| { |
| "best_global_step": 600, |
| "best_metric": 0.2925775945186615, |
| "best_model_checkpoint": "/workspace/anchor/qwen-compassion/train/checkpoint-600", |
| "epoch": 1.0, |
| "eval_steps": 50, |
| "global_step": 614, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.008143322475570033, |
| "grad_norm": 22.442302703857422, |
| "learning_rate": 2.5806451612903226e-05, |
| "loss": 3.3662, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.016286644951140065, |
| "grad_norm": 14.803105354309082, |
| "learning_rate": 5.8064516129032266e-05, |
| "loss": 3.436, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.024429967426710098, |
| "grad_norm": 9.32910442352295, |
| "learning_rate": 9.032258064516129e-05, |
| "loss": 2.2124, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.03257328990228013, |
| "grad_norm": 2.5130510330200195, |
| "learning_rate": 0.00012258064516129034, |
| "loss": 1.4505, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.04071661237785016, |
| "grad_norm": 18.144712448120117, |
| "learning_rate": 0.00015483870967741937, |
| "loss": 1.1231, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.048859934853420196, |
| "grad_norm": 3.0365991592407227, |
| "learning_rate": 0.0001870967741935484, |
| "loss": 0.872, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.057003257328990226, |
| "grad_norm": 2.5769875049591064, |
| "learning_rate": 0.00019998693330688282, |
| "loss": 0.8906, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.06514657980456026, |
| "grad_norm": 3.9422550201416016, |
| "learning_rate": 0.00019990709365929677, |
| "loss": 0.7195, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.0732899022801303, |
| "grad_norm": 2.395726442337036, |
| "learning_rate": 0.00019975473152335726, |
| "loss": 0.6868, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.08143322475570032, |
| "grad_norm": 1.726025104522705, |
| "learning_rate": 0.0001995299574986306, |
| "loss": 0.6497, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.08143322475570032, |
| "eval_loss": 0.4577290415763855, |
| "eval_runtime": 214.9578, |
| "eval_samples_per_second": 1.172, |
| "eval_steps_per_second": 0.586, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.08957654723127036, |
| "grad_norm": 1.482474684715271, |
| "learning_rate": 0.00019923293474842174, |
| "loss": 0.6646, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.09771986970684039, |
| "grad_norm": 1.236023187637329, |
| "learning_rate": 0.00019886387888133413, |
| "loss": 0.5642, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.10586319218241043, |
| "grad_norm": 1.1658198833465576, |
| "learning_rate": 0.00019842305779475968, |
| "loss": 0.4603, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.11400651465798045, |
| "grad_norm": 1.2695726156234741, |
| "learning_rate": 0.0001979107914804122, |
| "loss": 0.6375, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.12214983713355049, |
| "grad_norm": 1.4793083667755127, |
| "learning_rate": 0.00019732745179204552, |
| "loss": 0.4823, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.13029315960912052, |
| "grad_norm": 1.7086373567581177, |
| "learning_rate": 0.00019667346217552527, |
| "loss": 0.4646, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.13843648208469056, |
| "grad_norm": 1.3679929971694946, |
| "learning_rate": 0.00019594929736144976, |
| "loss": 0.6899, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.1465798045602606, |
| "grad_norm": 1.4010539054870605, |
| "learning_rate": 0.00019515548302054335, |
| "loss": 0.5442, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.15472312703583063, |
| "grad_norm": 1.1517243385314941, |
| "learning_rate": 0.0001942925953820725, |
| "loss": 0.5184, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.16286644951140064, |
| "grad_norm": 1.058547019958496, |
| "learning_rate": 0.00019336126081556134, |
| "loss": 0.6193, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.16286644951140064, |
| "eval_loss": 0.35910215973854065, |
| "eval_runtime": 120.7216, |
| "eval_samples_per_second": 2.087, |
| "eval_steps_per_second": 1.044, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.17100977198697068, |
| "grad_norm": 1.1780120134353638, |
| "learning_rate": 0.00019236215537611046, |
| "loss": 0.5177, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.1791530944625407, |
| "grad_norm": 1.2721914052963257, |
| "learning_rate": 0.00019129600431364897, |
| "loss": 0.5419, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.18729641693811075, |
| "grad_norm": 1.379878282546997, |
| "learning_rate": 0.00019016358154647618, |
| "loss": 0.5167, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.19543973941368079, |
| "grad_norm": 1.2618638277053833, |
| "learning_rate": 0.00018896570909947475, |
| "loss": 0.4613, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.20358306188925082, |
| "grad_norm": 1.0391523838043213, |
| "learning_rate": 0.00018770325650740345, |
| "loss": 0.4745, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.21172638436482086, |
| "grad_norm": 1.2194364070892334, |
| "learning_rate": 0.00018637714018370253, |
| "loss": 0.5386, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.21986970684039087, |
| "grad_norm": 0.7502462267875671, |
| "learning_rate": 0.00018498832275526988, |
| "loss": 0.4939, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.2280130293159609, |
| "grad_norm": 0.6979483366012573, |
| "learning_rate": 0.00018353781236369064, |
| "loss": 0.4941, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.23615635179153094, |
| "grad_norm": 1.1954421997070312, |
| "learning_rate": 0.00018202666193342833, |
| "loss": 0.5449, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.24429967426710097, |
| "grad_norm": 1.3286981582641602, |
| "learning_rate": 0.00018045596840750723, |
| "loss": 0.5998, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.24429967426710097, |
| "eval_loss": 0.3380209803581238, |
| "eval_runtime": 121.0869, |
| "eval_samples_per_second": 2.081, |
| "eval_steps_per_second": 1.041, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.252442996742671, |
| "grad_norm": 0.9305353760719299, |
| "learning_rate": 0.0001788268719512427, |
| "loss": 0.5237, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.26058631921824105, |
| "grad_norm": 1.387276530265808, |
| "learning_rate": 0.00017714055512459565, |
| "loss": 0.4911, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.2687296416938111, |
| "grad_norm": 0.9299679398536682, |
| "learning_rate": 0.0001753982420237533, |
| "loss": 0.5527, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.2768729641693811, |
| "grad_norm": 1.2602843046188354, |
| "learning_rate": 0.00017360119739255852, |
| "loss": 0.3972, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.28501628664495116, |
| "grad_norm": 0.7769179344177246, |
| "learning_rate": 0.00017175072570443312, |
| "loss": 0.5526, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.2931596091205212, |
| "grad_norm": 0.7283039093017578, |
| "learning_rate": 0.00016984817021546177, |
| "loss": 0.4304, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.30130293159609123, |
| "grad_norm": 1.4225355386734009, |
| "learning_rate": 0.000167894911989323, |
| "loss": 0.4737, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.30944625407166126, |
| "grad_norm": 1.0882941484451294, |
| "learning_rate": 0.00016589236889477646, |
| "loss": 0.4035, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.31758957654723124, |
| "grad_norm": 1.1490302085876465, |
| "learning_rate": 0.00016384199457643262, |
| "loss": 0.5222, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.3257328990228013, |
| "grad_norm": 0.9430150985717773, |
| "learning_rate": 0.00016174527739955342, |
| "loss": 0.6425, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.3257328990228013, |
| "eval_loss": 0.3253232538700104, |
| "eval_runtime": 120.2962, |
| "eval_samples_per_second": 2.095, |
| "eval_steps_per_second": 1.047, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.3338762214983713, |
| "grad_norm": 0.6319507360458374, |
| "learning_rate": 0.00015960373936964892, |
| "loss": 0.4485, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.34201954397394135, |
| "grad_norm": 1.1035504341125488, |
| "learning_rate": 0.0001574189350276545, |
| "loss": 0.5942, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.3501628664495114, |
| "grad_norm": 1.207892894744873, |
| "learning_rate": 0.00015519245032149083, |
| "loss": 0.558, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.3583061889250814, |
| "grad_norm": 0.8965429067611694, |
| "learning_rate": 0.0001529259014548253, |
| "loss": 0.478, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.36644951140065146, |
| "grad_norm": 1.1375535726547241, |
| "learning_rate": 0.00015062093371387097, |
| "loss": 0.4909, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.3745928338762215, |
| "grad_norm": 1.2835142612457275, |
| "learning_rate": 0.00014827922027307451, |
| "loss": 0.4748, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.38273615635179153, |
| "grad_norm": 1.0070925951004028, |
| "learning_rate": 0.00014590246098055996, |
| "loss": 0.5095, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.39087947882736157, |
| "grad_norm": 0.7836953401565552, |
| "learning_rate": 0.00014349238112421024, |
| "loss": 0.3591, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.3990228013029316, |
| "grad_norm": 0.8382024765014648, |
| "learning_rate": 0.000141050730179282, |
| "loss": 0.4465, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.40716612377850164, |
| "grad_norm": 1.106084942817688, |
| "learning_rate": 0.0001385792805384625, |
| "loss": 0.6015, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.40716612377850164, |
| "eval_loss": 0.3180280327796936, |
| "eval_runtime": 120.1946, |
| "eval_samples_per_second": 2.097, |
| "eval_steps_per_second": 1.048, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.4153094462540717, |
| "grad_norm": 1.028218150138855, |
| "learning_rate": 0.00013607982622529133, |
| "loss": 0.4443, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.4234527687296417, |
| "grad_norm": 3.6740164756774902, |
| "learning_rate": 0.00013355418159187985, |
| "loss": 0.5305, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.43159609120521175, |
| "grad_norm": 0.7290945053100586, |
| "learning_rate": 0.00013100418000187419, |
| "loss": 0.5237, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.43973941368078173, |
| "grad_norm": 1.0659680366516113, |
| "learning_rate": 0.0001284316724996181, |
| "loss": 0.4935, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.44788273615635177, |
| "grad_norm": 1.0143028497695923, |
| "learning_rate": 0.00012583852646648095, |
| "loss": 0.6135, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.4560260586319218, |
| "grad_norm": 1.121451735496521, |
| "learning_rate": 0.00012322662426532708, |
| "loss": 0.5242, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.46416938110749184, |
| "grad_norm": 0.6729561686515808, |
| "learning_rate": 0.00012059786187410984, |
| "loss": 0.3619, |
| "step": 285 |
| }, |
| { |
| "epoch": 0.4723127035830619, |
| "grad_norm": 0.6204145550727844, |
| "learning_rate": 0.00011795414750958265, |
| "loss": 0.3198, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.4804560260586319, |
| "grad_norm": 0.9363518953323364, |
| "learning_rate": 0.00011529740024212565, |
| "loss": 0.4189, |
| "step": 295 |
| }, |
| { |
| "epoch": 0.48859934853420195, |
| "grad_norm": 1.1184571981430054, |
| "learning_rate": 0.00011262954860269399, |
| "loss": 0.587, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.48859934853420195, |
| "eval_loss": 0.311747670173645, |
| "eval_runtime": 120.3226, |
| "eval_samples_per_second": 2.094, |
| "eval_steps_per_second": 1.047, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.496742671009772, |
| "grad_norm": 1.0931415557861328, |
| "learning_rate": 0.0001099525291828986, |
| "loss": 0.4254, |
| "step": 305 |
| }, |
| { |
| "epoch": 0.504885993485342, |
| "grad_norm": 1.0900492668151855, |
| "learning_rate": 0.00010726828522923562, |
| "loss": 0.4637, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.5130293159609121, |
| "grad_norm": 1.081187129020691, |
| "learning_rate": 0.00010457876523248518, |
| "loss": 0.574, |
| "step": 315 |
| }, |
| { |
| "epoch": 0.5211726384364821, |
| "grad_norm": 0.5426526665687561, |
| "learning_rate": 0.00010188592151330343, |
| "loss": 0.4139, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.5293159609120521, |
| "grad_norm": 0.6170647144317627, |
| "learning_rate": 9.919170880503415e-05, |
| "loss": 0.4787, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.5374592833876222, |
| "grad_norm": 0.9634841084480286, |
| "learning_rate": 9.649808283476941e-05, |
| "loss": 0.2533, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.5456026058631922, |
| "grad_norm": 0.6538633108139038, |
| "learning_rate": 9.38069989036886e-05, |
| "loss": 0.4913, |
| "step": 335 |
| }, |
| { |
| "epoch": 0.5537459283387622, |
| "grad_norm": 0.6944330334663391, |
| "learning_rate": 9.112041046770653e-05, |
| "loss": 0.2606, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.5618892508143323, |
| "grad_norm": 0.9925371408462524, |
| "learning_rate": 8.844026771946147e-05, |
| "loss": 0.4495, |
| "step": 345 |
| }, |
| { |
| "epoch": 0.5700325732899023, |
| "grad_norm": 1.0751277208328247, |
| "learning_rate": 8.57685161726715e-05, |
| "loss": 0.5536, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.5700325732899023, |
| "eval_loss": 0.3054664731025696, |
| "eval_runtime": 120.3383, |
| "eval_samples_per_second": 2.094, |
| "eval_steps_per_second": 1.047, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.5781758957654723, |
| "grad_norm": 0.8132654428482056, |
| "learning_rate": 8.310709524988743e-05, |
| "loss": 0.4582, |
| "step": 355 |
| }, |
| { |
| "epoch": 0.5863192182410424, |
| "grad_norm": 0.9619295001029968, |
| "learning_rate": 8.045793687466757e-05, |
| "loss": 0.3663, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.5944625407166124, |
| "grad_norm": 0.9950692057609558, |
| "learning_rate": 7.782296406919557e-05, |
| "loss": 0.5359, |
| "step": 365 |
| }, |
| { |
| "epoch": 0.6026058631921825, |
| "grad_norm": 1.0357235670089722, |
| "learning_rate": 7.520408955836007e-05, |
| "loss": 0.3589, |
| "step": 370 |
| }, |
| { |
| "epoch": 0.6107491856677525, |
| "grad_norm": 0.6220430731773376, |
| "learning_rate": 7.260321438130913e-05, |
| "loss": 0.3868, |
| "step": 375 |
| }, |
| { |
| "epoch": 0.6188925081433225, |
| "grad_norm": 0.7960296869277954, |
| "learning_rate": 7.002222651148714e-05, |
| "loss": 0.2924, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.6270358306188925, |
| "grad_norm": 1.227491855621338, |
| "learning_rate": 6.746299948615631e-05, |
| "loss": 0.6195, |
| "step": 385 |
| }, |
| { |
| "epoch": 0.6351791530944625, |
| "grad_norm": 0.7982805371284485, |
| "learning_rate": 6.492739104639727e-05, |
| "loss": 0.364, |
| "step": 390 |
| }, |
| { |
| "epoch": 0.6433224755700325, |
| "grad_norm": 0.7055819034576416, |
| "learning_rate": 6.24172417885762e-05, |
| "loss": 0.4519, |
| "step": 395 |
| }, |
| { |
| "epoch": 0.6514657980456026, |
| "grad_norm": 1.0085915327072144, |
| "learning_rate": 5.9934373828257106e-05, |
| "loss": 0.369, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.6514657980456026, |
| "eval_loss": 0.3009154200553894, |
| "eval_runtime": 120.3254, |
| "eval_samples_per_second": 2.094, |
| "eval_steps_per_second": 1.047, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.6596091205211726, |
| "grad_norm": 0.6098869442939758, |
| "learning_rate": 5.7480589477529545e-05, |
| "loss": 0.3941, |
| "step": 405 |
| }, |
| { |
| "epoch": 0.6677524429967426, |
| "grad_norm": 0.86444491147995, |
| "learning_rate": 5.505766993671162e-05, |
| "loss": 0.4796, |
| "step": 410 |
| }, |
| { |
| "epoch": 0.6758957654723127, |
| "grad_norm": 1.1165945529937744, |
| "learning_rate": 5.266737400137765e-05, |
| "loss": 0.3816, |
| "step": 415 |
| }, |
| { |
| "epoch": 0.6840390879478827, |
| "grad_norm": 1.1898753643035889, |
| "learning_rate": 5.031143678565005e-05, |
| "loss": 0.3584, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.6921824104234527, |
| "grad_norm": 0.5651485919952393, |
| "learning_rate": 4.799156846268095e-05, |
| "loss": 0.3323, |
| "step": 425 |
| }, |
| { |
| "epoch": 0.7003257328990228, |
| "grad_norm": 1.103471279144287, |
| "learning_rate": 4.57094530232389e-05, |
| "loss": 0.4991, |
| "step": 430 |
| }, |
| { |
| "epoch": 0.7084690553745928, |
| "grad_norm": 0.9560770988464355, |
| "learning_rate": 4.346674705330117e-05, |
| "loss": 0.4969, |
| "step": 435 |
| }, |
| { |
| "epoch": 0.7166123778501629, |
| "grad_norm": 1.0413203239440918, |
| "learning_rate": 4.1265078531538916e-05, |
| "loss": 0.3927, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.7247557003257329, |
| "grad_norm": 1.0384825468063354, |
| "learning_rate": 3.9106045647569e-05, |
| "loss": 0.4469, |
| "step": 445 |
| }, |
| { |
| "epoch": 0.7328990228013029, |
| "grad_norm": 1.2701938152313232, |
| "learning_rate": 3.69912156418289e-05, |
| "loss": 0.5587, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.7328990228013029, |
| "eval_loss": 0.29790911078453064, |
| "eval_runtime": 120.1996, |
| "eval_samples_per_second": 2.097, |
| "eval_steps_per_second": 1.048, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.741042345276873, |
| "grad_norm": 1.088590383529663, |
| "learning_rate": 3.492212366791831e-05, |
| "loss": 0.4347, |
| "step": 455 |
| }, |
| { |
| "epoch": 0.749185667752443, |
| "grad_norm": 0.9670878648757935, |
| "learning_rate": 3.290027167823204e-05, |
| "loss": 0.4226, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.757328990228013, |
| "grad_norm": 0.9960106015205383, |
| "learning_rate": 3.092712733369387e-05, |
| "loss": 0.3849, |
| "step": 465 |
| }, |
| { |
| "epoch": 0.7654723127035831, |
| "grad_norm": 0.6322615742683411, |
| "learning_rate": 2.9004122938382617e-05, |
| "loss": 0.3709, |
| "step": 470 |
| }, |
| { |
| "epoch": 0.7736156351791531, |
| "grad_norm": 0.8384042978286743, |
| "learning_rate": 2.7132654399823444e-05, |
| "loss": 0.4528, |
| "step": 475 |
| }, |
| { |
| "epoch": 0.7817589576547231, |
| "grad_norm": 0.8728402256965637, |
| "learning_rate": 2.531408021569982e-05, |
| "loss": 0.4617, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.7899022801302932, |
| "grad_norm": 0.9127026796340942, |
| "learning_rate": 2.3549720487720738e-05, |
| "loss": 0.5415, |
| "step": 485 |
| }, |
| { |
| "epoch": 0.7980456026058632, |
| "grad_norm": 0.9375382661819458, |
| "learning_rate": 2.184085596336011e-05, |
| "loss": 0.4591, |
| "step": 490 |
| }, |
| { |
| "epoch": 0.8061889250814332, |
| "grad_norm": 1.062622308731079, |
| "learning_rate": 2.0188727106162874e-05, |
| "loss": 0.4813, |
| "step": 495 |
| }, |
| { |
| "epoch": 0.8143322475570033, |
| "grad_norm": 0.8038843870162964, |
| "learning_rate": 1.8594533195293427e-05, |
| "loss": 0.4651, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8143322475570033, |
| "eval_loss": 0.2944928705692291, |
| "eval_runtime": 120.3478, |
| "eval_samples_per_second": 2.094, |
| "eval_steps_per_second": 1.047, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.8224755700325733, |
| "grad_norm": 0.9246629476547241, |
| "learning_rate": 1.7059431454979824e-05, |
| "loss": 0.4701, |
| "step": 505 |
| }, |
| { |
| "epoch": 0.8306188925081434, |
| "grad_norm": 1.2579699754714966, |
| "learning_rate": 1.5584536214485457e-05, |
| "loss": 0.4114, |
| "step": 510 |
| }, |
| { |
| "epoch": 0.8387622149837134, |
| "grad_norm": 1.0765819549560547, |
| "learning_rate": 1.4170918099218166e-05, |
| "loss": 0.4497, |
| "step": 515 |
| }, |
| { |
| "epoch": 0.8469055374592834, |
| "grad_norm": 0.8294546604156494, |
| "learning_rate": 1.2819603253564205e-05, |
| "loss": 0.3338, |
| "step": 520 |
| }, |
| { |
| "epoch": 0.8550488599348535, |
| "grad_norm": 0.8488652110099792, |
| "learning_rate": 1.153157259601062e-05, |
| "loss": 0.4017, |
| "step": 525 |
| }, |
| { |
| "epoch": 0.8631921824104235, |
| "grad_norm": 0.9699320197105408, |
| "learning_rate": 1.030776110709718e-05, |
| "loss": 0.4934, |
| "step": 530 |
| }, |
| { |
| "epoch": 0.8713355048859935, |
| "grad_norm": 0.939552903175354, |
| "learning_rate": 9.149057150714801e-06, |
| "loss": 0.4814, |
| "step": 535 |
| }, |
| { |
| "epoch": 0.8794788273615635, |
| "grad_norm": 0.970380961894989, |
| "learning_rate": 8.056301829242784e-06, |
| "loss": 0.4135, |
| "step": 540 |
| }, |
| { |
| "epoch": 0.8876221498371335, |
| "grad_norm": 0.5573286414146423, |
| "learning_rate": 7.030288372993066e-06, |
| "loss": 0.2724, |
| "step": 545 |
| }, |
| { |
| "epoch": 0.8957654723127035, |
| "grad_norm": 0.9565848112106323, |
| "learning_rate": 6.0717615644051206e-06, |
| "loss": 0.4606, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.8957654723127035, |
| "eval_loss": 0.2927910387516022, |
| "eval_runtime": 120.4502, |
| "eval_samples_per_second": 2.092, |
| "eval_steps_per_second": 1.046, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.9039087947882736, |
| "grad_norm": 0.7690985798835754, |
| "learning_rate": 5.181417197408734e-06, |
| "loss": 0.4854, |
| "step": 555 |
| }, |
| { |
| "epoch": 0.9120521172638436, |
| "grad_norm": 0.7944011688232422, |
| "learning_rate": 4.359901572347758e-06, |
| "loss": 0.425, |
| "step": 560 |
| }, |
| { |
| "epoch": 0.9201954397394136, |
| "grad_norm": 1.0082036256790161, |
| "learning_rate": 3.607811026831176e-06, |
| "loss": 0.5035, |
| "step": 565 |
| }, |
| { |
| "epoch": 0.9283387622149837, |
| "grad_norm": 0.9810898303985596, |
| "learning_rate": 2.9256915028519573e-06, |
| "loss": 0.3532, |
| "step": 570 |
| }, |
| { |
| "epoch": 0.9364820846905537, |
| "grad_norm": 0.8743959069252014, |
| "learning_rate": 2.3140381504882737e-06, |
| "loss": 0.4653, |
| "step": 575 |
| }, |
| { |
| "epoch": 0.9446254071661238, |
| "grad_norm": 0.8244614601135254, |
| "learning_rate": 1.7732949684743594e-06, |
| "loss": 0.351, |
| "step": 580 |
| }, |
| { |
| "epoch": 0.9527687296416938, |
| "grad_norm": 1.2245746850967407, |
| "learning_rate": 1.303854481902067e-06, |
| "loss": 0.4347, |
| "step": 585 |
| }, |
| { |
| "epoch": 0.9609120521172638, |
| "grad_norm": 0.9963445663452148, |
| "learning_rate": 9.060574572873237e-07, |
| "loss": 0.4403, |
| "step": 590 |
| }, |
| { |
| "epoch": 0.9690553745928339, |
| "grad_norm": 1.018866777420044, |
| "learning_rate": 5.801926552078563e-07, |
| "loss": 0.4273, |
| "step": 595 |
| }, |
| { |
| "epoch": 0.9771986970684039, |
| "grad_norm": 1.0799201726913452, |
| "learning_rate": 3.264966206921294e-07, |
| "loss": 0.4367, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.9771986970684039, |
| "eval_loss": 0.2925775945186615, |
| "eval_runtime": 120.683, |
| "eval_samples_per_second": 2.088, |
| "eval_steps_per_second": 1.044, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.9853420195439739, |
| "grad_norm": 1.1354663372039795, |
| "learning_rate": 1.451535115114866e-07, |
| "loss": 0.478, |
| "step": 605 |
| }, |
| { |
| "epoch": 0.993485342019544, |
| "grad_norm": 0.8019553422927856, |
| "learning_rate": 3.629496450011649e-08, |
| "loss": 0.3683, |
| "step": 610 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 614, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 100, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 2, |
| "early_stopping_threshold": 0.001 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 1 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.508719122176676e+18, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
