| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 100, | |
| "global_step": 610, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01639344262295082, | |
| "grad_norm": 1.9867063760757446, | |
| "learning_rate": 1.6393442622950818e-05, | |
| "loss": 0.3745, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03278688524590164, | |
| "grad_norm": 1.508779525756836, | |
| "learning_rate": 3.2786885245901635e-05, | |
| "loss": 0.2514, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04918032786885246, | |
| "grad_norm": 0.43040570616722107, | |
| "learning_rate": 4.918032786885246e-05, | |
| "loss": 0.1555, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06557377049180328, | |
| "grad_norm": 1.1166633367538452, | |
| "learning_rate": 6.557377049180327e-05, | |
| "loss": 0.1932, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08196721311475409, | |
| "grad_norm": 0.4990024268627167, | |
| "learning_rate": 8.19672131147541e-05, | |
| "loss": 0.1536, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09836065573770492, | |
| "grad_norm": 1.1272306442260742, | |
| "learning_rate": 9.836065573770493e-05, | |
| "loss": 0.1524, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.11475409836065574, | |
| "grad_norm": 0.4084867238998413, | |
| "learning_rate": 9.993370449424153e-05, | |
| "loss": 0.1391, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.13114754098360656, | |
| "grad_norm": 0.6679890751838684, | |
| "learning_rate": 9.970476054107763e-05, | |
| "loss": 0.1274, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.14754098360655737, | |
| "grad_norm": 1.0677056312561035, | |
| "learning_rate": 9.931309898856423e-05, | |
| "loss": 0.1283, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.16393442622950818, | |
| "grad_norm": 1.0797268152236938, | |
| "learning_rate": 9.876000201222912e-05, | |
| "loss": 0.1317, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16393442622950818, | |
| "eval_loss": 0.16001197695732117, | |
| "eval_runtime": 62.2223, | |
| "eval_samples_per_second": 1.093, | |
| "eval_steps_per_second": 1.093, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18032786885245902, | |
| "grad_norm": 0.9521363973617554, | |
| "learning_rate": 9.804728027590449e-05, | |
| "loss": 0.1882, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.19672131147540983, | |
| "grad_norm": 0.835111141204834, | |
| "learning_rate": 9.717726700418842e-05, | |
| "loss": 0.1557, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.21311475409836064, | |
| "grad_norm": 0.6795278787612915, | |
| "learning_rate": 9.61528103442088e-05, | |
| "loss": 0.161, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.22950819672131148, | |
| "grad_norm": 0.9582076072692871, | |
| "learning_rate": 9.497726404169412e-05, | |
| "loss": 0.1615, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.2459016393442623, | |
| "grad_norm": 0.9718400239944458, | |
| "learning_rate": 9.365447646187509e-05, | |
| "loss": 0.1473, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.26229508196721313, | |
| "grad_norm": 0.881258487701416, | |
| "learning_rate": 9.218877799115928e-05, | |
| "loss": 0.125, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.2786885245901639, | |
| "grad_norm": 0.9122475385665894, | |
| "learning_rate": 9.058496686082132e-05, | |
| "loss": 0.1401, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.29508196721311475, | |
| "grad_norm": 1.2337970733642578, | |
| "learning_rate": 8.884829343911762e-05, | |
| "loss": 0.145, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.3114754098360656, | |
| "grad_norm": 0.3640283942222595, | |
| "learning_rate": 8.698444304324835e-05, | |
| "loss": 0.1411, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.32786885245901637, | |
| "grad_norm": 0.9699270129203796, | |
| "learning_rate": 8.499951732743457e-05, | |
| "loss": 0.1605, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.32786885245901637, | |
| "eval_loss": 0.15475818514823914, | |
| "eval_runtime": 62.2552, | |
| "eval_samples_per_second": 1.092, | |
| "eval_steps_per_second": 1.092, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3442622950819672, | |
| "grad_norm": 1.2340621948242188, | |
| "learning_rate": 8.290001430804025e-05, | |
| "loss": 0.1302, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.36065573770491804, | |
| "grad_norm": 0.7683020234107971, | |
| "learning_rate": 8.06928070911306e-05, | |
| "loss": 0.1771, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.3770491803278688, | |
| "grad_norm": 1.2670581340789795, | |
| "learning_rate": 7.838512137210565e-05, | |
| "loss": 0.1442, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.39344262295081966, | |
| "grad_norm": 0.6768158078193665, | |
| "learning_rate": 7.598451178106857e-05, | |
| "loss": 0.1389, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.4098360655737705, | |
| "grad_norm": 1.129391074180603, | |
| "learning_rate": 7.3498837151366e-05, | |
| "loss": 0.1385, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.4262295081967213, | |
| "grad_norm": 1.101136326789856, | |
| "learning_rate": 7.093623479226385e-05, | |
| "loss": 0.1564, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.4426229508196721, | |
| "grad_norm": 0.2261950969696045, | |
| "learning_rate": 6.830509384998114e-05, | |
| "loss": 0.1431, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.45901639344262296, | |
| "grad_norm": 0.3458766043186188, | |
| "learning_rate": 6.561402784428974e-05, | |
| "loss": 0.1206, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.47540983606557374, | |
| "grad_norm": 1.1199126243591309, | |
| "learning_rate": 6.287184647058648e-05, | |
| "loss": 0.11, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.4918032786885246, | |
| "grad_norm": 0.45597025752067566, | |
| "learning_rate": 6.0087526759748304e-05, | |
| "loss": 0.1191, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.4918032786885246, | |
| "eval_loss": 0.1460057497024536, | |
| "eval_runtime": 62.2673, | |
| "eval_samples_per_second": 1.092, | |
| "eval_steps_per_second": 1.092, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5081967213114754, | |
| "grad_norm": 0.734933614730835, | |
| "learning_rate": 5.7270183690184495e-05, | |
| "loss": 0.1585, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.5245901639344263, | |
| "grad_norm": 0.8497341275215149, | |
| "learning_rate": 5.4429040348292256e-05, | |
| "loss": 0.1406, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.5409836065573771, | |
| "grad_norm": 0.8217471241950989, | |
| "learning_rate": 5.157339773500125e-05, | |
| "loss": 0.1388, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.5573770491803278, | |
| "grad_norm": 0.3617191016674042, | |
| "learning_rate": 4.8712604317250576e-05, | |
| "loss": 0.1493, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.5737704918032787, | |
| "grad_norm": 0.9715378880500793, | |
| "learning_rate": 4.585602542407722e-05, | |
| "loss": 0.1446, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.5901639344262295, | |
| "grad_norm": 0.8967556953430176, | |
| "learning_rate": 4.3013012587503254e-05, | |
| "loss": 0.1134, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.6065573770491803, | |
| "grad_norm": 1.1487077474594116, | |
| "learning_rate": 4.019287292859016e-05, | |
| "loss": 0.1419, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.6229508196721312, | |
| "grad_norm": 0.8955410718917847, | |
| "learning_rate": 3.7404838688880446e-05, | |
| "loss": 0.1339, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.639344262295082, | |
| "grad_norm": 0.7369993925094604, | |
| "learning_rate": 3.465803700697114e-05, | |
| "loss": 0.1435, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.6557377049180327, | |
| "grad_norm": 0.4338698089122772, | |
| "learning_rate": 3.196146003916084e-05, | |
| "loss": 0.1423, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6557377049180327, | |
| "eval_loss": 0.13493812084197998, | |
| "eval_runtime": 62.3222, | |
| "eval_samples_per_second": 1.091, | |
| "eval_steps_per_second": 1.091, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6721311475409836, | |
| "grad_norm": 1.00863516330719, | |
| "learning_rate": 2.932393552198597e-05, | |
| "loss": 0.1575, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.6885245901639344, | |
| "grad_norm": 1.1648294925689697, | |
| "learning_rate": 2.6754097873015148e-05, | |
| "loss": 0.1372, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.7049180327868853, | |
| "grad_norm": 0.5712903141975403, | |
| "learning_rate": 2.426035992450848e-05, | |
| "loss": 0.1293, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.7213114754098361, | |
| "grad_norm": 0.6494749188423157, | |
| "learning_rate": 2.1850885382476562e-05, | |
| "loss": 0.1347, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.7377049180327869, | |
| "grad_norm": 0.7962918281555176, | |
| "learning_rate": 1.9533562101300097e-05, | |
| "loss": 0.1157, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.7540983606557377, | |
| "grad_norm": 0.7322007417678833, | |
| "learning_rate": 1.7315976261399696e-05, | |
| "loss": 0.1339, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.7704918032786885, | |
| "grad_norm": 1.0486736297607422, | |
| "learning_rate": 1.5205387534490806e-05, | |
| "loss": 0.1327, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.7868852459016393, | |
| "grad_norm": 0.28778010606765747, | |
| "learning_rate": 1.3208705317724006e-05, | |
| "loss": 0.0872, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.8032786885245902, | |
| "grad_norm": 1.064375638961792, | |
| "learning_rate": 1.1332466114513512e-05, | |
| "loss": 0.1348, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.819672131147541, | |
| "grad_norm": 0.7255303263664246, | |
| "learning_rate": 9.582812136100783e-06, | |
| "loss": 0.1065, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.819672131147541, | |
| "eval_loss": 0.1266276091337204, | |
| "eval_runtime": 62.2443, | |
| "eval_samples_per_second": 1.092, | |
| "eval_steps_per_second": 1.092, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8360655737704918, | |
| "grad_norm": 0.8413804173469543, | |
| "learning_rate": 7.965471193905954e-06, | |
| "loss": 0.114, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.8524590163934426, | |
| "grad_norm": 0.9656268954277039, | |
| "learning_rate": 6.4857379484922375e-06, | |
| "loss": 0.1114, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.8688524590163934, | |
| "grad_norm": 0.6842993497848511, | |
| "learning_rate": 5.148456576529081e-06, | |
| "loss": 0.1265, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.8852459016393442, | |
| "grad_norm": 1.1006067991256714, | |
| "learning_rate": 3.958004912496127e-06, | |
| "loss": 0.1418, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.9016393442622951, | |
| "grad_norm": 0.8723937273025513, | |
| "learning_rate": 2.918280117043709e-06, | |
| "loss": 0.1555, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.9180327868852459, | |
| "grad_norm": 0.810757577419281, | |
| "learning_rate": 2.032685918926508e-06, | |
| "loss": 0.1195, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.9344262295081968, | |
| "grad_norm": 1.3418116569519043, | |
| "learning_rate": 1.3041214722768035e-06, | |
| "loss": 0.1395, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.9508196721311475, | |
| "grad_norm": 0.7165635228157043, | |
| "learning_rate": 7.349718656945504e-07, | |
| "loss": 0.1296, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.9672131147540983, | |
| "grad_norm": 0.7757364511489868, | |
| "learning_rate": 3.271003142248652e-07, | |
| "loss": 0.1411, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.9836065573770492, | |
| "grad_norm": 0.8983824253082275, | |
| "learning_rate": 8.184205978370996e-08, | |
| "loss": 0.1174, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9836065573770492, | |
| "eval_loss": 0.12527307868003845, | |
| "eval_runtime": 62.1952, | |
| "eval_samples_per_second": 1.093, | |
| "eval_steps_per_second": 1.093, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.8321591019630432, | |
| "learning_rate": 0.0, | |
| "loss": 0.1184, | |
| "step": 610 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 610, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 1.134915283869696e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |