Upload folder using huggingface_hub

.gitattributes

@@ -34,3 +34,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 tokenizer.json filter=lfs diff=lfs merge=lfs -text
+onnx/model.onnx_data filter=lfs diff=lfs merge=lfs -text
+onnx/model_fp16.onnx_data filter=lfs diff=lfs merge=lfs -text

config.json

@@ -35,6 +35,12 @@
   "tie_word_embeddings": true,
   "torch_dtype": "float32",
   "transformers_version": "4.47.1",
+  "transformers.js_config": {
+    "kv_cache_dtype": {
+      "q4f16": "float16",
+      "fp16": "float16"
+    }
+  },
   "use_cache": true,
   "vocab_size": 128256
-}
+}

onnx/model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b1860a6bda206918f22bc4700a803196de792979a961a649d684b35a091d5b9
+size 83590

onnx/model.onnx_data

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec00849ff44be3e1cfebae18aaaf06c9126da53e5daa1de285e740aab31c2716
+size 4976812032

onnx/model_bnb4.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9a9db6cd7d88964dfa5fb7e89be9ae1b5371e9f6058fdd0998ca83fa52e6d98
+size 1631947377

onnx/model_fp16.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba1237166f263f7e28df2d91f9e0ec6e635df1e2b63f88c172409efb3081aff4
+size 398674945

onnx/model_fp16.onnx_data

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa6fe8cb21aa7b2a7f53a0f0b855c015a24e5ddd5fe1e60ac39c509b84b67572
+size 2089811968

onnx/model_int8.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27f15664651c68f7ab03396778198743bcb0394e34933f6fe499a8a916909ffb
+size 1269746162

onnx/model_q4.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d6ab9eee4b33ae0d47ee4c1a8bed6df5b0fb9cd245a234b54071c424222f847
+size 1692764177

onnx/model_q4f16.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc8b4d8c2ea5eba115f6ddf99badb7b3690a7b98566fd161259a96074eb3a187
+size 1089697758

onnx/model_quantized.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:950a03f7cdfb3de8da8c1222ba262f2e00d309588e43765634d1b8ea4b19af78
+size 1269746207

onnx/model_uint8.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:950a03f7cdfb3de8da8c1222ba262f2e00d309588e43765634d1b8ea4b19af78
+size 1269746207

special_tokens_map.json

@@ -13,5 +13,11 @@
     "rstrip": false,
     "single_word": false
   },
-  "pad_token": "<|finetune_right_pad_id|>"
+  "pad_token": {
+    "content": "<|finetune_right_pad_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
 }

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb9b85b70a1b34e497c3476c315b656613e7ed0b2db8c6cf62ec83d381c98737
-size 17209986
+oid sha256:2c02b5c0c3fbb4b776c5ad9ede41ef8421ce0a8be84fa0bf5812fbbc80557e21
+size 11574684

tokenizer_config.json

@@ -2054,11 +2054,15 @@
   "clean_up_tokenization_spaces": true,
   "eos_token": "<|eot_id|>",
   "extra_special_tokens": {},
+  "max_length": 32000,
   "model_input_names": [
     "input_ids",
     "attention_mask"
   ],
   "model_max_length": 131072,
   "pad_token": "<|finetune_right_pad_id|>",
-  "tokenizer_class": "PreTrainedTokenizerFast"
+  "stride": 0,
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "truncation_side": "right",
+  "truncation_strategy": "longest_first"
 }