| { | |
| "best_metric": 0.9445157152020526, | |
| "best_model_checkpoint": "results/facebook/hubert-base-ls960/42/checkpoint-2800", | |
| "epoch": 3.872752420470263, | |
| "eval_steps": 400, | |
| "global_step": 2800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13831258644536654, | |
| "grad_norm": 13.194905281066895, | |
| "learning_rate": 0.00017857142857142857, | |
| "loss": 3.2443, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2766251728907331, | |
| "grad_norm": 30.376169204711914, | |
| "learning_rate": 0.00035714285714285714, | |
| "loss": 2.3348, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4149377593360996, | |
| "grad_norm": 27.163475036621094, | |
| "learning_rate": 0.000496031746031746, | |
| "loss": 1.7975, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5532503457814661, | |
| "grad_norm": 36.59109115600586, | |
| "learning_rate": 0.0004761904761904762, | |
| "loss": 2.0247, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5532503457814661, | |
| "eval_accuracy": 0.3563181526619628, | |
| "eval_f1_macro": 0.22129907013787337, | |
| "eval_loss": 1.99261474609375, | |
| "eval_runtime": 12.572, | |
| "eval_samples_per_second": 248.011, | |
| "eval_steps_per_second": 31.021, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6915629322268326, | |
| "grad_norm": 35.70507049560547, | |
| "learning_rate": 0.0004563492063492063, | |
| "loss": 1.7503, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8298755186721992, | |
| "grad_norm": 39.92527389526367, | |
| "learning_rate": 0.0004365079365079365, | |
| "loss": 1.5947, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9681881051175657, | |
| "grad_norm": 56.568458557128906, | |
| "learning_rate": 0.0004166666666666667, | |
| "loss": 1.5551, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1065006915629323, | |
| "grad_norm": 13.754646301269531, | |
| "learning_rate": 0.0003968253968253968, | |
| "loss": 1.4236, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.1065006915629323, | |
| "eval_accuracy": 0.47241821680564466, | |
| "eval_f1_macro": 0.347385003863519, | |
| "eval_loss": 1.1670863628387451, | |
| "eval_runtime": 12.5271, | |
| "eval_samples_per_second": 248.899, | |
| "eval_steps_per_second": 31.132, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.2448132780082988, | |
| "grad_norm": 71.1982421875, | |
| "learning_rate": 0.000376984126984127, | |
| "loss": 1.3711, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.3831258644536653, | |
| "grad_norm": 9.379792213439941, | |
| "learning_rate": 0.00035714285714285714, | |
| "loss": 1.3099, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.5214384508990317, | |
| "grad_norm": 90.52821350097656, | |
| "learning_rate": 0.00033730158730158733, | |
| "loss": 1.2128, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.6597510373443982, | |
| "grad_norm": 42.78731155395508, | |
| "learning_rate": 0.00031746031746031746, | |
| "loss": 1.1781, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.6597510373443982, | |
| "eval_accuracy": 0.5016035920461834, | |
| "eval_f1_macro": 0.37959579544420224, | |
| "eval_loss": 1.1146905422210693, | |
| "eval_runtime": 12.5371, | |
| "eval_samples_per_second": 248.703, | |
| "eval_steps_per_second": 31.108, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.798063623789765, | |
| "grad_norm": 8.767428398132324, | |
| "learning_rate": 0.00029761904761904765, | |
| "loss": 1.0924, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.9363762102351314, | |
| "grad_norm": 35.30613708496094, | |
| "learning_rate": 0.0002777777777777778, | |
| "loss": 1.0372, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.074688796680498, | |
| "grad_norm": 13.835848808288574, | |
| "learning_rate": 0.00025793650793650796, | |
| "loss": 0.9892, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.2130013831258646, | |
| "grad_norm": 22.654861450195312, | |
| "learning_rate": 0.0002380952380952381, | |
| "loss": 0.9702, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.2130013831258646, | |
| "eval_accuracy": 0.5651058370750481, | |
| "eval_f1_macro": 0.4853545303034886, | |
| "eval_loss": 0.9121200442314148, | |
| "eval_runtime": 12.5089, | |
| "eval_samples_per_second": 249.263, | |
| "eval_steps_per_second": 31.178, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.351313969571231, | |
| "grad_norm": 14.196218490600586, | |
| "learning_rate": 0.00021825396825396825, | |
| "loss": 0.9265, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.4896265560165975, | |
| "grad_norm": 16.53809928894043, | |
| "learning_rate": 0.0001984126984126984, | |
| "loss": 0.8034, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.627939142461964, | |
| "grad_norm": 36.136844635009766, | |
| "learning_rate": 0.00017857142857142857, | |
| "loss": 0.6985, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.7662517289073305, | |
| "grad_norm": 17.420682907104492, | |
| "learning_rate": 0.00015873015873015873, | |
| "loss": 0.67, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.7662517289073305, | |
| "eval_accuracy": 0.7514432328415651, | |
| "eval_f1_macro": 0.763988729870444, | |
| "eval_loss": 0.8275340795516968, | |
| "eval_runtime": 12.5495, | |
| "eval_samples_per_second": 248.457, | |
| "eval_steps_per_second": 31.077, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.904564315352697, | |
| "grad_norm": 18.59613800048828, | |
| "learning_rate": 0.0001388888888888889, | |
| "loss": 0.6103, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.0428769017980635, | |
| "grad_norm": 65.93004608154297, | |
| "learning_rate": 0.00011904761904761905, | |
| "loss": 0.4783, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.18118948824343, | |
| "grad_norm": 36.451961517333984, | |
| "learning_rate": 9.92063492063492e-05, | |
| "loss": 0.3912, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.3195020746887964, | |
| "grad_norm": 27.90683364868164, | |
| "learning_rate": 7.936507936507937e-05, | |
| "loss": 0.3462, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.3195020746887964, | |
| "eval_accuracy": 0.9313662604233482, | |
| "eval_f1_macro": 0.9296400757951072, | |
| "eval_loss": 0.31274116039276123, | |
| "eval_runtime": 12.7469, | |
| "eval_samples_per_second": 244.608, | |
| "eval_steps_per_second": 30.596, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.4578146611341634, | |
| "grad_norm": 22.907320022583008, | |
| "learning_rate": 5.9523809523809524e-05, | |
| "loss": 0.2888, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.59612724757953, | |
| "grad_norm": 40.974483489990234, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 0.2711, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.7344398340248963, | |
| "grad_norm": 1.7515863180160522, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.2292, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.872752420470263, | |
| "grad_norm": 26.21529197692871, | |
| "learning_rate": 0.0, | |
| "loss": 0.2277, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.872752420470263, | |
| "eval_accuracy": 0.9445157152020526, | |
| "eval_f1_macro": 0.9452855408900767, | |
| "eval_loss": 0.28722265362739563, | |
| "eval_runtime": 12.7658, | |
| "eval_samples_per_second": 244.245, | |
| "eval_steps_per_second": 30.55, | |
| "step": 2800 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 2800, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 2800, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.253602560365056e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |