| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 220, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 2.2727272727272728e-06, | |
| "loss": 2.3833, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 2.4762, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 2.2934, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 2.2509, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 2.1881, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 2.2269, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.9987413559579636e-05, | |
| "loss": 2.1204, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 4.988679806432712e-05, | |
| "loss": 2.1803, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.968597221690986e-05, | |
| "loss": 2.0278, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 4.938574467213518e-05, | |
| "loss": 1.9672, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.898732434036244e-05, | |
| "loss": 1.8425, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.849231551964771e-05, | |
| "loss": 1.9323, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 4.790271143580174e-05, | |
| "loss": 2.0277, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 4.722088621637309e-05, | |
| "loss": 1.9256, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "gpt4_scores": 0.6833333333333332, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.878501534461975, | |
| "eval_runtime": 4.9427, | |
| "eval_samples_per_second": 4.653, | |
| "eval_steps_per_second": 1.214, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 4.644958533087443e-05, | |
| "loss": 1.7957, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 4.559191453574582e-05, | |
| "loss": 1.8135, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.465132736856969e-05, | |
| "loss": 1.7996, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 4.3631611241893874e-05, | |
| "loss": 1.7751, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 4.2536872192658036e-05, | |
| "loss": 1.8436, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 4.137151834863213e-05, | |
| "loss": 1.7179, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 4.014024217844167e-05, | |
| "loss": 1.721, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 3.884800159665276e-05, | |
| "loss": 1.8576, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.8062, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 3.610166531514436e-05, | |
| "loss": 1.6772, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 3.465862814232822e-05, | |
| "loss": 1.7685, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 3.3176699082935545e-05, | |
| "loss": 1.721, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 3.166184534225087e-05, | |
| "loss": 1.6726, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 3.012016670162977e-05, | |
| "loss": 1.6738, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "gpt4_scores": 0.6666666666666666, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.8228907585144043, | |
| "eval_runtime": 4.9969, | |
| "eval_samples_per_second": 4.603, | |
| "eval_steps_per_second": 1.201, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 2.8557870956832132e-05, | |
| "loss": 1.6321, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 2.698124892141971e-05, | |
| "loss": 1.6618, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 2.5396649095870202e-05, | |
| "loss": 1.5755, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 2.3810452104406444e-05, | |
| "loss": 1.6999, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 2.222904500247473e-05, | |
| "loss": 1.6391, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 1.583, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 1.9106026612264316e-05, | |
| "loss": 1.6833, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 1.7576990616793137e-05, | |
| "loss": 1.6736, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 1.6077844460203206e-05, | |
| "loss": 1.5683, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 1.4614624674952842e-05, | |
| "loss": 1.5429, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 1.3193223130682936e-05, | |
| "loss": 1.6129, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 1.181936330973744e-05, | |
| "loss": 1.6105, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 1.049857726072005e-05, | |
| "loss": 1.5663, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 9.236183322886945e-06, | |
| "loss": 1.6181, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "gpt4_scores": 0.7000000000000001, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.8283250331878662, | |
| "eval_runtime": 4.9725, | |
| "eval_samples_per_second": 4.625, | |
| "eval_steps_per_second": 1.207, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 8.0372647110717e-06, | |
| "loss": 1.5877, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 6.906649047373246e-06, | |
| "loss": 1.3838, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 5.848888922025553e-06, | |
| "loss": 1.5406, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 4.868243561723535e-06, | |
| "loss": 1.497, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 3.968661679220468e-06, | |
| "loss": 1.5702, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 3.1537655732553768e-06, | |
| "loss": 1.6095, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 2.4268365428344736e-06, | |
| "loss": 1.4712, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 1.790801674598186e-06, | |
| "loss": 1.5552, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 1.248222056476367e-06, | |
| "loss": 1.5805, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 8.012824650910938e-07, | |
| "loss": 1.5588, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 4.517825684323324e-07, | |
| "loss": 1.64, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 2.011296792301165e-07, | |
| "loss": 1.5666, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 5.033308820289184e-08, | |
| "loss": 1.4496, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0, | |
| "loss": 1.5007, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "gpt4_scores": 0.6666666666666666, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.8388779163360596, | |
| "eval_runtime": 4.9378, | |
| "eval_samples_per_second": 4.658, | |
| "eval_steps_per_second": 1.215, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 220, | |
| "total_flos": 7586259830267904.0, | |
| "train_loss": 1.7501471573656255, | |
| "train_runtime": 1644.9971, | |
| "train_samples_per_second": 0.528, | |
| "train_steps_per_second": 0.134 | |
| } | |
| ], | |
| "logging_steps": 4, | |
| "max_steps": 220, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 20, | |
| "total_flos": 7586259830267904.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |