| { | |
| "best_metric": 0.8068910256410257, | |
| "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/canine/canine-base-finetuned-pos-ud-Tamil-TTB/checkpoint-1000", | |
| "epoch": 269.2307692307692, | |
| "global_step": 3500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 8e-05, | |
| "loss": 1.509, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 7.946308724832215e-05, | |
| "loss": 0.274, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 23.08, | |
| "learning_rate": 7.89261744966443e-05, | |
| "loss": 0.0679, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 30.77, | |
| "learning_rate": 7.838926174496645e-05, | |
| "loss": 0.0307, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "learning_rate": 7.78523489932886e-05, | |
| "loss": 0.025, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "eval_accuracy": 0.7892628205128205, | |
| "eval_loss": 1.4348915815353394, | |
| "eval_runtime": 0.3441, | |
| "eval_samples_per_second": 232.482, | |
| "eval_steps_per_second": 29.06, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 46.15, | |
| "learning_rate": 7.731543624161075e-05, | |
| "loss": 0.018, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 53.85, | |
| "learning_rate": 7.677852348993288e-05, | |
| "loss": 0.0152, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 61.54, | |
| "learning_rate": 7.624161073825503e-05, | |
| "loss": 0.0113, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 69.23, | |
| "learning_rate": 7.570469798657718e-05, | |
| "loss": 0.0081, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 76.92, | |
| "learning_rate": 7.516778523489933e-05, | |
| "loss": 0.0092, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 76.92, | |
| "eval_accuracy": 0.8068910256410257, | |
| "eval_loss": 1.4408690929412842, | |
| "eval_runtime": 0.3248, | |
| "eval_samples_per_second": 246.316, | |
| "eval_steps_per_second": 30.79, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 84.62, | |
| "learning_rate": 7.463087248322148e-05, | |
| "loss": 0.0069, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 92.31, | |
| "learning_rate": 7.409395973154362e-05, | |
| "loss": 0.0067, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 7.355704697986577e-05, | |
| "loss": 0.0069, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 107.69, | |
| "learning_rate": 7.302013422818792e-05, | |
| "loss": 0.0053, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 115.38, | |
| "learning_rate": 7.248322147651007e-05, | |
| "loss": 0.0046, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 115.38, | |
| "eval_accuracy": 0.780448717948718, | |
| "eval_loss": 1.9734928607940674, | |
| "eval_runtime": 0.3276, | |
| "eval_samples_per_second": 244.204, | |
| "eval_steps_per_second": 30.525, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 123.08, | |
| "learning_rate": 7.194630872483222e-05, | |
| "loss": 0.0049, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 130.77, | |
| "learning_rate": 7.140939597315438e-05, | |
| "loss": 0.0045, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 138.46, | |
| "learning_rate": 7.087248322147653e-05, | |
| "loss": 0.0052, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 146.15, | |
| "learning_rate": 7.033557046979866e-05, | |
| "loss": 0.004, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 153.85, | |
| "learning_rate": 6.979865771812081e-05, | |
| "loss": 0.0036, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 153.85, | |
| "eval_accuracy": 0.7996794871794872, | |
| "eval_loss": 1.8069559335708618, | |
| "eval_runtime": 0.3276, | |
| "eval_samples_per_second": 244.205, | |
| "eval_steps_per_second": 30.526, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 161.54, | |
| "learning_rate": 6.926174496644296e-05, | |
| "loss": 0.0057, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 169.23, | |
| "learning_rate": 6.87248322147651e-05, | |
| "loss": 0.0055, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 176.92, | |
| "learning_rate": 6.818791946308725e-05, | |
| "loss": 0.0049, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 184.62, | |
| "learning_rate": 6.76510067114094e-05, | |
| "loss": 0.0021, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 192.31, | |
| "learning_rate": 6.711409395973155e-05, | |
| "loss": 0.0042, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 192.31, | |
| "eval_accuracy": 0.7780448717948718, | |
| "eval_loss": 1.728083610534668, | |
| "eval_runtime": 0.3279, | |
| "eval_samples_per_second": 243.965, | |
| "eval_steps_per_second": 30.496, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "learning_rate": 6.65771812080537e-05, | |
| "loss": 0.0044, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 207.69, | |
| "learning_rate": 6.604026845637585e-05, | |
| "loss": 0.0035, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 215.38, | |
| "learning_rate": 6.5503355704698e-05, | |
| "loss": 0.0032, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 223.08, | |
| "learning_rate": 6.496644295302014e-05, | |
| "loss": 0.0033, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 230.77, | |
| "learning_rate": 6.442953020134228e-05, | |
| "loss": 0.0027, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 230.77, | |
| "eval_accuracy": 0.8004807692307693, | |
| "eval_loss": 1.849981665611267, | |
| "eval_runtime": 0.3277, | |
| "eval_samples_per_second": 244.159, | |
| "eval_steps_per_second": 30.52, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 238.46, | |
| "learning_rate": 6.389261744966443e-05, | |
| "loss": 0.0027, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 246.15, | |
| "learning_rate": 6.335570469798657e-05, | |
| "loss": 0.0017, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 253.85, | |
| "learning_rate": 6.281879194630872e-05, | |
| "loss": 0.0016, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 261.54, | |
| "learning_rate": 6.228187919463087e-05, | |
| "loss": 0.0016, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 269.23, | |
| "learning_rate": 6.174496644295302e-05, | |
| "loss": 0.0022, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 269.23, | |
| "eval_accuracy": 0.8060897435897436, | |
| "eval_loss": 1.7650340795516968, | |
| "eval_runtime": 0.3308, | |
| "eval_samples_per_second": 241.835, | |
| "eval_steps_per_second": 30.229, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 269.23, | |
| "step": 3500, | |
| "total_flos": 1.768647275040768e+16, | |
| "train_loss": 0.05915338551998139, | |
| "train_runtime": 490.6715, | |
| "train_samples_per_second": 978.251, | |
| "train_steps_per_second": 30.57 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 1154, | |
| "total_flos": 1.768647275040768e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |