| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.073446327683616, | |
| "eval_steps": 25, | |
| "global_step": 34, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0903954802259887, | |
| "grad_norm": 40.79306411743164, | |
| "learning_rate": 5e-05, | |
| "loss": 8.2235, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0903954802259887, | |
| "eval_loss": 8.391592025756836, | |
| "eval_runtime": 1.4474, | |
| "eval_samples_per_second": 34.544, | |
| "eval_steps_per_second": 8.981, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1807909604519774, | |
| "grad_norm": 40.161956787109375, | |
| "learning_rate": 0.0001, | |
| "loss": 8.3582, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.2711864406779661, | |
| "grad_norm": 37.882598876953125, | |
| "learning_rate": 9.978331270024886e-05, | |
| "loss": 6.2219, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.3615819209039548, | |
| "grad_norm": 27.498767852783203, | |
| "learning_rate": 9.913533761814537e-05, | |
| "loss": 2.5466, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.4519774011299435, | |
| "grad_norm": 23.556495666503906, | |
| "learning_rate": 9.80623151079494e-05, | |
| "loss": 1.3454, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5423728813559322, | |
| "grad_norm": 7.306985378265381, | |
| "learning_rate": 9.657457896300791e-05, | |
| "loss": 0.6438, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.632768361581921, | |
| "grad_norm": 5.81193208694458, | |
| "learning_rate": 9.468645689567598e-05, | |
| "loss": 0.4981, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.7231638418079096, | |
| "grad_norm": 5.2688069343566895, | |
| "learning_rate": 9.241613255361455e-05, | |
| "loss": 0.4217, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.8135593220338984, | |
| "grad_norm": 1.8101367950439453, | |
| "learning_rate": 8.978547040132317e-05, | |
| "loss": 0.3688, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.903954802259887, | |
| "grad_norm": 1.2399615049362183, | |
| "learning_rate": 8.681980515339464e-05, | |
| "loss": 0.3541, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.9943502824858758, | |
| "grad_norm": 4.346222877502441, | |
| "learning_rate": 8.354769778736406e-05, | |
| "loss": 0.3879, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.0847457627118644, | |
| "grad_norm": 4.47180700302124, | |
| "learning_rate": 8.000066048588211e-05, | |
| "loss": 0.6776, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.1751412429378532, | |
| "grad_norm": 3.507833480834961, | |
| "learning_rate": 7.62128531571699e-05, | |
| "loss": 0.3694, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.2655367231638417, | |
| "grad_norm": 0.4435901939868927, | |
| "learning_rate": 7.222075445642904e-05, | |
| "loss": 0.3389, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.3559322033898304, | |
| "grad_norm": 3.5629358291625977, | |
| "learning_rate": 6.80628104764508e-05, | |
| "loss": 0.3625, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.4463276836158192, | |
| "grad_norm": 1.2974547147750854, | |
| "learning_rate": 6.377906449072578e-05, | |
| "loss": 0.3493, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.536723163841808, | |
| "grad_norm": 0.8176577091217041, | |
| "learning_rate": 5.941077131483025e-05, | |
| "loss": 0.3511, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.6271186440677967, | |
| "grad_norm": 0.29680728912353516, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 0.3463, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.7175141242937855, | |
| "grad_norm": 0.7749252915382385, | |
| "learning_rate": 5.058922868516978e-05, | |
| "loss": 0.3578, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.807909604519774, | |
| "grad_norm": 0.6304458975791931, | |
| "learning_rate": 4.6220935509274235e-05, | |
| "loss": 0.3435, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.8983050847457628, | |
| "grad_norm": 1.5261859893798828, | |
| "learning_rate": 4.19371895235492e-05, | |
| "loss": 0.3424, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.9887005649717513, | |
| "grad_norm": 0.31700193881988525, | |
| "learning_rate": 3.777924554357096e-05, | |
| "loss": 0.3539, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.07909604519774, | |
| "grad_norm": 5.0429182052612305, | |
| "learning_rate": 3.378714684283011e-05, | |
| "loss": 0.6626, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.169491525423729, | |
| "grad_norm": 0.2609831690788269, | |
| "learning_rate": 2.9999339514117912e-05, | |
| "loss": 0.3353, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.2598870056497176, | |
| "grad_norm": 0.5319680571556091, | |
| "learning_rate": 2.645230221263596e-05, | |
| "loss": 0.3373, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.2598870056497176, | |
| "eval_loss": 0.35731905698776245, | |
| "eval_runtime": 1.4835, | |
| "eval_samples_per_second": 33.704, | |
| "eval_steps_per_second": 8.763, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.3502824858757063, | |
| "grad_norm": 1.6147388219833374, | |
| "learning_rate": 2.3180194846605367e-05, | |
| "loss": 0.3578, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.440677966101695, | |
| "grad_norm": 0.581788182258606, | |
| "learning_rate": 2.0214529598676836e-05, | |
| "loss": 0.3577, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.5310734463276834, | |
| "grad_norm": 1.152259111404419, | |
| "learning_rate": 1.758386744638546e-05, | |
| "loss": 0.3402, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.621468926553672, | |
| "grad_norm": 0.992540717124939, | |
| "learning_rate": 1.531354310432403e-05, | |
| "loss": 0.3486, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.711864406779661, | |
| "grad_norm": 1.0127880573272705, | |
| "learning_rate": 1.3425421036992098e-05, | |
| "loss": 0.3496, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.8022598870056497, | |
| "grad_norm": 0.23949310183525085, | |
| "learning_rate": 1.1937684892050604e-05, | |
| "loss": 0.3407, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.8926553672316384, | |
| "grad_norm": 0.46344828605651855, | |
| "learning_rate": 1.0864662381854632e-05, | |
| "loss": 0.3595, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.983050847457627, | |
| "grad_norm": 0.9140104651451111, | |
| "learning_rate": 1.0216687299751144e-05, | |
| "loss": 0.3471, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.073446327683616, | |
| "grad_norm": 1.3315057754516602, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6613, | |
| "step": 34 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 34, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0808353705112371e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |