| {"eval_loss":1.5189617872,"eval_binary":0.0,"eval_rouge":0.587340139,"eval_llm_as_a_judge":0.0,"eval_runtime":29.8597,"eval_samples_per_second":0.804,"eval_steps_per_second":0.402,"epoch":0.0,"step":0,"loss":null,"grad_norm":null,"learning_rate":null,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.0492902208,"step":250,"loss":0.8132,"grad_norm":0.5922675133,"learning_rate":0.000012266,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.0985804416,"step":500,"loss":0.4523,"grad_norm":0.5253532529,"learning_rate":0.0000245813,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.1478706625,"step":750,"loss":0.2817,"grad_norm":0.7233960629,"learning_rate":0.0000368966,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.1971608833,"step":1000,"loss":0.2174,"grad_norm":0.7034159899,"learning_rate":0.0000492118,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.2464511041,"step":1250,"loss":0.1833,"grad_norm":1.0380692482,"learning_rate":0.0000493929,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.2957413249,"step":1500,"loss":0.1602,"grad_norm":0.6863846779,"learning_rate":0.0000487444,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.3450315457,"step":1750,"loss":0.142,"grad_norm":0.6993585825,"learning_rate":0.0000480958,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.3943217666,"step":2000,"loss":0.1335,"grad_norm":0.6277371645,"learning_rate":0.0000474472,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.4436119874,"step":2250,"loss":0.1193,"grad_norm":0.462072134,"learning_rate":0.0000467986,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.4929022082,"step":2500,"loss":0.1168,"grad_norm":0.834663868,"learning_rate":0.0000461501,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.542192429,"step":2750,"loss":0.1119,"grad_norm":0.6872285008,"learning_rate":0.0000455015,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.5914826498,"step":3000,"loss":0.1069,"grad_norm":0.7022917271,"learning_rate":0.0000448529,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.6407728707,"step":3250,"loss":0.0984,"grad_norm":1.0676043034,"learning_rate":0.0000442043,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.6900630915,"step":3500,"loss":0.105,"grad_norm":0.9036217332,"learning_rate":0.0000435558,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.7393533123,"step":3750,"loss":0.1015,"grad_norm":0.3344025016,"learning_rate":0.0000429072,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.7886435331,"step":4000,"loss":0.0945,"grad_norm":0.5108351111,"learning_rate":0.0000422586,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.8379337539,"step":4250,"loss":0.0946,"grad_norm":0.5982660651,"learning_rate":0.00004161,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.8872239748,"step":4500,"loss":0.0939,"grad_norm":0.526283443,"learning_rate":0.0000409614,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.9365141956,"step":4750,"loss":0.0885,"grad_norm":0.5514108539,"learning_rate":0.0000403129,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":0.9858044164,"step":5000,"loss":0.0893,"grad_norm":0.431202352,"learning_rate":0.0000396643,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":0.1728554219,"eval_binary":0.0416666667,"eval_rouge":0.9288875691,"eval_llm_as_a_judge":0.0833333333,"eval_runtime":43.2241,"eval_samples_per_second":0.555,"eval_steps_per_second":0.278,"epoch":1.0,"step":5072,"loss":null,"grad_norm":null,"learning_rate":null,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.0350946372,"step":5250,"loss":0.0851,"grad_norm":0.666297853,"learning_rate":0.0000390157,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.084384858,"step":5500,"loss":0.0779,"grad_norm":0.5015407205,"learning_rate":0.0000383671,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.1336750789,"step":5750,"loss":0.0849,"grad_norm":0.8671593666,"learning_rate":0.0000377186,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.1829652997,"step":6000,"loss":0.0793,"grad_norm":0.7724654078,"learning_rate":0.00003707,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.2322555205,"step":6250,"loss":0.0843,"grad_norm":0.5280709267,"learning_rate":0.0000364214,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.2815457413,"step":6500,"loss":0.0754,"grad_norm":0.4885677993,"learning_rate":0.0000357728,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.3308359621,"step":6750,"loss":0.0844,"grad_norm":0.4332463145,"learning_rate":0.0000351243,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.380126183,"step":7000,"loss":0.0771,"grad_norm":0.6162127256,"learning_rate":0.0000344757,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.4294164038,"step":7250,"loss":0.0782,"grad_norm":0.547347486,"learning_rate":0.0000338271,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.4787066246,"step":7500,"loss":0.0802,"grad_norm":0.7598977089,"learning_rate":0.0000331785,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.5279968454,"step":7750,"loss":0.0713,"grad_norm":0.4129500687,"learning_rate":0.00003253,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.5772870662,"step":8000,"loss":0.0752,"grad_norm":0.4954877496,"learning_rate":0.0000318814,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.6265772871,"step":8250,"loss":0.0724,"grad_norm":0.5709223747,"learning_rate":0.0000312328,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.6758675079,"step":8500,"loss":0.079,"grad_norm":0.6234525442,"learning_rate":0.0000305842,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.7251577287,"step":8750,"loss":0.067,"grad_norm":0.4840267897,"learning_rate":0.0000299357,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.7744479495,"step":9000,"loss":0.0722,"grad_norm":0.581608355,"learning_rate":0.0000292871,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.8237381703,"step":9250,"loss":0.0672,"grad_norm":0.4147002101,"learning_rate":0.0000286385,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.8730283912,"step":9500,"loss":0.0732,"grad_norm":0.3916976452,"learning_rate":0.0000279899,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.922318612,"step":9750,"loss":0.0672,"grad_norm":0.3156816959,"learning_rate":0.0000273414,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":1.9716088328,"step":10000,"loss":0.0661,"grad_norm":0.4419353604,"learning_rate":0.0000266928,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":0.1462780684,"eval_binary":0.125,"eval_rouge":0.937830131,"eval_llm_as_a_judge":0.2083333333,"eval_runtime":43.0598,"eval_samples_per_second":0.557,"eval_steps_per_second":0.279,"epoch":2.0,"step":10144,"loss":null,"grad_norm":null,"learning_rate":null,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.0208990536,"step":10250,"loss":0.0648,"grad_norm":0.5617667437,"learning_rate":0.0000260442,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.0701892744,"step":10500,"loss":0.0672,"grad_norm":0.186208874,"learning_rate":0.0000253956,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.1194794953,"step":10750,"loss":0.0662,"grad_norm":0.4603845477,"learning_rate":0.0000247471,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.1687697161,"step":11000,"loss":0.0665,"grad_norm":0.4628971517,"learning_rate":0.0000240985,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.2180599369,"step":11250,"loss":0.0639,"grad_norm":0.4327046871,"learning_rate":0.0000234499,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.2673501577,"step":11500,"loss":0.0665,"grad_norm":0.3963807523,"learning_rate":0.0000228013,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.3166403785,"step":11750,"loss":0.065,"grad_norm":0.4147162139,"learning_rate":0.0000221528,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.3659305994,"step":12000,"loss":0.0629,"grad_norm":0.5985996127,"learning_rate":0.0000215042,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.4152208202,"step":12250,"loss":0.0677,"grad_norm":0.7621842027,"learning_rate":0.0000208556,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.464511041,"step":12500,"loss":0.0639,"grad_norm":0.7110651731,"learning_rate":0.000020207,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.5138012618,"step":12750,"loss":0.0611,"grad_norm":0.4381565154,"learning_rate":0.0000195584,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.5630914826,"step":13000,"loss":0.0623,"grad_norm":0.4191595912,"learning_rate":0.0000189099,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.6123817035,"step":13250,"loss":0.0622,"grad_norm":0.4833663404,"learning_rate":0.0000182613,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.6616719243,"step":13500,"loss":0.0657,"grad_norm":0.4338901937,"learning_rate":0.0000176127,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.7109621451,"step":13750,"loss":0.0611,"grad_norm":0.4739664495,"learning_rate":0.0000169641,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.7602523659,"step":14000,"loss":0.0676,"grad_norm":0.6134439707,"learning_rate":0.0000163156,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.8095425868,"step":14250,"loss":0.0625,"grad_norm":0.5598188639,"learning_rate":0.000015667,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.8588328076,"step":14500,"loss":0.0615,"grad_norm":1.0826495886,"learning_rate":0.0000150184,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.9081230284,"step":14750,"loss":0.0614,"grad_norm":0.6580705643,"learning_rate":0.0000143698,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":2.9574132492,"step":15000,"loss":0.0621,"grad_norm":0.4322754741,"learning_rate":0.0000137213,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":0.1389010996,"eval_binary":0.1666666667,"eval_rouge":0.938387172,"eval_llm_as_a_judge":0.2083333333,"eval_runtime":41.1025,"eval_samples_per_second":0.584,"eval_steps_per_second":0.292,"epoch":3.0,"step":15216,"loss":null,"grad_norm":null,"learning_rate":null,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.00670347,"step":15250,"loss":0.0594,"grad_norm":0.5666684508,"learning_rate":0.0000130727,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.0559936909,"step":15500,"loss":0.0589,"grad_norm":0.4245981574,"learning_rate":0.0000124241,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.1052839117,"step":15750,"loss":0.0566,"grad_norm":0.6567876339,"learning_rate":0.0000117755,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.1545741325,"step":16000,"loss":0.059,"grad_norm":0.4455921352,"learning_rate":0.000011127,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.2038643533,"step":16250,"loss":0.0579,"grad_norm":0.3630923033,"learning_rate":0.0000104784,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.2531545741,"step":16500,"loss":0.0565,"grad_norm":0.351392597,"learning_rate":0.0000098298,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.302444795,"step":16750,"loss":0.0602,"grad_norm":0.3952837586,"learning_rate":0.0000091812,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.3517350158,"step":17000,"loss":0.0575,"grad_norm":0.4846098721,"learning_rate":0.0000085327,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.4010252366,"step":17250,"loss":0.0563,"grad_norm":0.3543721437,"learning_rate":0.0000078841,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.4503154574,"step":17500,"loss":0.06,"grad_norm":0.4508172572,"learning_rate":0.0000072355,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.4996056782,"step":17750,"loss":0.0578,"grad_norm":0.237613067,"learning_rate":0.0000065869,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.5488958991,"step":18000,"loss":0.0572,"grad_norm":0.466924876,"learning_rate":0.0000059384,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.5981861199,"step":18250,"loss":0.0547,"grad_norm":0.5315561295,"learning_rate":0.0000052898,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.6474763407,"step":18500,"loss":0.0583,"grad_norm":0.8466533422,"learning_rate":0.0000046412,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.6967665615,"step":18750,"loss":0.0546,"grad_norm":1.0143963099,"learning_rate":0.0000039926,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.7460567823,"step":19000,"loss":0.0593,"grad_norm":0.3126162589,"learning_rate":0.0000033441,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.7953470032,"step":19250,"loss":0.0587,"grad_norm":0.6343043447,"learning_rate":0.0000026955,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.844637224,"step":19500,"loss":0.0569,"grad_norm":0.3089529574,"learning_rate":0.0000020469,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.8939274448,"step":19750,"loss":0.059,"grad_norm":0.9085928202,"learning_rate":0.0000013983,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.9432176656,"step":20000,"loss":0.0614,"grad_norm":0.3573316336,"learning_rate":0.0000007498,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":3.9925078864,"step":20250,"loss":0.0563,"grad_norm":0.3578336835,"learning_rate":0.0000001012,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":0.1394440532,"eval_binary":0.25,"eval_rouge":0.9403008615,"eval_llm_as_a_judge":0.2916666667,"eval_runtime":35.7534,"eval_samples_per_second":0.671,"eval_steps_per_second":0.336,"epoch":4.0,"step":20288,"loss":null,"grad_norm":null,"learning_rate":null,"train_runtime":null,"train_samples_per_second":null,"train_steps_per_second":null,"total_flos":null,"train_loss":null} | |
| {"eval_loss":null,"eval_binary":null,"eval_rouge":null,"eval_llm_as_a_judge":null,"eval_runtime":null,"eval_samples_per_second":null,"eval_steps_per_second":null,"epoch":4.0,"step":20288,"loss":null,"grad_norm":null,"learning_rate":null,"train_runtime":2349.2891,"train_samples_per_second":17.27,"train_steps_per_second":8.636,"total_flos":3.828478116e+16,"train_loss":0.0939987556} | |