| { | |
| "best_global_step": 30, | |
| "best_metric": 1.3498777151107788, | |
| "best_model_checkpoint": "outputs\\gaialab-naija-adapter-v0.4\\checkpoint-30", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17777777777777778, | |
| "grad_norm": 16.430423736572266, | |
| "learning_rate": 0.0, | |
| "loss": 4.572, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 15.095035552978516, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 4.4529, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "grad_norm": 8.647032737731934, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 4.1566, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 6.524106979370117, | |
| "learning_rate": 0.0002, | |
| "loss": 3.2851, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 5.279809474945068, | |
| "learning_rate": 0.0001925925925925926, | |
| "loss": 3.0107, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.055668354034424, | |
| "learning_rate": 0.0001851851851851852, | |
| "loss": 2.2786, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.4303853511810303, | |
| "eval_runtime": 3.2399, | |
| "eval_samples_per_second": 1.543, | |
| "eval_steps_per_second": 1.543, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.1777777777777778, | |
| "grad_norm": 3.808993101119995, | |
| "learning_rate": 0.00017777777777777779, | |
| "loss": 2.1329, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.3555555555555556, | |
| "grad_norm": 5.183055400848389, | |
| "learning_rate": 0.00017037037037037037, | |
| "loss": 1.9554, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.5333333333333332, | |
| "grad_norm": 4.297043323516846, | |
| "learning_rate": 0.00016296296296296295, | |
| "loss": 1.5536, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.7111111111111112, | |
| "grad_norm": 6.999485969543457, | |
| "learning_rate": 0.00015555555555555556, | |
| "loss": 1.4798, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.8888888888888888, | |
| "grad_norm": 2.94558048248291, | |
| "learning_rate": 0.00014814814814814815, | |
| "loss": 1.2903, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.834883689880371, | |
| "learning_rate": 0.00014074074074074076, | |
| "loss": 1.4292, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.6124086380004883, | |
| "eval_runtime": 2.189, | |
| "eval_samples_per_second": 2.284, | |
| "eval_steps_per_second": 2.284, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.1777777777777776, | |
| "grad_norm": 1.9701344966888428, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 1.2039, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.3555555555555556, | |
| "grad_norm": 1.6061532497406006, | |
| "learning_rate": 0.00012592592592592592, | |
| "loss": 1.1494, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.533333333333333, | |
| "grad_norm": 2.38248348236084, | |
| "learning_rate": 0.00011851851851851852, | |
| "loss": 1.3483, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.7111111111111112, | |
| "grad_norm": 1.7200688123703003, | |
| "learning_rate": 0.00011111111111111112, | |
| "loss": 0.9927, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.888888888888889, | |
| "grad_norm": 1.625122308731079, | |
| "learning_rate": 0.0001037037037037037, | |
| "loss": 1.1136, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.7844679355621338, | |
| "learning_rate": 9.62962962962963e-05, | |
| "loss": 0.9103, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.4323580265045166, | |
| "eval_runtime": 3.0258, | |
| "eval_samples_per_second": 1.652, | |
| "eval_steps_per_second": 1.652, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.1777777777777776, | |
| "grad_norm": 1.5016974210739136, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.8749, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 3.3555555555555556, | |
| "grad_norm": 1.3555611371994019, | |
| "learning_rate": 8.148148148148148e-05, | |
| "loss": 0.9289, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.533333333333333, | |
| "grad_norm": 1.7811331748962402, | |
| "learning_rate": 7.407407407407407e-05, | |
| "loss": 1.099, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 3.7111111111111112, | |
| "grad_norm": 1.5052400827407837, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.8184, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 3.888888888888889, | |
| "grad_norm": 1.7020034790039062, | |
| "learning_rate": 5.925925925925926e-05, | |
| "loss": 0.9686, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.030837297439575, | |
| "learning_rate": 5.185185185185185e-05, | |
| "loss": 0.8103, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.3614479303359985, | |
| "eval_runtime": 2.2939, | |
| "eval_samples_per_second": 2.18, | |
| "eval_steps_per_second": 2.18, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.177777777777778, | |
| "grad_norm": 1.6491711139678955, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.7975, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 4.355555555555555, | |
| "grad_norm": 1.469295859336853, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.8944, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.533333333333333, | |
| "grad_norm": 1.6997331380844116, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.7977, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 4.711111111111111, | |
| "grad_norm": 1.5283434391021729, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.7814, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.888888888888889, | |
| "grad_norm": 1.6913104057312012, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.7772, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.0519070625305176, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.7568, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.3498777151107788, | |
| "eval_runtime": 2.3264, | |
| "eval_samples_per_second": 2.149, | |
| "eval_steps_per_second": 2.149, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 30, | |
| "total_flos": 45576685612800.0, | |
| "train_loss": 1.6206799745559692, | |
| "train_runtime": 362.9377, | |
| "train_samples_per_second": 0.62, | |
| "train_steps_per_second": 0.083 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 45576685612800.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |