Upload folder using huggingface_hub

adapter_config.json

@@ -14,7 +14,7 @@
   "loftq_config": {},
   "lora_alpha": 112,
   "lora_bias": false,
-  "lora_dropout": 0.05,
+  "lora_dropout": 0,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:79ccd2b0725d5caaf6fde82e92d52927391ad95545839ba41f7843cc6b907d55
+oid sha256:9686a3feb3b7a157ea7b5fc7a245269c9b57e3de6162222864bfe150a7d5e29d
 size 322996232

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:019d1ca091101294bfbf2f45057dc925f538fd6c74c7db413e7331adcc9317d9
+oid sha256:625b5c93979fc045c41d22b46dab4ee40e52a0209c0e395bd81442d281308580
 size 646142667

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a8c0aa69084bb9d94d9a6de1d8d9b79dc3923ff47dece5164ccf871ba1b2abc
+oid sha256:2534e434cd5abbb8f7668d3eab0549db0ef95d6a797a3efa86b712e8e32266a7
 size 14645

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ffb01ee3f21af64a917386cd80a40a480326ab2f88c05fd143f4f5d49daa613b
+oid sha256:030ce4e896dde4e5df8ae1c595b2e24f4e2c44ea8055948beb08188b9a047014
 size 1465

trainer_state.json

@@ -2,636 +2,376 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 2.9876543209876543,
+  "epoch": 3.0,
   "eval_steps": 500,
-  "global_step": 62,
+  "global_step": 36,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "entropy": 2.1045966744422913,
-      "epoch": 0.04938271604938271,
-      "grad_norm": 16.678590774536133,
+      "entropy": 2.0738155096769333,
+      "epoch": 0.08602150537634409,
+      "grad_norm": 13.394003868103027,
       "learning_rate": 0.0,
-      "loss": 2.4474,
-      "mean_token_accuracy": 0.44178610295057297,
-      "num_tokens": 2014.0,
+      "loss": 2.2382,
+      "mean_token_accuracy": 0.45138081163167953,
+      "num_tokens": 5228.0,
       "step": 1
     },
     {
-      "entropy": 1.944445788860321,
-      "epoch": 0.09876543209876543,
-      "grad_norm": 12.955353736877441,
-      "learning_rate": 3.3333333333333333e-06,
-      "loss": 2.153,
-      "mean_token_accuracy": 0.4885662570595741,
-      "num_tokens": 4384.0,
+      "entropy": 2.0733328089118004,
+      "epoch": 0.17204301075268819,
+      "grad_norm": 15.095857620239258,
+      "learning_rate": 2e-06,
+      "loss": 2.3757,
+      "mean_token_accuracy": 0.4545624293386936,
+      "num_tokens": 9601.0,
       "step": 2
     },
     {
-      "entropy": 2.133327841758728,
-      "epoch": 0.14814814814814814,
-      "grad_norm": 12.755569458007812,
-      "learning_rate": 6.666666666666667e-06,
-      "loss": 2.3206,
-      "mean_token_accuracy": 0.45916735380887985,
-      "num_tokens": 6448.0,
+      "entropy": 2.137907035648823,
+      "epoch": 0.25806451612903225,
+      "grad_norm": 15.085412979125977,
+      "learning_rate": 4e-06,
+      "loss": 2.4367,
+      "mean_token_accuracy": 0.43796562775969505,
+      "num_tokens": 13215.0,
       "step": 3
     },
     {
-      "entropy": 2.04616542160511,
-      "epoch": 0.19753086419753085,
-      "grad_norm": 12.627863883972168,
-      "learning_rate": 1e-05,
-      "loss": 2.0161,
-      "mean_token_accuracy": 0.4829055927693844,
-      "num_tokens": 8532.0,
+      "entropy": 2.2101941779255867,
+      "epoch": 0.34408602150537637,
+      "grad_norm": 13.478177070617676,
+      "learning_rate": 6e-06,
+      "loss": 2.3352,
+      "mean_token_accuracy": 0.463874289765954,
+      "num_tokens": 16905.0,
       "step": 4
     },
     {
-      "entropy": 2.2228081226348877,
-      "epoch": 0.24691358024691357,
-      "grad_norm": 6.164249420166016,
-      "learning_rate": 1.3333333333333333e-05,
-      "loss": 1.8315,
-      "mean_token_accuracy": 0.4869111068546772,
-      "num_tokens": 11973.0,
+      "entropy": 2.2711042016744614,
+      "epoch": 0.43010752688172044,
+      "grad_norm": 9.6048583984375,
+      "learning_rate": 8e-06,
+      "loss": 2.1429,
+      "mean_token_accuracy": 0.46069133654236794,
+      "num_tokens": 21282.0,
       "step": 5
     },
     {
-      "entropy": 2.113390877842903,
-      "epoch": 0.2962962962962963,
-      "grad_norm": 6.344396591186523,
-      "learning_rate": 1.6666666666666667e-05,
-      "loss": 1.9057,
-      "mean_token_accuracy": 0.5254073962569237,
-      "num_tokens": 14982.0,
+      "entropy": 2.187708094716072,
+      "epoch": 0.5161290322580645,
+      "grad_norm": 7.296608924865723,
+      "learning_rate": 9.999999999999999e-06,
+      "loss": 2.0322,
+      "mean_token_accuracy": 0.4980849325656891,
+      "num_tokens": 25592.0,
       "step": 6
     },
     {
-      "entropy": 2.207372799515724,
-      "epoch": 0.345679012345679,
-      "grad_norm": 6.901336669921875,
-      "learning_rate": 2e-05,
-      "loss": 1.8379,
-      "mean_token_accuracy": 0.5119036361575127,
-      "num_tokens": 17566.0,
+      "entropy": 2.3398054093122482,
+      "epoch": 0.6021505376344086,
+      "grad_norm": 7.0159196853637695,
+      "learning_rate": 1.2e-05,
+      "loss": 2.006,
+      "mean_token_accuracy": 0.5018948670476675,
+      "num_tokens": 29734.0,
       "step": 7
     },
     {
-      "entropy": 2.2413517236709595,
-      "epoch": 0.3950617283950617,
-      "grad_norm": 6.814880847930908,
-      "learning_rate": 2.3333333333333336e-05,
-      "loss": 1.9594,
-      "mean_token_accuracy": 0.49952177330851555,
-      "num_tokens": 19825.0,
+      "entropy": 2.3085582107305527,
+      "epoch": 0.6881720430107527,
+      "grad_norm": 5.860436916351318,
+      "learning_rate": 1.4e-05,
+      "loss": 1.925,
+      "mean_token_accuracy": 0.5114013608545065,
+      "num_tokens": 34217.0,
       "step": 8
     },
     {
-      "entropy": 2.2734450548887253,
-      "epoch": 0.4444444444444444,
-      "grad_norm": 6.185186386108398,
-      "learning_rate": 2.6666666666666667e-05,
-      "loss": 1.9202,
-      "mean_token_accuracy": 0.5213926732540131,
-      "num_tokens": 21855.0,
+      "entropy": 2.2911834865808487,
+      "epoch": 0.7741935483870968,
+      "grad_norm": 5.837175369262695,
+      "learning_rate": 1.6e-05,
+      "loss": 1.9054,
+      "mean_token_accuracy": 0.5077734142541885,
+      "num_tokens": 38422.0,
       "step": 9
     },
     {
-      "entropy": 2.1350340843200684,
-      "epoch": 0.49382716049382713,
-      "grad_norm": 5.997181415557861,
-      "learning_rate": 3e-05,
-      "loss": 1.7231,
-      "mean_token_accuracy": 0.5484383963048458,
-      "num_tokens": 24405.0,
+      "entropy": 2.2874183356761932,
+      "epoch": 0.8602150537634409,
+      "grad_norm": 5.6462531089782715,
+      "learning_rate": 1.8e-05,
+      "loss": 1.8862,
+      "mean_token_accuracy": 0.5302404221147299,
+      "num_tokens": 42537.0,
       "step": 10
     },
     {
-      "entropy": 2.1509426087141037,
-      "epoch": 0.5432098765432098,
-      "grad_norm": 6.595911502838135,
-      "learning_rate": 3.3333333333333335e-05,
-      "loss": 1.7791,
-      "mean_token_accuracy": 0.5505530126392841,
-      "num_tokens": 26247.0,
+      "entropy": 2.249822184443474,
+      "epoch": 0.946236559139785,
+      "grad_norm": 5.423551082611084,
+      "learning_rate": 1.9999999999999998e-05,
+      "loss": 1.8544,
+      "mean_token_accuracy": 0.531640600413084,
+      "num_tokens": 46211.0,
       "step": 11
     },
     {
-      "entropy": 2.018960952758789,
-      "epoch": 0.5925925925925926,
-      "grad_norm": 5.737451076507568,
-      "learning_rate": 3.6666666666666666e-05,
-      "loss": 1.7496,
-      "mean_token_accuracy": 0.5482935383915901,
-      "num_tokens": 28315.0,
+      "entropy": 2.1156012296676634,
+      "epoch": 1.0,
+      "grad_norm": 5.370146751403809,
+      "learning_rate": 2.2e-05,
+      "loss": 1.7695,
+      "mean_token_accuracy": 0.5295870065689087,
+      "num_tokens": 49569.0,
       "step": 12
     },
     {
-      "entropy": 2.05392025411129,
-      "epoch": 0.6419753086419753,
-      "grad_norm": 6.572509765625,
-      "learning_rate": 4e-05,
-      "loss": 1.7735,
-      "mean_token_accuracy": 0.5521397292613983,
-      "num_tokens": 30028.0,
+      "entropy": 2.140422873198986,
+      "epoch": 1.086021505376344,
+      "grad_norm": 4.676677227020264,
+      "learning_rate": 2.4e-05,
+      "loss": 1.6377,
+      "mean_token_accuracy": 0.5523776523768902,
+      "num_tokens": 53425.0,
       "step": 13
     },
     {
-      "entropy": 1.830256313085556,
-      "epoch": 0.691358024691358,
-      "grad_norm": 4.667781352996826,
-      "learning_rate": 4.3333333333333334e-05,
-      "loss": 1.6375,
-      "mean_token_accuracy": 0.5683313086628914,
-      "num_tokens": 33151.0,
+      "entropy": 1.9793160483241081,
+      "epoch": 1.1720430107526882,
+      "grad_norm": 4.116484642028809,
+      "learning_rate": 2.6000000000000002e-05,
+      "loss": 1.5781,
+      "mean_token_accuracy": 0.5712024457752705,
+      "num_tokens": 59013.0,
       "step": 14
     },
     {
-      "entropy": 1.9952753484249115,
-      "epoch": 0.7407407407407407,
-      "grad_norm": 5.7879743576049805,
-      "learning_rate": 4.666666666666667e-05,
-      "loss": 1.6752,
-      "mean_token_accuracy": 0.5599516853690147,
-      "num_tokens": 35005.0,
+      "entropy": 2.038477085530758,
+      "epoch": 1.2580645161290323,
+      "grad_norm": 4.282267093658447,
+      "learning_rate": 2.8e-05,
+      "loss": 1.6025,
+      "mean_token_accuracy": 0.5666980221867561,
+      "num_tokens": 62876.0,
       "step": 15
     },
     {
-      "entropy": 2.2150524109601974,
-      "epoch": 0.7901234567901234,
-      "grad_norm": 6.519161224365234,
-      "learning_rate": 5e-05,
-      "loss": 1.7735,
-      "mean_token_accuracy": 0.5492940992116928,
-      "num_tokens": 36505.0,
+      "entropy": 1.9149735793471336,
+      "epoch": 1.3440860215053765,
+      "grad_norm": 4.070222854614258,
+      "learning_rate": 3e-05,
+      "loss": 1.5752,
+      "mean_token_accuracy": 0.5903957597911358,
+      "num_tokens": 66979.0,
       "step": 16
     },
     {
-      "entropy": 1.9987490773200989,
-      "epoch": 0.8395061728395061,
-      "grad_norm": 5.76025915145874,
-      "learning_rate": 4.994647308096509e-05,
-      "loss": 1.6564,
-      "mean_token_accuracy": 0.5628282427787781,
-      "num_tokens": 38467.0,
+      "entropy": 1.9767527654767036,
+      "epoch": 1.4301075268817205,
+      "grad_norm": 4.559730052947998,
+      "learning_rate": 2.9832462393376926e-05,
+      "loss": 1.6377,
+      "mean_token_accuracy": 0.5771633833646774,
+      "num_tokens": 70656.0,
       "step": 17
     },
     {
-      "entropy": 1.949499949812889,
-      "epoch": 0.8888888888888888,
-      "grad_norm": 4.927590370178223,
-      "learning_rate": 4.9786121534345265e-05,
-      "loss": 1.6235,
-      "mean_token_accuracy": 0.5548277460038662,
-      "num_tokens": 40889.0,
+      "entropy": 1.9485119134187698,
+      "epoch": 1.5161290322580645,
+      "grad_norm": 4.107751369476318,
+      "learning_rate": 2.9333592086792113e-05,
+      "loss": 1.5696,
+      "mean_token_accuracy": 0.5801203902810812,
+      "num_tokens": 75014.0,
       "step": 18
     },
     {
-      "entropy": 2.0500358641147614,
-      "epoch": 0.9382716049382716,
-      "grad_norm": 4.964672088623047,
-      "learning_rate": 4.951963201008076e-05,
-      "loss": 1.5668,
-      "mean_token_accuracy": 0.5864963307976723,
-      "num_tokens": 43002.0,
+      "entropy": 1.8602509573101997,
+      "epoch": 1.6021505376344085,
+      "grad_norm": 3.88903546333313,
+      "learning_rate": 2.8514533018536286e-05,
+      "loss": 1.4748,
+      "mean_token_accuracy": 0.5875820219516754,
+      "num_tokens": 79187.0,
       "step": 19
     },
     {
-      "entropy": 2.043752282857895,
-      "epoch": 0.9876543209876543,
-      "grad_norm": 4.934137344360352,
-      "learning_rate": 4.914814565722671e-05,
-      "loss": 1.5961,
-      "mean_token_accuracy": 0.5691769942641258,
-      "num_tokens": 45100.0,
+      "entropy": 1.8616999238729477,
+      "epoch": 1.6881720430107527,
+      "grad_norm": 3.7189409732818604,
+      "learning_rate": 2.7393581614739924e-05,
+      "loss": 1.4486,
+      "mean_token_accuracy": 0.6038471832871437,
+      "num_tokens": 83540.0,
       "step": 20
     },
     {
-      "entropy": 1.6749918460845947,
-      "epoch": 1.0,
-      "grad_norm": 10.593884468078613,
-      "learning_rate": 4.867325323737765e-05,
-      "loss": 1.418,
-      "mean_token_accuracy": 0.6043022274971008,
-      "num_tokens": 45620.0,
+      "entropy": 1.9136511757969856,
+      "epoch": 1.7741935483870968,
+      "grad_norm": 4.047863483428955,
+      "learning_rate": 2.5995778077447393e-05,
+      "loss": 1.516,
+      "mean_token_accuracy": 0.5893837288022041,
+      "num_tokens": 87678.0,
       "step": 21
     },
     {
-      "entropy": 1.888138085603714,
-      "epoch": 1.0493827160493827,
-      "grad_norm": 4.542983531951904,
-      "learning_rate": 4.8096988312782174e-05,
-      "loss": 1.2251,
-      "mean_token_accuracy": 0.6697289943695068,
-      "num_tokens": 47479.0,
+      "entropy": 1.916043758392334,
+      "epoch": 1.860215053763441,
+      "grad_norm": 4.67380952835083,
+      "learning_rate": 2.4352347027881003e-05,
+      "loss": 1.5726,
+      "mean_token_accuracy": 0.574837826192379,
+      "num_tokens": 91314.0,
       "step": 22
     },
     {
-      "entropy": 1.7415399253368378,
-      "epoch": 1.0987654320987654,
-      "grad_norm": 4.551285266876221,
-      "learning_rate": 4.742181853831721e-05,
-      "loss": 1.2521,
-      "mean_token_accuracy": 0.6552469357848167,
-      "num_tokens": 49530.0,
+      "entropy": 1.7887573465704918,
+      "epoch": 1.946236559139785,
+      "grad_norm": 3.5782926082611084,
+      "learning_rate": 2.25e-05,
+      "loss": 1.4336,
+      "mean_token_accuracy": 0.6035086195915937,
+      "num_tokens": 96051.0,
       "step": 23
     },
     {
-      "entropy": 1.6383564174175262,
-      "epoch": 1.1481481481481481,
-      "grad_norm": 4.755136013031006,
-      "learning_rate": 4.665063509461097e-05,
-      "loss": 1.2116,
-      "mean_token_accuracy": 0.6445794999599457,
-      "num_tokens": 51795.0,
+      "entropy": 1.7804332613945006,
+      "epoch": 2.0,
+      "grad_norm": 4.529049873352051,
+      "learning_rate": 2.0480115365495928e-05,
+      "loss": 1.4109,
+      "mean_token_accuracy": 0.597580885887146,
+      "num_tokens": 99138.0,
       "step": 24
     },
     {
-      "entropy": 1.6066707968711853,
-      "epoch": 1.1975308641975309,
-      "grad_norm": 4.747236728668213,
-      "learning_rate": 4.5786740307563636e-05,
-      "loss": 1.2268,
-      "mean_token_accuracy": 0.6532000750303268,
-      "num_tokens": 53775.0,
+      "entropy": 1.6290615797042847,
+      "epoch": 2.086021505376344,
+      "grad_norm": 2.889042854309082,
+      "learning_rate": 1.8337814009344716e-05,
+      "loss": 1.1525,
+      "mean_token_accuracy": 0.66117724776268,
+      "num_tokens": 105299.0,
       "step": 25
     },
     {
-      "entropy": 1.7001492828130722,
-      "epoch": 1.2469135802469136,
-      "grad_norm": 4.45648193359375,
-      "learning_rate": 4.4833833507280884e-05,
-      "loss": 1.2339,
-      "mean_token_accuracy": 0.6452045813202858,
-      "num_tokens": 56079.0,
+      "entropy": 1.8329174295067787,
+      "epoch": 2.172043010752688,
+      "grad_norm": 3.5605225563049316,
+      "learning_rate": 1.6120951403796367e-05,
+      "loss": 1.2644,
+      "mean_token_accuracy": 0.6234499476850033,
+      "num_tokens": 109812.0,
       "step": 26
     },
     {
-      "entropy": 1.5360765606164932,
-      "epoch": 1.2962962962962963,
-      "grad_norm": 4.499372482299805,
-      "learning_rate": 4.379599518697444e-05,
-      "loss": 1.219,
-      "mean_token_accuracy": 0.6569549441337585,
-      "num_tokens": 58349.0,
+      "entropy": 1.8179172724485397,
+      "epoch": 2.258064516129032,
+      "grad_norm": 3.6976897716522217,
+      "learning_rate": 1.3879048596203637e-05,
+      "loss": 1.2585,
+      "mean_token_accuracy": 0.6302964184433222,
+      "num_tokens": 114027.0,
       "step": 27
     },
     {
-      "entropy": 1.550552174448967,
-      "epoch": 1.345679012345679,
-      "grad_norm": 4.507914066314697,
-      "learning_rate": 4.267766952966369e-05,
-      "loss": 1.1618,
-      "mean_token_accuracy": 0.6673635169863701,
-      "num_tokens": 60570.0,
+      "entropy": 1.8265853449702263,
+      "epoch": 2.3440860215053765,
+      "grad_norm": 4.188506126403809,
+      "learning_rate": 1.1662185990655285e-05,
+      "loss": 1.2727,
+      "mean_token_accuracy": 0.6514358706772327,
+      "num_tokens": 117575.0,
       "step": 28
     },
     {
-      "entropy": 1.612103745341301,
-      "epoch": 1.3950617283950617,
-      "grad_norm": 5.213642597198486,
-      "learning_rate": 4.148364537750172e-05,
-      "loss": 1.2609,
-      "mean_token_accuracy": 0.6316312029957771,
-      "num_tokens": 62650.0,
+      "entropy": 1.709010697901249,
+      "epoch": 2.4301075268817205,
+      "grad_norm": 3.463564157485962,
+      "learning_rate": 9.519884634504074e-06,
+      "loss": 1.2256,
+      "mean_token_accuracy": 0.6496482565999031,
+      "num_tokens": 122001.0,
       "step": 29
     },
     {
-      "entropy": 1.4736015498638153,
-      "epoch": 1.4444444444444444,
-      "grad_norm": 4.291496753692627,
-      "learning_rate": 4.021903572521802e-05,
-      "loss": 1.1779,
-      "mean_token_accuracy": 0.6751257181167603,
-      "num_tokens": 65419.0,
+      "entropy": 1.6990807726979256,
+      "epoch": 2.5161290322580645,
+      "grad_norm": 3.3997740745544434,
+      "learning_rate": 7.500000000000004e-06,
+      "loss": 1.2212,
+      "mean_token_accuracy": 0.6523125059902668,
+      "num_tokens": 126120.0,
       "step": 30
     },
     {
-      "entropy": 1.5430797338485718,
-      "epoch": 1.4938271604938271,
-      "grad_norm": 4.683901309967041,
-      "learning_rate": 3.888925582549006e-05,
-      "loss": 1.0674,
-      "mean_token_accuracy": 0.6916785389184952,
-      "num_tokens": 67644.0,
+      "entropy": 1.6268121972680092,
+      "epoch": 2.6021505376344085,
+      "grad_norm": 3.414900541305542,
+      "learning_rate": 5.647652972118998e-06,
+      "loss": 1.192,
+      "mean_token_accuracy": 0.6545833833515644,
+      "num_tokens": 130586.0,
       "step": 31
     },
     {
-      "entropy": 1.5857491940259933,
-      "epoch": 1.5432098765432098,
-      "grad_norm": 4.896679878234863,
-      "learning_rate": 3.7500000000000003e-05,
-      "loss": 1.1883,
-      "mean_token_accuracy": 0.6605922058224678,
-      "num_tokens": 69764.0,
+      "entropy": 1.6691016554832458,
+      "epoch": 2.688172043010753,
+      "grad_norm": 3.3262569904327393,
+      "learning_rate": 4.004221922552608e-06,
+      "loss": 1.165,
+      "mean_token_accuracy": 0.6606670804321766,
+      "num_tokens": 135032.0,
       "step": 32
     },
     {
-      "entropy": 1.6383006423711777,
-      "epoch": 1.5925925925925926,
-      "grad_norm": 5.426114559173584,
-      "learning_rate": 3.6057217255475034e-05,
-      "loss": 1.1589,
-      "mean_token_accuracy": 0.6661425679922104,
-      "num_tokens": 71383.0,
+      "entropy": 1.8299531117081642,
+      "epoch": 2.774193548387097,
+      "grad_norm": 4.140204429626465,
+      "learning_rate": 2.60641838526008e-06,
+      "loss": 1.3193,
+      "mean_token_accuracy": 0.6374913081526756,
+      "num_tokens": 138140.0,
       "step": 33
     },
     {
-      "entropy": 1.5539519935846329,
-      "epoch": 1.6419753086419753,
-      "grad_norm": 5.141643524169922,
-      "learning_rate": 3.456708580912725e-05,
-      "loss": 1.1704,
-      "mean_token_accuracy": 0.6650657132267952,
-      "num_tokens": 73330.0,
+      "entropy": 1.6389653831720352,
+      "epoch": 2.860215053763441,
+      "grad_norm": 3.637709140777588,
+      "learning_rate": 1.4854669814637145e-06,
+      "loss": 1.1786,
+      "mean_token_accuracy": 0.663432490080595,
+      "num_tokens": 141988.0,
       "step": 34
     },
     {
-      "entropy": 1.4722215682268143,
-      "epoch": 1.691358024691358,
-      "grad_norm": 4.59281587600708,
-      "learning_rate": 3.303598663257904e-05,
-      "loss": 1.0203,
-      "mean_token_accuracy": 0.6820086091756821,
-      "num_tokens": 75429.0,
+      "entropy": 1.7234127074480057,
+      "epoch": 2.946236559139785,
+      "grad_norm": 3.819674253463745,
+      "learning_rate": 6.664079132078881e-07,
+      "loss": 1.2491,
+      "mean_token_accuracy": 0.6412298493087292,
+      "num_tokens": 145552.0,
       "step": 35
     },
     {
-      "entropy": 1.4010685980319977,
-      "epoch": 1.7407407407407407,
-      "grad_norm": 4.383392333984375,
-      "learning_rate": 3.147047612756302e-05,
-      "loss": 1.0248,
-      "mean_token_accuracy": 0.6845208778977394,
-      "num_tokens": 78459.0,
+      "entropy": 1.5378701567649842,
+      "epoch": 3.0,
+      "grad_norm": 3.6004512310028076,
+      "learning_rate": 1.6753760662307217e-07,
+      "loss": 1.0771,
+      "mean_token_accuracy": 0.6773547649383544,
+      "num_tokens": 148707.0,
       "step": 36
-    },
-    {
-      "entropy": 1.5117781907320023,
-      "epoch": 1.7901234567901234,
-      "grad_norm": 5.551262378692627,
-      "learning_rate": 2.9877258050403212e-05,
-      "loss": 1.135,
-      "mean_token_accuracy": 0.6859684735536575,
-      "num_tokens": 80102.0,
-      "step": 37
-    },
-    {
-      "entropy": 1.4906423836946487,
-      "epoch": 1.8395061728395061,
-      "grad_norm": 4.837874412536621,
-      "learning_rate": 2.8263154805501297e-05,
-      "loss": 1.1213,
-      "mean_token_accuracy": 0.6715603843331337,
-      "num_tokens": 82277.0,
-      "step": 38
-    },
-    {
-      "entropy": 1.2754715234041214,
-      "epoch": 1.8888888888888888,
-      "grad_norm": 3.6830785274505615,
-      "learning_rate": 2.663507823075358e-05,
-      "loss": 0.9663,
-      "mean_token_accuracy": 0.7010771930217743,
-      "num_tokens": 85933.0,
-      "step": 39
-    },
-    {
-      "entropy": 1.4681971222162247,
-      "epoch": 1.9382716049382716,
-      "grad_norm": 4.174588680267334,
-      "learning_rate": 2.5e-05,
-      "loss": 1.154,
-      "mean_token_accuracy": 0.6745116859674454,
-      "num_tokens": 88612.0,
-      "step": 40
-    },
-    {
-      "entropy": 1.4022098779678345,
-      "epoch": 1.9876543209876543,
-      "grad_norm": 4.754171848297119,
-      "learning_rate": 2.3364921769246423e-05,
-      "loss": 1.1365,
-      "mean_token_accuracy": 0.6754558682441711,
-      "num_tokens": 90823.0,
-      "step": 41
-    },
-    {
-      "entropy": 1.588551640510559,
-      "epoch": 2.0,
-      "grad_norm": 10.379326820373535,
-      "learning_rate": 2.173684519449872e-05,
-      "loss": 1.0527,
-      "mean_token_accuracy": 0.7145833373069763,
-      "num_tokens": 91240.0,
-      "step": 42
-    },
-    {
-      "entropy": 1.3871477395296097,
-      "epoch": 2.049382716049383,
-      "grad_norm": 3.7378275394439697,
-      "learning_rate": 2.0122741949596797e-05,
-      "loss": 0.7188,
-      "mean_token_accuracy": 0.7726586982607841,
-      "num_tokens": 93342.0,
-      "step": 43
-    },
-    {
-      "entropy": 1.4595267176628113,
-      "epoch": 2.0987654320987654,
-      "grad_norm": 4.306966781616211,
-      "learning_rate": 1.852952387243698e-05,
-      "loss": 0.7157,
-      "mean_token_accuracy": 0.7996886223554611,
-      "num_tokens": 95250.0,
-      "step": 44
-    },
-    {
-      "entropy": 1.2950524985790253,
-      "epoch": 2.148148148148148,
-      "grad_norm": 3.7582972049713135,
-      "learning_rate": 1.6964013367420966e-05,
-      "loss": 0.6962,
-      "mean_token_accuracy": 0.7863549143075943,
-      "num_tokens": 97864.0,
-      "step": 45
-    },
-    {
-      "entropy": 1.2319310307502747,
-      "epoch": 2.197530864197531,
-      "grad_norm": 3.64840030670166,
-      "learning_rate": 1.5432914190872757e-05,
-      "loss": 0.6659,
-      "mean_token_accuracy": 0.8138639628887177,
-      "num_tokens": 100323.0,
-      "step": 46
-    },
-    {
-      "entropy": 1.19608373939991,
-      "epoch": 2.246913580246914,
-      "grad_norm": 3.90803599357605,
-      "learning_rate": 1.3942782744524973e-05,
-      "loss": 0.6014,
-      "mean_token_accuracy": 0.8257261663675308,
-      "num_tokens": 102615.0,
-      "step": 47
-    },
-    {
-      "entropy": 1.2897031158208847,
-      "epoch": 2.2962962962962963,
-      "grad_norm": 4.506102561950684,
-      "learning_rate": 1.2500000000000006e-05,
-      "loss": 0.7683,
-      "mean_token_accuracy": 0.7799497023224831,
-      "num_tokens": 104543.0,
-      "step": 48
-    },
-    {
-      "entropy": 1.209401585161686,
-      "epoch": 2.3456790123456788,
-      "grad_norm": 3.957465887069702,
-      "learning_rate": 1.1110744174509952e-05,
-      "loss": 0.5976,
-      "mean_token_accuracy": 0.8096398636698723,
-      "num_tokens": 106982.0,
-      "step": 49
-    },
-    {
-      "entropy": 1.203671395778656,
-      "epoch": 2.3950617283950617,
-      "grad_norm": 4.684906005859375,
-      "learning_rate": 9.780964274781984e-06,
-      "loss": 0.6821,
-      "mean_token_accuracy": 0.8014075681567192,
-      "num_tokens": 108907.0,
-      "step": 50
-    },
-    {
-      "entropy": 1.21255362033844,
-      "epoch": 2.4444444444444446,
-      "grad_norm": 5.399158954620361,
-      "learning_rate": 8.51635462249828e-06,
-      "loss": 0.7489,
-      "mean_token_accuracy": 0.7789968624711037,
-      "num_tokens": 110649.0,
-      "step": 51
-    },
-    {
-      "entropy": 1.1142661720514297,
-      "epoch": 2.493827160493827,
-      "grad_norm": 4.029307842254639,
-      "learning_rate": 7.3223304703363135e-06,
-      "loss": 0.6049,
-      "mean_token_accuracy": 0.8112852945923805,
-      "num_tokens": 113000.0,
-      "step": 52
-    },
-    {
-      "entropy": 1.0388604775071144,
-      "epoch": 2.5432098765432096,
-      "grad_norm": 4.7035369873046875,
-      "learning_rate": 6.204004813025568e-06,
-      "loss": 0.6827,
-      "mean_token_accuracy": 0.8062720224261284,
-      "num_tokens": 115252.0,
-      "step": 53
-    },
-    {
-      "entropy": 0.9865575134754181,
-      "epoch": 2.5925925925925926,
-      "grad_norm": 4.302714824676514,
-      "learning_rate": 5.166166492719124e-06,
-      "loss": 0.5451,
-      "mean_token_accuracy": 0.8259979337453842,
-      "num_tokens": 117622.0,
-      "step": 54
-    },
-    {
-      "entropy": 0.9599115997552872,
-      "epoch": 2.6419753086419755,
-      "grad_norm": 4.7989349365234375,
-      "learning_rate": 4.213259692436367e-06,
-      "loss": 0.5905,
-      "mean_token_accuracy": 0.8102814555168152,
-      "num_tokens": 120361.0,
-      "step": 55
-    },
-    {
-      "entropy": 1.072813868522644,
-      "epoch": 2.691358024691358,
-      "grad_norm": 4.888575553894043,
-      "learning_rate": 3.3493649053890326e-06,
-      "loss": 0.689,
-      "mean_token_accuracy": 0.7948291003704071,
-      "num_tokens": 122610.0,
-      "step": 56
-    },
-    {
-      "entropy": 1.0314068496227264,
-      "epoch": 2.7407407407407405,
-      "grad_norm": 5.251216411590576,
-      "learning_rate": 2.578181461682794e-06,
-      "loss": 0.5865,
-      "mean_token_accuracy": 0.8308965489268303,
-      "num_tokens": 124571.0,
-      "step": 57
-    },
-    {
-      "entropy": 1.0771163925528526,
-      "epoch": 2.7901234567901234,
-      "grad_norm": 5.723106384277344,
-      "learning_rate": 1.9030116872178316e-06,
-      "loss": 0.7885,
-      "mean_token_accuracy": 0.7869210541248322,
-      "num_tokens": 126741.0,
-      "step": 58
-    },
-    {
-      "entropy": 1.0658350437879562,
-      "epoch": 2.8395061728395063,
-      "grad_norm": 4.829237937927246,
-      "learning_rate": 1.3267467626223606e-06,
-      "loss": 0.6772,
-      "mean_token_accuracy": 0.8049523532390594,
-      "num_tokens": 129201.0,
-      "step": 59
-    },
-    {
-      "entropy": 1.0056668892502785,
-      "epoch": 2.888888888888889,
-      "grad_norm": 5.5694780349731445,
-      "learning_rate": 8.51854342773295e-07,
-      "loss": 0.6802,
-      "mean_token_accuracy": 0.8025626465678215,
-      "num_tokens": 131344.0,
-      "step": 60
-    },
-    {
-      "entropy": 1.0122049376368523,
-      "epoch": 2.9382716049382713,
-      "grad_norm": 4.719507217407227,
-      "learning_rate": 4.803679899192392e-07,
-      "loss": 0.6776,
-      "mean_token_accuracy": 0.79056666046381,
-      "num_tokens": 133881.0,
-      "step": 61
-    },
-    {
-      "entropy": 1.0161257833242416,
-      "epoch": 2.9876543209876543,
-      "grad_norm": 5.256667613983154,
-      "learning_rate": 2.1387846565474045e-07,
-      "loss": 0.6248,
-      "mean_token_accuracy": 0.812793679535389,
-      "num_tokens": 136287.0,
-      "step": 62
     }
   ],
   "logging_steps": 1,
-  "max_steps": 63,
+  "max_steps": 36,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 3,
   "save_steps": 1,
@@ -642,12 +382,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": false
+        "should_training_stop": true
       },
       "attributes": {}
     }
   },
-  "total_flos": 3668970048184320.0,
+  "total_flos": 3948689883697152.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:857e29b87cbe8e931f9efcd116edb33ab62db020fffc73980185f4578c5fed6f
+oid sha256:53d0f078cee8357b221829a4c8eb0a57bcfaea48e8a35db45a7c304eadd3444e
 size 6353