913043478261, | |
| "eval_loss": 0.8330853581428528, | |
| "eval_runtime": 12.9877, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 18.08695652173913, | |
| "grad_norm": 0.2738870084285736, | |
| "learning_rate": 9.83707609731432e-06, | |
| "loss": 0.788, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 18.782608695652176, | |
| "grad_norm": 0.27857592701911926, | |
| "learning_rate": 9.806308479691595e-06, | |
| "loss": 0.7954, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 18.782608695652176, | |
| "eval_loss": 0.8223147988319397, | |
| "eval_runtime": 12.9802, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 19.47826086956522, | |
| "grad_norm": 0.2908647656440735, | |
| "learning_rate": 9.77293816123866e-06, | |
| "loss": 0.8235, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 19.82608695652174, | |
| "eval_loss": 0.8121759295463562, | |
| "eval_runtime": 12.98, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 20.17391304347826, | |
| "grad_norm": 0.26751863956451416, | |
| "learning_rate": 9.736983212571646e-06, | |
| "loss": 0.7656, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 20.869565217391305, | |
| "grad_norm": 0.25762423872947693, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.7896, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 20.869565217391305, | |
| "eval_loss": 0.8017132878303528, | |
| "eval_runtime": 12.9778, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 21.565217391304348, | |
| "grad_norm": 0.2695617973804474, | |
| "learning_rate": 9.657398694630713e-06, | |
| "loss": 0.7775, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 21.91304347826087, | |
| "eval_loss": 0.7932851910591125, | |
| "eval_runtime": 12.9771, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 22.26086956521739, | |
| "grad_norm": 0.2556411027908325, | |
| "learning_rate": 9.613812221777212e-06, | |
| "loss": 0.7745, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 22.956521739130434, | |
| "grad_norm": 0.23346319794654846, | |
| "learning_rate": 9.567727288213005e-06, | |
| "loss": 0.7315, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 22.956521739130434, | |
| "eval_loss": 0.7861952185630798, | |
| "eval_runtime": 12.9804, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 23.652173913043477, | |
| "grad_norm": 0.22954562306404114, | |
| "learning_rate": 9.519168849742603e-06, | |
| "loss": 0.7702, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.7799764275550842, | |
| "eval_runtime": 12.9807, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 24.347826086956523, | |
| "grad_norm": 0.21833382546901703, | |
| "learning_rate": 9.468163201617063e-06, | |
| "loss": 0.7262, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 24.695652173913043, | |
| "eval_loss": 0.7756370902061462, | |
| "eval_runtime": 12.9905, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 25.043478260869566, | |
| "grad_norm": 0.23636947572231293, | |
| "learning_rate": 9.414737964294636e-06, | |
| "loss": 0.749, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 25.73913043478261, | |
| "grad_norm": 0.21684788167476654, | |
| "learning_rate": 9.358922068483813e-06, | |
| "loss": 0.7683, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 25.73913043478261, | |
| "eval_loss": 0.7715494632720947, | |
| "eval_runtime": 12.9818, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 26.434782608695652, | |
| "grad_norm": 0.2162179797887802, | |
| "learning_rate": 9.30074573947683e-06, | |
| "loss": 0.7043, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 26.782608695652176, | |
| "eval_loss": 0.7655512690544128, | |
| "eval_runtime": 12.9823, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 27.130434782608695, | |
| "grad_norm": 0.2316512167453766, | |
| "learning_rate": 9.24024048078213e-06, | |
| "loss": 0.7234, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 27.82608695652174, | |
| "grad_norm": 0.21031486988067627, | |
| "learning_rate": 9.177439057064684e-06, | |
| "loss": 0.7314, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 27.82608695652174, | |
| "eval_loss": 0.7621185779571533, | |
| "eval_runtime": 12.9789, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 28.52173913043478, | |
| "grad_norm": 0.20802505314350128, | |
| "learning_rate": 9.112375476403313e-06, | |
| "loss": 0.7093, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 28.869565217391305, | |
| "eval_loss": 0.7585732936859131, | |
| "eval_runtime": 12.9843, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 29.217391304347824, | |
| "grad_norm": 0.20854762196540833, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 0.7138, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 29.91304347826087, | |
| "grad_norm": 0.21915678679943085, | |
| "learning_rate": 8.97560398247424e-06, | |
| "loss": 0.7047, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 29.91304347826087, | |
| "eval_loss": 0.7542468905448914, | |
| "eval_runtime": 12.9809, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 30.608695652173914, | |
| "grad_norm": 0.20012685656547546, | |
| "learning_rate": 8.903970133383297e-06, | |
| "loss": 0.707, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 30.956521739130434, | |
| "eval_loss": 0.7506493926048279, | |
| "eval_runtime": 12.9843, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 31.304347826086957, | |
| "grad_norm": 0.2392660677433014, | |
| "learning_rate": 8.83022221559489e-06, | |
| "loss": 0.6921, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 0.20580413937568665, | |
| "learning_rate": 8.754400164907496e-06, | |
| "loss": 0.7128, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 0.7475277781486511, | |
| "eval_runtime": 12.9807, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 32.69565217391305, | |
| "grad_norm": 0.21809720993041992, | |
| "learning_rate": 8.676545040299145e-06, | |
| "loss": 0.676, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 32.69565217391305, | |
| "eval_loss": 0.7451496124267578, | |
| "eval_runtime": 12.9817, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 33.391304347826086, | |
| "grad_norm": 0.22367319464683533, | |
| "learning_rate": 8.596699001693257e-06, | |
| "loss": 0.7113, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 33.73913043478261, | |
| "eval_loss": 0.741963803768158, | |
| "eval_runtime": 12.988, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 34.08695652173913, | |
| "grad_norm": 0.22299408912658691, | |
| "learning_rate": 8.51490528712831e-06, | |
| "loss": 0.68, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 34.78260869565217, | |
| "grad_norm": 0.23509928584098816, | |
| "learning_rate": 8.43120818934367e-06, | |
| "loss": 0.6733, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 34.78260869565217, | |
| "eval_loss": 0.7395928502082825, | |
| "eval_runtime": 12.9789, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 35.47826086956522, | |
| "grad_norm": 0.22088173031806946, | |
| "learning_rate": 8.345653031794292e-06, | |
| "loss": 0.698, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 35.82608695652174, | |
| "eval_loss": 0.7369885444641113, | |
| "eval_runtime": 12.9801, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 36.17391304347826, | |
| "grad_norm": 0.2378576695919037, | |
| "learning_rate": 8.258286144107277e-06, | |
| "loss": 0.6446, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 36.869565217391305, | |
| "grad_norm": 0.22731085121631622, | |
| "learning_rate": 8.16915483699355e-06, | |
| "loss": 0.6868, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 36.869565217391305, | |
| "eval_loss": 0.7338590621948242, | |
| "eval_runtime": 12.9803, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 37.56521739130435, | |
| "grad_norm": 0.21938204765319824, | |
| "learning_rate": 8.078307376628292e-06, | |
| "loss": 0.6633, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 37.91304347826087, | |
| "eval_loss": 0.7310268878936768, | |
| "eval_runtime": 12.9795, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 38.26086956521739, | |
| "grad_norm": 0.21967856585979462, | |
| "learning_rate": 7.985792958513932e-06, | |
| "loss": 0.6498, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 38.95652173913044, | |
| "grad_norm": 0.22135666012763977, | |
| "learning_rate": 7.891661680839932e-06, | |
| "loss": 0.675, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 38.95652173913044, | |
| "eval_loss": 0.7296469211578369, | |
| "eval_runtime": 12.9811, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 39.65217391304348, | |
| "grad_norm": 0.2276638001203537, | |
| "learning_rate": 7.795964517353734e-06, | |
| "loss": 0.6563, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 0.7270292639732361, | |
| "eval_runtime": 12.9804, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 40.34782608695652, | |
| "grad_norm": 0.23463626205921173, | |
| "learning_rate": 7.698753289757565e-06, | |
| "loss": 0.64, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 40.69565217391305, | |
| "eval_loss": 0.725705087184906, | |
| "eval_runtime": 12.9813, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 41.04347826086956, | |
| "grad_norm": 0.22769178450107574, | |
| "learning_rate": 7.600080639646077e-06, | |
| "loss": 0.657, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 41.73913043478261, | |
| "grad_norm": 0.22011993825435638, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.6314, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 41.73913043478261, | |
| "eval_loss": 0.7242138981819153, | |
| "eval_runtime": 12.9801, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 42.43478260869565, | |
| "grad_norm": 0.2390049695968628, | |
| "learning_rate": 7.398565566251232e-06, | |
| "loss": 0.619, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 42.78260869565217, | |
| "eval_loss": 0.7225276827812195, | |
| "eval_runtime": 12.9807, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 43.130434782608695, | |
| "grad_norm": 0.2800259292125702, | |
| "learning_rate": 7.295832266935059e-06, | |
| "loss": 0.6532, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 43.82608695652174, | |
| "grad_norm": 0.22223666310310364, | |
| "learning_rate": 7.191855733945388e-06, | |
| "loss": 0.6256, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 43.82608695652174, | |
| "eval_loss": 0.7211093902587891, | |
| "eval_runtime": 12.9815, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 44.52173913043478, | |
| "grad_norm": 0.2679072320461273, | |
| "learning_rate": 7.08669227240909e-06, | |
| "loss": 0.634, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 44.869565217391305, | |
| "eval_loss": 0.7197514176368713, | |
| "eval_runtime": 12.9829, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 45.21739130434783, | |
| "grad_norm": 0.2466752678155899, | |
| "learning_rate": 6.980398830195785e-06, | |
| "loss": 0.6245, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 45.91304347826087, | |
| "grad_norm": 0.2207140475511551, | |
| "learning_rate": 6.873032967079562e-06, | |
| "loss": 0.5984, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 45.91304347826087, | |
| "eval_loss": 0.7184772491455078, | |
| "eval_runtime": 12.9845, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 46.608695652173914, | |
| "grad_norm": 0.24495841562747955, | |
| "learning_rate": 6.7646528235693445e-06, | |
| "loss": 0.636, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 46.95652173913044, | |
| "eval_loss": 0.7176054120063782, | |
| "eval_runtime": 12.9825, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 47.30434782608695, | |
| "grad_norm": 0.24388910830020905, | |
| "learning_rate": 6.655317089424791e-06, | |
| "loss": 0.6088, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "grad_norm": 0.2629048824310303, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 0.6084, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 0.7173163890838623, | |
| "eval_runtime": 12.9805, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 48.69565217391305, | |
| "grad_norm": 0.23480680584907532, | |
| "learning_rate": 6.434016163555452e-06, | |
| "loss": 0.6068, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 48.69565217391305, | |
| "eval_loss": 0.716775119304657, | |
| "eval_runtime": 12.9805, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 49.391304347826086, | |
| "grad_norm": 0.24330304563045502, | |
| "learning_rate": 6.322170810186013e-06, | |
| "loss": 0.5982, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 49.73913043478261, | |
| "eval_loss": 0.7166422009468079, | |
| "eval_runtime": 12.9831, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 50.08695652173913, | |
| "grad_norm": 0.2731475234031677, | |
| "learning_rate": 6.209609477998339e-06, | |
| "loss": 0.5933, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 50.78260869565217, | |
| "grad_norm": 0.28438258171081543, | |
| "learning_rate": 6.0963931209395165e-06, | |
| "loss": 0.6024, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 50.78260869565217, | |
| "eval_loss": 0.7170538306236267, | |
| "eval_runtime": 12.9821, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 51.47826086956522, | |
| "grad_norm": 0.26069560647010803, | |
| "learning_rate": 5.982583047664151e-06, | |
| "loss": 0.5876, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 51.82608695652174, | |
| "eval_loss": 0.7170079350471497, | |
| "eval_runtime": 12.9793, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 52.17391304347826, | |
| "grad_norm": 0.30849406123161316, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.5845, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 52.869565217391305, | |
| "grad_norm": 0.2836209833621979, | |
| "learning_rate": 5.753428561247416e-06, | |
| "loss": 0.5852, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 52.869565217391305, | |
| "eval_loss": 0.7168818116188049, | |
| "eval_runtime": 12.9839, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 53.56521739130435, | |
| "grad_norm": 0.2691332995891571, | |
| "learning_rate": 5.638208239302975e-06, | |
| "loss": 0.5803, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 53.91304347826087, | |
| "eval_loss": 0.7174960970878601, | |
| "eval_runtime": 12.9787, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 54.26086956521739, | |
| "grad_norm": 0.2947712242603302, | |
| "learning_rate": 5.522642316338268e-06, | |
| "loss": 0.5606, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 54.95652173913044, | |
| "grad_norm": 0.28288403153419495, | |
| "learning_rate": 5.406793373339292e-06, | |
| "loss": 0.5794, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 54.95652173913044, | |
| "eval_loss": 0.717235803604126, | |
| "eval_runtime": 12.9824, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 55.65217391304348, | |
| "grad_norm": 0.2836114168167114, | |
| "learning_rate": 5.290724144552379e-06, | |
| "loss": 0.5699, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_loss": 0.7187849879264832, | |
| "eval_runtime": 12.9808, | |
| "eval_samples_per_second": 0.924, | |
| "eval_steps_per_second": 0.924, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 56.34782608695652, | |
| "grad_norm": 0.30464258790016174, | |
| "learning_rate": 5.174497483512506e-06, | |
| "loss": 0.5722, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 56.69565217391305, | |
| "eval_loss": 0.719237744808197, | |
| "eval_runtime": 12.9776, | |
| "eval_samples_per_second": 0.925, | |
| "eval_steps_per_second": 0.925, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 56.69565217391305, | |
| "step": 163, | |
| "total_flos": 2.5340798505582592e+17, | |
| "train_loss": 0.767207163243206, | |
| "train_runtime": 9109.3845, | |
| "train_samples_per_second": 0.757, | |
| "train_steps_per_second": 0.033 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 150, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 7, | |
| "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": 2.5340798505582592e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |