Upload folder using huggingface_hub

README.md

@@ -4,7 +4,7 @@ license_name: katanemo-research
 license_link: >-
   https://huggingface.co/katanemolabs/Arch-Function-1.5B/blob/main/LICENSE
 base_model:
-- Qwen/Qwen2.5-1.5B-Instruct
+- Qwen/Qwen2.5-Coder-1.5B-Instruct
 language:
 - en
 pipeline_tag: text-generation
@@ -84,6 +84,17 @@ We evaluate Katanemo Arch-Function series on the [Berkeley Function-Calling Lead
     <td>63.41%</td>
     <td>82.93%</td>
   </tr>
+  <tr style="text-align: center; vertical-align: middle; font-weight: bold;">
+    <td> </td>
+    <td>Arch-Function-7B</td>
+    <td>59.62%</td>
+    <td>86.83%</td>
+    <td>88.07%</td>
+    <td>71.57%</td>
+    <td>21.00%</td>
+    <td>95.12%</td>
+    <td>73.63%</td>
+  </tr>
   <tr style="text-align: center; vertical-align: middle;">
     <td>6</td>
     <td>o1-preview-2024-09-12 (Prompt)</td>
@@ -95,17 +106,6 @@ We evaluate Katanemo Arch-Function series on the [Berkeley Function-Calling Lead
     <td>73.17%</td>
     <td>74.60%</td>
   </tr>
-  <tr style="text-align: center; vertical-align: middle; font-weight: bold;">
-    <td> </td>
-    <td>Arch-Function-7B</td>
-    <td>58.44%</td>
-    <td>85.58%</td>
-    <td>88.14%</td>
-    <td>69.08%</td>
-    <td>20.50%</td>
-    <td>92.68%</td>
-    <td>74.05%</td>
-  </tr>
   <tr style="text-align: center; vertical-align: middle; ">
     <td>9</td>
     <td>Gemini-1.5-Flash-002 (Prompt)</td>
@@ -117,6 +117,17 @@ We evaluate Katanemo Arch-Function series on the [Berkeley Function-Calling Lead
     <td>85.37%</td>
     <td>78.54%</td>
   </tr>
+  <tr style="text-align: center; vertical-align: middle; font-weight: bold;">
+    <td> </td>
+    <td>Arch-Function-3B</td>
+    <td>57.69%</td>
+    <td>85.19%</td>
+    <td>86.18%</td>
+    <td>71.21%</td>
+    <td>17.50%</td>
+    <td>90.24%</td>
+    <td>72.88%</td>
+  </tr>
   <tr style="text-align: center; vertical-align: middle; ">
     <td>12</td>
     <td>Claude-3.5-Sonnet-20240620 (FC)</td>
@@ -139,30 +150,17 @@ We evaluate Katanemo Arch-Function series on the [Berkeley Function-Calling Lead
     <td>75.61%</td>
     <td>49.44%</td>
   </tr>
-  <tr style="text-align: center; vertical-align: middle; font-weight: bold;">
-    <td> </td>
-    <td>Arch-Function-3B</td>
-    <td>56.57%</td>
-    <td>83.62%</td>
-    <td>85.36%</td>
-    <td>66.90%</td>
-    <td>19.50%</td>
-    <td>97.56%</td>
-    <td>70.99%</td>
-  </tr>
-  </tr>
   <tr style="text-align: center; vertical-align: middle; font-weight: bold;">
     <td> </td>
     <td>Arch-Function-1.5B</td>
-    <td>54.52%</td>
-    <td>80.31%</td>
-    <td>82.04%</td>
-    <td>66.19%</td>
-    <td>17.25%</td>
-    <td>97.56%</td>
-    <td>69.95%</td>
+    <td>56.20%</td>
+    <td>84.40%</td>
+    <td>83.96%</td>
+    <td>69.36%</td>
+    <td>15.88%</td>
+    <td>87.80%</td>
+    <td>74.39%</td>
   </tr>
-
   <tr style="text-align: center; vertical-align: middle; ">
     <td>21</td>
     <td>Llama-3.1-70B-Instruct (Prompt)</td>

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "Qwen/Qwen2.5-1.5B-Instruct",
+  "_name_or_path": "Qwen/Qwen2.5-Coder-1.5B-Instruct",
   "architectures": [
     "Qwen2ForCausalLM"
   ],
@@ -11,7 +11,7 @@
   "initializer_range": 0.02,
   "intermediate_size": 8960,
   "max_position_embeddings": 32768,
-  "max_window_layers": 21,
+  "max_window_layers": 28,
   "model_type": "qwen2",
   "num_attention_heads": 12,
   "num_hidden_layers": 28,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:277363bbda7db5c37e1f533b74b621a3ff362764486b8001ffdf868e6e36364e
+oid sha256:3440dc10971cff500578021d8f87b978786b39379963fb4056bab0fb8d085630
 size 3087467144

tokenizer_config.json

@@ -199,9 +199,9 @@
   "clean_up_tokenization_spaces": false,
   "eos_token": "<|im_end|>",
   "errors": "replace",
-  "model_max_length": 131072,
+  "model_max_length": 32768,
   "pad_token": "<|endoftext|>",
   "split_special_tokens": false,
   "tokenizer_class": "Qwen2Tokenizer",
   "unk_token": null
-}
+}