Add files using upload-large-folder tool
Browse files- .gitattributes +1 -0
- .idea/.gitignore +8 -0
- .idea/inspectionProfiles/Project_Default.xml +6 -0
- .idea/inspectionProfiles/profiles_settings.xml +6 -0
- .idea/mellum-kotlin-sft.iml +8 -0
- .idea/misc.xml +7 -0
- .idea/modules.xml +8 -0
- .idea/workspace.xml +51 -0
- README.md +52 -3
- mellum-4b-sft-kotlin.Q8_0.gguf +3 -0
.gitattributes
CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
mellum-4b-sft-kotlin.Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
|
.idea/.gitignore
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# Default ignored files
|
2 |
+
/shelf/
|
3 |
+
/workspace.xml
|
4 |
+
# Editor-based HTTP Client requests
|
5 |
+
/httpRequests/
|
6 |
+
# Datasource local storage ignored files
|
7 |
+
/dataSources/
|
8 |
+
/dataSources.local.xml
|
.idea/inspectionProfiles/Project_Default.xml
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
<component name="InspectionProjectProfileManager">
|
2 |
+
<profile version="1.0">
|
3 |
+
<option name="myName" value="Project Default" />
|
4 |
+
<inspection_tool class="PyPackageRequirementsInspection" enabled="false" level="WARNING" enabled_by_default="false" />
|
5 |
+
</profile>
|
6 |
+
</component>
|
.idea/inspectionProfiles/profiles_settings.xml
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
<component name="InspectionProjectProfileManager">
|
2 |
+
<settings>
|
3 |
+
<option name="USE_PROJECT_PROFILE" value="false" />
|
4 |
+
<version value="1.0" />
|
5 |
+
</settings>
|
6 |
+
</component>
|
.idea/mellum-kotlin-sft.iml
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
<?xml version="1.0" encoding="UTF-8"?>
|
2 |
+
<module type="PYTHON_MODULE" version="4">
|
3 |
+
<component name="NewModuleRootManager">
|
4 |
+
<content url="file://$MODULE_DIR$" />
|
5 |
+
<orderEntry type="jdk" jdkName="Python 3.10 virtualenv at ~/Dev/COLING2025/code/evalica/.venv" jdkType="Python SDK" />
|
6 |
+
<orderEntry type="sourceFolder" forTests="false" />
|
7 |
+
</component>
|
8 |
+
</module>
|
.idea/misc.xml
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
<?xml version="1.0" encoding="UTF-8"?>
|
2 |
+
<project version="4">
|
3 |
+
<component name="Black">
|
4 |
+
<option name="sdkName" value="Python 3.10 virtualenv at ~/Dev/COLING2025/code/evalica/.venv" />
|
5 |
+
</component>
|
6 |
+
<component name="ProjectRootManager" version="2" project-jdk-name="Python 3.10 virtualenv at ~/Dev/COLING2025/code/evalica/.venv" project-jdk-type="Python SDK" />
|
7 |
+
</project>
|
.idea/modules.xml
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
<?xml version="1.0" encoding="UTF-8"?>
|
2 |
+
<project version="4">
|
3 |
+
<component name="ProjectModuleManager">
|
4 |
+
<modules>
|
5 |
+
<module fileurl="file://$PROJECT_DIR$/.idea/mellum-kotlin-sft.iml" filepath="$PROJECT_DIR$/.idea/mellum-kotlin-sft.iml" />
|
6 |
+
</modules>
|
7 |
+
</component>
|
8 |
+
</project>
|
.idea/workspace.xml
ADDED
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
<?xml version="1.0" encoding="UTF-8"?>
|
2 |
+
<project version="4">
|
3 |
+
<component name="ChangeListManager">
|
4 |
+
<list default="true" id="87b26083-bdbe-4b86-a4b7-7dbe725f6576" name="Changes" comment="" />
|
5 |
+
<option name="SHOW_DIALOG" value="false" />
|
6 |
+
<option name="HIGHLIGHT_CONFLICTS" value="true" />
|
7 |
+
<option name="HIGHLIGHT_NON_ACTIVE_CHANGELIST" value="false" />
|
8 |
+
<option name="LAST_RESOLUTION" value="IGNORE" />
|
9 |
+
</component>
|
10 |
+
<component name="ProjectColorInfo"><![CDATA[{
|
11 |
+
"associatedIndex": 1
|
12 |
+
}]]></component>
|
13 |
+
<component name="ProjectId" id="2zPejhdNra6cXP7DuLMYAYWSgVX" />
|
14 |
+
<component name="ProjectViewState">
|
15 |
+
<option name="hideEmptyMiddlePackages" value="true" />
|
16 |
+
<option name="showLibraryContents" value="true" />
|
17 |
+
</component>
|
18 |
+
<component name="PropertiesComponent"><![CDATA[{
|
19 |
+
"keyToString": {
|
20 |
+
"ModuleVcsDetector.initialDetectionPerformed": "true",
|
21 |
+
"RunOnceActivity.ShowReadmeOnStart": "true",
|
22 |
+
"junie.onboarding.icon.badge.shown": "true",
|
23 |
+
"last_opened_file_path": "/Users/Uladzislau.Sazanovich/Downloads/mellum-kotlin-sft",
|
24 |
+
"nodejs_package_manager_path": "npm",
|
25 |
+
"settings.editor.selected.configurable": "settings.space",
|
26 |
+
"vue.rearranger.settings.migration": "true"
|
27 |
+
}
|
28 |
+
}]]></component>
|
29 |
+
<component name="SharedIndexes">
|
30 |
+
<attachedChunks>
|
31 |
+
<set>
|
32 |
+
<option value="bundled-js-predefined-d6986cc7102b-09060db00ec0-JavaScript-PY-251.26927.74" />
|
33 |
+
<option value="bundled-python-sdk-657d8234b839-64d779b69b7a-com.jetbrains.pycharm.pro.sharedIndexes.bundled-PY-251.26927.74" />
|
34 |
+
</set>
|
35 |
+
</attachedChunks>
|
36 |
+
</component>
|
37 |
+
<component name="TaskManager">
|
38 |
+
<task active="true" id="Default" summary="Default task">
|
39 |
+
<changelist id="87b26083-bdbe-4b86-a4b7-7dbe725f6576" name="Changes" comment="" />
|
40 |
+
<created>1751637272035</created>
|
41 |
+
<option name="number" value="Default" />
|
42 |
+
<option name="presentableId" value="Default" />
|
43 |
+
<updated>1751637272035</updated>
|
44 |
+
<workItem from="1751637273082" duration="138000" />
|
45 |
+
</task>
|
46 |
+
<servers />
|
47 |
+
</component>
|
48 |
+
<component name="TypeScriptGeneratedFilesManager">
|
49 |
+
<option name="version" value="3" />
|
50 |
+
</component>
|
51 |
+
</project>
|
README.md
CHANGED
@@ -1,3 +1,52 @@
|
|
1 |
-
---
|
2 |
-
license: apache-2.0
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
datasets:
|
4 |
+
- bigcode/the-stack
|
5 |
+
- bigcode/the-stack-v2
|
6 |
+
- bigcode/starcoderdata
|
7 |
+
- bigcode/commitpack
|
8 |
+
library_name: transformers
|
9 |
+
tags:
|
10 |
+
- code
|
11 |
+
base_model:
|
12 |
+
- JetBrains/Mellum-4b-sft-kotlin
|
13 |
+
---
|
14 |
+
|
15 |
+
# Model Description
|
16 |
+
Mellum-4b-sft-kotlin is a fine-tuned version of JetBrains' first open-source large language model (LLM) optimized for code-related tasks.
|
17 |
+
|
18 |
+
Pre-trained on over 4 trillion tokens with a context window of 8192 tokens across multiple programming languages, and then fine-tuned, Mellum-4b-sft-python is tailored specifically for code completion in Python.
|
19 |
+
The model follows a LLaMA-style architecture with 4 billion parameters, making it efficient for both cloud inference (e.g., via vLLM) and local deployment (e.g., using llama.cpp or Ollama).
|
20 |
+
|
21 |
+
Mellum was trained using Automatic Mixed Precision (AMP) with bf16 precision.
|
22 |
+
The uploaded version on Hugging Face retains the bf16 format for public use.
|
23 |
+
|
24 |
+
Designed for integration into professional developer tooling (e.g., intelligent code suggestions in IDEs), AI-powered coding assistants, and research on code understanding and generation, Mellum is also well-suited for educational applications and fine-tuning experiments.
|
25 |
+
|
26 |
+
# Limitations
|
27 |
+
- Biases: May reflect biases present in public codebases. For example it will likely produce code which is similar in style to the open-source repositories.
|
28 |
+
- Security: Code suggestions should not be assumed to be secure or free of vulnerabilities.
|
29 |
+
- Format: This model is suitable mostly for FIM Completion objective with context's files.
|
30 |
+
|
31 |
+
# Sample Usage
|
32 |
+
Here are examples of how to run and sample from the model.
|
33 |
+
|
34 |
+
## Fill-in-the-middle example
|
35 |
+
|
36 |
+
```bash
|
37 |
+
llama-cli -m mellum-4b-sft-kotlin.Q8_0.gguf --temp 0 -p $'<filename>Utils.kt\npackage utils\n\nfun multiply(x: Int, y: Int): Int {\n return x * y\n}\n\n<filename>Config.kt\npackage config\n\nobject Config {\n const val DEBUG = true\n const val MAX_VALUE = 100\n}\n\n<filename>Example.kt\n<fim_suffix>\nfun main() {\n val result = calculateSum(5, 10)\n println(result)\n}\n<fim_prefix>fun calculateSum(a: Int, b: Int): Int {\n<fim_middle>'
|
38 |
+
```
|
39 |
+
|
40 |
+
# Citation
|
41 |
+
If you use this model, please cite:
|
42 |
+
|
43 |
+
```bibtex
|
44 |
+
@misc{Mellum-4b-base,
|
45 |
+
title = {Mellum-4b-base},
|
46 |
+
author = {Pavlichenko, Nikita and Nazarov, Iurii and Dolgov, Ivan and Garanina, Ekaterina and Lasocki, Karol and Reshetnikova, Julia and Boitsov, Sergei and Bondyrev, Ivan and Karaeva, Dariia and Sheptyakov, Maksim and Ustalov, Dmitry and Mukhin, Artem and Proshev, Semyon and Abramov, Nikita and Kolomyttseva, Olga and Lysaniuk, Kseniia and Zavidnyi, Ilia and Semenkin, Anton and Tankov, Vladislav and Sazanovich, Uladzislau},
|
47 |
+
year = {2025},
|
48 |
+
}
|
49 |
+
```
|
50 |
+
|
51 |
+
# Contact
|
52 |
+
For questions, collaborations and requests reach us out via [email protected]
|
mellum-4b-sft-kotlin.Q8_0.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f2e10ba80657c353421a49b4e818895ded6c662e4dcd5ffed0edee823bdc81f4
|
3 |
+
size 4274519360
|