| { | |
| "best_metric": 0.9354831576347351, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-1_60-samples_config-4_full/checkpoint-287", | |
| "epoch": 104.34782608695652, | |
| "eval_steps": 500, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 2.058788776397705, | |
| "learning_rate": 3.3333333333333335e-07, | |
| "loss": 2.5057, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 2.0099380016326904, | |
| "learning_rate": 6.666666666666667e-07, | |
| "loss": 2.5391, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "eval_loss": 2.416827917098999, | |
| "eval_runtime": 4.8263, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 2.0508503913879395, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 2.5182, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "eval_loss": 2.4065160751342773, | |
| "eval_runtime": 4.8317, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 1.9739177227020264, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 2.5073, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 1.7424209117889404, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 2.4879, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "eval_loss": 2.3912572860717773, | |
| "eval_runtime": 4.8313, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 1.7085793018341064, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 2.4947, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "eval_loss": 2.372018575668335, | |
| "eval_runtime": 4.8224, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 1.7649006843566895, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.4911, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 1.5436650514602661, | |
| "learning_rate": 4.666666666666667e-06, | |
| "loss": 2.4335, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 2.3479087352752686, | |
| "eval_runtime": 4.8287, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 1.5427696704864502, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 2.4424, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 2.310939073562622, | |
| "eval_runtime": 4.8306, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 1.6715421676635742, | |
| "learning_rate": 6e-06, | |
| "loss": 2.3843, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 1.8118067979812622, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 2.3698, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 2.2672276496887207, | |
| "eval_runtime": 4.8229, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 1.769518494606018, | |
| "learning_rate": 7.333333333333333e-06, | |
| "loss": 2.3512, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.2129247188568115, | |
| "eval_runtime": 4.8207, | |
| "eval_samples_per_second": 2.489, | |
| "eval_steps_per_second": 2.489, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 1.4598925113677979, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 2.32, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "eval_loss": 2.182978391647339, | |
| "eval_runtime": 4.8234, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 1.3492408990859985, | |
| "learning_rate": 8.666666666666668e-06, | |
| "loss": 2.2635, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 1.289293646812439, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 2.2555, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "eval_loss": 2.12662935256958, | |
| "eval_runtime": 4.8292, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 1.3162202835083008, | |
| "learning_rate": 1e-05, | |
| "loss": 2.1681, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "eval_loss": 2.0537397861480713, | |
| "eval_runtime": 4.8349, | |
| "eval_samples_per_second": 2.482, | |
| "eval_steps_per_second": 2.482, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 1.3937517404556274, | |
| "learning_rate": 9.99864620589731e-06, | |
| "loss": 2.1476, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 1.0678613185882568, | |
| "learning_rate": 9.994585556692624e-06, | |
| "loss": 2.0737, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "eval_loss": 1.9880471229553223, | |
| "eval_runtime": 4.8264, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 1.7729694843292236, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 2.0403, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 1.9276618957519531, | |
| "eval_runtime": 4.8258, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 13.217391304347826, | |
| "grad_norm": 1.1538618803024292, | |
| "learning_rate": 9.978353953249023e-06, | |
| "loss": 2.005, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 0.9281337261199951, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 1.9476, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_loss": 1.871132254600525, | |
| "eval_runtime": 4.8266, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 14.608695652173914, | |
| "grad_norm": 0.8572821617126465, | |
| "learning_rate": 9.951340343707852e-06, | |
| "loss": 1.9204, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 14.956521739130435, | |
| "eval_loss": 1.8154711723327637, | |
| "eval_runtime": 4.8222, | |
| "eval_samples_per_second": 2.489, | |
| "eval_steps_per_second": 2.489, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 15.304347826086957, | |
| "grad_norm": 0.8229424953460693, | |
| "learning_rate": 9.933807660562898e-06, | |
| "loss": 1.8583, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.8312278985977173, | |
| "learning_rate": 9.913603233532067e-06, | |
| "loss": 1.8461, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.7615426778793335, | |
| "eval_runtime": 4.8193, | |
| "eval_samples_per_second": 2.49, | |
| "eval_steps_per_second": 2.49, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 16.695652173913043, | |
| "grad_norm": 0.8826591372489929, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 1.8095, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 16.695652173913043, | |
| "eval_loss": 1.7235876321792603, | |
| "eval_runtime": 4.8227, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 17.391304347826086, | |
| "grad_norm": 0.9071628451347351, | |
| "learning_rate": 9.86522435289912e-06, | |
| "loss": 1.7597, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 17.73913043478261, | |
| "eval_loss": 1.6579965353012085, | |
| "eval_runtime": 4.8233, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 18.08695652173913, | |
| "grad_norm": 0.9293504357337952, | |
| "learning_rate": 9.83707609731432e-06, | |
| "loss": 1.6805, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 18.782608695652176, | |
| "grad_norm": 0.8878646492958069, | |
| "learning_rate": 9.806308479691595e-06, | |
| "loss": 1.6484, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 18.782608695652176, | |
| "eval_loss": 1.5918583869934082, | |
| "eval_runtime": 4.8278, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 19.47826086956522, | |
| "grad_norm": 1.0276228189468384, | |
| "learning_rate": 9.77293816123866e-06, | |
| "loss": 1.6443, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 19.82608695652174, | |
| "eval_loss": 1.5261772871017456, | |
| "eval_runtime": 4.8278, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 20.17391304347826, | |
| "grad_norm": 1.2222720384597778, | |
| "learning_rate": 9.736983212571646e-06, | |
| "loss": 1.5605, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 20.869565217391305, | |
| "grad_norm": 1.040975570678711, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 1.5204, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 20.869565217391305, | |
| "eval_loss": 1.4560518264770508, | |
| "eval_runtime": 4.8273, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 21.565217391304348, | |
| "grad_norm": 0.8610687255859375, | |
| "learning_rate": 9.657398694630713e-06, | |
| "loss": 1.463, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 21.91304347826087, | |
| "eval_loss": 1.3959513902664185, | |
| "eval_runtime": 4.8279, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 22.26086956521739, | |
| "grad_norm": 0.9347313642501831, | |
| "learning_rate": 9.613812221777212e-06, | |
| "loss": 1.4249, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 22.956521739130434, | |
| "grad_norm": 0.911578357219696, | |
| "learning_rate": 9.567727288213005e-06, | |
| "loss": 1.3833, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 22.956521739130434, | |
| "eval_loss": 1.3404473066329956, | |
| "eval_runtime": 4.8261, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 23.652173913043477, | |
| "grad_norm": 0.8542620539665222, | |
| "learning_rate": 9.519168849742603e-06, | |
| "loss": 1.3385, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 1.2874761819839478, | |
| "eval_runtime": 4.8231, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 24.347826086956523, | |
| "grad_norm": 0.90443354845047, | |
| "learning_rate": 9.468163201617063e-06, | |
| "loss": 1.3094, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 24.695652173913043, | |
| "eval_loss": 1.2503851652145386, | |
| "eval_runtime": 4.8253, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 25.043478260869566, | |
| "grad_norm": 0.8763517141342163, | |
| "learning_rate": 9.414737964294636e-06, | |
| "loss": 1.2694, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 25.73913043478261, | |
| "grad_norm": 0.7862639427185059, | |
| "learning_rate": 9.358922068483813e-06, | |
| "loss": 1.2303, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 25.73913043478261, | |
| "eval_loss": 1.200731873512268, | |
| "eval_runtime": 4.8257, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 26.434782608695652, | |
| "grad_norm": 0.7432467341423035, | |
| "learning_rate": 9.30074573947683e-06, | |
| "loss": 1.1677, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 26.782608695652176, | |
| "eval_loss": 1.1599854230880737, | |
| "eval_runtime": 4.8291, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 27.130434782608695, | |
| "grad_norm": 0.6284440159797668, | |
| "learning_rate": 9.24024048078213e-06, | |
| "loss": 1.175, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 27.82608695652174, | |
| "grad_norm": 0.5942031145095825, | |
| "learning_rate": 9.177439057064684e-06, | |
| "loss": 1.1674, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 27.82608695652174, | |
| "eval_loss": 1.1331721544265747, | |
| "eval_runtime": 4.8209, | |
| "eval_samples_per_second": 2.489, | |
| "eval_steps_per_second": 2.489, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 28.52173913043478, | |
| "grad_norm": 0.5919143557548523, | |
| "learning_rate": 9.112375476403313e-06, | |
| "loss": 1.1068, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 28.869565217391305, | |
| "eval_loss": 1.1100302934646606, | |
| "eval_runtime": 4.8285, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 29.217391304347824, | |
| "grad_norm": 0.653171181678772, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 1.0762, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 29.91304347826087, | |
| "grad_norm": 0.6148922443389893, | |
| "learning_rate": 8.97560398247424e-06, | |
| "loss": 1.104, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 29.91304347826087, | |
| "eval_loss": 1.0883766412734985, | |
| "eval_runtime": 4.8276, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 30.608695652173914, | |
| "grad_norm": 0.5734871029853821, | |
| "learning_rate": 8.903970133383297e-06, | |
| "loss": 1.0617, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 30.956521739130434, | |
| "eval_loss": 1.0717229843139648, | |
| "eval_runtime": 4.8304, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 31.304347826086957, | |
| "grad_norm": 0.5561685562133789, | |
| "learning_rate": 8.83022221559489e-06, | |
| "loss": 1.0674, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "grad_norm": 0.522966206073761, | |
| "learning_rate": 8.754400164907496e-06, | |
| "loss": 1.0354, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 1.057722568511963, | |
| "eval_runtime": 4.8513, | |
| "eval_samples_per_second": 2.474, | |
| "eval_steps_per_second": 2.474, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 32.69565217391305, | |
| "grad_norm": 0.46999314427375793, | |
| "learning_rate": 8.676545040299145e-06, | |
| "loss": 1.0195, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 32.69565217391305, | |
| "eval_loss": 1.049912452697754, | |
| "eval_runtime": 4.823, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 33.391304347826086, | |
| "grad_norm": 0.49655812978744507, | |
| "learning_rate": 8.596699001693257e-06, | |
| "loss": 1.0659, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 33.73913043478261, | |
| "eval_loss": 1.0396418571472168, | |
| "eval_runtime": 4.8306, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 34.08695652173913, | |
| "grad_norm": 0.44858285784721375, | |
| "learning_rate": 8.51490528712831e-06, | |
| "loss": 0.9917, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 34.78260869565217, | |
| "grad_norm": 0.44484949111938477, | |
| "learning_rate": 8.43120818934367e-06, | |
| "loss": 1.0118, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 34.78260869565217, | |
| "eval_loss": 1.031006932258606, | |
| "eval_runtime": 4.8275, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 35.47826086956522, | |
| "grad_norm": 0.3746640086174011, | |
| "learning_rate": 8.345653031794292e-06, | |
| "loss": 1.0009, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 35.82608695652174, | |
| "eval_loss": 1.0246673822402954, | |
| "eval_runtime": 4.8365, | |
| "eval_samples_per_second": 2.481, | |
| "eval_steps_per_second": 2.481, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 36.17391304347826, | |
| "grad_norm": 0.4000430107116699, | |
| "learning_rate": 8.258286144107277e-06, | |
| "loss": 0.9993, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 36.869565217391305, | |
| "grad_norm": 0.44598138332366943, | |
| "learning_rate": 8.16915483699355e-06, | |
| "loss": 0.9938, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 36.869565217391305, | |
| "eval_loss": 1.0180528163909912, | |
| "eval_runtime": 4.8232, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 37.56521739130435, | |
| "grad_norm": 0.40717804431915283, | |
| "learning_rate": 8.078307376628292e-06, | |
| "loss": 0.9736, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 37.91304347826087, | |
| "eval_loss": 1.0123671293258667, | |
| "eval_runtime": 4.8227, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 38.26086956521739, | |
| "grad_norm": 0.4021812975406647, | |
| "learning_rate": 7.985792958513932e-06, | |
| "loss": 0.9937, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 38.95652173913044, | |
| "grad_norm": 0.3653680086135864, | |
| "learning_rate": 7.891661680839932e-06, | |
| "loss": 0.9888, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 38.95652173913044, | |
| "eval_loss": 1.0076369047164917, | |
| "eval_runtime": 4.8197, | |
| "eval_samples_per_second": 2.49, | |
| "eval_steps_per_second": 2.49, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 39.65217391304348, | |
| "grad_norm": 0.36845943331718445, | |
| "learning_rate": 7.795964517353734e-06, | |
| "loss": 0.9637, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 1.0018855333328247, | |
| "eval_runtime": 4.8188, | |
| "eval_samples_per_second": 2.49, | |
| "eval_steps_per_second": 2.49, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 40.34782608695652, | |
| "grad_norm": 0.3739126920700073, | |
| "learning_rate": 7.698753289757565e-06, | |
| "loss": 0.9769, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 40.69565217391305, | |
| "eval_loss": 0.998671293258667, | |
| "eval_runtime": 4.8236, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 41.04347826086956, | |
| "grad_norm": 0.36004525423049927, | |
| "learning_rate": 7.600080639646077e-06, | |
| "loss": 0.9648, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 41.73913043478261, | |
| "grad_norm": 0.3854582607746124, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.936, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 41.73913043478261, | |
| "eval_loss": 0.9939180016517639, | |
| "eval_runtime": 4.8226, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 42.43478260869565, | |
| "grad_norm": 0.3872157335281372, | |
| "learning_rate": 7.398565566251232e-06, | |
| "loss": 0.9863, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 42.78260869565217, | |
| "eval_loss": 0.9906224608421326, | |
| "eval_runtime": 4.8249, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 43.130434782608695, | |
| "grad_norm": 0.45275992155075073, | |
| "learning_rate": 7.295832266935059e-06, | |
| "loss": 0.9381, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 43.82608695652174, | |
| "grad_norm": 0.3777986466884613, | |
| "learning_rate": 7.191855733945388e-06, | |
| "loss": 0.9626, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 43.82608695652174, | |
| "eval_loss": 0.9862757325172424, | |
| "eval_runtime": 4.8256, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 44.52173913043478, | |
| "grad_norm": 0.3554269075393677, | |
| "learning_rate": 7.08669227240909e-06, | |
| "loss": 0.9438, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 44.869565217391305, | |
| "eval_loss": 0.9825396537780762, | |
| "eval_runtime": 4.8278, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 45.21739130434783, | |
| "grad_norm": 0.4316815435886383, | |
| "learning_rate": 6.980398830195785e-06, | |
| "loss": 0.9621, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 45.91304347826087, | |
| "grad_norm": 0.3970458209514618, | |
| "learning_rate": 6.873032967079562e-06, | |
| "loss": 0.9034, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 45.91304347826087, | |
| "eval_loss": 0.980421781539917, | |
| "eval_runtime": 4.8261, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 46.608695652173914, | |
| "grad_norm": 0.36759161949157715, | |
| "learning_rate": 6.7646528235693445e-06, | |
| "loss": 0.9398, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 46.95652173913044, | |
| "eval_loss": 0.976270854473114, | |
| "eval_runtime": 4.8284, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 47.30434782608695, | |
| "grad_norm": 0.4169844388961792, | |
| "learning_rate": 6.655317089424791e-06, | |
| "loss": 0.9681, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "grad_norm": 0.3838629126548767, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 0.9206, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 0.9739949703216553, | |
| "eval_runtime": 4.8201, | |
| "eval_samples_per_second": 2.49, | |
| "eval_steps_per_second": 2.49, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 48.69565217391305, | |
| "grad_norm": 0.43010860681533813, | |
| "learning_rate": 6.434016163555452e-06, | |
| "loss": 0.9251, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 48.69565217391305, | |
| "eval_loss": 0.9728464484214783, | |
| "eval_runtime": 4.8239, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 49.391304347826086, | |
| "grad_norm": 0.3847574293613434, | |
| "learning_rate": 6.322170810186013e-06, | |
| "loss": 0.9245, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 49.73913043478261, | |
| "eval_loss": 0.9704022407531738, | |
| "eval_runtime": 4.8259, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 50.08695652173913, | |
| "grad_norm": 0.519916832447052, | |
| "learning_rate": 6.209609477998339e-06, | |
| "loss": 0.9363, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 50.78260869565217, | |
| "grad_norm": 0.46963444352149963, | |
| "learning_rate": 6.0963931209395165e-06, | |
| "loss": 0.9332, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 50.78260869565217, | |
| "eval_loss": 0.9670591354370117, | |
| "eval_runtime": 4.8259, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 51.47826086956522, | |
| "grad_norm": 0.39990323781967163, | |
| "learning_rate": 5.982583047664151e-06, | |
| "loss": 0.9012, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 51.82608695652174, | |
| "eval_loss": 0.9651336669921875, | |
| "eval_runtime": 4.822, | |
| "eval_samples_per_second": 2.489, | |
| "eval_steps_per_second": 2.489, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 52.17391304347826, | |
| "grad_norm": 0.6043806076049805, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.915, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 52.869565217391305, | |
| "grad_norm": 0.4192598760128021, | |
| "learning_rate": 5.753428561247416e-06, | |
| "loss": 0.9075, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 52.869565217391305, | |
| "eval_loss": 0.9627308249473572, | |
| "eval_runtime": 4.8216, | |
| "eval_samples_per_second": 2.489, | |
| "eval_steps_per_second": 2.489, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 53.56521739130435, | |
| "grad_norm": 0.42425036430358887, | |
| "learning_rate": 5.638208239302975e-06, | |
| "loss": 0.9031, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 53.91304347826087, | |
| "eval_loss": 0.9614337086677551, | |
| "eval_runtime": 4.8281, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 54.26086956521739, | |
| "grad_norm": 0.49185582995414734, | |
| "learning_rate": 5.522642316338268e-06, | |
| "loss": 0.9089, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 54.95652173913044, | |
| "grad_norm": 0.45466744899749756, | |
| "learning_rate": 5.406793373339292e-06, | |
| "loss": 0.8969, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 54.95652173913044, | |
| "eval_loss": 0.9591866135597229, | |
| "eval_runtime": 4.827, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 55.65217391304348, | |
| "grad_norm": 0.4060223698616028, | |
| "learning_rate": 5.290724144552379e-06, | |
| "loss": 0.9102, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_loss": 0.9583068490028381, | |
| "eval_runtime": 4.8182, | |
| "eval_samples_per_second": 2.491, | |
| "eval_steps_per_second": 2.491, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 56.34782608695652, | |
| "grad_norm": 0.46647414565086365, | |
| "learning_rate": 5.174497483512506e-06, | |
| "loss": 0.8955, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 56.69565217391305, | |
| "eval_loss": 0.9563028216362, | |
| "eval_runtime": 4.8258, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 57.04347826086956, | |
| "grad_norm": 0.5661187767982483, | |
| "learning_rate": 5.0581763290069865e-06, | |
| "loss": 0.9039, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 57.73913043478261, | |
| "grad_norm": 0.4606133699417114, | |
| "learning_rate": 4.941823670993016e-06, | |
| "loss": 0.8775, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 57.73913043478261, | |
| "eval_loss": 0.9546887278556824, | |
| "eval_runtime": 4.8258, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 58.43478260869565, | |
| "grad_norm": 0.4152405261993408, | |
| "learning_rate": 4.825502516487497e-06, | |
| "loss": 0.8879, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 58.78260869565217, | |
| "eval_loss": 0.9539990425109863, | |
| "eval_runtime": 4.8323, | |
| "eval_samples_per_second": 2.483, | |
| "eval_steps_per_second": 2.483, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 59.130434782608695, | |
| "grad_norm": 0.46841880679130554, | |
| "learning_rate": 4.7092758554476215e-06, | |
| "loss": 0.9008, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 59.82608695652174, | |
| "grad_norm": 0.5221750736236572, | |
| "learning_rate": 4.59320662666071e-06, | |
| "loss": 0.8805, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 59.82608695652174, | |
| "eval_loss": 0.9510453343391418, | |
| "eval_runtime": 4.8267, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 60.52173913043478, | |
| "grad_norm": 0.45134589076042175, | |
| "learning_rate": 4.477357683661734e-06, | |
| "loss": 0.8982, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 60.869565217391305, | |
| "eval_loss": 0.9505268931388855, | |
| "eval_runtime": 4.8236, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 61.21739130434783, | |
| "grad_norm": 0.5287215113639832, | |
| "learning_rate": 4.361791760697027e-06, | |
| "loss": 0.8613, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 61.91304347826087, | |
| "grad_norm": 0.4524611532688141, | |
| "learning_rate": 4.246571438752585e-06, | |
| "loss": 0.8897, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 61.91304347826087, | |
| "eval_loss": 0.9494467377662659, | |
| "eval_runtime": 4.8277, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 62.608695652173914, | |
| "grad_norm": 0.4399959146976471, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 0.8515, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 62.95652173913044, | |
| "eval_loss": 0.9478859305381775, | |
| "eval_runtime": 4.8225, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 63.30434782608695, | |
| "grad_norm": 0.5381987690925598, | |
| "learning_rate": 4.017416952335849e-06, | |
| "loss": 0.9205, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "grad_norm": 0.5035319924354553, | |
| "learning_rate": 3.903606879060483e-06, | |
| "loss": 0.8637, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_loss": 0.9468648433685303, | |
| "eval_runtime": 4.8249, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 64.69565217391305, | |
| "grad_norm": 0.4583827555179596, | |
| "learning_rate": 3.790390522001662e-06, | |
| "loss": 0.8719, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 64.69565217391305, | |
| "eval_loss": 0.9470892548561096, | |
| "eval_runtime": 4.824, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 65.3913043478261, | |
| "grad_norm": 0.5146140456199646, | |
| "learning_rate": 3.6778291898139907e-06, | |
| "loss": 0.8635, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 65.73913043478261, | |
| "eval_loss": 0.9452293515205383, | |
| "eval_runtime": 4.8297, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 66.08695652173913, | |
| "grad_norm": 0.5240075588226318, | |
| "learning_rate": 3.5659838364445505e-06, | |
| "loss": 0.8809, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 66.78260869565217, | |
| "grad_norm": 0.4892469048500061, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 0.8579, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 66.78260869565217, | |
| "eval_loss": 0.9444965720176697, | |
| "eval_runtime": 4.8242, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 67.47826086956522, | |
| "grad_norm": 0.5614312291145325, | |
| "learning_rate": 3.3446829105752103e-06, | |
| "loss": 0.8465, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 67.82608695652173, | |
| "eval_loss": 0.9434299468994141, | |
| "eval_runtime": 4.8252, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 68.17391304347827, | |
| "grad_norm": 0.5231534242630005, | |
| "learning_rate": 3.2353471764306567e-06, | |
| "loss": 0.8791, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 68.8695652173913, | |
| "grad_norm": 0.5704526305198669, | |
| "learning_rate": 3.12696703292044e-06, | |
| "loss": 0.8588, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 68.8695652173913, | |
| "eval_loss": 0.9436232447624207, | |
| "eval_runtime": 4.8241, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 69.56521739130434, | |
| "grad_norm": 0.5715120434761047, | |
| "learning_rate": 3.019601169804216e-06, | |
| "loss": 0.868, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 69.91304347826087, | |
| "eval_loss": 0.9421423077583313, | |
| "eval_runtime": 4.8217, | |
| "eval_samples_per_second": 2.489, | |
| "eval_steps_per_second": 2.489, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 70.26086956521739, | |
| "grad_norm": 0.7542048692703247, | |
| "learning_rate": 2.9133077275909112e-06, | |
| "loss": 0.8691, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 70.95652173913044, | |
| "grad_norm": 0.5780174732208252, | |
| "learning_rate": 2.8081442660546126e-06, | |
| "loss": 0.8523, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 70.95652173913044, | |
| "eval_loss": 0.9417908191680908, | |
| "eval_runtime": 4.8236, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 71.65217391304348, | |
| "grad_norm": 0.5677995681762695, | |
| "learning_rate": 2.7041677330649408e-06, | |
| "loss": 0.8654, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_loss": 0.9404106140136719, | |
| "eval_runtime": 4.8254, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 72.34782608695652, | |
| "grad_norm": 0.5781915187835693, | |
| "learning_rate": 2.601434433748771e-06, | |
| "loss": 0.8525, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 72.69565217391305, | |
| "eval_loss": 0.9405445456504822, | |
| "eval_runtime": 4.8227, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 73.04347826086956, | |
| "grad_norm": 0.5282586812973022, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 0.8419, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 73.73913043478261, | |
| "grad_norm": 0.7152076363563538, | |
| "learning_rate": 2.3999193603539234e-06, | |
| "loss": 0.8565, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 73.73913043478261, | |
| "eval_loss": 0.9400214552879333, | |
| "eval_runtime": 4.8268, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 74.43478260869566, | |
| "grad_norm": 0.5548727512359619, | |
| "learning_rate": 2.3012467102424373e-06, | |
| "loss": 0.8424, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 74.78260869565217, | |
| "eval_loss": 0.9406850337982178, | |
| "eval_runtime": 4.8319, | |
| "eval_samples_per_second": 2.483, | |
| "eval_steps_per_second": 2.483, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 75.1304347826087, | |
| "grad_norm": 0.6737377047538757, | |
| "learning_rate": 2.204035482646267e-06, | |
| "loss": 0.8541, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 75.82608695652173, | |
| "grad_norm": 0.5421648621559143, | |
| "learning_rate": 2.1083383191600676e-06, | |
| "loss": 0.8342, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 75.82608695652173, | |
| "eval_loss": 0.9395294189453125, | |
| "eval_runtime": 4.8312, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 76.52173913043478, | |
| "grad_norm": 0.6473485827445984, | |
| "learning_rate": 2.0142070414860704e-06, | |
| "loss": 0.8539, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 76.8695652173913, | |
| "eval_loss": 0.9393353462219238, | |
| "eval_runtime": 4.8273, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 77.21739130434783, | |
| "grad_norm": 0.6593698263168335, | |
| "learning_rate": 1.9216926233717087e-06, | |
| "loss": 0.8531, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 77.91304347826087, | |
| "grad_norm": 0.612883448600769, | |
| "learning_rate": 1.8308451630064484e-06, | |
| "loss": 0.8413, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 77.91304347826087, | |
| "eval_loss": 0.9383335113525391, | |
| "eval_runtime": 4.8256, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 78.6086956521739, | |
| "grad_norm": 0.600571870803833, | |
| "learning_rate": 1.7417138558927244e-06, | |
| "loss": 0.8488, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 78.95652173913044, | |
| "eval_loss": 0.938234806060791, | |
| "eval_runtime": 4.8279, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 227 | |
| }, | |
| { | |
| "epoch": 79.30434782608695, | |
| "grad_norm": 0.5200878977775574, | |
| "learning_rate": 1.6543469682057105e-06, | |
| "loss": 0.8469, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 0.5332892537117004, | |
| "learning_rate": 1.5687918106563326e-06, | |
| "loss": 0.8319, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_loss": 0.9395381808280945, | |
| "eval_runtime": 4.8185, | |
| "eval_samples_per_second": 2.49, | |
| "eval_steps_per_second": 2.49, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 80.69565217391305, | |
| "grad_norm": 0.5492921471595764, | |
| "learning_rate": 1.4850947128716914e-06, | |
| "loss": 0.8402, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 80.69565217391305, | |
| "eval_loss": 0.9382343292236328, | |
| "eval_runtime": 4.8345, | |
| "eval_samples_per_second": 2.482, | |
| "eval_steps_per_second": 2.482, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 81.3913043478261, | |
| "grad_norm": 0.6471025943756104, | |
| "learning_rate": 1.4033009983067454e-06, | |
| "loss": 0.8604, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 81.73913043478261, | |
| "eval_loss": 0.9375772476196289, | |
| "eval_runtime": 4.8261, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 82.08695652173913, | |
| "grad_norm": 0.6458178758621216, | |
| "learning_rate": 1.3234549597008572e-06, | |
| "loss": 0.8182, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 82.78260869565217, | |
| "grad_norm": 0.639596164226532, | |
| "learning_rate": 1.2455998350925042e-06, | |
| "loss": 0.8516, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 82.78260869565217, | |
| "eval_loss": 0.9374153017997742, | |
| "eval_runtime": 4.8221, | |
| "eval_samples_per_second": 2.489, | |
| "eval_steps_per_second": 2.489, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 83.47826086956522, | |
| "grad_norm": 0.5777437686920166, | |
| "learning_rate": 1.1697777844051105e-06, | |
| "loss": 0.8195, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 83.82608695652173, | |
| "eval_loss": 0.9378395676612854, | |
| "eval_runtime": 4.822, | |
| "eval_samples_per_second": 2.489, | |
| "eval_steps_per_second": 2.489, | |
| "step": 241 | |
| }, | |
| { | |
| "epoch": 84.17391304347827, | |
| "grad_norm": 0.6369026303291321, | |
| "learning_rate": 1.096029866616704e-06, | |
| "loss": 0.8491, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 84.8695652173913, | |
| "grad_norm": 0.6049113273620605, | |
| "learning_rate": 1.0243960175257605e-06, | |
| "loss": 0.8456, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 84.8695652173913, | |
| "eval_loss": 0.9380698204040527, | |
| "eval_runtime": 4.8285, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 85.56521739130434, | |
| "grad_norm": 0.6273989677429199, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 0.8313, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 85.91304347826087, | |
| "eval_loss": 0.9373624324798584, | |
| "eval_runtime": 4.8293, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 86.26086956521739, | |
| "grad_norm": 0.6432866454124451, | |
| "learning_rate": 8.876245235966884e-07, | |
| "loss": 0.8236, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 86.95652173913044, | |
| "grad_norm": 0.6188175082206726, | |
| "learning_rate": 8.225609429353187e-07, | |
| "loss": 0.8415, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 86.95652173913044, | |
| "eval_loss": 0.9369013905525208, | |
| "eval_runtime": 4.8226, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 87.65217391304348, | |
| "grad_norm": 0.5841935276985168, | |
| "learning_rate": 7.597595192178702e-07, | |
| "loss": 0.8318, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_loss": 0.9364583492279053, | |
| "eval_runtime": 4.8193, | |
| "eval_samples_per_second": 2.49, | |
| "eval_steps_per_second": 2.49, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 88.34782608695652, | |
| "grad_norm": 0.6125827431678772, | |
| "learning_rate": 6.992542605231739e-07, | |
| "loss": 0.8271, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 88.69565217391305, | |
| "eval_loss": 0.9369893670082092, | |
| "eval_runtime": 4.8228, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 89.04347826086956, | |
| "grad_norm": 0.5938271284103394, | |
| "learning_rate": 6.410779315161885e-07, | |
| "loss": 0.8479, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 89.73913043478261, | |
| "grad_norm": 0.6235251426696777, | |
| "learning_rate": 5.852620357053651e-07, | |
| "loss": 0.8361, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 89.73913043478261, | |
| "eval_loss": 0.9364290237426758, | |
| "eval_runtime": 4.8251, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 90.43478260869566, | |
| "grad_norm": 0.6185177564620972, | |
| "learning_rate": 5.318367983829393e-07, | |
| "loss": 0.8216, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 90.78260869565217, | |
| "eval_loss": 0.9365130066871643, | |
| "eval_runtime": 4.8285, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 261 | |
| }, | |
| { | |
| "epoch": 91.1304347826087, | |
| "grad_norm": 0.6606243848800659, | |
| "learning_rate": 4.808311502573976e-07, | |
| "loss": 0.8269, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 91.82608695652173, | |
| "grad_norm": 0.6094285845756531, | |
| "learning_rate": 4.322727117869951e-07, | |
| "loss": 0.8387, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 91.82608695652173, | |
| "eval_loss": 0.936617910861969, | |
| "eval_runtime": 4.8322, | |
| "eval_samples_per_second": 2.483, | |
| "eval_steps_per_second": 2.483, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 92.52173913043478, | |
| "grad_norm": 0.6140238642692566, | |
| "learning_rate": 3.8618777822278854e-07, | |
| "loss": 0.8457, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 92.8695652173913, | |
| "eval_loss": 0.9366480708122253, | |
| "eval_runtime": 4.826, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 267 | |
| }, | |
| { | |
| "epoch": 93.21739130434783, | |
| "grad_norm": 0.7491313219070435, | |
| "learning_rate": 3.426013053692878e-07, | |
| "loss": 0.822, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 93.91304347826087, | |
| "grad_norm": 0.6457171440124512, | |
| "learning_rate": 3.015368960704584e-07, | |
| "loss": 0.8491, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 93.91304347826087, | |
| "eval_loss": 0.9367270469665527, | |
| "eval_runtime": 4.8289, | |
| "eval_samples_per_second": 2.485, | |
| "eval_steps_per_second": 2.485, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 94.6086956521739, | |
| "grad_norm": 0.6033779382705688, | |
| "learning_rate": 2.63016787428354e-07, | |
| "loss": 0.8171, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 94.95652173913044, | |
| "eval_loss": 0.9357192516326904, | |
| "eval_runtime": 4.8259, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 95.30434782608695, | |
| "grad_norm": 0.6807251572608948, | |
| "learning_rate": 2.2706183876134047e-07, | |
| "loss": 0.8514, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "grad_norm": 0.6259418725967407, | |
| "learning_rate": 1.9369152030840553e-07, | |
| "loss": 0.8168, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_loss": 0.9367218017578125, | |
| "eval_runtime": 4.8235, | |
| "eval_samples_per_second": 2.488, | |
| "eval_steps_per_second": 2.488, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 96.69565217391305, | |
| "grad_norm": 0.673136293888092, | |
| "learning_rate": 1.6292390268568103e-07, | |
| "loss": 0.8161, | |
| "step": 278 | |
| }, | |
| { | |
| "epoch": 96.69565217391305, | |
| "eval_loss": 0.9363929629325867, | |
| "eval_runtime": 4.8214, | |
| "eval_samples_per_second": 2.489, | |
| "eval_steps_per_second": 2.489, | |
| "step": 278 | |
| }, | |
| { | |
| "epoch": 97.3913043478261, | |
| "grad_norm": 0.5515984296798706, | |
| "learning_rate": 1.3477564710088097e-07, | |
| "loss": 0.8442, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 97.73913043478261, | |
| "eval_loss": 0.9356262683868408, | |
| "eval_runtime": 4.8255, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 281 | |
| }, | |
| { | |
| "epoch": 98.08695652173913, | |
| "grad_norm": 0.6790144443511963, | |
| "learning_rate": 1.0926199633097156e-07, | |
| "loss": 0.8332, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 98.78260869565217, | |
| "grad_norm": 0.5827218890190125, | |
| "learning_rate": 8.639676646793382e-08, | |
| "loss": 0.8388, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 98.78260869565217, | |
| "eval_loss": 0.936260461807251, | |
| "eval_runtime": 4.8222, | |
| "eval_samples_per_second": 2.489, | |
| "eval_steps_per_second": 2.489, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 99.47826086956522, | |
| "grad_norm": 0.5916593670845032, | |
| "learning_rate": 6.61923394371039e-08, | |
| "loss": 0.8365, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 99.82608695652173, | |
| "eval_loss": 0.9354831576347351, | |
| "eval_runtime": 4.831, | |
| "eval_samples_per_second": 2.484, | |
| "eval_steps_per_second": 2.484, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 100.17391304347827, | |
| "grad_norm": 0.9281457662582397, | |
| "learning_rate": 4.865965629214819e-08, | |
| "loss": 0.8107, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 100.8695652173913, | |
| "grad_norm": 0.6146421432495117, | |
| "learning_rate": 3.3808211290284886e-08, | |
| "loss": 0.8493, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 100.8695652173913, | |
| "eval_loss": 0.9359912276268005, | |
| "eval_runtime": 4.826, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 101.56521739130434, | |
| "grad_norm": 0.6010165214538574, | |
| "learning_rate": 2.1646046750978255e-08, | |
| "loss": 0.8267, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 101.91304347826087, | |
| "eval_loss": 0.9354932904243469, | |
| "eval_runtime": 4.8205, | |
| "eval_samples_per_second": 2.489, | |
| "eval_steps_per_second": 2.489, | |
| "step": 293 | |
| }, | |
| { | |
| "epoch": 102.26086956521739, | |
| "grad_norm": 0.6717774868011475, | |
| "learning_rate": 1.2179748700879013e-08, | |
| "loss": 0.8293, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 102.95652173913044, | |
| "grad_norm": 0.5756195783615112, | |
| "learning_rate": 5.414443307377171e-09, | |
| "loss": 0.8304, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 102.95652173913044, | |
| "eval_loss": 0.9361304640769958, | |
| "eval_runtime": 4.8264, | |
| "eval_samples_per_second": 2.486, | |
| "eval_steps_per_second": 2.486, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 103.65217391304348, | |
| "grad_norm": 0.6329156160354614, | |
| "learning_rate": 1.3537941026914302e-09, | |
| "loss": 0.8216, | |
| "step": 298 | |
| }, | |
| { | |
| "epoch": 104.0, | |
| "eval_loss": 0.9361245632171631, | |
| "eval_runtime": 4.8193, | |
| "eval_samples_per_second": 2.49, | |
| "eval_steps_per_second": 2.49, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 104.34782608695652, | |
| "grad_norm": 0.6687307953834534, | |
| "learning_rate": 0.0, | |
| "loss": 0.8436, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 104.34782608695652, | |
| "eval_loss": 0.9358415007591248, | |
| "eval_runtime": 4.8451, | |
| "eval_samples_per_second": 2.477, | |
| "eval_steps_per_second": 2.477, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 104.34782608695652, | |
| "step": 300, | |
| "total_flos": 7.184388637222502e+16, | |
| "train_loss": 1.1654362523555755, | |
| "train_runtime": 5630.1569, | |
| "train_samples_per_second": 1.226, | |
| "train_steps_per_second": 0.053 | |
| } | |
| ], | |
| "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": 7.184388637222502e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |