| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 500, |
| "global_step": 103, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.4083243012428284, |
| "epoch": 0.009708737864077669, |
| "grad_norm": 0.94761723279953, |
| "learning_rate": 0.0, |
| "loss": 1.9706486463546753, |
| "mean_token_accuracy": 0.5420544818043709, |
| "num_tokens": 97105.0, |
| "step": 1 |
| }, |
| { |
| "entropy": 1.421248883008957, |
| "epoch": 0.019417475728155338, |
| "grad_norm": 1.021393060684204, |
| "learning_rate": 3.125e-06, |
| "loss": 2.03873872756958, |
| "mean_token_accuracy": 0.5355476140975952, |
| "num_tokens": 192679.0, |
| "step": 2 |
| }, |
| { |
| "entropy": 1.4670909941196442, |
| "epoch": 0.02912621359223301, |
| "grad_norm": 0.8904154896736145, |
| "learning_rate": 6.25e-06, |
| "loss": 1.994919776916504, |
| "mean_token_accuracy": 0.5325814262032509, |
| "num_tokens": 290599.0, |
| "step": 3 |
| }, |
| { |
| "entropy": 1.5951472371816635, |
| "epoch": 0.038834951456310676, |
| "grad_norm": 0.6973261833190918, |
| "learning_rate": 9.375000000000001e-06, |
| "loss": 1.9966566562652588, |
| "mean_token_accuracy": 0.5266260281205177, |
| "num_tokens": 388130.0, |
| "step": 4 |
| }, |
| { |
| "entropy": 1.6103769689798355, |
| "epoch": 0.04854368932038835, |
| "grad_norm": 0.5392856597900391, |
| "learning_rate": 1.25e-05, |
| "loss": 1.8900949954986572, |
| "mean_token_accuracy": 0.5452185645699501, |
| "num_tokens": 485983.0, |
| "step": 5 |
| }, |
| { |
| "entropy": 1.7284279018640518, |
| "epoch": 0.05825242718446602, |
| "grad_norm": 0.4281271696090698, |
| "learning_rate": 1.5625e-05, |
| "loss": 1.8690071105957031, |
| "mean_token_accuracy": 0.5448430627584457, |
| "num_tokens": 583257.0, |
| "step": 6 |
| }, |
| { |
| "entropy": 1.79091015458107, |
| "epoch": 0.06796116504854369, |
| "grad_norm": 0.4108770191669464, |
| "learning_rate": 1.8750000000000002e-05, |
| "loss": 1.8907827138900757, |
| "mean_token_accuracy": 0.5405718386173248, |
| "num_tokens": 678553.0, |
| "step": 7 |
| }, |
| { |
| "entropy": 1.8500836789608002, |
| "epoch": 0.07766990291262135, |
| "grad_norm": 0.35201919078826904, |
| "learning_rate": 2.1875e-05, |
| "loss": 1.8045190572738647, |
| "mean_token_accuracy": 0.5530312657356262, |
| "num_tokens": 774991.0, |
| "step": 8 |
| }, |
| { |
| "entropy": 1.9261595904827118, |
| "epoch": 0.08737864077669903, |
| "grad_norm": 0.36849886178970337, |
| "learning_rate": 2.5e-05, |
| "loss": 1.7962449789047241, |
| "mean_token_accuracy": 0.5506185740232468, |
| "num_tokens": 871685.0, |
| "step": 9 |
| }, |
| { |
| "entropy": 1.875639259815216, |
| "epoch": 0.0970873786407767, |
| "grad_norm": 0.36218366026878357, |
| "learning_rate": 2.8125000000000003e-05, |
| "loss": 1.703657627105713, |
| "mean_token_accuracy": 0.5688499510288239, |
| "num_tokens": 968934.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.8956582099199295, |
| "epoch": 0.10679611650485436, |
| "grad_norm": 0.38436606526374817, |
| "learning_rate": 3.125e-05, |
| "loss": 1.6929868459701538, |
| "mean_token_accuracy": 0.5671246349811554, |
| "num_tokens": 1066190.0, |
| "step": 11 |
| }, |
| { |
| "entropy": 1.9003087729215622, |
| "epoch": 0.11650485436893204, |
| "grad_norm": 0.38454198837280273, |
| "learning_rate": 3.4375e-05, |
| "loss": 1.6784961223602295, |
| "mean_token_accuracy": 0.5702497884631157, |
| "num_tokens": 1163717.0, |
| "step": 12 |
| }, |
| { |
| "entropy": 1.7804109752178192, |
| "epoch": 0.1262135922330097, |
| "grad_norm": 0.34963515400886536, |
| "learning_rate": 3.7500000000000003e-05, |
| "loss": 1.609951376914978, |
| "mean_token_accuracy": 0.5796908214688301, |
| "num_tokens": 1261260.0, |
| "step": 13 |
| }, |
| { |
| "entropy": 1.6776917278766632, |
| "epoch": 0.13592233009708737, |
| "grad_norm": 0.2801634669303894, |
| "learning_rate": 4.0625000000000005e-05, |
| "loss": 1.5957839488983154, |
| "mean_token_accuracy": 0.5835207626223564, |
| "num_tokens": 1358315.0, |
| "step": 14 |
| }, |
| { |
| "entropy": 1.5199688822031021, |
| "epoch": 0.14563106796116504, |
| "grad_norm": 0.27892422676086426, |
| "learning_rate": 4.375e-05, |
| "loss": 1.5371708869934082, |
| "mean_token_accuracy": 0.5937215387821198, |
| "num_tokens": 1455424.0, |
| "step": 15 |
| }, |
| { |
| "entropy": 1.42391437292099, |
| "epoch": 0.1553398058252427, |
| "grad_norm": 0.36465972661972046, |
| "learning_rate": 4.6875e-05, |
| "loss": 1.5261038541793823, |
| "mean_token_accuracy": 0.5960521325469017, |
| "num_tokens": 1552741.0, |
| "step": 16 |
| }, |
| { |
| "entropy": 1.4002629518508911, |
| "epoch": 0.1650485436893204, |
| "grad_norm": 0.33470022678375244, |
| "learning_rate": 5e-05, |
| "loss": 1.506011962890625, |
| "mean_token_accuracy": 0.5978913903236389, |
| "num_tokens": 1650044.0, |
| "step": 17 |
| }, |
| { |
| "entropy": 1.4228239804506302, |
| "epoch": 0.17475728155339806, |
| "grad_norm": 0.24347813427448273, |
| "learning_rate": 4.999856295503635e-05, |
| "loss": 1.4823517799377441, |
| "mean_token_accuracy": 0.599531851708889, |
| "num_tokens": 1747782.0, |
| "step": 18 |
| }, |
| { |
| "entropy": 1.448434516787529, |
| "epoch": 0.18446601941747573, |
| "grad_norm": 0.20727963745594025, |
| "learning_rate": 4.999425198535325e-05, |
| "loss": 1.463187575340271, |
| "mean_token_accuracy": 0.6064818128943443, |
| "num_tokens": 1845724.0, |
| "step": 19 |
| }, |
| { |
| "entropy": 1.4923068583011627, |
| "epoch": 0.1941747572815534, |
| "grad_norm": 0.2315221130847931, |
| "learning_rate": 4.998706758655528e-05, |
| "loss": 1.4465688467025757, |
| "mean_token_accuracy": 0.6043036803603172, |
| "num_tokens": 1943411.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.472718134522438, |
| "epoch": 0.20388349514563106, |
| "grad_norm": 0.24738778173923492, |
| "learning_rate": 4.997701058458676e-05, |
| "loss": 1.4063471555709839, |
| "mean_token_accuracy": 0.6147852465510368, |
| "num_tokens": 2041045.0, |
| "step": 21 |
| }, |
| { |
| "entropy": 1.4919932037591934, |
| "epoch": 0.21359223300970873, |
| "grad_norm": 0.21912068128585815, |
| "learning_rate": 4.996408213563684e-05, |
| "loss": 1.4053853750228882, |
| "mean_token_accuracy": 0.6167104989290237, |
| "num_tokens": 2138535.0, |
| "step": 22 |
| }, |
| { |
| "entropy": 1.4844385087490082, |
| "epoch": 0.22330097087378642, |
| "grad_norm": 0.22786955535411835, |
| "learning_rate": 4.994828372600649e-05, |
| "loss": 1.3828628063201904, |
| "mean_token_accuracy": 0.6216765120625496, |
| "num_tokens": 2235538.0, |
| "step": 23 |
| }, |
| { |
| "entropy": 1.4436479061841965, |
| "epoch": 0.23300970873786409, |
| "grad_norm": 0.19272480905056, |
| "learning_rate": 4.9929617171937724e-05, |
| "loss": 1.3706296682357788, |
| "mean_token_accuracy": 0.6209864541888237, |
| "num_tokens": 2332954.0, |
| "step": 24 |
| }, |
| { |
| "entropy": 1.4084927588701248, |
| "epoch": 0.24271844660194175, |
| "grad_norm": 0.17803719639778137, |
| "learning_rate": 4.9908084619404735e-05, |
| "loss": 1.3561553955078125, |
| "mean_token_accuracy": 0.6240377500653267, |
| "num_tokens": 2431031.0, |
| "step": 25 |
| }, |
| { |
| "entropy": 1.3695293813943863, |
| "epoch": 0.2524271844660194, |
| "grad_norm": 0.1710546761751175, |
| "learning_rate": 4.988368854386722e-05, |
| "loss": 1.3544114828109741, |
| "mean_token_accuracy": 0.6251663640141487, |
| "num_tokens": 2527933.0, |
| "step": 26 |
| }, |
| { |
| "entropy": 1.2857100814580917, |
| "epoch": 0.2621359223300971, |
| "grad_norm": 0.18847821652889252, |
| "learning_rate": 4.985643174998578e-05, |
| "loss": 1.3139867782592773, |
| "mean_token_accuracy": 0.6356627941131592, |
| "num_tokens": 2625130.0, |
| "step": 27 |
| }, |
| { |
| "entropy": 1.280667930841446, |
| "epoch": 0.27184466019417475, |
| "grad_norm": 0.19685013592243195, |
| "learning_rate": 4.982631737129948e-05, |
| "loss": 1.3132545948028564, |
| "mean_token_accuracy": 0.6352921947836876, |
| "num_tokens": 2722735.0, |
| "step": 28 |
| }, |
| { |
| "entropy": 1.3028772920370102, |
| "epoch": 0.2815533980582524, |
| "grad_norm": 0.19354666769504547, |
| "learning_rate": 4.9793348869865616e-05, |
| "loss": 1.3400448560714722, |
| "mean_token_accuracy": 0.6271930709481239, |
| "num_tokens": 2820721.0, |
| "step": 29 |
| }, |
| { |
| "entropy": 1.3173751831054688, |
| "epoch": 0.2912621359223301, |
| "grad_norm": 0.16872914135456085, |
| "learning_rate": 4.9757530035861716e-05, |
| "loss": 1.3304100036621094, |
| "mean_token_accuracy": 0.6323504224419594, |
| "num_tokens": 2918531.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.284382238984108, |
| "epoch": 0.30097087378640774, |
| "grad_norm": 0.17004726827144623, |
| "learning_rate": 4.971886498714977e-05, |
| "loss": 1.2796059846878052, |
| "mean_token_accuracy": 0.6409082934260368, |
| "num_tokens": 3014027.0, |
| "step": 31 |
| }, |
| { |
| "entropy": 1.3059177845716476, |
| "epoch": 0.3106796116504854, |
| "grad_norm": 0.17118602991104126, |
| "learning_rate": 4.967735816880286e-05, |
| "loss": 1.2877871990203857, |
| "mean_token_accuracy": 0.6396790593862534, |
| "num_tokens": 3111172.0, |
| "step": 32 |
| }, |
| { |
| "entropy": 1.3147014528512955, |
| "epoch": 0.32038834951456313, |
| "grad_norm": 0.16340142488479614, |
| "learning_rate": 4.963301435259413e-05, |
| "loss": 1.2802796363830566, |
| "mean_token_accuracy": 0.6386668905615807, |
| "num_tokens": 3208860.0, |
| "step": 33 |
| }, |
| { |
| "entropy": 1.3065699636936188, |
| "epoch": 0.3300970873786408, |
| "grad_norm": 0.1647864431142807, |
| "learning_rate": 4.95858386364482e-05, |
| "loss": 1.2798354625701904, |
| "mean_token_accuracy": 0.6393692642450333, |
| "num_tokens": 3305964.0, |
| "step": 34 |
| }, |
| { |
| "entropy": 1.2940323948860168, |
| "epoch": 0.33980582524271846, |
| "grad_norm": 0.15510039031505585, |
| "learning_rate": 4.9535836443855096e-05, |
| "loss": 1.2712514400482178, |
| "mean_token_accuracy": 0.6405108198523521, |
| "num_tokens": 3403337.0, |
| "step": 35 |
| }, |
| { |
| "entropy": 1.2636306136846542, |
| "epoch": 0.34951456310679613, |
| "grad_norm": 0.16528593003749847, |
| "learning_rate": 4.948301352324673e-05, |
| "loss": 1.2383002042770386, |
| "mean_token_accuracy": 0.6498897895216942, |
| "num_tokens": 3500500.0, |
| "step": 36 |
| }, |
| { |
| "entropy": 1.2304436564445496, |
| "epoch": 0.3592233009708738, |
| "grad_norm": 0.15354204177856445, |
| "learning_rate": 4.942737594733609e-05, |
| "loss": 1.2233033180236816, |
| "mean_token_accuracy": 0.6514501944184303, |
| "num_tokens": 3597306.0, |
| "step": 37 |
| }, |
| { |
| "entropy": 1.2565466612577438, |
| "epoch": 0.36893203883495146, |
| "grad_norm": 0.15500810742378235, |
| "learning_rate": 4.9368930112419e-05, |
| "loss": 1.2482693195343018, |
| "mean_token_accuracy": 0.645413227379322, |
| "num_tokens": 3694352.0, |
| "step": 38 |
| }, |
| { |
| "entropy": 1.2443159818649292, |
| "epoch": 0.3786407766990291, |
| "grad_norm": 0.15541698038578033, |
| "learning_rate": 4.930768273763889e-05, |
| "loss": 1.2520265579223633, |
| "mean_token_accuracy": 0.6438535004854202, |
| "num_tokens": 3792464.0, |
| "step": 39 |
| }, |
| { |
| "entropy": 1.2296875715255737, |
| "epoch": 0.3883495145631068, |
| "grad_norm": 0.15958574414253235, |
| "learning_rate": 4.92436408642143e-05, |
| "loss": 1.2351443767547607, |
| "mean_token_accuracy": 0.6492253467440605, |
| "num_tokens": 3889619.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.2195448875427246, |
| "epoch": 0.39805825242718446, |
| "grad_norm": 0.1615312397480011, |
| "learning_rate": 4.917681185462934e-05, |
| "loss": 1.2214652299880981, |
| "mean_token_accuracy": 0.6524631232023239, |
| "num_tokens": 3986630.0, |
| "step": 41 |
| }, |
| { |
| "entropy": 1.2140728384256363, |
| "epoch": 0.4077669902912621, |
| "grad_norm": 0.15110917389392853, |
| "learning_rate": 4.910720339178735e-05, |
| "loss": 1.2222949266433716, |
| "mean_token_accuracy": 0.6521077156066895, |
| "num_tokens": 4083691.0, |
| "step": 42 |
| }, |
| { |
| "entropy": 1.2051743566989899, |
| "epoch": 0.4174757281553398, |
| "grad_norm": 0.1491296887397766, |
| "learning_rate": 4.90348234781276e-05, |
| "loss": 1.2039450407028198, |
| "mean_token_accuracy": 0.654543362557888, |
| "num_tokens": 4181870.0, |
| "step": 43 |
| }, |
| { |
| "entropy": 1.2167726755142212, |
| "epoch": 0.42718446601941745, |
| "grad_norm": 0.1601928472518921, |
| "learning_rate": 4.895968043470532e-05, |
| "loss": 1.2167367935180664, |
| "mean_token_accuracy": 0.65306556224823, |
| "num_tokens": 4277672.0, |
| "step": 44 |
| }, |
| { |
| "entropy": 1.1795921474695206, |
| "epoch": 0.4368932038834951, |
| "grad_norm": 0.14628776907920837, |
| "learning_rate": 4.8881782900235094e-05, |
| "loss": 1.1818878650665283, |
| "mean_token_accuracy": 0.6603984162211418, |
| "num_tokens": 4375365.0, |
| "step": 45 |
| }, |
| { |
| "entropy": 1.2080871313810349, |
| "epoch": 0.44660194174757284, |
| "grad_norm": 0.15034087002277374, |
| "learning_rate": 4.880113983009768e-05, |
| "loss": 1.2044813632965088, |
| "mean_token_accuracy": 0.6534772142767906, |
| "num_tokens": 4472689.0, |
| "step": 46 |
| }, |
| { |
| "entropy": 1.2424661964178085, |
| "epoch": 0.4563106796116505, |
| "grad_norm": 0.14910851418972015, |
| "learning_rate": 4.87177604953105e-05, |
| "loss": 1.2322750091552734, |
| "mean_token_accuracy": 0.6469002515077591, |
| "num_tokens": 4569961.0, |
| "step": 47 |
| }, |
| { |
| "entropy": 1.209616169333458, |
| "epoch": 0.46601941747572817, |
| "grad_norm": 0.14911304414272308, |
| "learning_rate": 4.86316544814618e-05, |
| "loss": 1.2000218629837036, |
| "mean_token_accuracy": 0.6570775136351585, |
| "num_tokens": 4667359.0, |
| "step": 48 |
| }, |
| { |
| "entropy": 1.201382964849472, |
| "epoch": 0.47572815533980584, |
| "grad_norm": 0.17708911001682281, |
| "learning_rate": 4.854283168760868e-05, |
| "loss": 1.1916768550872803, |
| "mean_token_accuracy": 0.6561535075306892, |
| "num_tokens": 4765400.0, |
| "step": 49 |
| }, |
| { |
| "entropy": 1.1911841928958893, |
| "epoch": 0.4854368932038835, |
| "grad_norm": 0.1523158699274063, |
| "learning_rate": 4.8451302325139004e-05, |
| "loss": 1.1832383871078491, |
| "mean_token_accuracy": 0.6573370844125748, |
| "num_tokens": 4863204.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 1.1881461590528488, |
| "epoch": 0.49514563106796117, |
| "grad_norm": 0.15072602033615112, |
| "learning_rate": 4.835707691659753e-05, |
| "loss": 1.1761033535003662, |
| "mean_token_accuracy": 0.6590301543474197, |
| "num_tokens": 4960926.0, |
| "step": 51 |
| }, |
| { |
| "entropy": 1.1763849258422852, |
| "epoch": 0.5048543689320388, |
| "grad_norm": 0.15600641071796417, |
| "learning_rate": 4.8260166294476164e-05, |
| "loss": 1.1728320121765137, |
| "mean_token_accuracy": 0.6621676161885262, |
| "num_tokens": 5058013.0, |
| "step": 52 |
| }, |
| { |
| "entropy": 1.1916624307632446, |
| "epoch": 0.5145631067961165, |
| "grad_norm": 0.15378843247890472, |
| "learning_rate": 4.8160581599968625e-05, |
| "loss": 1.184821367263794, |
| "mean_token_accuracy": 0.6572789996862411, |
| "num_tokens": 5153970.0, |
| "step": 53 |
| }, |
| { |
| "entropy": 1.1805166453123093, |
| "epoch": 0.5242718446601942, |
| "grad_norm": 0.15211763978004456, |
| "learning_rate": 4.8058334281689595e-05, |
| "loss": 1.1789004802703857, |
| "mean_token_accuracy": 0.658557578921318, |
| "num_tokens": 5251276.0, |
| "step": 54 |
| }, |
| { |
| "entropy": 1.1733617633581161, |
| "epoch": 0.5339805825242718, |
| "grad_norm": 0.15024270117282867, |
| "learning_rate": 4.7953436094358595e-05, |
| "loss": 1.1704764366149902, |
| "mean_token_accuracy": 0.6585692763328552, |
| "num_tokens": 5348605.0, |
| "step": 55 |
| }, |
| { |
| "entropy": 1.1367567777633667, |
| "epoch": 0.5436893203883495, |
| "grad_norm": 0.14717121422290802, |
| "learning_rate": 4.784589909744855e-05, |
| "loss": 1.1350934505462646, |
| "mean_token_accuracy": 0.6707980632781982, |
| "num_tokens": 5446513.0, |
| "step": 56 |
| }, |
| { |
| "entropy": 1.1633624881505966, |
| "epoch": 0.5533980582524272, |
| "grad_norm": 0.15170669555664062, |
| "learning_rate": 4.7735735653799463e-05, |
| "loss": 1.1654052734375, |
| "mean_token_accuracy": 0.6653316840529442, |
| "num_tokens": 5543963.0, |
| "step": 57 |
| }, |
| { |
| "entropy": 1.1694072782993317, |
| "epoch": 0.5631067961165048, |
| "grad_norm": 0.14897707104682922, |
| "learning_rate": 4.762295842819707e-05, |
| "loss": 1.1644500494003296, |
| "mean_token_accuracy": 0.6641355082392693, |
| "num_tokens": 5641181.0, |
| "step": 58 |
| }, |
| { |
| "entropy": 1.1587684005498886, |
| "epoch": 0.5728155339805825, |
| "grad_norm": 0.14866051077842712, |
| "learning_rate": 4.75075803859169e-05, |
| "loss": 1.1571202278137207, |
| "mean_token_accuracy": 0.6645473390817642, |
| "num_tokens": 5738020.0, |
| "step": 59 |
| }, |
| { |
| "entropy": 1.1717360317707062, |
| "epoch": 0.5825242718446602, |
| "grad_norm": 0.15497078001499176, |
| "learning_rate": 4.7389614791233726e-05, |
| "loss": 1.1673715114593506, |
| "mean_token_accuracy": 0.662636861205101, |
| "num_tokens": 5835528.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 1.1693744212388992, |
| "epoch": 0.5922330097087378, |
| "grad_norm": 0.15867234766483307, |
| "learning_rate": 4.726907520589664e-05, |
| "loss": 1.1548094749450684, |
| "mean_token_accuracy": 0.6624177768826485, |
| "num_tokens": 5933622.0, |
| "step": 61 |
| }, |
| { |
| "entropy": 1.1339199990034103, |
| "epoch": 0.6019417475728155, |
| "grad_norm": 0.15212447941303253, |
| "learning_rate": 4.714597548756996e-05, |
| "loss": 1.1325181722640991, |
| "mean_token_accuracy": 0.6707736179232597, |
| "num_tokens": 6030609.0, |
| "step": 62 |
| }, |
| { |
| "entropy": 1.1296663135290146, |
| "epoch": 0.6116504854368932, |
| "grad_norm": 0.15575166046619415, |
| "learning_rate": 4.702032978824011e-05, |
| "loss": 1.1267575025558472, |
| "mean_token_accuracy": 0.6701638028025627, |
| "num_tokens": 6127908.0, |
| "step": 63 |
| }, |
| { |
| "entropy": 1.1318050771951675, |
| "epoch": 0.6213592233009708, |
| "grad_norm": 0.15813998878002167, |
| "learning_rate": 4.6892152552588655e-05, |
| "loss": 1.1390020847320557, |
| "mean_token_accuracy": 0.6672858372330666, |
| "num_tokens": 6225260.0, |
| "step": 64 |
| }, |
| { |
| "entropy": 1.1477476954460144, |
| "epoch": 0.6310679611650486, |
| "grad_norm": 0.15706756711006165, |
| "learning_rate": 4.6761458516331655e-05, |
| "loss": 1.153120994567871, |
| "mean_token_accuracy": 0.663187600672245, |
| "num_tokens": 6322811.0, |
| "step": 65 |
| }, |
| { |
| "entropy": 1.133908674120903, |
| "epoch": 0.6407766990291263, |
| "grad_norm": 0.1579563319683075, |
| "learning_rate": 4.662826270452565e-05, |
| "loss": 1.1408581733703613, |
| "mean_token_accuracy": 0.6672231107950211, |
| "num_tokens": 6419134.0, |
| "step": 66 |
| }, |
| { |
| "entropy": 1.150125876069069, |
| "epoch": 0.6504854368932039, |
| "grad_norm": 0.15973767638206482, |
| "learning_rate": 4.649258042984026e-05, |
| "loss": 1.1557530164718628, |
| "mean_token_accuracy": 0.6634695678949356, |
| "num_tokens": 6516137.0, |
| "step": 67 |
| }, |
| { |
| "entropy": 1.124921292066574, |
| "epoch": 0.6601941747572816, |
| "grad_norm": 0.15293626487255096, |
| "learning_rate": 4.6354427290797875e-05, |
| "loss": 1.1272459030151367, |
| "mean_token_accuracy": 0.6693216189742088, |
| "num_tokens": 6613796.0, |
| "step": 68 |
| }, |
| { |
| "entropy": 1.1551329046487808, |
| "epoch": 0.6699029126213593, |
| "grad_norm": 0.15234892070293427, |
| "learning_rate": 4.621381916998029e-05, |
| "loss": 1.1462955474853516, |
| "mean_token_accuracy": 0.6653169468045235, |
| "num_tokens": 6711660.0, |
| "step": 69 |
| }, |
| { |
| "entropy": 1.1397110670804977, |
| "epoch": 0.6796116504854369, |
| "grad_norm": 0.15561261773109436, |
| "learning_rate": 4.607077223220285e-05, |
| "loss": 1.1248576641082764, |
| "mean_token_accuracy": 0.6719892099499702, |
| "num_tokens": 6808587.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 1.1522215157747269, |
| "epoch": 0.6893203883495146, |
| "grad_norm": 0.1659662127494812, |
| "learning_rate": 4.592530292265609e-05, |
| "loss": 1.1409823894500732, |
| "mean_token_accuracy": 0.6674581989645958, |
| "num_tokens": 6906120.0, |
| "step": 71 |
| }, |
| { |
| "entropy": 1.1350633352994919, |
| "epoch": 0.6990291262135923, |
| "grad_norm": 0.15686708688735962, |
| "learning_rate": 4.5777427965015096e-05, |
| "loss": 1.128638744354248, |
| "mean_token_accuracy": 0.6704377010464668, |
| "num_tokens": 7004058.0, |
| "step": 72 |
| }, |
| { |
| "entropy": 1.1182153522968292, |
| "epoch": 0.7087378640776699, |
| "grad_norm": 0.15111538767814636, |
| "learning_rate": 4.562716435951692e-05, |
| "loss": 1.1166245937347412, |
| "mean_token_accuracy": 0.6737237125635147, |
| "num_tokens": 7101920.0, |
| "step": 73 |
| }, |
| { |
| "entropy": 1.1188742965459824, |
| "epoch": 0.7184466019417476, |
| "grad_norm": 0.15409265458583832, |
| "learning_rate": 4.5474529381006146e-05, |
| "loss": 1.118614673614502, |
| "mean_token_accuracy": 0.6710415109992027, |
| "num_tokens": 7198756.0, |
| "step": 74 |
| }, |
| { |
| "entropy": 1.1149023473262787, |
| "epoch": 0.7281553398058253, |
| "grad_norm": 0.16151487827301025, |
| "learning_rate": 4.531954057694897e-05, |
| "loss": 1.1311017274856567, |
| "mean_token_accuracy": 0.6720475554466248, |
| "num_tokens": 7296339.0, |
| "step": 75 |
| }, |
| { |
| "entropy": 1.1312980204820633, |
| "epoch": 0.7378640776699029, |
| "grad_norm": 0.1549658626317978, |
| "learning_rate": 4.516221576541581e-05, |
| "loss": 1.1340723037719727, |
| "mean_token_accuracy": 0.6688062027096748, |
| "num_tokens": 7393950.0, |
| "step": 76 |
| }, |
| { |
| "entropy": 1.1147898435592651, |
| "epoch": 0.7475728155339806, |
| "grad_norm": 0.15624849498271942, |
| "learning_rate": 4.5002573033032904e-05, |
| "loss": 1.1091203689575195, |
| "mean_token_accuracy": 0.674225278198719, |
| "num_tokens": 7489205.0, |
| "step": 77 |
| }, |
| { |
| "entropy": 1.137469932436943, |
| "epoch": 0.7572815533980582, |
| "grad_norm": 0.16516146063804626, |
| "learning_rate": 4.484063073290301e-05, |
| "loss": 1.1193323135375977, |
| "mean_token_accuracy": 0.6739235296845436, |
| "num_tokens": 7584161.0, |
| "step": 78 |
| }, |
| { |
| "entropy": 1.1054588705301285, |
| "epoch": 0.7669902912621359, |
| "grad_norm": 0.15215378999710083, |
| "learning_rate": 4.467640748249548e-05, |
| "loss": 1.1007866859436035, |
| "mean_token_accuracy": 0.6783005967736244, |
| "num_tokens": 7681353.0, |
| "step": 79 |
| }, |
| { |
| "entropy": 1.0961224287748337, |
| "epoch": 0.7766990291262136, |
| "grad_norm": 0.15287643671035767, |
| "learning_rate": 4.4509922161505926e-05, |
| "loss": 1.0958147048950195, |
| "mean_token_accuracy": 0.676163524389267, |
| "num_tokens": 7778582.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 1.1254196166992188, |
| "epoch": 0.7864077669902912, |
| "grad_norm": 0.16113276779651642, |
| "learning_rate": 4.4341193909685686e-05, |
| "loss": 1.130951166152954, |
| "mean_token_accuracy": 0.6697586253285408, |
| "num_tokens": 7874950.0, |
| "step": 81 |
| }, |
| { |
| "entropy": 1.0993820428848267, |
| "epoch": 0.7961165048543689, |
| "grad_norm": 0.1620422899723053, |
| "learning_rate": 4.417024212464152e-05, |
| "loss": 1.1005873680114746, |
| "mean_token_accuracy": 0.6766335293650627, |
| "num_tokens": 7971956.0, |
| "step": 82 |
| }, |
| { |
| "entropy": 1.1041197627782822, |
| "epoch": 0.8058252427184466, |
| "grad_norm": 0.15546955168247223, |
| "learning_rate": 4.399708645960559e-05, |
| "loss": 1.10465407371521, |
| "mean_token_accuracy": 0.674575038254261, |
| "num_tokens": 8068461.0, |
| "step": 83 |
| }, |
| { |
| "entropy": 1.0924506783485413, |
| "epoch": 0.8155339805825242, |
| "grad_norm": 0.16429384052753448, |
| "learning_rate": 4.382174682117598e-05, |
| "loss": 1.093902587890625, |
| "mean_token_accuracy": 0.6806196495890617, |
| "num_tokens": 8166250.0, |
| "step": 84 |
| }, |
| { |
| "entropy": 1.1175890564918518, |
| "epoch": 0.8252427184466019, |
| "grad_norm": 0.1590225100517273, |
| "learning_rate": 4.364424336702825e-05, |
| "loss": 1.1183996200561523, |
| "mean_token_accuracy": 0.6712751016020775, |
| "num_tokens": 8263412.0, |
| "step": 85 |
| }, |
| { |
| "entropy": 1.091995120048523, |
| "epoch": 0.8349514563106796, |
| "grad_norm": 0.16027086973190308, |
| "learning_rate": 4.346459650359798e-05, |
| "loss": 1.089324951171875, |
| "mean_token_accuracy": 0.6779745444655418, |
| "num_tokens": 8361054.0, |
| "step": 86 |
| }, |
| { |
| "entropy": 1.1091380715370178, |
| "epoch": 0.8446601941747572, |
| "grad_norm": 0.16367396712303162, |
| "learning_rate": 4.328282688373479e-05, |
| "loss": 1.1101912260055542, |
| "mean_token_accuracy": 0.6726596802473068, |
| "num_tokens": 8458058.0, |
| "step": 87 |
| }, |
| { |
| "entropy": 1.1332892328500748, |
| "epoch": 0.8543689320388349, |
| "grad_norm": 0.16062098741531372, |
| "learning_rate": 4.3098955404328045e-05, |
| "loss": 1.1345933675765991, |
| "mean_token_accuracy": 0.669020764529705, |
| "num_tokens": 8552707.0, |
| "step": 88 |
| }, |
| { |
| "entropy": 1.124135971069336, |
| "epoch": 0.8640776699029126, |
| "grad_norm": 0.15737071633338928, |
| "learning_rate": 4.29130032039044e-05, |
| "loss": 1.1120331287384033, |
| "mean_token_accuracy": 0.6742131784558296, |
| "num_tokens": 8648715.0, |
| "step": 89 |
| }, |
| { |
| "entropy": 1.11882945895195, |
| "epoch": 0.8737864077669902, |
| "grad_norm": 0.15959268808364868, |
| "learning_rate": 4.272499166019771e-05, |
| "loss": 1.117668628692627, |
| "mean_token_accuracy": 0.6719657182693481, |
| "num_tokens": 8744790.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 1.1137337386608124, |
| "epoch": 0.883495145631068, |
| "grad_norm": 0.1567354053258896, |
| "learning_rate": 4.253494238769134e-05, |
| "loss": 1.10910165309906, |
| "mean_token_accuracy": 0.6751311644911766, |
| "num_tokens": 8841963.0, |
| "step": 91 |
| }, |
| { |
| "entropy": 1.0963155031204224, |
| "epoch": 0.8932038834951457, |
| "grad_norm": 0.16550451517105103, |
| "learning_rate": 4.234287723513326e-05, |
| "loss": 1.0936603546142578, |
| "mean_token_accuracy": 0.6774952635169029, |
| "num_tokens": 8938823.0, |
| "step": 92 |
| }, |
| { |
| "entropy": 1.1137651056051254, |
| "epoch": 0.9029126213592233, |
| "grad_norm": 0.15746837854385376, |
| "learning_rate": 4.2148818283024304e-05, |
| "loss": 1.11457097530365, |
| "mean_token_accuracy": 0.6719932034611702, |
| "num_tokens": 9036976.0, |
| "step": 93 |
| }, |
| { |
| "entropy": 1.085622027516365, |
| "epoch": 0.912621359223301, |
| "grad_norm": 0.16476289927959442, |
| "learning_rate": 4.195278784107964e-05, |
| "loss": 1.086329460144043, |
| "mean_token_accuracy": 0.677251435816288, |
| "num_tokens": 9134903.0, |
| "step": 94 |
| }, |
| { |
| "entropy": 1.1134615540504456, |
| "epoch": 0.9223300970873787, |
| "grad_norm": 0.16442131996154785, |
| "learning_rate": 4.175480844566404e-05, |
| "loss": 1.1127569675445557, |
| "mean_token_accuracy": 0.6718188151717186, |
| "num_tokens": 9231494.0, |
| "step": 95 |
| }, |
| { |
| "entropy": 1.1019631028175354, |
| "epoch": 0.9320388349514563, |
| "grad_norm": 0.15546253323554993, |
| "learning_rate": 4.1554902857200924e-05, |
| "loss": 1.0951738357543945, |
| "mean_token_accuracy": 0.6784623265266418, |
| "num_tokens": 9328418.0, |
| "step": 96 |
| }, |
| { |
| "entropy": 1.0921340584754944, |
| "epoch": 0.941747572815534, |
| "grad_norm": 0.16863901913166046, |
| "learning_rate": 4.135309405755583e-05, |
| "loss": 1.0979890823364258, |
| "mean_token_accuracy": 0.6765805631875992, |
| "num_tokens": 9426256.0, |
| "step": 97 |
| }, |
| { |
| "entropy": 1.104684755206108, |
| "epoch": 0.9514563106796117, |
| "grad_norm": 0.16420894861221313, |
| "learning_rate": 4.11494052473943e-05, |
| "loss": 1.106919765472412, |
| "mean_token_accuracy": 0.6736476495862007, |
| "num_tokens": 9523841.0, |
| "step": 98 |
| }, |
| { |
| "entropy": 1.0703193694353104, |
| "epoch": 0.9611650485436893, |
| "grad_norm": 0.1605386883020401, |
| "learning_rate": 4.094385984351462e-05, |
| "loss": 1.0785481929779053, |
| "mean_token_accuracy": 0.6812796592712402, |
| "num_tokens": 9621225.0, |
| "step": 99 |
| }, |
| { |
| "entropy": 1.090438723564148, |
| "epoch": 0.970873786407767, |
| "grad_norm": 0.15666428208351135, |
| "learning_rate": 4.073648147615579e-05, |
| "loss": 1.0910429954528809, |
| "mean_token_accuracy": 0.6792959794402122, |
| "num_tokens": 9718805.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.1004114151000977, |
| "epoch": 0.9805825242718447, |
| "grad_norm": 0.16704976558685303, |
| "learning_rate": 4.052729398628089e-05, |
| "loss": 1.0998764038085938, |
| "mean_token_accuracy": 0.6745198145508766, |
| "num_tokens": 9816814.0, |
| "step": 101 |
| }, |
| { |
| "entropy": 1.0953031778335571, |
| "epoch": 0.9902912621359223, |
| "grad_norm": 0.16081705689430237, |
| "learning_rate": 4.031632142283622e-05, |
| "loss": 1.0844577550888062, |
| "mean_token_accuracy": 0.6796873956918716, |
| "num_tokens": 9914411.0, |
| "step": 102 |
| }, |
| { |
| "entropy": 1.0845869928598404, |
| "epoch": 1.0, |
| "grad_norm": 0.16136308014392853, |
| "learning_rate": 4.0103588039986556e-05, |
| "loss": 1.0797569751739502, |
| "mean_token_accuracy": 0.6803081929683685, |
| "num_tokens": 10010251.0, |
| "step": 103 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 309, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.4475135272615936e+18, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
