| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.9932104752667313, |
| "eval_steps": 1024, |
| "global_step": 21504, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.011823934229365849, |
| "grad_norm": 0.526302695274353, |
| "learning_rate": 0.000498046875, |
| "loss": 9.227896690368652, |
| "step": 256 |
| }, |
| { |
| "epoch": 0.023647868458731697, |
| "grad_norm": 0.40557485818862915, |
| "learning_rate": 0.000998046875, |
| "loss": 6.491486072540283, |
| "step": 512 |
| }, |
| { |
| "epoch": 0.03547180268809755, |
| "grad_norm": 0.2611554265022278, |
| "learning_rate": 0.000999640996023194, |
| "loss": 3.871699333190918, |
| "step": 768 |
| }, |
| { |
| "epoch": 0.047295736917463395, |
| "grad_norm": 0.3245032727718353, |
| "learning_rate": 0.0009985588674043958, |
| "loss": 2.9131131172180176, |
| "step": 1024 |
| }, |
| { |
| "epoch": 0.047295736917463395, |
| "eval_bleu": 0.37680876061744384, |
| "eval_ce_loss": 2.7332340827271273, |
| "eval_loss": 2.7332340827271273, |
| "step": 1024 |
| }, |
| { |
| "epoch": 0.047295736917463395, |
| "eval_bleu": 0.37680876061744384, |
| "eval_ce_loss": 2.7332340827271273, |
| "eval_loss": 2.7332340827271273, |
| "eval_runtime": 123.8633, |
| "eval_samples_per_second": 225.999, |
| "eval_steps_per_second": 3.536, |
| "step": 1024 |
| }, |
| { |
| "epoch": 0.05911967114682925, |
| "grad_norm": 0.40771251916885376, |
| "learning_rate": 0.0009967551747861387, |
| "loss": 2.60915207862854, |
| "step": 1280 |
| }, |
| { |
| "epoch": 0.0709436053761951, |
| "grad_norm": 0.456460177898407, |
| "learning_rate": 0.000994232528651847, |
| "loss": 2.4321346282958984, |
| "step": 1536 |
| }, |
| { |
| "epoch": 0.08276753960556095, |
| "grad_norm": 0.583853542804718, |
| "learning_rate": 0.0009909945800260092, |
| "loss": 2.3184077739715576, |
| "step": 1792 |
| }, |
| { |
| "epoch": 0.09459147383492679, |
| "grad_norm": 0.4923252463340759, |
| "learning_rate": 0.0009870460151900522, |
| "loss": 2.2322187423706055, |
| "step": 2048 |
| }, |
| { |
| "epoch": 0.09459147383492679, |
| "eval_bleu": 0.44342304632288854, |
| "eval_ce_loss": 2.180436749164372, |
| "eval_loss": 2.180436749164372, |
| "step": 2048 |
| }, |
| { |
| "epoch": 0.09459147383492679, |
| "eval_bleu": 0.44342304632288854, |
| "eval_ce_loss": 2.180436749164372, |
| "eval_loss": 2.180436749164372, |
| "eval_runtime": 112.2247, |
| "eval_samples_per_second": 249.437, |
| "eval_steps_per_second": 3.903, |
| "step": 2048 |
| }, |
| { |
| "epoch": 0.10641540806429264, |
| "grad_norm": 0.4688430726528168, |
| "learning_rate": 0.0009823925488998885, |
| "loss": 2.177077531814575, |
| "step": 2304 |
| }, |
| { |
| "epoch": 0.1182393422936585, |
| "grad_norm": 0.559630274772644, |
| "learning_rate": 0.0009770409161149525, |
| "loss": 2.114898920059204, |
| "step": 2560 |
| }, |
| { |
| "epoch": 0.13006327652302435, |
| "grad_norm": 0.5434157848358154, |
| "learning_rate": 0.0009709988622506973, |
| "loss": 2.0660085678100586, |
| "step": 2816 |
| }, |
| { |
| "epoch": 0.1418872107523902, |
| "grad_norm": 0.6108939051628113, |
| "learning_rate": 0.000964275131968659, |
| "loss": 2.0337276458740234, |
| "step": 3072 |
| }, |
| { |
| "epoch": 0.1418872107523902, |
| "eval_bleu": 0.45990036074581603, |
| "eval_ce_loss": 2.022843875841463, |
| "eval_loss": 2.022843875841463, |
| "step": 3072 |
| }, |
| { |
| "epoch": 0.1418872107523902, |
| "eval_bleu": 0.45990036074581603, |
| "eval_ce_loss": 2.022843875841463, |
| "eval_loss": 2.022843875841463, |
| "eval_runtime": 115.2151, |
| "eval_samples_per_second": 242.963, |
| "eval_steps_per_second": 3.802, |
| "step": 3072 |
| }, |
| { |
| "epoch": 0.15371114498175603, |
| "grad_norm": 0.784342348575592, |
| "learning_rate": 0.0009568794565203123, |
| "loss": 2.0136122703552246, |
| "step": 3328 |
| }, |
| { |
| "epoch": 0.1655350792111219, |
| "grad_norm": 0.5167074203491211, |
| "learning_rate": 0.0009488225396630347, |
| "loss": 1.9863924980163574, |
| "step": 3584 |
| }, |
| { |
| "epoch": 0.17735901344048774, |
| "grad_norm": 0.6403853297233582, |
| "learning_rate": 0.0009401160421685646, |
| "loss": 1.9703547954559326, |
| "step": 3840 |
| }, |
| { |
| "epoch": 0.18918294766985358, |
| "grad_norm": 0.5935908555984497, |
| "learning_rate": 0.0009307725649463714, |
| "loss": 1.9429030418395996, |
| "step": 4096 |
| }, |
| { |
| "epoch": 0.18918294766985358, |
| "eval_bleu": 0.47210647699382713, |
| "eval_ce_loss": 1.938350352522445, |
| "eval_loss": 1.938350352522445, |
| "step": 4096 |
| }, |
| { |
| "epoch": 0.18918294766985358, |
| "eval_bleu": 0.47210647699382713, |
| "eval_ce_loss": 1.938350352522445, |
| "eval_loss": 1.938350352522445, |
| "eval_runtime": 115.0632, |
| "eval_samples_per_second": 243.284, |
| "eval_steps_per_second": 3.807, |
| "step": 4096 |
| }, |
| { |
| "epoch": 0.20100688189921945, |
| "grad_norm": 0.6215645670890808, |
| "learning_rate": 0.0009208056308063659, |
| "loss": 1.93471097946167, |
| "step": 4352 |
| }, |
| { |
| "epoch": 0.2128308161285853, |
| "grad_norm": 0.6367761492729187, |
| "learning_rate": 0.0009102296648873445, |
| "loss": 1.9198952913284302, |
| "step": 4608 |
| }, |
| { |
| "epoch": 0.22465475035795113, |
| "grad_norm": 0.686975359916687, |
| "learning_rate": 0.0008990599737794927, |
| "loss": 1.905943512916565, |
| "step": 4864 |
| }, |
| { |
| "epoch": 0.236478684587317, |
| "grad_norm": 0.5950681567192078, |
| "learning_rate": 0.0008873127233711644, |
| "loss": 1.8912562131881714, |
| "step": 5120 |
| }, |
| { |
| "epoch": 0.236478684587317, |
| "eval_bleu": 0.4805682328833919, |
| "eval_ce_loss": 1.8848687097362187, |
| "eval_loss": 1.8848687097362187, |
| "step": 5120 |
| }, |
| { |
| "epoch": 0.236478684587317, |
| "eval_bleu": 0.4805682328833919, |
| "eval_ce_loss": 1.8848687097362187, |
| "eval_loss": 1.8848687097362187, |
| "eval_runtime": 115.9176, |
| "eval_samples_per_second": 241.49, |
| "eval_steps_per_second": 3.779, |
| "step": 5120 |
| }, |
| { |
| "epoch": 0.24830261881668284, |
| "grad_norm": 0.5887701511383057, |
| "learning_rate": 0.0008750049154520011, |
| "loss": 1.868552327156067, |
| "step": 5376 |
| }, |
| { |
| "epoch": 0.2601265530460487, |
| "grad_norm": 0.6690143942832947, |
| "learning_rate": 0.0008621543631062487, |
| "loss": 1.8647550344467163, |
| "step": 5632 |
| }, |
| { |
| "epoch": 0.27195048727541454, |
| "grad_norm": 0.6140768527984619, |
| "learning_rate": 0.0008487796649318904, |
| "loss": 1.8564214706420898, |
| "step": 5888 |
| }, |
| { |
| "epoch": 0.2837744215047804, |
| "grad_norm": 0.720294713973999, |
| "learning_rate": 0.0008349001781229053, |
| "loss": 1.852386236190796, |
| "step": 6144 |
| }, |
| { |
| "epoch": 0.2837744215047804, |
| "eval_bleu": 0.48331258984762954, |
| "eval_ce_loss": 1.8427181676642534, |
| "eval_loss": 1.8427181676642534, |
| "step": 6144 |
| }, |
| { |
| "epoch": 0.2837744215047804, |
| "eval_bleu": 0.48331258984762954, |
| "eval_ce_loss": 1.8427181676642534, |
| "eval_loss": 1.8427181676642534, |
| "eval_runtime": 114.2068, |
| "eval_samples_per_second": 245.108, |
| "eval_steps_per_second": 3.835, |
| "step": 6144 |
| }, |
| { |
| "epoch": 0.2955983557341462, |
| "grad_norm": 0.6018796563148499, |
| "learning_rate": 0.0008205359904536107, |
| "loss": 1.8348900079727173, |
| "step": 6400 |
| }, |
| { |
| "epoch": 0.30742228996351206, |
| "grad_norm": 0.5634647607803345, |
| "learning_rate": 0.0008057078912056363, |
| "loss": 1.8397186994552612, |
| "step": 6656 |
| }, |
| { |
| "epoch": 0.3192462241928779, |
| "grad_norm": 0.7408234477043152, |
| "learning_rate": 0.0007904373410796086, |
| "loss": 1.8375225067138672, |
| "step": 6912 |
| }, |
| { |
| "epoch": 0.3310701584222438, |
| "grad_norm": 0.7126314043998718, |
| "learning_rate": 0.0007747464411350876, |
| "loss": 1.8180835247039795, |
| "step": 7168 |
| }, |
| { |
| "epoch": 0.3310701584222438, |
| "eval_bleu": 0.4866456652902785, |
| "eval_ce_loss": 1.8144619418605823, |
| "eval_loss": 1.8144619418605823, |
| "step": 7168 |
| }, |
| { |
| "epoch": 0.3310701584222438, |
| "eval_bleu": 0.4866456652902785, |
| "eval_ce_loss": 1.8144619418605823, |
| "eval_loss": 1.8144619418605823, |
| "eval_runtime": 113.3842, |
| "eval_samples_per_second": 246.886, |
| "eval_steps_per_second": 3.863, |
| "step": 7168 |
| }, |
| { |
| "epoch": 0.34289409265160964, |
| "grad_norm": 0.7098305225372314, |
| "learning_rate": 0.000758657900803716, |
| "loss": 1.8147518634796143, |
| "step": 7424 |
| }, |
| { |
| "epoch": 0.3547180268809755, |
| "grad_norm": 0.7013034820556641, |
| "learning_rate": 0.000742195005021869, |
| "loss": 1.8021713495254517, |
| "step": 7680 |
| }, |
| { |
| "epoch": 0.3665419611103413, |
| "grad_norm": 0.7947620749473572, |
| "learning_rate": 0.0007253815805303786, |
| "loss": 1.8060873746871948, |
| "step": 7936 |
| }, |
| { |
| "epoch": 0.37836589533970716, |
| "grad_norm": 0.6340846419334412, |
| "learning_rate": 0.0007082419613901028, |
| "loss": 1.7940701246261597, |
| "step": 8192 |
| }, |
| { |
| "epoch": 0.37836589533970716, |
| "eval_bleu": 0.4905398670578081, |
| "eval_ce_loss": 1.7928550904744291, |
| "eval_loss": 1.7928550904744291, |
| "step": 8192 |
| }, |
| { |
| "epoch": 0.37836589533970716, |
| "eval_bleu": 0.4905398670578081, |
| "eval_ce_loss": 1.7928550904744291, |
| "eval_loss": 1.7928550904744291, |
| "eval_runtime": 117.9382, |
| "eval_samples_per_second": 237.353, |
| "eval_steps_per_second": 3.714, |
| "step": 8192 |
| }, |
| { |
| "epoch": 0.390189829569073, |
| "grad_norm": 0.7005776166915894, |
| "learning_rate": 0.0006908009537632514, |
| "loss": 1.7938249111175537, |
| "step": 8448 |
| }, |
| { |
| "epoch": 0.4020137637984389, |
| "grad_norm": 0.6635662317276001, |
| "learning_rate": 0.0006730838000114403, |
| "loss": 1.7853164672851562, |
| "step": 8704 |
| }, |
| { |
| "epoch": 0.41383769802780473, |
| "grad_norm": 0.6025717854499817, |
| "learning_rate": 0.0006551161421624341, |
| "loss": 1.7877322435379028, |
| "step": 8960 |
| }, |
| { |
| "epoch": 0.4256616322571706, |
| "grad_norm": 0.6658259034156799, |
| "learning_rate": 0.0006369239847984517, |
| "loss": 1.7728495597839355, |
| "step": 9216 |
| }, |
| { |
| "epoch": 0.4256616322571706, |
| "eval_bleu": 0.491521411484475, |
| "eval_ce_loss": 1.7803823493387056, |
| "eval_loss": 1.7803823493387056, |
| "step": 9216 |
| }, |
| { |
| "epoch": 0.4256616322571706, |
| "eval_bleu": 0.491521411484475, |
| "eval_ce_loss": 1.7803823493387056, |
| "eval_loss": 1.7803823493387056, |
| "eval_runtime": 113.1853, |
| "eval_samples_per_second": 247.32, |
| "eval_steps_per_second": 3.87, |
| "step": 9216 |
| }, |
| { |
| "epoch": 0.4374855664865364, |
| "grad_norm": 0.6417582035064697, |
| "learning_rate": 0.0006185336574197479, |
| "loss": 1.767223596572876, |
| "step": 9472 |
| }, |
| { |
| "epoch": 0.44930950071590225, |
| "grad_norm": 0.6002334356307983, |
| "learning_rate": 0.0005999717763379407, |
| "loss": 1.7725470066070557, |
| "step": 9728 |
| }, |
| { |
| "epoch": 0.4611334349452681, |
| "grad_norm": 0.6244451403617859, |
| "learning_rate": 0.0005812652061542363, |
| "loss": 1.7692821025848389, |
| "step": 9984 |
| }, |
| { |
| "epoch": 0.472957369174634, |
| "grad_norm": 0.8042502999305725, |
| "learning_rate": 0.0005624410208783071, |
| "loss": 1.7654914855957031, |
| "step": 10240 |
| }, |
| { |
| "epoch": 0.472957369174634, |
| "eval_bleu": 0.4928508058708409, |
| "eval_ce_loss": 1.7702069639070939, |
| "eval_loss": 1.7702069639070939, |
| "step": 10240 |
| }, |
| { |
| "epoch": 0.472957369174634, |
| "eval_bleu": 0.4928508058708409, |
| "eval_ce_loss": 1.7702069639070939, |
| "eval_loss": 1.7702069639070939, |
| "eval_runtime": 111.2062, |
| "eval_samples_per_second": 251.722, |
| "eval_steps_per_second": 3.939, |
| "step": 10240 |
| }, |
| { |
| "epoch": 0.48478130340399983, |
| "grad_norm": 0.6481178998947144, |
| "learning_rate": 0.0005435264647440881, |
| "loss": 1.7591522932052612, |
| "step": 10496 |
| }, |
| { |
| "epoch": 0.49660523763336567, |
| "grad_norm": 0.6686874628067017, |
| "learning_rate": 0.000524548912779213, |
| "loss": 1.7500008344650269, |
| "step": 10752 |
| }, |
| { |
| "epoch": 0.5084291718627315, |
| "grad_norm": 0.6231724619865417, |
| "learning_rate": 0.0005055358311851499, |
| "loss": 1.7510260343551636, |
| "step": 11008 |
| }, |
| { |
| "epoch": 0.5202531060920974, |
| "grad_norm": 0.6039224863052368, |
| "learning_rate": 0.0004865147375853812, |
| "loss": 1.752295732498169, |
| "step": 11264 |
| }, |
| { |
| "epoch": 0.5202531060920974, |
| "eval_bleu": 0.49508567529295305, |
| "eval_ce_loss": 1.7585640699351759, |
| "eval_loss": 1.7585640699351759, |
| "step": 11264 |
| }, |
| { |
| "epoch": 0.5202531060920974, |
| "eval_bleu": 0.49508567529295305, |
| "eval_ce_loss": 1.7585640699351759, |
| "eval_loss": 1.7585640699351759, |
| "eval_runtime": 112.7663, |
| "eval_samples_per_second": 248.239, |
| "eval_steps_per_second": 3.884, |
| "step": 11264 |
| }, |
| { |
| "epoch": 0.5320770403214632, |
| "grad_norm": 0.6699737310409546, |
| "learning_rate": 0.0004675131611991607, |
| "loss": 1.7415655851364136, |
| "step": 11520 |
| }, |
| { |
| "epoch": 0.5439009745508291, |
| "grad_norm": 0.608718991279602, |
| "learning_rate": 0.0004485586029984899, |
| "loss": 1.7422733306884766, |
| "step": 11776 |
| }, |
| { |
| "epoch": 0.5557249087801949, |
| "grad_norm": 0.6244316697120667, |
| "learning_rate": 0.00042967849590597266, |
| "loss": 1.729731559753418, |
| "step": 12032 |
| }, |
| { |
| "epoch": 0.5675488430095608, |
| "grad_norm": 0.6394057869911194, |
| "learning_rate": 0.0004109001650911621, |
| "loss": 1.7405071258544922, |
| "step": 12288 |
| }, |
| { |
| "epoch": 0.5675488430095608, |
| "eval_bleu": 0.4976184098974849, |
| "eval_ce_loss": 1.7458034745634419, |
| "eval_loss": 1.7458034745634419, |
| "step": 12288 |
| }, |
| { |
| "epoch": 0.5675488430095608, |
| "eval_bleu": 0.4976184098974849, |
| "eval_ce_loss": 1.7458034745634419, |
| "eval_loss": 1.7458034745634419, |
| "eval_runtime": 113.3537, |
| "eval_samples_per_second": 246.953, |
| "eval_steps_per_second": 3.864, |
| "step": 12288 |
| }, |
| { |
| "epoch": 0.5793727772389267, |
| "grad_norm": 0.6351113319396973, |
| "learning_rate": 0.0003922507884228551, |
| "loss": 1.7407402992248535, |
| "step": 12544 |
| }, |
| { |
| "epoch": 0.5911967114682924, |
| "grad_norm": 0.6591783761978149, |
| "learning_rate": 0.00037375735713457723, |
| "loss": 1.7183029651641846, |
| "step": 12800 |
| }, |
| { |
| "epoch": 0.6030206456976583, |
| "grad_norm": 0.6365231275558472, |
| "learning_rate": 0.00035544663676018276, |
| "loss": 1.7422548532485962, |
| "step": 13056 |
| }, |
| { |
| "epoch": 0.6148445799270241, |
| "grad_norm": 0.6042783260345459, |
| "learning_rate": 0.00033734512839611255, |
| "loss": 1.731939673423767, |
| "step": 13312 |
| }, |
| { |
| "epoch": 0.6148445799270241, |
| "eval_bleu": 0.5030280015017363, |
| "eval_ce_loss": 1.7251083521538129, |
| "eval_loss": 1.7251083521538129, |
| "step": 13312 |
| }, |
| { |
| "epoch": 0.6148445799270241, |
| "eval_bleu": 0.5030280015017363, |
| "eval_ce_loss": 1.7251083521538129, |
| "eval_loss": 1.7251083521538129, |
| "eval_runtime": 114.7776, |
| "eval_samples_per_second": 243.889, |
| "eval_steps_per_second": 3.816, |
| "step": 13312 |
| }, |
| { |
| "epoch": 0.62666851415639, |
| "grad_norm": 0.8445199131965637, |
| "learning_rate": 0.0003194790303463687, |
| "loss": 1.7291183471679688, |
| "step": 13568 |
| }, |
| { |
| "epoch": 0.6384924483857558, |
| "grad_norm": 0.6880821585655212, |
| "learning_rate": 0.00030187420020572406, |
| "loss": 1.7337849140167236, |
| "step": 13824 |
| }, |
| { |
| "epoch": 0.6503163826151217, |
| "grad_norm": 0.6618293523788452, |
| "learning_rate": 0.00028455611743603626, |
| "loss": 1.7243127822875977, |
| "step": 14080 |
| }, |
| { |
| "epoch": 0.6621403168444876, |
| "grad_norm": 0.6698670387268066, |
| "learning_rate": 0.0002675498464898373, |
| "loss": 1.7110638618469238, |
| "step": 14336 |
| }, |
| { |
| "epoch": 0.6621403168444876, |
| "eval_bleu": 0.5033082348172044, |
| "eval_ce_loss": 1.7181274387390102, |
| "eval_loss": 1.7181274387390102, |
| "step": 14336 |
| }, |
| { |
| "epoch": 0.6621403168444876, |
| "eval_bleu": 0.5033082348172044, |
| "eval_ce_loss": 1.7181274387390102, |
| "eval_loss": 1.7181274387390102, |
| "eval_runtime": 113.7527, |
| "eval_samples_per_second": 246.086, |
| "eval_steps_per_second": 3.85, |
| "step": 14336 |
| }, |
| { |
| "epoch": 0.6739642510738534, |
| "grad_norm": 0.7639990448951721, |
| "learning_rate": 0.0002508800005345623, |
| "loss": 1.7315621376037598, |
| "step": 14592 |
| }, |
| { |
| "epoch": 0.6857881853032193, |
| "grad_norm": 0.6297742128372192, |
| "learning_rate": 0.00023457070582992562, |
| "loss": 1.7225186824798584, |
| "step": 14848 |
| }, |
| { |
| "epoch": 0.6976121195325851, |
| "grad_norm": 0.6663972735404968, |
| "learning_rate": 0.00021864556680999692, |
| "loss": 1.726626992225647, |
| "step": 15104 |
| }, |
| { |
| "epoch": 0.709436053761951, |
| "grad_norm": 0.6332765221595764, |
| "learning_rate": 0.0002031276319205152, |
| "loss": 1.7189596891403198, |
| "step": 15360 |
| }, |
| { |
| "epoch": 0.709436053761951, |
| "eval_bleu": 0.499889278094126, |
| "eval_ce_loss": 1.7232733518565626, |
| "eval_loss": 1.7232733518565626, |
| "step": 15360 |
| }, |
| { |
| "epoch": 0.709436053761951, |
| "eval_bleu": 0.499889278094126, |
| "eval_ce_loss": 1.7232733518565626, |
| "eval_loss": 1.7232733518565626, |
| "eval_runtime": 112.4417, |
| "eval_samples_per_second": 248.956, |
| "eval_steps_per_second": 3.895, |
| "step": 15360 |
| }, |
| { |
| "epoch": 0.7212599879913169, |
| "grad_norm": 0.7163941264152527, |
| "learning_rate": 0.00018803936026088542, |
| "loss": 1.7184514999389648, |
| "step": 15616 |
| }, |
| { |
| "epoch": 0.7330839222206826, |
| "grad_norm": 0.6824806332588196, |
| "learning_rate": 0.00017340258907913464, |
| "loss": 1.715112566947937, |
| "step": 15872 |
| }, |
| { |
| "epoch": 0.7449078564500485, |
| "grad_norm": 0.6306845545768738, |
| "learning_rate": 0.0001592385021668743, |
| "loss": 1.7062709331512451, |
| "step": 16128 |
| }, |
| { |
| "epoch": 0.7567317906794143, |
| "grad_norm": 0.6257657408714294, |
| "learning_rate": 0.0001455675992000087, |
| "loss": 1.7213497161865234, |
| "step": 16384 |
| }, |
| { |
| "epoch": 0.7567317906794143, |
| "eval_bleu": 0.5018863282352667, |
| "eval_ce_loss": 1.712352893123888, |
| "eval_loss": 1.712352893123888, |
| "step": 16384 |
| }, |
| { |
| "epoch": 0.7567317906794143, |
| "eval_bleu": 0.5018863282352667, |
| "eval_ce_loss": 1.712352893123888, |
| "eval_loss": 1.712352893123888, |
| "eval_runtime": 114.1951, |
| "eval_samples_per_second": 245.133, |
| "eval_steps_per_second": 3.836, |
| "step": 16384 |
| }, |
| { |
| "epoch": 0.7685557249087802, |
| "grad_norm": 0.660374104976654, |
| "learning_rate": 0.000132409666069565, |
| "loss": 1.7006453275680542, |
| "step": 16640 |
| }, |
| { |
| "epoch": 0.780379659138146, |
| "grad_norm": 0.6620026230812073, |
| "learning_rate": 0.0001197837462455823, |
| "loss": 1.708181381225586, |
| "step": 16896 |
| }, |
| { |
| "epoch": 0.7922035933675119, |
| "grad_norm": 0.6751854419708252, |
| "learning_rate": 0.00010770811321550749, |
| "loss": 1.6984927654266357, |
| "step": 17152 |
| }, |
| { |
| "epoch": 0.8040275275968778, |
| "grad_norm": 0.6999450325965881, |
| "learning_rate": 9.620024403698591e-05, |
| "loss": 1.7066656351089478, |
| "step": 17408 |
| }, |
| { |
| "epoch": 0.8040275275968778, |
| "eval_bleu": 0.5057314692968334, |
| "eval_ce_loss": 1.6962808190959773, |
| "eval_loss": 1.6962808190959773, |
| "step": 17408 |
| }, |
| { |
| "epoch": 0.8040275275968778, |
| "eval_bleu": 0.5057314692968334, |
| "eval_ce_loss": 1.6962808190959773, |
| "eval_loss": 1.6962808190959773, |
| "eval_runtime": 113.9552, |
| "eval_samples_per_second": 245.649, |
| "eval_steps_per_second": 3.844, |
| "step": 17408 |
| }, |
| { |
| "epoch": 0.8158514618262436, |
| "grad_norm": 0.6427358388900757, |
| "learning_rate": 8.527679404332429e-05, |
| "loss": 1.701072096824646, |
| "step": 17664 |
| }, |
| { |
| "epoch": 0.8276753960556095, |
| "grad_norm": 0.6666444540023804, |
| "learning_rate": 7.495357273823544e-05, |
| "loss": 1.7122679948806763, |
| "step": 17920 |
| }, |
| { |
| "epoch": 0.8394993302849753, |
| "grad_norm": 0.6453679203987122, |
| "learning_rate": 6.524552091475183e-05, |
| "loss": 1.7073798179626465, |
| "step": 18176 |
| }, |
| { |
| "epoch": 0.8513232645143411, |
| "grad_norm": 0.648821234703064, |
| "learning_rate": 5.6166689031422024e-05, |
| "loss": 1.6983073949813843, |
| "step": 18432 |
| }, |
| { |
| "epoch": 0.8513232645143411, |
| "eval_bleu": 0.5035658750943952, |
| "eval_ce_loss": 1.7023935086650936, |
| "eval_loss": 1.7023935086650936, |
| "step": 18432 |
| }, |
| { |
| "epoch": 0.8513232645143411, |
| "eval_bleu": 0.5035658750943952, |
| "eval_ce_loss": 1.7023935086650936, |
| "eval_loss": 1.7023935086650936, |
| "eval_runtime": 112.2723, |
| "eval_samples_per_second": 249.331, |
| "eval_steps_per_second": 3.901, |
| "step": 18432 |
| }, |
| { |
| "epoch": 0.8631471987437069, |
| "grad_norm": 0.6233495473861694, |
| "learning_rate": 4.773021687709067e-05, |
| "loss": 1.7038770914077759, |
| "step": 18688 |
| }, |
| { |
| "epoch": 0.8749711329730728, |
| "grad_norm": 0.6371004581451416, |
| "learning_rate": 3.994831455368719e-05, |
| "loss": 1.70210599899292, |
| "step": 18944 |
| }, |
| { |
| "epoch": 0.8867950672024387, |
| "grad_norm": 0.6629026532173157, |
| "learning_rate": 3.283224480455282e-05, |
| "loss": 1.7117823362350464, |
| "step": 19200 |
| }, |
| { |
| "epoch": 0.8986190014318045, |
| "grad_norm": 0.640360414981842, |
| "learning_rate": 2.639230671387627e-05, |
| "loss": 1.6973387002944946, |
| "step": 19456 |
| }, |
| { |
| "epoch": 0.8986190014318045, |
| "eval_bleu": 0.5034783964085036, |
| "eval_ce_loss": 1.7032495644538914, |
| "eval_loss": 1.7032495644538914, |
| "step": 19456 |
| }, |
| { |
| "epoch": 0.8986190014318045, |
| "eval_bleu": 0.5034783964085036, |
| "eval_ce_loss": 1.7032495644538914, |
| "eval_loss": 1.7032495644538914, |
| "eval_runtime": 113.7629, |
| "eval_samples_per_second": 246.064, |
| "eval_steps_per_second": 3.85, |
| "step": 19456 |
| }, |
| { |
| "epoch": 0.9104429356611704, |
| "grad_norm": 0.625438392162323, |
| "learning_rate": 2.063782080083576e-05, |
| "loss": 1.695639967918396, |
| "step": 19712 |
| }, |
| { |
| "epoch": 0.9222668698905362, |
| "grad_norm": 0.6405285000801086, |
| "learning_rate": 1.557711553001523e-05, |
| "loss": 1.697039246559143, |
| "step": 19968 |
| }, |
| { |
| "epoch": 0.9340908041199021, |
| "grad_norm": 0.6282166838645935, |
| "learning_rate": 1.1217515257622269e-05, |
| "loss": 1.6970669031143188, |
| "step": 20224 |
| }, |
| { |
| "epoch": 0.945914738349268, |
| "grad_norm": 0.6324545741081238, |
| "learning_rate": 7.565329630950746e-06, |
| "loss": 1.7013015747070312, |
| "step": 20480 |
| }, |
| { |
| "epoch": 0.945914738349268, |
| "eval_bleu": 0.5066154325587271, |
| "eval_ce_loss": 1.6894163007605565, |
| "eval_loss": 1.6894163007605565, |
| "step": 20480 |
| }, |
| { |
| "epoch": 0.945914738349268, |
| "eval_bleu": 0.5066154325587271, |
| "eval_ce_loss": 1.6894163007605565, |
| "eval_loss": 1.6894163007605565, |
| "eval_runtime": 112.9102, |
| "eval_samples_per_second": 247.923, |
| "eval_steps_per_second": 3.879, |
| "step": 20480 |
| }, |
| { |
| "epoch": 0.9577386725786338, |
| "grad_norm": 0.6203088164329529, |
| "learning_rate": 4.62584445643166e-06, |
| "loss": 1.6999309062957764, |
| "step": 20736 |
| }, |
| { |
| "epoch": 0.9695626068079997, |
| "grad_norm": 0.6161953806877136, |
| "learning_rate": 2.40331404948807e-06, |
| "loss": 1.7076772451400757, |
| "step": 20992 |
| }, |
| { |
| "epoch": 0.9813865410373654, |
| "grad_norm": 0.6234349012374878, |
| "learning_rate": 9.009550772663965e-07, |
| "loss": 1.6992141008377075, |
| "step": 21248 |
| }, |
| { |
| "epoch": 0.9932104752667313, |
| "grad_norm": 0.6052360534667969, |
| "learning_rate": 1.2094190315575791e-07, |
| "loss": 1.7013704776763916, |
| "step": 21504 |
| }, |
| { |
| "epoch": 0.9932104752667313, |
| "eval_bleu": 0.5054799917933074, |
| "eval_ce_loss": 1.699716367678011, |
| "eval_loss": 1.699716367678011, |
| "step": 21504 |
| }, |
| { |
| "epoch": 0.9932104752667313, |
| "eval_bleu": 0.5054799917933074, |
| "eval_ce_loss": 1.699716367678011, |
| "eval_loss": 1.699716367678011, |
| "eval_runtime": 113.8908, |
| "eval_samples_per_second": 245.788, |
| "eval_steps_per_second": 3.846, |
| "step": 21504 |
| } |
| ], |
| "logging_steps": 256, |
| "max_steps": 21651, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 1024, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
