Training in progress, step 5600

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:86c2b324b8049a18ce1ba034f6aba65985584aa14901fa58e73e8ab119745c91
+oid sha256:7e6bca69b60cb9bd68a36bbce794f14bfd7178662bdf36126c832f4b88b52a6e
 size 14764744

last-checkpoint/adapter_config.json

@@ -32,9 +32,9 @@
   "revision": null,
   "target_modules": [
     "q_proj",
+    "v_proj",
     "o_proj",
-    "k_proj",
-    "v_proj"
+    "k_proj"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:86c2b324b8049a18ce1ba034f6aba65985584aa14901fa58e73e8ab119745c91
+oid sha256:52217ccaabbb2c9a520ff6c890c15d98aa47a4c28fad51432c095d328a2a29be
 size 14764744

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95d5dbf3ccb5eae5ac9fbedaf3a3bf948c0bd9f0e56e9fa24be21b44b23bf9cb
+oid sha256:a6af5e95ef3e872afad1bdf1fdf3ea5f5155675c6586239ad660ccd39c37347e
 size 29614667

last-checkpoint/rng_state_0.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dcdb3748172857f29588339ceb5799f4474929a2fc74897129d647054ac61d53
+oid sha256:a82aeb7d859423fae1de432598feb0f08060628848416302157961cd81d904dc
 size 14917

last-checkpoint/rng_state_1.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:edff8728f15b32954602ef9d3cf6e0a55beee203f9b419667f25948eb333f5b0
+oid sha256:51933555f7d7492da069489fcbf900fea3e97204e288093912746efcc0733e6e
 size 14917

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8053fa83978164af3b1105d8792ddd5aaa5fa308eacd91789cad36b689c1f7e5
+oid sha256:7d223d319f4b8402bb6d092355111dac11a4bb47d6e23f14a8f4c5bc3c5b9c8f
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a1dc2197e9ad250585b3ac3735f65e4c4e30b977a4b11f1897913f7d241839d3
+oid sha256:dd465d73b04f6c59912627375bc7905b039595a268513d7eb113da343a1d81e2
 size 1465

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.406642787365679,
+  "epoch": 0.8267665676269216,
   "eval_steps": 100,
-  "global_step": 5400,
+  "global_step": 4800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -5336,678 +5336,12 @@
       "eval_samples_per_second": 5.161,
       "eval_steps_per_second": 1.291,
       "step": 4800
-    },
-    {
-      "entropy": 0.9813777908682824,
-      "epoch": 1.252946922826441,
-      "grad_norm": 0.44488945603370667,
-      "learning_rate": 7.473300338629852e-06,
-      "loss": 0.9760551452636719,
-      "mean_token_accuracy": 0.7828159511089325,
-      "num_tokens": 129388.0,
-      "step": 4810
-    },
-    {
-      "entropy": 1.028295949101448,
-      "epoch": 1.2555519374796482,
-      "grad_norm": 0.476211279630661,
-      "learning_rate": 7.447251888512634e-06,
-      "loss": 1.026865005493164,
-      "mean_token_accuracy": 0.7739313766360283,
-      "num_tokens": 256578.0,
-      "step": 4820
-    },
-    {
-      "entropy": 0.993380619585514,
-      "epoch": 1.2581569521328557,
-      "grad_norm": 0.5022156834602356,
-      "learning_rate": 7.4212034383954155e-06,
-      "loss": 0.9900455474853516,
-      "mean_token_accuracy": 0.7809839487075806,
-      "num_tokens": 379760.0,
-      "step": 4830
-    },
-    {
-      "entropy": 0.9588794216513634,
-      "epoch": 1.2607619667860632,
-      "grad_norm": 0.43847885727882385,
-      "learning_rate": 7.395154988278198e-06,
-      "loss": 0.9578507423400879,
-      "mean_token_accuracy": 0.7875250384211541,
-      "num_tokens": 506218.0,
-      "step": 4840
-    },
-    {
-      "entropy": 0.981762844324112,
-      "epoch": 1.2633669814392707,
-      "grad_norm": 0.44927820563316345,
-      "learning_rate": 7.36910653816098e-06,
-      "loss": 0.9703155517578125,
-      "mean_token_accuracy": 0.7809026747941971,
-      "num_tokens": 630994.0,
-      "step": 4850
-    },
-    {
-      "entropy": 0.9776830449700356,
-      "epoch": 1.265971996092478,
-      "grad_norm": 0.4348759055137634,
-      "learning_rate": 7.343058088043762e-06,
-      "loss": 0.9691827774047852,
-      "mean_token_accuracy": 0.7856391072273254,
-      "num_tokens": 757625.0,
-      "step": 4860
-    },
-    {
-      "entropy": 0.9467781752347946,
-      "epoch": 1.2685770107456855,
-      "grad_norm": 0.4643247127532959,
-      "learning_rate": 7.3170096379265445e-06,
-      "loss": 0.9435815811157227,
-      "mean_token_accuracy": 0.7907302647829055,
-      "num_tokens": 881836.0,
-      "step": 4870
-    },
-    {
-      "entropy": 0.9590273395180702,
-      "epoch": 1.2711820253988928,
-      "grad_norm": 0.45526987314224243,
-      "learning_rate": 7.290961187809326e-06,
-      "loss": 0.9674400329589844,
-      "mean_token_accuracy": 0.785734324157238,
-      "num_tokens": 1011144.0,
-      "step": 4880
-    },
-    {
-      "entropy": 1.017872877418995,
-      "epoch": 1.2737870400521003,
-      "grad_norm": 0.42657098174095154,
-      "learning_rate": 7.264912737692108e-06,
-      "loss": 1.0152156829833985,
-      "mean_token_accuracy": 0.7740847021341324,
-      "num_tokens": 1136047.0,
-      "step": 4890
-    },
-    {
-      "entropy": 0.993815653026104,
-      "epoch": 1.2763920547053078,
-      "grad_norm": 0.4295169711112976,
-      "learning_rate": 7.23886428757489e-06,
-      "loss": 0.9871231079101562,
-      "mean_token_accuracy": 0.7813043370842934,
-      "num_tokens": 1258878.0,
-      "step": 4900
-    },
-    {
-      "epoch": 1.2763920547053078,
-      "eval_entropy": 0.9970914693472669,
-      "eval_loss": 0.9884027242660522,
-      "eval_mean_token_accuracy": 0.7808390200208133,
-      "eval_num_tokens": 1258878.0,
-      "eval_runtime": 1248.5811,
-      "eval_samples_per_second": 5.451,
-      "eval_steps_per_second": 1.363,
-      "step": 4900
-    },
-    {
-      "entropy": 1.0345384523272514,
-      "epoch": 1.278997069358515,
-      "grad_norm": 0.46928712725639343,
-      "learning_rate": 7.212815837457672e-06,
-      "loss": 1.0300195693969727,
-      "mean_token_accuracy": 0.771189559996128,
-      "num_tokens": 1386649.0,
-      "step": 4910
-    },
-    {
-      "entropy": 1.0029998540878295,
-      "epoch": 1.2816020840117226,
-      "grad_norm": 0.48189452290534973,
-      "learning_rate": 7.186767387340454e-06,
-      "loss": 0.9906700134277344,
-      "mean_token_accuracy": 0.7785976707935334,
-      "num_tokens": 1512338.0,
-      "step": 4920
-    },
-    {
-      "entropy": 1.0033065855503083,
-      "epoch": 1.2842070986649299,
-      "grad_norm": 0.4588858187198639,
-      "learning_rate": 7.160718937223236e-06,
-      "loss": 0.9927248001098633,
-      "mean_token_accuracy": 0.7797123953700066,
-      "num_tokens": 1642624.0,
-      "step": 4930
-    },
-    {
-      "entropy": 0.9549823224544525,
-      "epoch": 1.2868121133181374,
-      "grad_norm": 0.43219390511512756,
-      "learning_rate": 7.134670487106018e-06,
-      "loss": 0.9539813041687012,
-      "mean_token_accuracy": 0.7869110450148582,
-      "num_tokens": 1771847.0,
-      "step": 4940
-    },
-    {
-      "entropy": 1.0237484708428384,
-      "epoch": 1.289417127971345,
-      "grad_norm": 0.4907780885696411,
-      "learning_rate": 7.108622036988799e-06,
-      "loss": 1.0278969764709474,
-      "mean_token_accuracy": 0.7715816020965576,
-      "num_tokens": 1900902.0,
-      "step": 4950
-    },
-    {
-      "entropy": 1.006957694888115,
-      "epoch": 1.2920221426245524,
-      "grad_norm": 0.4545274078845978,
-      "learning_rate": 7.082573586871582e-06,
-      "loss": 1.0091020584106445,
-      "mean_token_accuracy": 0.7763800770044327,
-      "num_tokens": 2031777.0,
-      "step": 4960
-    },
-    {
-      "entropy": 0.9983339875936508,
-      "epoch": 1.2946271572777597,
-      "grad_norm": 0.4333556592464447,
-      "learning_rate": 7.056525136754364e-06,
-      "loss": 0.984560203552246,
-      "mean_token_accuracy": 0.7822976499795914,
-      "num_tokens": 2156527.0,
-      "step": 4970
-    },
-    {
-      "entropy": 0.9773911476135254,
-      "epoch": 1.2972321719309672,
-      "grad_norm": 0.4380306303501129,
-      "learning_rate": 7.030476686637145e-06,
-      "loss": 0.9723411560058594,
-      "mean_token_accuracy": 0.7831587105989456,
-      "num_tokens": 2280664.0,
-      "step": 4980
-    },
-    {
-      "entropy": 1.0176811680197715,
-      "epoch": 1.2998371865841745,
-      "grad_norm": 0.49647819995880127,
-      "learning_rate": 7.004428236519928e-06,
-      "loss": 1.0118232727050782,
-      "mean_token_accuracy": 0.7755638837814331,
-      "num_tokens": 2408726.0,
-      "step": 4990
-    },
-    {
-      "entropy": 1.003194873034954,
-      "epoch": 1.302442201237382,
-      "grad_norm": 0.4423670470714569,
-      "learning_rate": 6.978379786402709e-06,
-      "loss": 0.9979192733764648,
-      "mean_token_accuracy": 0.7790161743760109,
-      "num_tokens": 2538321.0,
-      "step": 5000
-    },
-    {
-      "epoch": 1.302442201237382,
-      "eval_entropy": 0.9890262984991354,
-      "eval_loss": 0.9879328608512878,
-      "eval_mean_token_accuracy": 0.7809134237774671,
-      "eval_num_tokens": 2538321.0,
-      "eval_runtime": 1248.3546,
-      "eval_samples_per_second": 5.452,
-      "eval_steps_per_second": 1.363,
-      "step": 5000
-    },
-    {
-      "entropy": 0.9911179274320603,
-      "epoch": 1.3050472158905895,
-      "grad_norm": 0.4474965035915375,
-      "learning_rate": 6.952331336285491e-06,
-      "loss": 0.9846304893493653,
-      "mean_token_accuracy": 0.7800104588270187,
-      "num_tokens": 2664423.0,
-      "step": 5010
-    },
-    {
-      "entropy": 0.9809451669454574,
-      "epoch": 1.3076522305437968,
-      "grad_norm": 0.4665301442146301,
-      "learning_rate": 6.926282886168274e-06,
-      "loss": 0.9809820175170898,
-      "mean_token_accuracy": 0.7841726258397103,
-      "num_tokens": 2790315.0,
-      "step": 5020
-    },
-    {
-      "entropy": 0.9852373987436295,
-      "epoch": 1.3102572451970043,
-      "grad_norm": 0.429098516702652,
-      "learning_rate": 6.900234436051055e-06,
-      "loss": 0.9816387176513672,
-      "mean_token_accuracy": 0.7819439515471458,
-      "num_tokens": 2915995.0,
-      "step": 5030
-    },
-    {
-      "entropy": 0.986951345205307,
-      "epoch": 1.3128622598502115,
-      "grad_norm": 0.45305219292640686,
-      "learning_rate": 6.874185985933837e-06,
-      "loss": 0.9891517639160157,
-      "mean_token_accuracy": 0.780052900314331,
-      "num_tokens": 3041830.0,
-      "step": 5040
-    },
-    {
-      "entropy": 1.0265516340732574,
-      "epoch": 1.315467274503419,
-      "grad_norm": 0.4666403532028198,
-      "learning_rate": 6.848137535816619e-06,
-      "loss": 1.0243612289428712,
-      "mean_token_accuracy": 0.7726057410240174,
-      "num_tokens": 3167296.0,
-      "step": 5050
-    },
-    {
-      "entropy": 1.0306162551045417,
-      "epoch": 1.3180722891566266,
-      "grad_norm": 0.39735814929008484,
-      "learning_rate": 6.8220890856994014e-06,
-      "loss": 1.0236353874206543,
-      "mean_token_accuracy": 0.7734616458415985,
-      "num_tokens": 3292887.0,
-      "step": 5060
-    },
-    {
-      "entropy": 1.0042320653796195,
-      "epoch": 1.320677303809834,
-      "grad_norm": 0.39348119497299194,
-      "learning_rate": 6.796040635582184e-06,
-      "loss": 1.0001495361328125,
-      "mean_token_accuracy": 0.7790237024426461,
-      "num_tokens": 3418112.0,
-      "step": 5070
-    },
-    {
-      "entropy": 1.0145965054631234,
-      "epoch": 1.3232823184630413,
-      "grad_norm": 0.4478921592235565,
-      "learning_rate": 6.7699921854649655e-06,
-      "loss": 1.0137309074401855,
-      "mean_token_accuracy": 0.7771003857254982,
-      "num_tokens": 3548473.0,
-      "step": 5080
-    },
-    {
-      "entropy": 1.0103399023413657,
-      "epoch": 1.3258873331162488,
-      "grad_norm": 0.5050984621047974,
-      "learning_rate": 6.7439437353477475e-06,
-      "loss": 0.999553108215332,
-      "mean_token_accuracy": 0.7785449594259262,
-      "num_tokens": 3673127.0,
-      "step": 5090
-    },
-    {
-      "entropy": 0.966312225162983,
-      "epoch": 1.3284923477694561,
-      "grad_norm": 0.43490949273109436,
-      "learning_rate": 6.717895285230529e-06,
-      "loss": 0.9707491874694825,
-      "mean_token_accuracy": 0.7834598645567894,
-      "num_tokens": 3802026.0,
-      "step": 5100
-    },
-    {
-      "epoch": 1.3284923477694561,
-      "eval_entropy": 0.9884775571971607,
-      "eval_loss": 0.9877337217330933,
-      "eval_mean_token_accuracy": 0.7809631075057805,
-      "eval_num_tokens": 3802026.0,
-      "eval_runtime": 1249.8046,
-      "eval_samples_per_second": 5.446,
-      "eval_steps_per_second": 1.362,
-      "step": 5100
-    },
-    {
-      "entropy": 0.9976687163114548,
-      "epoch": 1.3310973624226636,
-      "grad_norm": 0.43964383006095886,
-      "learning_rate": 6.691846835113312e-06,
-      "loss": 0.9837272644042969,
-      "mean_token_accuracy": 0.7808806389570236,
-      "num_tokens": 3930388.0,
-      "step": 5110
-    },
-    {
-      "entropy": 0.983721150457859,
-      "epoch": 1.3337023770758711,
-      "grad_norm": 0.46595776081085205,
-      "learning_rate": 6.665798384996093e-06,
-      "loss": 0.9791852951049804,
-      "mean_token_accuracy": 0.7797340914607048,
-      "num_tokens": 4057438.0,
-      "step": 5120
-    },
-    {
-      "entropy": 1.0189708486199378,
-      "epoch": 1.3363073917290784,
-      "grad_norm": 0.4384433329105377,
-      "learning_rate": 6.639749934878875e-06,
-      "loss": 1.0155632019042968,
-      "mean_token_accuracy": 0.7752848476171493,
-      "num_tokens": 4184226.0,
-      "step": 5130
-    },
-    {
-      "entropy": 0.9738220989704132,
-      "epoch": 1.338912406382286,
-      "grad_norm": 0.4892994463443756,
-      "learning_rate": 6.613701484761658e-06,
-      "loss": 0.9773561477661132,
-      "mean_token_accuracy": 0.7810323163866997,
-      "num_tokens": 4315468.0,
-      "step": 5140
-    },
-    {
-      "entropy": 0.9959368094801903,
-      "epoch": 1.3415174210354932,
-      "grad_norm": 0.4051533043384552,
-      "learning_rate": 6.587653034644439e-06,
-      "loss": 0.9895273208618164,
-      "mean_token_accuracy": 0.7815038353204727,
-      "num_tokens": 4439960.0,
-      "step": 5150
-    },
-    {
-      "entropy": 0.9800890773534775,
-      "epoch": 1.3441224356887007,
-      "grad_norm": 0.4586341679096222,
-      "learning_rate": 6.561604584527222e-06,
-      "loss": 0.9737453460693359,
-      "mean_token_accuracy": 0.7814203977584839,
-      "num_tokens": 4566056.0,
-      "step": 5160
-    },
-    {
-      "entropy": 1.0000663980841638,
-      "epoch": 1.3467274503419082,
-      "grad_norm": 0.4102465510368347,
-      "learning_rate": 6.535556134410003e-06,
-      "loss": 0.9939519882202148,
-      "mean_token_accuracy": 0.7790467858314514,
-      "num_tokens": 4694925.0,
-      "step": 5170
-    },
-    {
-      "entropy": 1.0039643123745918,
-      "epoch": 1.3493324649951157,
-      "grad_norm": 0.4231637418270111,
-      "learning_rate": 6.509507684292785e-06,
-      "loss": 1.001540756225586,
-      "mean_token_accuracy": 0.7798496216535569,
-      "num_tokens": 4817681.0,
-      "step": 5180
-    },
-    {
-      "entropy": 0.9824537605047226,
-      "epoch": 1.351937479648323,
-      "grad_norm": 0.4193444550037384,
-      "learning_rate": 6.483459234175568e-06,
-      "loss": 0.9902861595153809,
-      "mean_token_accuracy": 0.7809780746698379,
-      "num_tokens": 4946004.0,
-      "step": 5190
-    },
-    {
-      "entropy": 0.9934640824794769,
-      "epoch": 1.3545424943015305,
-      "grad_norm": 0.5171741843223572,
-      "learning_rate": 6.457410784058349e-06,
-      "loss": 0.9897693634033203,
-      "mean_token_accuracy": 0.7789523780345917,
-      "num_tokens": 5071552.0,
-      "step": 5200
-    },
-    {
-      "epoch": 1.3545424943015305,
-      "eval_entropy": 0.9874165067941686,
-      "eval_loss": 0.9873093962669373,
-      "eval_mean_token_accuracy": 0.7809983336757409,
-      "eval_num_tokens": 5071552.0,
-      "eval_runtime": 1248.5888,
-      "eval_samples_per_second": 5.451,
-      "eval_steps_per_second": 1.363,
-      "step": 5200
-    },
-    {
-      "entropy": 0.9975464969873429,
-      "epoch": 1.3571475089547378,
-      "grad_norm": 0.43779075145721436,
-      "learning_rate": 6.431362333941131e-06,
-      "loss": 0.9847593307495117,
-      "mean_token_accuracy": 0.7821353122591972,
-      "num_tokens": 5199409.0,
-      "step": 5210
-    },
-    {
-      "entropy": 0.9890801250934601,
-      "epoch": 1.3597525236079453,
-      "grad_norm": 0.46792900562286377,
-      "learning_rate": 6.405313883823912e-06,
-      "loss": 0.9886545181274414,
-      "mean_token_accuracy": 0.7808847963809967,
-      "num_tokens": 5327018.0,
-      "step": 5220
-    },
-    {
-      "entropy": 1.0165737330913545,
-      "epoch": 1.3623575382611528,
-      "grad_norm": 0.422018438577652,
-      "learning_rate": 6.379265433706695e-06,
-      "loss": 1.0161245346069336,
-      "mean_token_accuracy": 0.7733713835477829,
-      "num_tokens": 5453326.0,
-      "step": 5230
-    },
-    {
-      "entropy": 0.9966152131557464,
-      "epoch": 1.36496255291436,
-      "grad_norm": 0.46802228689193726,
-      "learning_rate": 6.353216983589477e-06,
-      "loss": 0.9952875137329101,
-      "mean_token_accuracy": 0.7801799476146698,
-      "num_tokens": 5579340.0,
-      "step": 5240
-    },
-    {
-      "entropy": 1.013704551756382,
-      "epoch": 1.3675675675675676,
-      "grad_norm": 0.4642789363861084,
-      "learning_rate": 6.327168533472258e-06,
-      "loss": 1.0044668197631836,
-      "mean_token_accuracy": 0.7750307902693748,
-      "num_tokens": 5704206.0,
-      "step": 5250
-    },
-    {
-      "entropy": 0.9664336591959,
-      "epoch": 1.3701725822207749,
-      "grad_norm": 0.45772185921669006,
-      "learning_rate": 6.301120083355041e-06,
-      "loss": 0.9622204780578614,
-      "mean_token_accuracy": 0.7862247511744499,
-      "num_tokens": 5832067.0,
-      "step": 5260
-    },
-    {
-      "entropy": 0.9767679080367089,
-      "epoch": 1.3727775968739824,
-      "grad_norm": 0.4938032627105713,
-      "learning_rate": 6.2750716332378225e-06,
-      "loss": 0.9772611618041992,
-      "mean_token_accuracy": 0.7834762305021286,
-      "num_tokens": 5956416.0,
-      "step": 5270
-    },
-    {
-      "entropy": 0.997693844139576,
-      "epoch": 1.3753826115271899,
-      "grad_norm": 0.4085782766342163,
-      "learning_rate": 6.249023183120605e-06,
-      "loss": 0.993937873840332,
-      "mean_token_accuracy": 0.7790059551596642,
-      "num_tokens": 6080387.0,
-      "step": 5280
-    },
-    {
-      "entropy": 0.9551204591989517,
-      "epoch": 1.3779876261803974,
-      "grad_norm": 0.47124868631362915,
-      "learning_rate": 6.2229747330033865e-06,
-      "loss": 0.9600910186767578,
-      "mean_token_accuracy": 0.7857999742031098,
-      "num_tokens": 6210495.0,
-      "step": 5290
-    },
-    {
-      "entropy": 0.9895543649792671,
-      "epoch": 1.3805926408336047,
-      "grad_norm": 0.4795711934566498,
-      "learning_rate": 6.196926282886169e-06,
-      "loss": 0.980061149597168,
-      "mean_token_accuracy": 0.7814426705241203,
-      "num_tokens": 6338971.0,
-      "step": 5300
-    },
-    {
-      "epoch": 1.3805926408336047,
-      "eval_entropy": 0.9903583783310532,
-      "eval_loss": 0.9870790839195251,
-      "eval_mean_token_accuracy": 0.781017405095868,
-      "eval_num_tokens": 6338971.0,
-      "eval_runtime": 1249.0996,
-      "eval_samples_per_second": 5.449,
-      "eval_steps_per_second": 1.363,
-      "step": 5300
-    },
-    {
-      "entropy": 1.0024344369769096,
-      "epoch": 1.3831976554868122,
-      "grad_norm": 0.44125476479530334,
-      "learning_rate": 6.1708778327689515e-06,
-      "loss": 1.0059074401855468,
-      "mean_token_accuracy": 0.7774184092879295,
-      "num_tokens": 6465148.0,
-      "step": 5310
-    },
-    {
-      "entropy": 0.9910042881965637,
-      "epoch": 1.3858026701400195,
-      "grad_norm": 0.4707449972629547,
-      "learning_rate": 6.144829382651733e-06,
-      "loss": 0.987660026550293,
-      "mean_token_accuracy": 0.7800036445260048,
-      "num_tokens": 6592528.0,
-      "step": 5320
-    },
-    {
-      "entropy": 0.9776637852191925,
-      "epoch": 1.388407684793227,
-      "grad_norm": 0.40415525436401367,
-      "learning_rate": 6.118780932534515e-06,
-      "loss": 0.9776216506958008,
-      "mean_token_accuracy": 0.7837252721190453,
-      "num_tokens": 6718090.0,
-      "step": 5330
-    },
-    {
-      "entropy": 0.9891742005944252,
-      "epoch": 1.3910126994464345,
-      "grad_norm": 0.4758605659008026,
-      "learning_rate": 6.092732482417296e-06,
-      "loss": 0.9910886764526368,
-      "mean_token_accuracy": 0.7781824618577957,
-      "num_tokens": 6845948.0,
-      "step": 5340
-    },
-    {
-      "entropy": 1.007858295738697,
-      "epoch": 1.3936177140996417,
-      "grad_norm": 0.47813597321510315,
-      "learning_rate": 6.066684032300079e-06,
-      "loss": 1.0086344718933105,
-      "mean_token_accuracy": 0.7754964306950569,
-      "num_tokens": 6973246.0,
-      "step": 5350
-    },
-    {
-      "entropy": 1.0145272225141526,
-      "epoch": 1.3962227287528493,
-      "grad_norm": 0.4583241939544678,
-      "learning_rate": 6.040635582182861e-06,
-      "loss": 1.0088977813720703,
-      "mean_token_accuracy": 0.775096683204174,
-      "num_tokens": 7099698.0,
-      "step": 5360
-    },
-    {
-      "entropy": 1.0434660747647286,
-      "epoch": 1.3988277434060565,
-      "grad_norm": 0.4353461265563965,
-      "learning_rate": 6.014587132065642e-06,
-      "loss": 1.0382314682006837,
-      "mean_token_accuracy": 0.7711483478546143,
-      "num_tokens": 7223734.0,
-      "step": 5370
-    },
-    {
-      "entropy": 0.990787972509861,
-      "epoch": 1.401432758059264,
-      "grad_norm": 0.47876155376434326,
-      "learning_rate": 5.988538681948425e-06,
-      "loss": 0.989743995666504,
-      "mean_token_accuracy": 0.7813950344920159,
-      "num_tokens": 7350371.0,
-      "step": 5380
-    },
-    {
-      "entropy": 1.0080674409866333,
-      "epoch": 1.4040377727124715,
-      "grad_norm": 0.5123191475868225,
-      "learning_rate": 5.962490231831206e-06,
-      "loss": 1.002880859375,
-      "mean_token_accuracy": 0.7779936477541923,
-      "num_tokens": 7477903.0,
-      "step": 5390
-    },
-    {
-      "entropy": 1.0048462897539139,
-      "epoch": 1.406642787365679,
-      "grad_norm": 0.4499814212322235,
-      "learning_rate": 5.936441781713989e-06,
-      "loss": 0.999144458770752,
-      "mean_token_accuracy": 0.7778741553425789,
-      "num_tokens": 7604005.0,
-      "step": 5400
-    },
-    {
-      "epoch": 1.406642787365679,
-      "eval_entropy": 0.9904807468068586,
-      "eval_loss": 0.9868780970573425,
-      "eval_mean_token_accuracy": 0.7810666585290754,
-      "eval_num_tokens": 7604005.0,
-      "eval_runtime": 1248.7827,
-      "eval_samples_per_second": 5.45,
-      "eval_steps_per_second": 1.363,
-      "step": 5400
     }
   ],
   "logging_steps": 10,
-  "max_steps": 7678,
+  "max_steps": 5806,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 2,
+  "num_train_epochs": 1,
   "save_steps": 200,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -6021,7 +5355,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 8.792910420571587e+17,
+  "total_flos": 7.887012992552796e+17,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9eb97127f3633f0387d329b923ad0716a30b1001fb17ad9a7247f8ed8a4f5dab
+oid sha256:e193b6114ef398d374d4a4437acc56d06a32e2263d65fa1d64344fb64467f82c
 size 5777