| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0031486146095717885, | |
| "eval_steps": 50, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.574307304785894e-05, | |
| "eval_loss": 2.3866379261016846, | |
| "eval_runtime": 1500.9459, | |
| "eval_samples_per_second": 17.819, | |
| "eval_steps_per_second": 8.91, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 7.871536523929471e-05, | |
| "grad_norm": 12.448644638061523, | |
| "learning_rate": 5e-05, | |
| "loss": 7.7391, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00015743073047858942, | |
| "grad_norm": 6.404262542724609, | |
| "learning_rate": 0.0001, | |
| "loss": 10.35, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00023614609571788412, | |
| "grad_norm": 6.720837593078613, | |
| "learning_rate": 9.98292246503335e-05, | |
| "loss": 5.6461, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00031486146095717883, | |
| "grad_norm": 5.995464324951172, | |
| "learning_rate": 9.931806517013612e-05, | |
| "loss": 2.7984, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00039357682619647357, | |
| "grad_norm": 9.630922317504883, | |
| "learning_rate": 9.847001329696653e-05, | |
| "loss": 5.0363, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.00047229219143576825, | |
| "grad_norm": 7.160970687866211, | |
| "learning_rate": 9.729086208503174e-05, | |
| "loss": 6.5517, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.000551007556675063, | |
| "grad_norm": 9.611334800720215, | |
| "learning_rate": 9.578866633275288e-05, | |
| "loss": 5.6197, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0006297229219143577, | |
| "grad_norm": 10.17956256866455, | |
| "learning_rate": 9.397368756032445e-05, | |
| "loss": 5.6486, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0007084382871536523, | |
| "grad_norm": 9.616400718688965, | |
| "learning_rate": 9.185832391312644e-05, | |
| "loss": 4.7867, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0007871536523929471, | |
| "grad_norm": 10.657410621643066, | |
| "learning_rate": 8.945702546981969e-05, | |
| "loss": 4.4847, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0007871536523929471, | |
| "eval_loss": 1.1559579372406006, | |
| "eval_runtime": 1500.0771, | |
| "eval_samples_per_second": 17.83, | |
| "eval_steps_per_second": 8.915, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0008658690176322418, | |
| "grad_norm": 8.259451866149902, | |
| "learning_rate": 8.678619553365659e-05, | |
| "loss": 7.1665, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.0009445843828715365, | |
| "grad_norm": 4.17130708694458, | |
| "learning_rate": 8.386407858128706e-05, | |
| "loss": 3.9723, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0010232997481108312, | |
| "grad_norm": 3.682018280029297, | |
| "learning_rate": 8.07106356344834e-05, | |
| "loss": 1.8443, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.001102015113350126, | |
| "grad_norm": 3.572805881500244, | |
| "learning_rate": 7.734740790612136e-05, | |
| "loss": 3.2193, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.0011807304785894208, | |
| "grad_norm": 5.4666523933410645, | |
| "learning_rate": 7.379736965185368e-05, | |
| "loss": 4.6715, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.0012594458438287153, | |
| "grad_norm": 7.5685200691223145, | |
| "learning_rate": 7.008477123264848e-05, | |
| "loss": 5.328, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.0013381612090680101, | |
| "grad_norm": 9.007378578186035, | |
| "learning_rate": 6.623497346023418e-05, | |
| "loss": 4.6684, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.0014168765743073047, | |
| "grad_norm": 5.461382865905762, | |
| "learning_rate": 6.227427435703997e-05, | |
| "loss": 4.786, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0014955919395465995, | |
| "grad_norm": 9.335121154785156, | |
| "learning_rate": 5.8229729514036705e-05, | |
| "loss": 5.2938, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.0015743073047858943, | |
| "grad_norm": 11.640975952148438, | |
| "learning_rate": 5.4128967273616625e-05, | |
| "loss": 4.8757, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0015743073047858943, | |
| "eval_loss": 1.0214067697525024, | |
| "eval_runtime": 1500.7708, | |
| "eval_samples_per_second": 17.822, | |
| "eval_steps_per_second": 8.911, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.0016530226700251888, | |
| "grad_norm": 8.590002059936523, | |
| "learning_rate": 5e-05, | |
| "loss": 4.4979, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.0017317380352644836, | |
| "grad_norm": 8.198857307434082, | |
| "learning_rate": 4.5871032726383386e-05, | |
| "loss": 2.8344, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.0018104534005037784, | |
| "grad_norm": 2.7320914268493652, | |
| "learning_rate": 4.17702704859633e-05, | |
| "loss": 2.9306, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.001889168765743073, | |
| "grad_norm": 3.878121852874756, | |
| "learning_rate": 3.772572564296005e-05, | |
| "loss": 2.7627, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.001967884130982368, | |
| "grad_norm": 9.245368957519531, | |
| "learning_rate": 3.3765026539765834e-05, | |
| "loss": 4.5011, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.0020465994962216624, | |
| "grad_norm": 7.411713123321533, | |
| "learning_rate": 2.991522876735154e-05, | |
| "loss": 5.2115, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.0021253148614609574, | |
| "grad_norm": 8.136418342590332, | |
| "learning_rate": 2.6202630348146324e-05, | |
| "loss": 4.8772, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.002204030226700252, | |
| "grad_norm": 8.70052719116211, | |
| "learning_rate": 2.2652592093878666e-05, | |
| "loss": 5.0952, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.0022827455919395465, | |
| "grad_norm": 7.519270420074463, | |
| "learning_rate": 1.928936436551661e-05, | |
| "loss": 4.8706, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.0023614609571788415, | |
| "grad_norm": 11.668645858764648, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 4.718, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.0023614609571788415, | |
| "eval_loss": 0.9602340459823608, | |
| "eval_runtime": 1507.1015, | |
| "eval_samples_per_second": 17.747, | |
| "eval_steps_per_second": 8.873, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.002440176322418136, | |
| "grad_norm": 6.727120876312256, | |
| "learning_rate": 1.3213804466343421e-05, | |
| "loss": 5.3968, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.0025188916876574307, | |
| "grad_norm": 2.8705267906188965, | |
| "learning_rate": 1.0542974530180327e-05, | |
| "loss": 2.9229, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.0025976070528967252, | |
| "grad_norm": 4.095298767089844, | |
| "learning_rate": 8.141676086873572e-06, | |
| "loss": 1.7476, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.0026763224181360202, | |
| "grad_norm": 5.578668117523193, | |
| "learning_rate": 6.026312439675552e-06, | |
| "loss": 3.0731, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.002755037783375315, | |
| "grad_norm": 3.9415435791015625, | |
| "learning_rate": 4.2113336672471245e-06, | |
| "loss": 3.6369, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.0028337531486146094, | |
| "grad_norm": 12.799013137817383, | |
| "learning_rate": 2.7091379149682685e-06, | |
| "loss": 4.7291, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.0029124685138539044, | |
| "grad_norm": 11.411933898925781, | |
| "learning_rate": 1.5299867030334814e-06, | |
| "loss": 4.3032, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.002991183879093199, | |
| "grad_norm": 8.253731727600098, | |
| "learning_rate": 6.819348298638839e-07, | |
| "loss": 4.2849, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.0030698992443324935, | |
| "grad_norm": 9.577661514282227, | |
| "learning_rate": 1.7077534966650766e-07, | |
| "loss": 4.268, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.0031486146095717885, | |
| "grad_norm": 10.62493896484375, | |
| "learning_rate": 0.0, | |
| "loss": 4.8146, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.0031486146095717885, | |
| "eval_loss": 0.9478885531425476, | |
| "eval_runtime": 1500.0374, | |
| "eval_samples_per_second": 17.83, | |
| "eval_steps_per_second": 8.915, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.951789291130061e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |