Update README.md
Browse files
README.md
CHANGED
|
@@ -5,7 +5,7 @@ datasets:
|
|
| 5 |
- bigcode/the-stack-v2
|
| 6 |
- bigcode/starcoderdata
|
| 7 |
- bigcode/commitpack
|
| 8 |
-
library_name:
|
| 9 |
tags:
|
| 10 |
- code
|
| 11 |
model-index:
|
|
@@ -33,7 +33,7 @@ model-index:
|
|
| 33 |
metrics:
|
| 34 |
- name: EM
|
| 35 |
type: exact_match
|
| 36 |
-
value: 0.
|
| 37 |
verified: false
|
| 38 |
- task:
|
| 39 |
type: text-generation
|
|
@@ -73,7 +73,7 @@ model-index:
|
|
| 73 |
metrics:
|
| 74 |
- name: EM
|
| 75 |
type: exact_match
|
| 76 |
-
value: 0.
|
| 77 |
verified: false
|
| 78 |
- task:
|
| 79 |
type: text-generation
|
|
@@ -117,7 +117,7 @@ model-index:
|
|
| 117 |
metrics:
|
| 118 |
- name: EM
|
| 119 |
type: exact_match
|
| 120 |
-
value: 0.
|
| 121 |
verified: false
|
| 122 |
- task:
|
| 123 |
type: text-generation
|
|
@@ -157,7 +157,7 @@ model-index:
|
|
| 157 |
metrics:
|
| 158 |
- name: pass@1
|
| 159 |
type: pass@1
|
| 160 |
-
value: 0.
|
| 161 |
verified: false
|
| 162 |
- task:
|
| 163 |
type: text-generation
|
|
@@ -209,8 +209,24 @@ model-index:
|
|
| 209 |
type: pass@1
|
| 210 |
value: 0.2969
|
| 211 |
verified: false
|
|
|
|
|
|
|
| 212 |
---
|
| 213 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 214 |
# Model Description
|
| 215 |
Mellum-4b-base is JetBrains' first open-source large language model (LLM) optimized for code-related tasks.
|
| 216 |
|
|
|
|
| 5 |
- bigcode/the-stack-v2
|
| 6 |
- bigcode/starcoderdata
|
| 7 |
- bigcode/commitpack
|
| 8 |
+
library_name: exllamav2
|
| 9 |
tags:
|
| 10 |
- code
|
| 11 |
model-index:
|
|
|
|
| 33 |
metrics:
|
| 34 |
- name: EM
|
| 35 |
type: exact_match
|
| 36 |
+
value: 0.282
|
| 37 |
verified: false
|
| 38 |
- task:
|
| 39 |
type: text-generation
|
|
|
|
| 73 |
metrics:
|
| 74 |
- name: EM
|
| 75 |
type: exact_match
|
| 76 |
+
value: 0.211
|
| 77 |
verified: false
|
| 78 |
- task:
|
| 79 |
type: text-generation
|
|
|
|
| 117 |
metrics:
|
| 118 |
- name: EM
|
| 119 |
type: exact_match
|
| 120 |
+
value: 0.291
|
| 121 |
verified: false
|
| 122 |
- task:
|
| 123 |
type: text-generation
|
|
|
|
| 157 |
metrics:
|
| 158 |
- name: pass@1
|
| 159 |
type: pass@1
|
| 160 |
+
value: 0.253
|
| 161 |
verified: false
|
| 162 |
- task:
|
| 163 |
type: text-generation
|
|
|
|
| 209 |
type: pass@1
|
| 210 |
value: 0.2969
|
| 211 |
verified: false
|
| 212 |
+
base_model:
|
| 213 |
+
- JetBrains/Mellum-4b-base
|
| 214 |
---
|
| 215 |
+
# Mellum-4b-base-exl2
|
| 216 |
+
Original model: [Mellum-4b-base](https://huggingface.co/JetBrains/Mellum-4b-base) by [JetBrains](https://huggingface.co/JetBrains)
|
| 217 |
+
## Quants
|
| 218 |
+
[4bpw h6 (main)](https://huggingface.co/cgus/Mellum-4b-base-exl2/tree/main)
|
| 219 |
+
[4.5bpw h6](https://huggingface.co/cgus/Mellum-4b-base-exl2/tree/4.5bpw-h6)
|
| 220 |
+
[5bpw h6](https://huggingface.co/cgus/Mellum-4b-base-exl2/tree/5bpw-h6)
|
| 221 |
+
[6bpw h6](https://huggingface.co/cgus/Mellum-4b-base-exl2/tree/6bpw-h6)
|
| 222 |
+
[8bpw h8](https://huggingface.co/cgus/Mellum-4b-base-exl2/tree/8bpw-h8)
|
| 223 |
+
## Quantization notes
|
| 224 |
+
Made with Exllamav2 0.2.9 dev branch with default dataset. But it should work with older versions too.
|
| 225 |
+
It can be used with RTX GPUs on Windows or RTX/ROCm on Linux.
|
| 226 |
+
Usable with TabbyAPI or full version of Text-Generation-WebUI (not portable).
|
| 227 |
+
It's a text-completion model meant to be used for code autocompletion with dev tools, e.g. with VSCode + Continue extension.
|
| 228 |
+
On TabbyAPI v1/chat/completions endpoint will be disabled, it's normal as it lacks instruction template and isn't meant to be used in this manner anyway.
|
| 229 |
+
# Original model card
|
| 230 |
# Model Description
|
| 231 |
Mellum-4b-base is JetBrains' first open-source large language model (LLM) optimized for code-related tasks.
|
| 232 |
|