Model save

README.md

@@ -27,7 +27,7 @@ print(output["generated_text"])
 
 ## Training procedure
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/yucaiwenzhenghao-xi-an-jiaotong-university-/huggingface/runs/t2qed9ky) 
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/yucaiwenzhenghao-xi-an-jiaotong-university-/huggingface/runs/6ef4hzdi) 
 
 
 This model was trained with SFT.

all_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 0.999259807549963,
     "total_flos": 76888336760832.0,
-    "train_loss": 0.7675497867442943,
-    "train_runtime": 4035.0195,
+    "train_loss": 0.7675483689484773,
+    "train_runtime": 4012.107,
     "train_samples": 16610,
-    "train_samples_per_second": 5.356,
-    "train_steps_per_second": 0.167
+    "train_samples_per_second": 5.386,
+    "train_steps_per_second": 0.168
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e81042c69ee0514cde5b9c6675b7857a75de87f237af783b48a8568714d4d52
+oid sha256:bd1e4ab52eace928ef1ac462684f417230daeed1323c2ef226d2be90846376ab
 size 3087467144

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 0.999259807549963,
     "total_flos": 76888336760832.0,
-    "train_loss": 0.7675497867442943,
-    "train_runtime": 4035.0195,
+    "train_loss": 0.7675483689484773,
+    "train_runtime": 4012.107,
     "train_samples": 16610,
-    "train_samples_per_second": 5.356,
-    "train_steps_per_second": 0.167
+    "train_samples_per_second": 5.386,
+    "train_steps_per_second": 0.168
 }

trainer_state.json

@@ -10,1005 +10,1005 @@
   "log_history": [
     {
       "epoch": 0.007401924500370096,
-      "grad_norm": 2.6781647247926057,
+      "grad_norm": 2.6769801199789844,
       "learning_rate": 1.4705882352941177e-06,
       "loss": 1.0786,
       "step": 5
     },
     {
       "epoch": 0.014803849000740192,
-      "grad_norm": 2.288401958011756,
+      "grad_norm": 2.289153229357034,
       "learning_rate": 2.9411764705882355e-06,
-      "loss": 1.0902,
+      "loss": 1.0901,
       "step": 10
     },
     {
       "epoch": 0.02220577350111029,
-      "grad_norm": 1.4174741515480414,
+      "grad_norm": 1.4151517309864259,
       "learning_rate": 4.411764705882353e-06,
       "loss": 1.062,
       "step": 15
     },
     {
       "epoch": 0.029607698001480384,
-      "grad_norm": 1.3747780673148666,
+      "grad_norm": 1.3753052911702501,
       "learning_rate": 5.882352941176471e-06,
       "loss": 1.0087,
       "step": 20
     },
     {
       "epoch": 0.037009622501850484,
-      "grad_norm": 1.1143835808010325,
+      "grad_norm": 1.117549735717063,
       "learning_rate": 7.352941176470589e-06,
       "loss": 0.9684,
       "step": 25
     },
     {
       "epoch": 0.04441154700222058,
-      "grad_norm": 0.9508184768376305,
+      "grad_norm": 0.9514445146454603,
       "learning_rate": 8.823529411764707e-06,
       "loss": 0.9217,
       "step": 30
     },
     {
       "epoch": 0.05181347150259067,
-      "grad_norm": 0.7597273167912698,
+      "grad_norm": 0.7594358797490329,
       "learning_rate": 1.0294117647058823e-05,
-      "loss": 0.8858,
+      "loss": 0.8859,
       "step": 35
     },
     {
       "epoch": 0.05921539600296077,
-      "grad_norm": 0.6842125108521274,
+      "grad_norm": 0.6853982401028429,
       "learning_rate": 1.1764705882352942e-05,
-      "loss": 0.863,
+      "loss": 0.8631,
       "step": 40
     },
     {
       "epoch": 0.06661732050333087,
-      "grad_norm": 0.7459216653551343,
+      "grad_norm": 0.7503960595270591,
       "learning_rate": 1.323529411764706e-05,
       "loss": 0.8486,
       "step": 45
     },
     {
       "epoch": 0.07401924500370097,
-      "grad_norm": 0.8145580582471774,
+      "grad_norm": 0.81742237765088,
       "learning_rate": 1.4705882352941179e-05,
       "loss": 0.861,
       "step": 50
     },
     {
       "epoch": 0.08142116950407106,
-      "grad_norm": 0.6247313023791736,
+      "grad_norm": 0.6232853094087558,
       "learning_rate": 1.6176470588235296e-05,
       "loss": 0.8545,
       "step": 55
     },
     {
       "epoch": 0.08882309400444116,
-      "grad_norm": 0.7347557433652043,
+      "grad_norm": 0.7312499750471685,
       "learning_rate": 1.7647058823529414e-05,
-      "loss": 0.8292,
+      "loss": 0.8293,
       "step": 60
     },
     {
       "epoch": 0.09622501850481126,
-      "grad_norm": 0.683279809341516,
+      "grad_norm": 0.6849629343658746,
       "learning_rate": 1.911764705882353e-05,
       "loss": 0.8259,
       "step": 65
     },
     {
       "epoch": 0.10362694300518134,
-      "grad_norm": 0.7678369445379591,
+      "grad_norm": 0.7670921820036198,
       "learning_rate": 1.9999464266898485e-05,
-      "loss": 0.821,
+      "loss": 0.8211,
       "step": 70
     },
     {
       "epoch": 0.11102886750555144,
-      "grad_norm": 0.9654917736747842,
+      "grad_norm": 0.9562639233109864,
       "learning_rate": 1.9993437928712977e-05,
-      "loss": 0.8164,
+      "loss": 0.8165,
       "step": 75
     },
     {
       "epoch": 0.11843079200592153,
-      "grad_norm": 0.8193410530544045,
+      "grad_norm": 0.8187251839834904,
       "learning_rate": 1.998071963486563e-05,
       "loss": 0.8062,
       "step": 80
     },
     {
       "epoch": 0.12583271650629163,
-      "grad_norm": 0.6566255946293276,
+      "grad_norm": 0.6607954272855435,
       "learning_rate": 1.9961317901970953e-05,
       "loss": 0.7945,
       "step": 85
     },
     {
       "epoch": 0.13323464100666174,
-      "grad_norm": 0.7534606513821027,
+      "grad_norm": 0.7512885304003232,
       "learning_rate": 1.993524572210807e-05,
       "loss": 0.7947,
       "step": 90
     },
     {
       "epoch": 0.14063656550703182,
-      "grad_norm": 0.7929209076815809,
+      "grad_norm": 0.7919859825103274,
       "learning_rate": 1.990252055412077e-05,
-      "loss": 0.7906,
+      "loss": 0.7905,
       "step": 95
     },
     {
       "epoch": 0.14803849000740193,
-      "grad_norm": 0.7734193761295386,
+      "grad_norm": 0.7738309429220392,
       "learning_rate": 1.9863164311926433e-05,
       "loss": 0.8172,
       "step": 100
     },
     {
       "epoch": 0.14803849000740193,
-      "eval_loss": 0.81982421875,
-      "eval_runtime": 5.3926,
-      "eval_samples_per_second": 23.736,
-      "eval_steps_per_second": 2.967,
+      "eval_loss": 0.8197109699249268,
+      "eval_runtime": 5.3703,
+      "eval_samples_per_second": 23.835,
+      "eval_steps_per_second": 2.979,
       "step": 100
     },
     {
       "epoch": 0.15544041450777202,
-      "grad_norm": 0.7144802532437275,
+      "grad_norm": 0.7130351365422478,
       "learning_rate": 1.981720334984174e-05,
       "loss": 0.7921,
       "step": 105
     },
     {
       "epoch": 0.16284233900814213,
-      "grad_norm": 0.7076920816159826,
+      "grad_norm": 0.7061699809541575,
       "learning_rate": 1.9764668444934853e-05,
       "loss": 0.7859,
       "step": 110
     },
     {
       "epoch": 0.1702442635085122,
-      "grad_norm": 0.7200029250965106,
+      "grad_norm": 0.7182947479746219,
       "learning_rate": 1.970559477641606e-05,
-      "loss": 0.7632,
+      "loss": 0.7631,
       "step": 115
     },
     {
       "epoch": 0.17764618800888232,
-      "grad_norm": 0.7290155858431464,
+      "grad_norm": 0.7311916453291761,
       "learning_rate": 1.9640021902080523e-05,
       "loss": 0.7929,
       "step": 120
     },
     {
       "epoch": 0.1850481125092524,
-      "grad_norm": 0.6265865520761479,
+      "grad_norm": 0.6285240054314736,
       "learning_rate": 1.9567993731818988e-05,
       "loss": 0.7916,
       "step": 125
     },
     {
       "epoch": 0.19245003700962252,
-      "grad_norm": 0.7409568184465672,
+      "grad_norm": 0.7404725250217843,
       "learning_rate": 1.9489558498214197e-05,
       "loss": 0.7842,
       "step": 130
     },
     {
       "epoch": 0.1998519615099926,
-      "grad_norm": 0.6474337479927059,
+      "grad_norm": 0.6486675105037224,
       "learning_rate": 1.9404768724242667e-05,
-      "loss": 0.7703,
+      "loss": 0.7704,
       "step": 135
     },
     {
       "epoch": 0.20725388601036268,
-      "grad_norm": 0.7295582072883805,
+      "grad_norm": 0.7306207383223645,
       "learning_rate": 1.931368118810346e-05,
-      "loss": 0.7946,
+      "loss": 0.7947,
       "step": 140
     },
     {
       "epoch": 0.2146558105107328,
-      "grad_norm": 0.7514683834432097,
+      "grad_norm": 0.7466637103159883,
       "learning_rate": 1.92163568851975e-05,
-      "loss": 0.7757,
+      "loss": 0.7756,
       "step": 145
     },
     {
       "epoch": 0.22205773501110287,
-      "grad_norm": 0.8025939179758401,
+      "grad_norm": 0.7971646525282183,
       "learning_rate": 1.911286098728296e-05,
       "loss": 0.772,
       "step": 150
     },
     {
       "epoch": 0.22945965951147299,
-      "grad_norm": 0.6984116896613638,
+      "grad_norm": 0.7009790395257299,
       "learning_rate": 1.900326279883392e-05,
-      "loss": 0.8016,
+      "loss": 0.8015,
       "step": 155
     },
     {
       "epoch": 0.23686158401184307,
-      "grad_norm": 0.7047019976319429,
+      "grad_norm": 0.7000968794313863,
       "learning_rate": 1.8887635710631716e-05,
-      "loss": 0.8044,
+      "loss": 0.8043,
       "step": 160
     },
     {
       "epoch": 0.24426350851221318,
-      "grad_norm": 0.6992015109851578,
+      "grad_norm": 0.6947279868702056,
       "learning_rate": 1.8766057150619865e-05,
       "loss": 0.7775,
       "step": 165
     },
     {
       "epoch": 0.25166543301258326,
-      "grad_norm": 0.705185843698171,
+      "grad_norm": 0.7054404867195952,
       "learning_rate": 1.8638608532055635e-05,
-      "loss": 0.7946,
+      "loss": 0.7947,
       "step": 170
     },
     {
       "epoch": 0.25906735751295334,
-      "grad_norm": 0.6603988530124156,
+      "grad_norm": 0.6591670530138999,
       "learning_rate": 1.8505375198992856e-05,
-      "loss": 0.7829,
+      "loss": 0.7831,
       "step": 175
     },
     {
       "epoch": 0.2664692820133235,
-      "grad_norm": 0.7165198506448356,
+      "grad_norm": 0.7106688805726791,
       "learning_rate": 1.836644636913258e-05,
-      "loss": 0.7541,
+      "loss": 0.7543,
       "step": 180
     },
     {
       "epoch": 0.27387120651369357,
-      "grad_norm": 0.7730262515380419,
+      "grad_norm": 0.7764398857270777,
       "learning_rate": 1.8221915074079764e-05,
-      "loss": 0.7778,
+      "loss": 0.7779,
       "step": 185
     },
     {
       "epoch": 0.28127313101406365,
-      "grad_norm": 0.8563593582048756,
+      "grad_norm": 0.8612641589834193,
       "learning_rate": 1.8071878097046064e-05,
-      "loss": 0.7563,
+      "loss": 0.7564,
       "step": 190
     },
     {
       "epoch": 0.28867505551443373,
-      "grad_norm": 0.6241995078162836,
+      "grad_norm": 0.6890423279187838,
       "learning_rate": 1.7916435908040413e-05,
-      "loss": 0.7725,
+      "loss": 0.7723,
       "step": 195
     },
     {
       "epoch": 0.29607698001480387,
-      "grad_norm": 0.6480078305074343,
+      "grad_norm": 0.6443921271623081,
       "learning_rate": 1.7755692596590778e-05,
       "loss": 0.7746,
       "step": 200
     },
     {
       "epoch": 0.29607698001480387,
-      "eval_loss": 0.7899575233459473,
-      "eval_runtime": 5.4002,
-      "eval_samples_per_second": 23.703,
-      "eval_steps_per_second": 2.963,
+      "eval_loss": 0.7898486852645874,
+      "eval_runtime": 5.365,
+      "eval_samples_per_second": 23.858,
+      "eval_steps_per_second": 2.982,
       "step": 200
     },
     {
       "epoch": 0.30347890451517395,
-      "grad_norm": 0.6919136091545958,
+      "grad_norm": 0.6925911374679931,
       "learning_rate": 1.7589755802042188e-05,
       "loss": 0.7729,
       "step": 205
     },
     {
       "epoch": 0.31088082901554404,
-      "grad_norm": 0.6856028013988533,
+      "grad_norm": 0.6892184297566977,
       "learning_rate": 1.7418736641477636e-05,
-      "loss": 0.7562,
+      "loss": 0.7561,
       "step": 210
     },
     {
       "epoch": 0.3182827535159141,
-      "grad_norm": 0.6252759549970119,
+      "grad_norm": 0.624988978642692,
       "learning_rate": 1.7242749635310222e-05,
       "loss": 0.7581,
       "step": 215
     },
     {
       "epoch": 0.32568467801628426,
-      "grad_norm": 0.6521044078734693,
+      "grad_norm": 0.6494026923603395,
       "learning_rate": 1.7061912630596252e-05,
       "loss": 0.7604,
       "step": 220
     },
     {
       "epoch": 0.33308660251665434,
-      "grad_norm": 0.7192237679886784,
+      "grad_norm": 0.7186278404090587,
       "learning_rate": 1.6876346722120747e-05,
       "loss": 0.7752,
       "step": 225
     },
     {
       "epoch": 0.3404885270170244,
-      "grad_norm": 0.6888594212614928,
+      "grad_norm": 0.6877574111727853,
       "learning_rate": 1.6686176171308125e-05,
       "loss": 0.7977,
       "step": 230
     },
     {
       "epoch": 0.3478904515173945,
-      "grad_norm": 0.6671822064087696,
+      "grad_norm": 0.6696578728231912,
       "learning_rate": 1.6491528323012412e-05,
       "loss": 0.7595,
       "step": 235
     },
     {
       "epoch": 0.35529237601776464,
-      "grad_norm": 0.6532924847521687,
+      "grad_norm": 0.6335948217388314,
       "learning_rate": 1.6292533520242663e-05,
-      "loss": 0.7623,
+      "loss": 0.7622,
       "step": 240
     },
     {
       "epoch": 0.3626943005181347,
-      "grad_norm": 0.6369988599906944,
+      "grad_norm": 0.640069468830107,
       "learning_rate": 1.6089325016880737e-05,
-      "loss": 0.7525,
+      "loss": 0.7526,
       "step": 245
     },
     {
       "epoch": 0.3700962250185048,
-      "grad_norm": 0.6869876881988707,
+      "grad_norm": 0.676843647805683,
       "learning_rate": 1.588203888844982e-05,
-      "loss": 0.7681,
+      "loss": 0.768,
       "step": 250
     },
     {
       "epoch": 0.3774981495188749,
-      "grad_norm": 0.6300847252583518,
+      "grad_norm": 0.6351408648447299,
       "learning_rate": 1.5670813940993504e-05,
       "loss": 0.741,
       "step": 255
     },
     {
       "epoch": 0.38490007401924503,
-      "grad_norm": 0.5905982241821089,
+      "grad_norm": 0.5889957497779358,
       "learning_rate": 1.5455791618126407e-05,
       "loss": 0.7332,
       "step": 260
     },
     {
       "epoch": 0.3923019985196151,
-      "grad_norm": 0.6259578486934526,
+      "grad_norm": 0.6263513470697065,
       "learning_rate": 1.5237115906318565e-05,
-      "loss": 0.7571,
+      "loss": 0.7572,
       "step": 265
     },
     {
       "epoch": 0.3997039230199852,
-      "grad_norm": 0.6527525521030698,
+      "grad_norm": 0.6463851097490673,
       "learning_rate": 1.5014933238477069e-05,
-      "loss": 0.7379,
+      "loss": 0.7378,
       "step": 270
     },
     {
       "epoch": 0.4071058475203553,
-      "grad_norm": 0.6914681891506994,
+      "grad_norm": 0.6864222070004046,
       "learning_rate": 1.4789392395889468e-05,
-      "loss": 0.7632,
+      "loss": 0.7633,
       "step": 275
     },
     {
       "epoch": 0.41450777202072536,
-      "grad_norm": 0.6437284417495679,
+      "grad_norm": 0.6463393459572199,
       "learning_rate": 1.4560644408594602e-05,
-      "loss": 0.744,
+      "loss": 0.7439,
       "step": 280
     },
     {
       "epoch": 0.4219096965210955,
-      "grad_norm": 0.6526633861445033,
+      "grad_norm": 0.6563326063729631,
       "learning_rate": 1.432884245424761e-05,
-      "loss": 0.7556,
+      "loss": 0.7555,
       "step": 285
     },
     {
       "epoch": 0.4293116210214656,
-      "grad_norm": 0.6585175519478199,
+      "grad_norm": 0.6577336494046302,
       "learning_rate": 1.4094141755546816e-05,
-      "loss": 0.7832,
+      "loss": 0.7831,
       "step": 290
     },
     {
       "epoch": 0.43671354552183567,
-      "grad_norm": 0.6245524428900567,
+      "grad_norm": 0.6230469159629758,
       "learning_rate": 1.3856699476291176e-05,
-      "loss": 0.7426,
+      "loss": 0.7427,
       "step": 295
     },
     {
       "epoch": 0.44411547002220575,
-      "grad_norm": 0.6836573073424697,
+      "grad_norm": 0.6803603436676453,
       "learning_rate": 1.3616674616137902e-05,
-      "loss": 0.7644,
+      "loss": 0.7643,
       "step": 300
     },
     {
       "epoch": 0.44411547002220575,
-      "eval_loss": 0.7754608392715454,
-      "eval_runtime": 5.3768,
-      "eval_samples_per_second": 23.806,
-      "eval_steps_per_second": 2.976,
+      "eval_loss": 0.7755386829376221,
+      "eval_runtime": 5.3686,
+      "eval_samples_per_second": 23.843,
+      "eval_steps_per_second": 2.98,
       "step": 300
     },
     {
       "epoch": 0.4515173945225759,
-      "grad_norm": 0.6493769918765077,
+      "grad_norm": 0.6538871821580508,
       "learning_rate": 1.3374227904130724e-05,
       "loss": 0.7548,
       "step": 305
     },
     {
       "epoch": 0.45891931902294597,
-      "grad_norm": 0.6438484172918043,
+      "grad_norm": 0.6412401526107059,
       "learning_rate": 1.3129521691070108e-05,
       "loss": 0.7327,
       "step": 310
     },
     {
       "epoch": 0.46632124352331605,
-      "grad_norm": 0.7116070089605211,
+      "grad_norm": 0.7175626595120815,
       "learning_rate": 1.2882719840797473e-05,
       "loss": 0.7513,
       "step": 315
     },
     {
       "epoch": 0.47372316802368614,
-      "grad_norm": 0.6771383036079919,
+      "grad_norm": 0.6865616455390183,
       "learning_rate": 1.2633987620466229e-05,
-      "loss": 0.7352,
+      "loss": 0.7353,
       "step": 320
     },
     {
       "epoch": 0.4811250925240563,
-      "grad_norm": 0.6345431158484713,
+      "grad_norm": 0.6349023571241349,
       "learning_rate": 1.2383491589873122e-05,
-      "loss": 0.7405,
+      "loss": 0.7404,
       "step": 325
     },
     {
       "epoch": 0.48852701702442636,
-      "grad_norm": 0.6060761585679342,
+      "grad_norm": 0.6069292373819503,
       "learning_rate": 1.213139948992394e-05,
-      "loss": 0.7497,
+      "loss": 0.7499,
       "step": 330
     },
     {
       "epoch": 0.49592894152479644,
-      "grad_norm": 0.6887701360104382,
+      "grad_norm": 0.6884470180718896,
       "learning_rate": 1.187788013030837e-05,
       "loss": 0.7468,
       "step": 335
     },
     {
       "epoch": 0.5033308660251665,
-      "grad_norm": 0.6105058268452026,
+      "grad_norm": 0.6109069389568509,
       "learning_rate": 1.1623103276459086e-05,
       "loss": 0.7506,
       "step": 340
     },
     {
       "epoch": 0.5107327905255367,
-      "grad_norm": 0.6373752567024148,
+      "grad_norm": 0.642078110839711,
       "learning_rate": 1.1367239535870913e-05,
       "loss": 0.7425,
       "step": 345
     },
     {
       "epoch": 0.5181347150259067,
-      "grad_norm": 0.6837419442122492,
+      "grad_norm": 0.6834108569853193,
       "learning_rate": 1.1110460243856051e-05,
       "loss": 0.7301,
       "step": 350
     },
     {
       "epoch": 0.5255366395262768,
-      "grad_norm": 0.658084823444436,
+      "grad_norm": 0.663649951052548,
       "learning_rate": 1.085293734881197e-05,
       "loss": 0.7468,
       "step": 355
     },
     {
       "epoch": 0.532938564026647,
-      "grad_norm": 0.5979260768516592,
+      "grad_norm": 0.5954162815689811,
       "learning_rate": 1.0594843297078736e-05,
-      "loss": 0.7659,
+      "loss": 0.7658,
       "step": 360
     },
     {
       "epoch": 0.540340488527017,
-      "grad_norm": 0.6879050891191374,
+      "grad_norm": 0.686716585388219,
       "learning_rate": 1.0336350917462925e-05,
-      "loss": 0.7557,
+      "loss": 0.7558,
       "step": 365
     },
     {
       "epoch": 0.5477424130273871,
-      "grad_norm": 0.601162386657378,
+      "grad_norm": 0.5992494474064747,
       "learning_rate": 1.0077633305505402e-05,
       "loss": 0.7432,
       "step": 370
     },
     {
       "epoch": 0.5551443375277573,
-      "grad_norm": 0.6301610633507608,
+      "grad_norm": 0.6301308510752842,
       "learning_rate": 9.818863707570476e-06,
       "loss": 0.7607,
       "step": 375
     },
     {
       "epoch": 0.5625462620281273,
-      "grad_norm": 0.6574643884953185,
+      "grad_norm": 0.6533374072399002,
       "learning_rate": 9.560215404834094e-06,
-      "loss": 0.7515,
+      "loss": 0.7514,
       "step": 380
     },
     {
       "epoch": 0.5699481865284974,
-      "grad_norm": 0.6326515269463117,
+      "grad_norm": 0.6339343966871546,
       "learning_rate": 9.30186159724869e-06,
       "loss": 0.7144,
       "step": 385
     },
     {
       "epoch": 0.5773501110288675,
-      "grad_norm": 0.6017001200572872,
+      "grad_norm": 0.6024127775100068,
       "learning_rate": 9.043975287562443e-06,
-      "loss": 0.7469,
+      "loss": 0.747,
       "step": 390
     },
     {
       "epoch": 0.5847520355292376,
-      "grad_norm": 0.677499413851095,
+      "grad_norm": 0.6705486605989903,
       "learning_rate": 8.786729165470584e-06,
-      "loss": 0.7253,
+      "loss": 0.7251,
       "step": 395
     },
     {
       "epoch": 0.5921539600296077,
-      "grad_norm": 0.6843777497552594,
+      "grad_norm": 0.6833286308389832,
       "learning_rate": 8.530295491976338e-06,
       "loss": 0.7306,
       "step": 400
     },
     {
       "epoch": 0.5921539600296077,
-      "eval_loss": 0.7637214660644531,
-      "eval_runtime": 5.3855,
-      "eval_samples_per_second": 23.768,
-      "eval_steps_per_second": 2.971,
+      "eval_loss": 0.7636929750442505,
+      "eval_runtime": 5.3692,
+      "eval_samples_per_second": 23.84,
+      "eval_steps_per_second": 2.98,
       "step": 400
     },
     {
       "epoch": 0.5995558845299778,
-      "grad_norm": 0.5899918694280646,
+      "grad_norm": 0.5926825289122618,
       "learning_rate": 8.274845984038916e-06,
       "loss": 0.7173,
       "step": 405
     },
     {
       "epoch": 0.6069578090303479,
-      "grad_norm": 0.6518517618816643,
+      "grad_norm": 0.6485786020572889,
       "learning_rate": 8.020551699585843e-06,
-      "loss": 0.7468,
+      "loss": 0.7469,
       "step": 410
     },
     {
       "epoch": 0.6143597335307179,
-      "grad_norm": 0.6059385090841897,
+      "grad_norm": 0.6038143644627321,
       "learning_rate": 7.76758292296659e-06,
       "loss": 0.7264,
       "step": 415
     },
     {
       "epoch": 0.6217616580310881,
-      "grad_norm": 0.6513276466511633,
+      "grad_norm": 0.6525648213164897,
       "learning_rate": 7.5161090509242005e-06,
       "loss": 0.7418,
       "step": 420
     },
     {
       "epoch": 0.6291635825314582,
-      "grad_norm": 0.6489500136832538,
+      "grad_norm": 0.6472092560370133,
       "learning_rate": 7.2662984791613186e-06,
-      "loss": 0.7344,
+      "loss": 0.7345,
       "step": 425
     },
     {
       "epoch": 0.6365655070318282,
-      "grad_norm": 0.6490633968151128,
+      "grad_norm": 0.6486952224924798,
       "learning_rate": 7.01831848957653e-06,
       "loss": 0.7488,
       "step": 430
     },
     {
       "epoch": 0.6439674315321984,
-      "grad_norm": 0.5991424360475788,
+      "grad_norm": 0.5995921591151809,
       "learning_rate": 6.772335138246548e-06,
       "loss": 0.7467,
       "step": 435
     },
     {
       "epoch": 0.6513693560325685,
-      "grad_norm": 0.6251007140054168,
+      "grad_norm": 0.6241512176448558,
       "learning_rate": 6.528513144229256e-06,
       "loss": 0.7427,
       "step": 440
     },
     {
       "epoch": 0.6587712805329385,
-      "grad_norm": 0.6090203477386716,
+      "grad_norm": 0.6111647376974468,
       "learning_rate": 6.287015779262064e-06,
       "loss": 0.7488,
       "step": 445
     },
     {
       "epoch": 0.6661732050333087,
-      "grad_norm": 0.6497593128690399,
+      "grad_norm": 0.6498096979963152,
       "learning_rate": 6.048004758429451e-06,
-      "loss": 0.7274,
+      "loss": 0.7273,
       "step": 450
     },
     {
       "epoch": 0.6735751295336787,
-      "grad_norm": 0.602361959825884,
+      "grad_norm": 0.6032913806566699,
       "learning_rate": 5.811640131872867e-06,
-      "loss": 0.7498,
+      "loss": 0.7497,
       "step": 455
     },
     {
       "epoch": 0.6809770540340488,
-      "grad_norm": 0.6064508470587194,
+      "grad_norm": 0.6076584860831334,
       "learning_rate": 5.578080177615575e-06,
-      "loss": 0.7202,
+      "loss": 0.7201,
       "step": 460
     },
     {
       "epoch": 0.688378978534419,
-      "grad_norm": 0.6225819526538398,
+      "grad_norm": 0.6233401316768133,
       "learning_rate": 5.347481295574141e-06,
       "loss": 0.717,
       "step": 465
     },
     {
       "epoch": 0.695780903034789,
-      "grad_norm": 0.6109601431815137,
+      "grad_norm": 0.6074432637453558,
       "learning_rate": 5.119997902827584e-06,
       "loss": 0.7285,
       "step": 470
     },
     {
       "epoch": 0.7031828275351591,
-      "grad_norm": 0.6062843935931566,
+      "grad_norm": 0.6053282667804308,
       "learning_rate": 4.8957823302142916e-06,
       "loss": 0.7353,
       "step": 475
     },
     {
       "epoch": 0.7105847520355293,
-      "grad_norm": 0.564845772481059,
+      "grad_norm": 0.5651527733504048,
       "learning_rate": 4.674984720325961e-06,
       "loss": 0.7207,
       "step": 480
     },
     {
       "epoch": 0.7179866765358993,
-      "grad_norm": 0.5840010916752476,
+      "grad_norm": 0.5860040588339941,
       "learning_rate": 4.457752926966888e-06,
       "loss": 0.7149,
       "step": 485
     },
     {
       "epoch": 0.7253886010362695,
-      "grad_norm": 0.580599582113682,
+      "grad_norm": 0.5804110480769434,
       "learning_rate": 4.244232416145839e-06,
       "loss": 0.7337,
       "step": 490
     },
     {
       "epoch": 0.7327905255366395,
-      "grad_norm": 0.5868998547392814,
+      "grad_norm": 0.5927598587788678,
       "learning_rate": 4.0345661686669745e-06,
       "loss": 0.7269,
       "step": 495
     },
     {
       "epoch": 0.7401924500370096,
-      "grad_norm": 0.621482167117223,
+      "grad_norm": 0.6147179659228765,
       "learning_rate": 3.828894584384867e-06,
       "loss": 0.7355,
       "step": 500
     },
     {
       "epoch": 0.7401924500370096,
-      "eval_loss": 0.7563306093215942,
-      "eval_runtime": 5.3879,
-      "eval_samples_per_second": 23.757,
-      "eval_steps_per_second": 2.97,
+      "eval_loss": 0.7562806010246277,
+      "eval_runtime": 5.3637,
+      "eval_samples_per_second": 23.864,
+      "eval_steps_per_second": 2.983,
       "step": 500
     },
     {
       "epoch": 0.7475943745373798,
-      "grad_norm": 0.5521085978482521,
+      "grad_norm": 0.5559997594301341,
       "learning_rate": 3.62735538818787e-06,
       "loss": 0.7196,
       "step": 505
     },
     {
       "epoch": 0.7549962990377498,
-      "grad_norm": 0.6287112506195603,
+      "grad_norm": 0.6309244918615329,
       "learning_rate": 3.4300835377726904e-06,
       "loss": 0.7233,
       "step": 510
     },
     {
       "epoch": 0.7623982235381199,
-      "grad_norm": 0.6180203465461854,
+      "grad_norm": 0.6184371610771912,
       "learning_rate": 3.2372111332720045e-06,
       "loss": 0.7587,
       "step": 515
     },
     {
       "epoch": 0.7698001480384901,
-      "grad_norm": 0.6063112930873026,
+      "grad_norm": 0.606915064634782,
       "learning_rate": 3.048867328795588e-06,
-      "loss": 0.7157,
+      "loss": 0.7156,
       "step": 520
     },
     {
       "epoch": 0.7772020725388601,
-      "grad_norm": 0.5997235418142394,
+      "grad_norm": 0.5970702305562683,
       "learning_rate": 2.865178245944218e-06,
-      "loss": 0.7143,
+      "loss": 0.7144,
       "step": 525
     },
     {
       "epoch": 0.7846039970392302,
-      "grad_norm": 0.5699540463370885,
+      "grad_norm": 0.5700255038227583,
       "learning_rate": 2.686266889354211e-06,
       "loss": 0.7374,
       "step": 530
     },
     {
       "epoch": 0.7920059215396003,
-      "grad_norm": 0.5881815010328638,
+      "grad_norm": 0.5871155109774989,
       "learning_rate": 2.5122530643292274e-06,
       "loss": 0.7428,
       "step": 535
     },
     {
       "epoch": 0.7994078460399704,
-      "grad_norm": 0.62807719166905,
+      "grad_norm": 0.6301777356538756,
       "learning_rate": 2.3432532966144526e-06,
-      "loss": 0.7324,
+      "loss": 0.7323,
       "step": 540
     },
     {
       "epoch": 0.8068097705403405,
-      "grad_norm": 0.5840252605297537,
+      "grad_norm": 0.5814869552487476,
       "learning_rate": 2.1793807543668857e-06,
-      "loss": 0.7339,
+      "loss": 0.7338,
       "step": 545
     },
     {
       "epoch": 0.8142116950407106,
-      "grad_norm": 0.5460055437727909,
+      "grad_norm": 0.5484281351456736,
       "learning_rate": 2.0207451723739633e-06,
       "loss": 0.7257,
       "step": 550
     },
     {
       "epoch": 0.8216136195410807,
-      "grad_norm": 0.5668656255730524,
+      "grad_norm": 0.5634326261626533,
       "learning_rate": 1.8674527785713247e-06,
-      "loss": 0.7324,
+      "loss": 0.7325,
       "step": 555
     },
     {
       "epoch": 0.8290155440414507,
-      "grad_norm": 0.6163708716263486,
+      "grad_norm": 0.6140394590528712,
       "learning_rate": 1.7196062229088606e-06,
       "loss": 0.6996,
       "step": 560
     },
     {
       "epoch": 0.8364174685418209,
-      "grad_norm": 0.6216420810406753,
+      "grad_norm": 0.6235750823611041,
       "learning_rate": 1.577304508612717e-06,
       "loss": 0.7297,
       "step": 565
     },
     {
       "epoch": 0.843819393042191,
-      "grad_norm": 0.5185684391833455,
+      "grad_norm": 0.5186238249593603,
       "learning_rate": 1.4406429258892762e-06,
       "loss": 0.7503,
       "step": 570
     },
     {
       "epoch": 0.851221317542561,
-      "grad_norm": 0.6120311845293285,
+      "grad_norm": 0.6162193205928901,
       "learning_rate": 1.3097129881154936e-06,
-      "loss": 0.7198,
+      "loss": 0.7199,
       "step": 575
     },
     {
       "epoch": 0.8586232420429312,
-      "grad_norm": 0.5849978417823803,
+      "grad_norm": 0.5804633378076518,
       "learning_rate": 1.1846023705583442e-06,
-      "loss": 0.7163,
+      "loss": 0.7162,
       "step": 580
     },
     {
       "epoch": 0.8660251665433013,
-      "grad_norm": 0.5761676526922828,
+      "grad_norm": 0.5772033426035238,
       "learning_rate": 1.065394851664394e-06,
-      "loss": 0.7343,
+      "loss": 0.7344,
       "step": 585
     },
     {
       "epoch": 0.8734270910436713,
-      "grad_norm": 0.544363784078604,
+      "grad_norm": 0.5426053523198552,
       "learning_rate": 9.521702569588199e-07,
-      "loss": 0.7535,
+      "loss": 0.7536,
       "step": 590
     },
     {
       "epoch": 0.8808290155440415,
-      "grad_norm": 0.574501510855254,
+      "grad_norm": 0.5722535438927212,
       "learning_rate": 8.450044055914497e-07,
       "loss": 0.722,
       "step": 595
     },
     {
       "epoch": 0.8882309400444115,
-      "grad_norm": 0.5591346956729883,
+      "grad_norm": 0.5616139648656882,
       "learning_rate": 7.439690595656013e-07,
-      "loss": 0.7446,
+      "loss": 0.7445,
       "step": 600
     },
     {
       "epoch": 0.8882309400444115,
-      "eval_loss": 0.7531599998474121,
-      "eval_runtime": 5.3736,
-      "eval_samples_per_second": 23.82,
-      "eval_steps_per_second": 2.977,
+      "eval_loss": 0.7532752156257629,
+      "eval_runtime": 5.3758,
+      "eval_samples_per_second": 23.81,
+      "eval_steps_per_second": 2.976,
       "step": 600
     },
     {
       "epoch": 0.8956328645447816,
-      "grad_norm": 0.6304431795586802,
+      "grad_norm": 0.6286499659968906,
       "learning_rate": 6.491318756837417e-07,
-      "loss": 0.7297,
+      "loss": 0.7298,
       "step": 605
     },
     {
       "epoch": 0.9030347890451518,
-      "grad_norm": 0.5107916832941353,
+      "grad_norm": 0.5116306332006605,
       "learning_rate": 5.605563602421149e-07,
-      "loss": 0.7059,
+      "loss": 0.7058,
       "step": 610
     },
     {
       "epoch": 0.9104367135455218,
-      "grad_norm": 0.5555038297086843,
+      "grad_norm": 0.5536125071383629,
       "learning_rate": 4.783018265047179e-07,
-      "loss": 0.7557,
+      "loss": 0.7556,
       "step": 615
     },
     {
       "epoch": 0.9178386380458919,
-      "grad_norm": 0.5812727386003688,
+      "grad_norm": 0.5813134121106769,
       "learning_rate": 4.024233549850509e-07,
-      "loss": 0.7436,
+      "loss": 0.7435,
       "step": 620
     },
     {
       "epoch": 0.9252405625462621,
-      "grad_norm": 0.5508995904728403,
+      "grad_norm": 0.5492708568069725,
       "learning_rate": 3.329717565622825e-07,
-      "loss": 0.7404,
+      "loss": 0.7403,
       "step": 625
     },
     {
       "epoch": 0.9326424870466321,
-      "grad_norm": 0.5490514571703399,
+      "grad_norm": 0.5511538126684848,
       "learning_rate": 2.6999353845651113e-07,
-      "loss": 0.7239,
+      "loss": 0.7241,
       "step": 630
     },
     {
       "epoch": 0.9400444115470022,
-      "grad_norm": 0.5827548037115096,
+      "grad_norm": 0.5858089750359664,
       "learning_rate": 2.1353087308590314e-07,
       "loss": 0.7389,
       "step": 635
     },
     {
       "epoch": 0.9474463360473723,
-      "grad_norm": 0.55809711437009,
+      "grad_norm": 0.5558966176000505,
       "learning_rate": 1.6362156982656085e-07,
-      "loss": 0.7291,
+      "loss": 0.729,
       "step": 640
     },
     {
       "epoch": 0.9548482605477424,
-      "grad_norm": 0.5227834824691131,
+      "grad_norm": 0.5220503199938477,
       "learning_rate": 1.2029904969404482e-07,
       "loss": 0.7126,
       "step": 645
     },
     {
       "epoch": 0.9622501850481125,
-      "grad_norm": 0.59929257188385,
+      "grad_norm": 0.6027507206871425,
       "learning_rate": 8.359232296349163e-08,
-      "loss": 0.7164,
+      "loss": 0.7163,
       "step": 650
     },
     {
       "epoch": 0.9696521095484826,
-      "grad_norm": 0.565007803417137,
+      "grad_norm": 0.5697946889297966,
       "learning_rate": 5.3525969743324356e-08,
       "loss": 0.7321,
       "step": 655
     },
     {
       "epoch": 0.9770540340488527,
-      "grad_norm": 0.5455273265621937,
+      "grad_norm": 0.5452210155995557,
       "learning_rate": 3.012012351554017e-08,
-      "loss": 0.7064,
+      "loss": 0.7063,
       "step": 660
     },
     {
       "epoch": 0.9844559585492227,
-      "grad_norm": 0.5811872933851167,
+      "grad_norm": 0.5740830386758149,
       "learning_rate": 1.3390457653639221e-08,
-      "loss": 0.7352,
+      "loss": 0.7354,
       "step": 665
     },
     {
       "epoch": 0.9918578830495929,
-      "grad_norm": 0.6397544103158143,
+      "grad_norm": 0.6403583337563233,
       "learning_rate": 3.3481749271768726e-09,
       "loss": 0.7462,
       "step": 670
     },
     {
       "epoch": 0.999259807549963,
-      "grad_norm": 0.5723897760097666,
+      "grad_norm": 0.5720756141987368,
       "learning_rate": 0.0,
-      "loss": 0.7157,
+      "loss": 0.7158,
       "step": 675
     },
     {
       "epoch": 0.999259807549963,
       "step": 675,
       "total_flos": 76888336760832.0,
-      "train_loss": 0.7675497867442943,
-      "train_runtime": 4035.0195,
-      "train_samples_per_second": 5.356,
-      "train_steps_per_second": 0.167
+      "train_loss": 0.7675483689484773,
+      "train_runtime": 4012.107,
+      "train_samples_per_second": 5.386,
+      "train_steps_per_second": 0.168
     }
   ],
   "logging_steps": 5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:884369941ee1c4e9dc85f44d8f02f759616d13cf01e3e36bf70b94767b7a0d62
+oid sha256:1c4d617f90564ed82bddf5ce73aeb0f6a8b6c7377ab9cd0833b61f25b6fa234d
 size 7416