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/kwai-intelligent/huggingface/runs/krliapjc) 
+[<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/kwai-intelligent/huggingface/runs/8uax40e0) 
 
 
 This model was trained with SFT.

all_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
     "total_flos": 76916824473600.0,
-    "train_loss": 0.8026438337811351,
-    "train_runtime": 513.2467,
+    "train_loss": 0.8026589176358556,
+    "train_runtime": 513.3168,
     "train_samples": 16610,
-    "train_samples_per_second": 42.105,
+    "train_samples_per_second": 42.099,
     "train_steps_per_second": 0.329
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:abfed090f74d5f377a4ad89f0cc2e8fcdbf4c04662157059d85a2cc9b613796c
+oid sha256:10a29e81ff1f9513bc4ddd1052cf1b76ec4fdb4152141a9e23e8a6bbf40051e0
 size 3087467144

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
     "total_flos": 76916824473600.0,
-    "train_loss": 0.8026438337811351,
-    "train_runtime": 513.2467,
+    "train_loss": 0.8026589176358556,
+    "train_runtime": 513.3168,
     "train_samples": 16610,
-    "train_samples_per_second": 42.105,
+    "train_samples_per_second": 42.099,
     "train_steps_per_second": 0.329
 }

trainer_state.json

@@ -10,239 +10,239 @@
   "log_history": [
     {
       "epoch": 0.029585798816568046,
-      "grad_norm": 2.3435169147496064,
+      "grad_norm": 2.345960428445246,
       "learning_rate": 5.882352941176471e-06,
       "loss": 1.0993,
       "step": 5
     },
     {
       "epoch": 0.05917159763313609,
-      "grad_norm": 1.6335495301840217,
+      "grad_norm": 1.633297171900171,
       "learning_rate": 1.1764705882352942e-05,
       "loss": 1.0402,
       "step": 10
     },
     {
       "epoch": 0.08875739644970414,
-      "grad_norm": 0.8866958174949716,
+      "grad_norm": 0.8867646581470677,
       "learning_rate": 1.7647058823529414e-05,
       "loss": 0.9533,
       "step": 15
     },
     {
       "epoch": 0.11834319526627218,
-      "grad_norm": 0.6055613212513322,
+      "grad_norm": 0.6040455522283187,
       "learning_rate": 1.9980782984658682e-05,
       "loss": 0.8923,
       "step": 20
     },
     {
       "epoch": 0.14792899408284024,
-      "grad_norm": 0.5406027515772251,
+      "grad_norm": 0.5392420757662921,
       "learning_rate": 1.9863613034027224e-05,
       "loss": 0.8552,
       "step": 25
     },
     {
       "epoch": 0.17751479289940827,
-      "grad_norm": 0.4254004034742397,
+      "grad_norm": 0.4256913969311469,
       "learning_rate": 1.9641197940012136e-05,
       "loss": 0.8283,
       "step": 30
     },
     {
       "epoch": 0.20710059171597633,
-      "grad_norm": 0.3760526547765697,
+      "grad_norm": 0.37685830827919437,
       "learning_rate": 1.9315910880512792e-05,
       "loss": 0.8229,
       "step": 35
     },
     {
       "epoch": 0.23668639053254437,
-      "grad_norm": 0.399426488535239,
+      "grad_norm": 0.4015521534729439,
       "learning_rate": 1.8891222681391853e-05,
       "loss": 0.8225,
       "step": 40
     },
     {
       "epoch": 0.26627218934911245,
-      "grad_norm": 0.3434313759700535,
+      "grad_norm": 0.34328981072821885,
       "learning_rate": 1.8371664782625287e-05,
       "loss": 0.8074,
       "step": 45
     },
     {
       "epoch": 0.2958579881656805,
-      "grad_norm": 0.33389109621919627,
+      "grad_norm": 0.33367336634248834,
       "learning_rate": 1.7762780887657576e-05,
       "loss": 0.7978,
       "step": 50
     },
     {
       "epoch": 0.3254437869822485,
-      "grad_norm": 0.3657842480065054,
+      "grad_norm": 0.36636527674416564,
       "learning_rate": 1.7071067811865477e-05,
       "loss": 0.7877,
       "step": 55
     },
     {
       "epoch": 0.35502958579881655,
-      "grad_norm": 0.368950139280336,
+      "grad_norm": 0.4076527323191576,
       "learning_rate": 1.6303906161279554e-05,
       "loss": 0.7981,
       "step": 60
     },
     {
       "epoch": 0.38461538461538464,
-      "grad_norm": 0.34413004217596443,
+      "grad_norm": 0.3430395445047715,
       "learning_rate": 1.5469481581224274e-05,
       "loss": 0.7721,
       "step": 65
     },
     {
       "epoch": 0.41420118343195267,
-      "grad_norm": 0.34731234215972995,
+      "grad_norm": 0.3443470062719444,
       "learning_rate": 1.4576697415156818e-05,
-      "loss": 0.7741,
+      "loss": 0.774,
       "step": 70
     },
     {
       "epoch": 0.4437869822485207,
-      "grad_norm": 0.335771953006324,
+      "grad_norm": 0.33825635000382087,
       "learning_rate": 1.3635079705638298e-05,
       "loss": 0.7852,
       "step": 75
     },
     {
       "epoch": 0.47337278106508873,
-      "grad_norm": 0.35153493118429924,
+      "grad_norm": 0.3516086889229178,
       "learning_rate": 1.2654675551080724e-05,
       "loss": 0.7666,
       "step": 80
     },
     {
       "epoch": 0.5029585798816568,
-      "grad_norm": 0.36146747609137597,
+      "grad_norm": 0.3597883894381192,
       "learning_rate": 1.164594590280734e-05,
       "loss": 0.7706,
       "step": 85
     },
     {
       "epoch": 0.5325443786982249,
-      "grad_norm": 0.3587139051889533,
+      "grad_norm": 0.3597678201693243,
       "learning_rate": 1.0619653946285948e-05,
       "loss": 0.77,
       "step": 90
     },
     {
       "epoch": 0.5621301775147929,
-      "grad_norm": 0.372613651402319,
+      "grad_norm": 0.36574059635601586,
       "learning_rate": 9.586750257511868e-06,
       "loss": 0.7771,
       "step": 95
     },
     {
       "epoch": 0.591715976331361,
-      "grad_norm": 0.33327026352448047,
+      "grad_norm": 0.33643101593032865,
       "learning_rate": 8.558255959926533e-06,
       "loss": 0.753,
       "step": 100
     },
     {
       "epoch": 0.591715976331361,
-      "eval_loss": 0.7879945039749146,
-      "eval_runtime": 0.8233,
-      "eval_samples_per_second": 155.47,
-      "eval_steps_per_second": 4.858,
+      "eval_loss": 0.7880026698112488,
+      "eval_runtime": 0.822,
+      "eval_samples_per_second": 155.708,
+      "eval_steps_per_second": 4.866,
       "step": 100
     },
     {
       "epoch": 0.621301775147929,
-      "grad_norm": 0.334920360429092,
+      "grad_norm": 0.33458441790355503,
       "learning_rate": 7.545145128592009e-06,
       "loss": 0.7569,
       "step": 105
     },
     {
       "epoch": 0.650887573964497,
-      "grad_norm": 0.34415232200036044,
+      "grad_norm": 0.34592641470618773,
       "learning_rate": 6.558227696373617e-06,
       "loss": 0.7681,
       "step": 110
     },
     {
       "epoch": 0.6804733727810651,
-      "grad_norm": 0.30933479401137964,
+      "grad_norm": 0.30769886773546357,
       "learning_rate": 5.608034111526298e-06,
       "loss": 0.7623,
       "step": 115
     },
     {
       "epoch": 0.7100591715976331,
-      "grad_norm": 0.29790825625309886,
+      "grad_norm": 0.2979314843836126,
       "learning_rate": 4.704702977392914e-06,
       "loss": 0.7513,
       "step": 120
     },
     {
       "epoch": 0.7396449704142012,
-      "grad_norm": 0.342701906369065,
+      "grad_norm": 0.3445603911116372,
       "learning_rate": 3.857872873103322e-06,
-      "loss": 0.7537,
+      "loss": 0.7538,
       "step": 125
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 0.31139159077861756,
+      "grad_norm": 0.3127072734721163,
       "learning_rate": 3.0765795095517026e-06,
       "loss": 0.7555,
       "step": 130
     },
     {
       "epoch": 0.7988165680473372,
-      "grad_norm": 0.32219403439145694,
+      "grad_norm": 0.32268011378307926,
       "learning_rate": 2.369159318001937e-06,
       "loss": 0.7584,
       "step": 135
     },
     {
       "epoch": 0.8284023668639053,
-      "grad_norm": 0.29872698957110416,
+      "grad_norm": 0.29889999351128643,
       "learning_rate": 1.743160500034443e-06,
       "loss": 0.7498,
       "step": 140
     },
     {
       "epoch": 0.8579881656804734,
-      "grad_norm": 0.3108030175496591,
+      "grad_norm": 0.3114879641169587,
       "learning_rate": 1.2052624879351105e-06,
       "loss": 0.7566,
       "step": 145
     },
     {
       "epoch": 0.8875739644970414,
-      "grad_norm": 0.2947968949117839,
+      "grad_norm": 0.2950379770777633,
       "learning_rate": 7.612046748871327e-07,
-      "loss": 0.7665,
+      "loss": 0.7666,
       "step": 150
     },
     {
       "epoch": 0.9171597633136095,
-      "grad_norm": 0.29717837362127947,
+      "grad_norm": 0.2981392642716918,
       "learning_rate": 4.1572517541747294e-07,
       "loss": 0.7613,
       "step": 155
     },
     {
       "epoch": 0.9467455621301775,
-      "grad_norm": 0.28592535456716145,
+      "grad_norm": 0.2852481942843539,
       "learning_rate": 1.7251026952640583e-07,
       "loss": 0.7607,
       "step": 160
     },
     {
       "epoch": 0.9763313609467456,
-      "grad_norm": 0.27622399826112193,
+      "grad_norm": 0.2764284922215319,
       "learning_rate": 3.4155069933301535e-08,
       "loss": 0.7445,
       "step": 165
@@ -251,9 +251,9 @@
       "epoch": 1.0,
       "step": 169,
       "total_flos": 76916824473600.0,
-      "train_loss": 0.8026438337811351,
-      "train_runtime": 513.2467,
-      "train_samples_per_second": 42.105,
+      "train_loss": 0.8026589176358556,
+      "train_runtime": 513.3168,
+      "train_samples_per_second": 42.099,
       "train_steps_per_second": 0.329
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c7aa35a75b0944de1940977769e0826c8da44b75f7262e4a2554bd4c6486e91f
+oid sha256:8d6e9e49805a2fcfa2aeb1bcc3ee7c2711d94477f5d0b3e155b69c2f861ce97f
 size 7416