| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.6797488226059656, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01674515960230246, | |
| "eval_loss": 1.4724388122558594, | |
| "eval_runtime": 73.0963, | |
| "eval_samples_per_second": 2.763, | |
| "eval_steps_per_second": 1.382, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05023547880690738, | |
| "grad_norm": 102.90454864501953, | |
| "learning_rate": 3e-05, | |
| "loss": 45.0108, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.10047095761381476, | |
| "grad_norm": 66.23411560058594, | |
| "learning_rate": 6e-05, | |
| "loss": 42.1127, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.15070643642072212, | |
| "grad_norm": 49.03892517089844, | |
| "learning_rate": 9e-05, | |
| "loss": 25.533, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.15070643642072212, | |
| "eval_loss": 0.48065412044525146, | |
| "eval_runtime": 73.1001, | |
| "eval_samples_per_second": 2.763, | |
| "eval_steps_per_second": 1.382, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.20094191522762953, | |
| "grad_norm": 40.06981658935547, | |
| "learning_rate": 0.00012, | |
| "loss": 12.9041, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.25117739403453687, | |
| "grad_norm": 18.526214599609375, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 8.4587, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.30141287284144425, | |
| "grad_norm": 26.23703384399414, | |
| "learning_rate": 0.00018, | |
| "loss": 6.8537, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.30141287284144425, | |
| "eval_loss": 0.16974753141403198, | |
| "eval_runtime": 73.0714, | |
| "eval_samples_per_second": 2.764, | |
| "eval_steps_per_second": 1.382, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3516483516483517, | |
| "grad_norm": 25.50685691833496, | |
| "learning_rate": 0.0001999229036240723, | |
| "loss": 5.2731, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.40188383045525905, | |
| "grad_norm": 16.794132232666016, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 3.9976, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.4521193092621664, | |
| "grad_norm": 16.295429229736328, | |
| "learning_rate": 0.00019624552364536473, | |
| "loss": 3.9352, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.4521193092621664, | |
| "eval_loss": 0.1071891188621521, | |
| "eval_runtime": 73.2333, | |
| "eval_samples_per_second": 2.758, | |
| "eval_steps_per_second": 1.379, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5023547880690737, | |
| "grad_norm": 12.938387870788574, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 3.7022, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5525902668759811, | |
| "grad_norm": 11.224324226379395, | |
| "learning_rate": 0.00018724960070727972, | |
| "loss": 2.9652, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.6028257456828885, | |
| "grad_norm": 8.89415454864502, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 3.0822, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.6028257456828885, | |
| "eval_loss": 0.09062672406435013, | |
| "eval_runtime": 73.1812, | |
| "eval_samples_per_second": 2.76, | |
| "eval_steps_per_second": 1.38, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.6530612244897959, | |
| "grad_norm": 9.960721969604492, | |
| "learning_rate": 0.00017343225094356855, | |
| "loss": 3.1298, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.7032967032967034, | |
| "grad_norm": 8.25685977935791, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 2.9661, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.7535321821036107, | |
| "grad_norm": 9.187298774719238, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 2.7679, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.7535321821036107, | |
| "eval_loss": 0.073814257979393, | |
| "eval_runtime": 73.168, | |
| "eval_samples_per_second": 2.761, | |
| "eval_steps_per_second": 1.38, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.8037676609105181, | |
| "grad_norm": 8.129315376281738, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 2.3404, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.8540031397174255, | |
| "grad_norm": 9.582749366760254, | |
| "learning_rate": 0.0001346117057077493, | |
| "loss": 2.556, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.9042386185243328, | |
| "grad_norm": 8.937057495117188, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 2.3485, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.9042386185243328, | |
| "eval_loss": 0.06895963847637177, | |
| "eval_runtime": 72.5181, | |
| "eval_samples_per_second": 2.786, | |
| "eval_steps_per_second": 1.393, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.9544740973312402, | |
| "grad_norm": 7.194057941436768, | |
| "learning_rate": 0.00011175373974578378, | |
| "loss": 2.2831, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.0099424385138671, | |
| "grad_norm": 6.12868070602417, | |
| "learning_rate": 0.0001, | |
| "loss": 2.2803, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.0601779173207744, | |
| "grad_norm": 6.514719009399414, | |
| "learning_rate": 8.824626025421626e-05, | |
| "loss": 1.8185, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.0601779173207744, | |
| "eval_loss": 0.058350082486867905, | |
| "eval_runtime": 73.106, | |
| "eval_samples_per_second": 2.763, | |
| "eval_steps_per_second": 1.382, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.1104133961276819, | |
| "grad_norm": 5.304825782775879, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 1.6933, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.1606488749345891, | |
| "grad_norm": 5.396218776702881, | |
| "learning_rate": 6.538829429225069e-05, | |
| "loss": 1.6576, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.2108843537414966, | |
| "grad_norm": 7.664739608764648, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 1.566, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.2108843537414966, | |
| "eval_loss": 0.05733713135123253, | |
| "eval_runtime": 73.1797, | |
| "eval_samples_per_second": 2.76, | |
| "eval_steps_per_second": 1.38, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.2611198325484039, | |
| "grad_norm": 8.14849853515625, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 1.6062, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.3113553113553114, | |
| "grad_norm": 6.613561153411865, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 1.7725, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.3615907901622188, | |
| "grad_norm": 6.738033771514893, | |
| "learning_rate": 2.6567749056431467e-05, | |
| "loss": 1.4988, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.3615907901622188, | |
| "eval_loss": 0.052346888929605484, | |
| "eval_runtime": 73.2318, | |
| "eval_samples_per_second": 2.758, | |
| "eval_steps_per_second": 1.379, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.411826268969126, | |
| "grad_norm": 5.425446033477783, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 1.4703, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.4620617477760334, | |
| "grad_norm": 5.082719802856445, | |
| "learning_rate": 1.2750399292720283e-05, | |
| "loss": 1.6151, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.5122972265829409, | |
| "grad_norm": 5.419787406921387, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 1.6353, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.5122972265829409, | |
| "eval_loss": 0.0507192499935627, | |
| "eval_runtime": 73.2046, | |
| "eval_samples_per_second": 2.759, | |
| "eval_steps_per_second": 1.38, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.5625327053898483, | |
| "grad_norm": 5.491358757019043, | |
| "learning_rate": 3.7544763546352834e-06, | |
| "loss": 1.5038, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.6127681841967556, | |
| "grad_norm": 4.776278495788574, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 1.4482, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.6630036630036629, | |
| "grad_norm": 4.944296836853027, | |
| "learning_rate": 7.709637592770991e-08, | |
| "loss": 1.5689, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.6630036630036629, | |
| "eval_loss": 0.05006199702620506, | |
| "eval_runtime": 73.1223, | |
| "eval_samples_per_second": 2.762, | |
| "eval_steps_per_second": 1.381, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.259104775831552e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |