| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 340, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.029411764705882353, | |
| "grad_norm": 2.488327741622925, | |
| "learning_rate": 5.882352941176471e-05, | |
| "loss": 1.9016, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.14705882352941177, | |
| "grad_norm": 1.427666425704956, | |
| "learning_rate": 0.00029411764705882356, | |
| "loss": 1.8391, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.29411764705882354, | |
| "grad_norm": 0.6930230259895325, | |
| "learning_rate": 0.0005882352941176471, | |
| "loss": 1.5625, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4411764705882353, | |
| "grad_norm": 0.2596357762813568, | |
| "learning_rate": 0.0008823529411764706, | |
| "loss": 1.3849, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 0.16956600546836853, | |
| "learning_rate": 0.0011764705882352942, | |
| "loss": 1.2969, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7352941176470589, | |
| "grad_norm": 0.12523840367794037, | |
| "learning_rate": 0.0014705882352941178, | |
| "loss": 1.2405, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8823529411764706, | |
| "grad_norm": 0.09968027472496033, | |
| "learning_rate": 0.0017647058823529412, | |
| "loss": 1.2024, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.7381153106689453, | |
| "eval_runtime": 0.828, | |
| "eval_samples_per_second": 4.831, | |
| "eval_steps_per_second": 1.208, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.0294117647058822, | |
| "grad_norm": 0.366025447845459, | |
| "learning_rate": 0.001999947298487173, | |
| "loss": 1.1724, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 0.12924526631832123, | |
| "learning_rate": 0.0019981033287370443, | |
| "loss": 1.1507, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3235294117647058, | |
| "grad_norm": 0.15898454189300537, | |
| "learning_rate": 0.0019936298356132177, | |
| "loss": 1.1306, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.4705882352941178, | |
| "grad_norm": 0.09307724982500076, | |
| "learning_rate": 0.0019865386046236597, | |
| "loss": 1.1156, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.6176470588235294, | |
| "grad_norm": 0.0881420224905014, | |
| "learning_rate": 0.001976848317759601, | |
| "loss": 1.1038, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.7647058823529411, | |
| "grad_norm": 0.08049994707107544, | |
| "learning_rate": 0.0019645845042774554, | |
| "loss": 1.0939, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.9117647058823528, | |
| "grad_norm": 0.07463372498750687, | |
| "learning_rate": 0.001949779473441478, | |
| "loss": 1.0846, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.6922645568847656, | |
| "eval_runtime": 0.8305, | |
| "eval_samples_per_second": 4.817, | |
| "eval_steps_per_second": 1.204, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.0588235294117645, | |
| "grad_norm": 0.09218032658100128, | |
| "learning_rate": 0.0019324722294043557, | |
| "loss": 1.0691, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.2058823529411766, | |
| "grad_norm": 0.09160086512565613, | |
| "learning_rate": 0.0019127083684499803, | |
| "loss": 1.0627, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 0.08734273165464401, | |
| "learning_rate": 0.0018905399588691164, | |
| "loss": 1.0554, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.16841983795166016, | |
| "learning_rate": 0.001866025403784439, | |
| "loss": 1.0504, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.6470588235294117, | |
| "grad_norm": 0.0975656732916832, | |
| "learning_rate": 0.0018392292872863268, | |
| "loss": 1.0538, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.7941176470588234, | |
| "grad_norm": 0.08383582532405853, | |
| "learning_rate": 0.0018102222042847736, | |
| "loss": 1.0443, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "grad_norm": 0.09132234007120132, | |
| "learning_rate": 0.0017790805745256705, | |
| "loss": 1.0447, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.6731293201446533, | |
| "eval_runtime": 0.8295, | |
| "eval_samples_per_second": 4.822, | |
| "eval_steps_per_second": 1.206, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 3.088235294117647, | |
| "grad_norm": 0.09759514033794403, | |
| "learning_rate": 0.0017458864412614435, | |
| "loss": 1.0375, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 3.235294117647059, | |
| "grad_norm": 0.09400475025177002, | |
| "learning_rate": 0.0017107272551064472, | |
| "loss": 1.0277, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.3823529411764706, | |
| "grad_norm": 0.07706195116043091, | |
| "learning_rate": 0.0016736956436465573, | |
| "loss": 1.0239, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 0.08647977560758591, | |
| "learning_rate": 0.0016348891674099228, | |
| "loss": 1.0229, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.6764705882352944, | |
| "grad_norm": 0.08289068937301636, | |
| "learning_rate": 0.0015944100628417868, | |
| "loss": 1.0238, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 3.8235294117647056, | |
| "grad_norm": 0.08053147792816162, | |
| "learning_rate": 0.0015523649729605059, | |
| "loss": 1.0195, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.9705882352941178, | |
| "grad_norm": 0.08788559585809708, | |
| "learning_rate": 0.001508864666404365, | |
| "loss": 1.0207, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.6660211086273193, | |
| "eval_runtime": 0.8294, | |
| "eval_samples_per_second": 4.823, | |
| "eval_steps_per_second": 1.206, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 4.117647058823529, | |
| "grad_norm": 0.0886395052075386, | |
| "learning_rate": 0.0014640237456093634, | |
| "loss": 1.007, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.264705882352941, | |
| "grad_norm": 0.08254272490739822, | |
| "learning_rate": 0.0014179603448867834, | |
| "loss": 1.0049, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 4.411764705882353, | |
| "grad_norm": 0.08430016040802002, | |
| "learning_rate": 0.0013707958191959608, | |
| "loss": 1.0103, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.5588235294117645, | |
| "grad_norm": 0.0816187709569931, | |
| "learning_rate": 0.001322654424432195, | |
| "loss": 1.0011, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 0.08980533480644226, | |
| "learning_rate": 0.0012736629900720832, | |
| "loss": 1.0107, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.852941176470588, | |
| "grad_norm": 0.08757297694683075, | |
| "learning_rate": 0.0012239505850387032, | |
| "loss": 1.0035, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.08535553514957428, | |
| "learning_rate": 0.0011736481776669307, | |
| "loss": 1.0039, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.668144941329956, | |
| "eval_runtime": 0.8286, | |
| "eval_samples_per_second": 4.827, | |
| "eval_steps_per_second": 1.207, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.147058823529412, | |
| "grad_norm": 0.08561732620000839, | |
| "learning_rate": 0.0011228882906647141, | |
| "loss": 0.9888, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 5.294117647058823, | |
| "grad_norm": 0.08237646520137787, | |
| "learning_rate": 0.0010718046519793277, | |
| "loss": 0.994, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.4411764705882355, | |
| "grad_norm": 0.07582972943782806, | |
| "learning_rate": 0.0010205318424883906, | |
| "loss": 0.9953, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 5.588235294117647, | |
| "grad_norm": 0.07832607626914978, | |
| "learning_rate": 0.0009692049414438299, | |
| "loss": 0.991, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 5.735294117647059, | |
| "grad_norm": 0.08117620646953583, | |
| "learning_rate": 0.0009179591706028624, | |
| "loss": 0.9931, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 5.882352941176471, | |
| "grad_norm": 0.07949723303318024, | |
| "learning_rate": 0.0008669295379835467, | |
| "loss": 0.9957, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.6619951725006104, | |
| "eval_runtime": 0.8305, | |
| "eval_samples_per_second": 4.817, | |
| "eval_steps_per_second": 1.204, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 6.029411764705882, | |
| "grad_norm": 0.07223918288946152, | |
| "learning_rate": 0.0008162504821834296, | |
| "loss": 0.9855, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 6.176470588235294, | |
| "grad_norm": 0.0734533816576004, | |
| "learning_rate": 0.0007660555181983517, | |
| "loss": 0.9822, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 6.323529411764706, | |
| "grad_norm": 0.07729926705360413, | |
| "learning_rate": 0.0007164768856744892, | |
| "loss": 0.9813, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 6.470588235294118, | |
| "grad_norm": 0.0778665617108345, | |
| "learning_rate": 0.0006676452005203405, | |
| "loss": 0.9853, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 6.617647058823529, | |
| "grad_norm": 0.06967757642269135, | |
| "learning_rate": 0.0006196891107964744, | |
| "loss": 0.9837, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 6.764705882352941, | |
| "grad_norm": 0.07201401889324188, | |
| "learning_rate": 0.0005727349577896194, | |
| "loss": 0.9823, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 6.911764705882353, | |
| "grad_norm": 0.07436826825141907, | |
| "learning_rate": 0.00052690644316399, | |
| "loss": 0.9793, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.6655919551849365, | |
| "eval_runtime": 0.8312, | |
| "eval_samples_per_second": 4.812, | |
| "eval_steps_per_second": 1.203, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 0.07127279043197632, | |
| "learning_rate": 0.0004823243030667576, | |
| "loss": 0.98, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 7.205882352941177, | |
| "grad_norm": 0.06584794819355011, | |
| "learning_rate": 0.0004391059900462304, | |
| "loss": 0.9738, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 7.352941176470588, | |
| "grad_norm": 0.06734811514616013, | |
| "learning_rate": 0.0003973653636207437, | |
| "loss": 0.9705, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.06853578239679337, | |
| "learning_rate": 0.0003572123903134606, | |
| "loss": 0.9778, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 7.647058823529412, | |
| "grad_norm": 0.0678030252456665, | |
| "learning_rate": 0.0003187528539433457, | |
| "loss": 0.9722, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 7.794117647058823, | |
| "grad_norm": 0.06483301520347595, | |
| "learning_rate": 0.0002820880769355582, | |
| "loss": 0.9782, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 7.9411764705882355, | |
| "grad_norm": 0.06444835662841797, | |
| "learning_rate": 0.00024731465338547555, | |
| "loss": 0.9761, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.670668125152588, | |
| "eval_runtime": 0.8308, | |
| "eval_samples_per_second": 4.815, | |
| "eval_steps_per_second": 1.204, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.088235294117647, | |
| "grad_norm": 0.06315235048532486, | |
| "learning_rate": 0.00021452419457960138, | |
| "loss": 0.9718, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 8.235294117647058, | |
| "grad_norm": 0.06345119327306747, | |
| "learning_rate": 0.0001838030876437784, | |
| "loss": 0.977, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 8.382352941176471, | |
| "grad_norm": 0.0637635886669159, | |
| "learning_rate": 0.00015523226795456348, | |
| "loss": 0.9741, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 8.529411764705882, | |
| "grad_norm": 0.06295765936374664, | |
| "learning_rate": 0.00012888700591334225, | |
| "loss": 0.9698, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 8.676470588235293, | |
| "grad_norm": 0.06005469709634781, | |
| "learning_rate": 0.00010483670864493777, | |
| "loss": 0.967, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 8.823529411764707, | |
| "grad_norm": 0.060096003115177155, | |
| "learning_rate": 8.31447371431372e-05, | |
| "loss": 0.9651, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 8.970588235294118, | |
| "grad_norm": 0.05939820781350136, | |
| "learning_rate": 6.386823934487617e-05, | |
| "loss": 0.9678, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.6740888357162476, | |
| "eval_runtime": 0.8293, | |
| "eval_samples_per_second": 4.824, | |
| "eval_steps_per_second": 1.206, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 9.117647058823529, | |
| "grad_norm": 0.058787424117326736, | |
| "learning_rate": 4.705799957284351e-05, | |
| "loss": 0.9715, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 9.264705882352942, | |
| "grad_norm": 0.057846549898386, | |
| "learning_rate": 3.275830474315855e-05, | |
| "loss": 0.9632, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 0.05796463415026665, | |
| "learning_rate": 2.1006827690595476e-05, | |
| "loss": 0.9622, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 9.558823529411764, | |
| "grad_norm": 0.057375218719244, | |
| "learning_rate": 1.1834527918740623e-05, | |
| "loss": 0.9712, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 9.705882352941176, | |
| "grad_norm": 0.05756480246782303, | |
| "learning_rate": 5.265570036553813e-06, | |
| "loss": 0.9672, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 9.852941176470589, | |
| "grad_norm": 0.05712839215993881, | |
| "learning_rate": 1.3172600962190196e-06, | |
| "loss": 0.9695, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.057304926216602325, | |
| "learning_rate": 0.0, | |
| "loss": 0.9709, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.6745353937149048, | |
| "eval_runtime": 0.8292, | |
| "eval_samples_per_second": 4.824, | |
| "eval_steps_per_second": 1.206, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 340, | |
| "total_flos": 1.0444708917333197e+18, | |
| "train_loss": 1.0473914388348073, | |
| "train_runtime": 1591.5084, | |
| "train_samples_per_second": 108.407, | |
| "train_steps_per_second": 0.214 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 340, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0444708917333197e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |