| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.10055304172951232, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001005530417295123, | |
| "eval_loss": 10.755413055419922, | |
| "eval_runtime": 11.149, | |
| "eval_samples_per_second": 150.237, | |
| "eval_steps_per_second": 18.836, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0030165912518853697, | |
| "grad_norm": 3.2384252548217773, | |
| "learning_rate": 1.5e-05, | |
| "loss": 10.7531, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.006033182503770739, | |
| "grad_norm": 3.522120952606201, | |
| "learning_rate": 3e-05, | |
| "loss": 10.7362, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00904977375565611, | |
| "grad_norm": 3.7233874797821045, | |
| "learning_rate": 4.5e-05, | |
| "loss": 10.6587, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00904977375565611, | |
| "eval_loss": 10.54302978515625, | |
| "eval_runtime": 11.1743, | |
| "eval_samples_per_second": 149.898, | |
| "eval_steps_per_second": 18.793, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.012066365007541479, | |
| "grad_norm": 4.279703617095947, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 10.4367, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.015082956259426848, | |
| "grad_norm": 5.507565021514893, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 10.1176, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.01809954751131222, | |
| "grad_norm": 6.325550079345703, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 9.7615, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.01809954751131222, | |
| "eval_loss": 9.461087226867676, | |
| "eval_runtime": 11.1633, | |
| "eval_samples_per_second": 150.045, | |
| "eval_steps_per_second": 18.812, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.021116138763197588, | |
| "grad_norm": 6.642955780029297, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 9.3199, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.024132730015082957, | |
| "grad_norm": 7.369113922119141, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 8.8135, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.027149321266968326, | |
| "grad_norm": 6.994415760040283, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 8.3456, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.027149321266968326, | |
| "eval_loss": 7.989864349365234, | |
| "eval_runtime": 11.0595, | |
| "eval_samples_per_second": 151.454, | |
| "eval_steps_per_second": 18.988, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.030165912518853696, | |
| "grad_norm": 6.771633148193359, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 7.8555, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.033182503770739065, | |
| "grad_norm": 6.242250442504883, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 7.5154, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.03619909502262444, | |
| "grad_norm": 6.238581657409668, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 7.0909, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.03619909502262444, | |
| "eval_loss": 6.755636215209961, | |
| "eval_runtime": 11.0524, | |
| "eval_samples_per_second": 151.551, | |
| "eval_steps_per_second": 19.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0392156862745098, | |
| "grad_norm": 6.453060626983643, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 6.5399, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.042232277526395176, | |
| "grad_norm": 5.792873859405518, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 6.2014, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.04524886877828054, | |
| "grad_norm": 5.358177661895752, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 6.046, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.04524886877828054, | |
| "eval_loss": 5.793038845062256, | |
| "eval_runtime": 11.0914, | |
| "eval_samples_per_second": 151.018, | |
| "eval_steps_per_second": 18.934, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.048265460030165915, | |
| "grad_norm": 5.039991855621338, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 5.6358, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.05128205128205128, | |
| "grad_norm": 4.412015438079834, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 5.4325, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.05429864253393665, | |
| "grad_norm": 4.06430721282959, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 5.3275, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.05429864253393665, | |
| "eval_loss": 5.077151298522949, | |
| "eval_runtime": 11.262, | |
| "eval_samples_per_second": 148.731, | |
| "eval_steps_per_second": 18.647, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.05731523378582202, | |
| "grad_norm": 4.049082279205322, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 4.9527, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.06033182503770739, | |
| "grad_norm": 3.6301987171173096, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 4.8217, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.06334841628959276, | |
| "grad_norm": 3.112250566482544, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 4.6221, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.06334841628959276, | |
| "eval_loss": 4.606419563293457, | |
| "eval_runtime": 11.0191, | |
| "eval_samples_per_second": 152.009, | |
| "eval_steps_per_second": 19.058, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.06636500754147813, | |
| "grad_norm": 3.022951364517212, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 4.6649, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.0693815987933635, | |
| "grad_norm": 2.6910369396209717, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 4.456, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.07239819004524888, | |
| "grad_norm": 2.79231858253479, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 4.3638, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.07239819004524888, | |
| "eval_loss": 4.356507301330566, | |
| "eval_runtime": 11.0721, | |
| "eval_samples_per_second": 151.282, | |
| "eval_steps_per_second": 18.967, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.07541478129713423, | |
| "grad_norm": 2.6221296787261963, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 4.3823, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.0784313725490196, | |
| "grad_norm": 2.882395029067993, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 4.5094, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.08144796380090498, | |
| "grad_norm": 2.5901410579681396, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 4.2592, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.08144796380090498, | |
| "eval_loss": 4.210515022277832, | |
| "eval_runtime": 11.0579, | |
| "eval_samples_per_second": 151.475, | |
| "eval_steps_per_second": 18.991, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.08446455505279035, | |
| "grad_norm": 2.67712140083313, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 4.168, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.08748114630467571, | |
| "grad_norm": 2.538766860961914, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 4.1359, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.09049773755656108, | |
| "grad_norm": 2.631545305252075, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 4.1228, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.09049773755656108, | |
| "eval_loss": 4.152175426483154, | |
| "eval_runtime": 11.1662, | |
| "eval_samples_per_second": 150.007, | |
| "eval_steps_per_second": 18.807, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.09351432880844646, | |
| "grad_norm": 2.2963476181030273, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 4.1301, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.09653092006033183, | |
| "grad_norm": 2.7444159984588623, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 4.3495, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.09954751131221719, | |
| "grad_norm": 2.268852472305298, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 4.1211, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.09954751131221719, | |
| "eval_loss": 4.140535354614258, | |
| "eval_runtime": 11.1432, | |
| "eval_samples_per_second": 150.315, | |
| "eval_steps_per_second": 18.846, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 8571248128819200.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |