commit c188cd5fdd4b4ff1c119aabe00a0e9950abd7fc0 Author: ModelHub XC Date: Mon Jun 22 16:57:13 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: QuantFactory/CodeLlama-7B-KStack-GGUF Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..a9f6ae5 --- /dev/null +++ b/.gitattributes @@ -0,0 +1,49 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text +CodeLlama-7B-KStack.Q8_0.gguf filter=lfs diff=lfs merge=lfs -text +CodeLlama-7B-KStack.Q6_K.gguf filter=lfs diff=lfs merge=lfs -text +CodeLlama-7B-KStack.Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text +CodeLlama-7B-KStack.Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text +CodeLlama-7B-KStack.Q5_0.gguf filter=lfs diff=lfs merge=lfs -text +CodeLlama-7B-KStack.Q5_1.gguf filter=lfs diff=lfs merge=lfs -text +CodeLlama-7B-KStack.Q4_1.gguf filter=lfs diff=lfs merge=lfs -text +CodeLlama-7B-KStack.Q4_0.gguf filter=lfs diff=lfs merge=lfs -text +CodeLlama-7B-KStack.Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text +CodeLlama-7B-KStack.Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text +CodeLlama-7B-KStack.Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text +CodeLlama-7B-KStack.Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text +CodeLlama-7B-KStack.Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text +CodeLlama-7B-KStack.Q2_K.gguf filter=lfs diff=lfs merge=lfs -text diff --git a/CodeLlama-7B-KStack.Q2_K.gguf b/CodeLlama-7B-KStack.Q2_K.gguf new file mode 100644 index 0000000..7e9c3c2 --- /dev/null +++ b/CodeLlama-7B-KStack.Q2_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8bfa07dc05bad273070d206de294e6749286798e9fef0408adfcc84efd36003a +size 2532939776 diff --git a/CodeLlama-7B-KStack.Q3_K_L.gguf b/CodeLlama-7B-KStack.Q3_K_L.gguf new file mode 100644 index 0000000..888c13f --- /dev/null +++ b/CodeLlama-7B-KStack.Q3_K_L.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ec5e0ee882177633e74a774ec48d685d562564154fcfe5786624769f86bfa871 +size 3597193728 diff --git a/CodeLlama-7B-KStack.Q3_K_M.gguf b/CodeLlama-7B-KStack.Q3_K_M.gguf new file mode 100644 index 0000000..fb99c41 --- /dev/null +++ b/CodeLlama-7B-KStack.Q3_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b759db68d0885add4f4b6620485c3628a512d5cc60e9ff8f5777843d74380b48 +size 3298087424 diff --git a/CodeLlama-7B-KStack.Q3_K_S.gguf b/CodeLlama-7B-KStack.Q3_K_S.gguf new file mode 100644 index 0000000..2b244bc --- /dev/null +++ b/CodeLlama-7B-KStack.Q3_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:196f2600d42350b3235f38def16e74c0f132c277d83dfd233c351a8e74b6e54c +size 2948387328 diff --git a/CodeLlama-7B-KStack.Q4_0.gguf b/CodeLlama-7B-KStack.Q4_0.gguf new file mode 100644 index 0000000..bcdf8ba --- /dev/null +++ b/CodeLlama-7B-KStack.Q4_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a497169229565793778fe149e1008b10e8de81114db709e544d9ad96b020beaf +size 3825898496 diff --git a/CodeLlama-7B-KStack.Q4_1.gguf b/CodeLlama-7B-KStack.Q4_1.gguf new file mode 100644 index 0000000..736dabd --- /dev/null +++ b/CodeLlama-7B-KStack.Q4_1.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ad1d12b5a0c03d34d4542fdd22a644bff75310f93851d34a3e43aaed955848d6 +size 4238844928 diff --git a/CodeLlama-7B-KStack.Q4_K_M.gguf b/CodeLlama-7B-KStack.Q4_K_M.gguf new file mode 100644 index 0000000..dbac679 --- /dev/null +++ b/CodeLlama-7B-KStack.Q4_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f5876008779c64cfbf8cb932905552c11d50d9ca941fff0375f650f2cc77bcd +size 4081095680 diff --git a/CodeLlama-7B-KStack.Q4_K_S.gguf b/CodeLlama-7B-KStack.Q4_K_S.gguf new file mode 100644 index 0000000..56da6af --- /dev/null +++ b/CodeLlama-7B-KStack.Q4_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c54150638c414d75c40b81c850fdb62978613483c214f5035b0f83e1240f14db +size 3856831488 diff --git a/CodeLlama-7B-KStack.Q5_0.gguf b/CodeLlama-7B-KStack.Q5_0.gguf new file mode 100644 index 0000000..d022b4d --- /dev/null +++ b/CodeLlama-7B-KStack.Q5_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e66ccea2b150d312f8d261781f0f58a51fddfc136fedabc0c112db69d45f37e1 +size 4651791360 diff --git a/CodeLlama-7B-KStack.Q5_1.gguf b/CodeLlama-7B-KStack.Q5_1.gguf new file mode 100644 index 0000000..818c83c --- /dev/null +++ b/CodeLlama-7B-KStack.Q5_1.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e6b3a6a9ee1b246dc5075693916cbf187dc85c6435e15cd8ec7c012ca82d0457 +size 5064737792 diff --git a/CodeLlama-7B-KStack.Q5_K_M.gguf b/CodeLlama-7B-KStack.Q5_K_M.gguf new file mode 100644 index 0000000..85affed --- /dev/null +++ b/CodeLlama-7B-KStack.Q5_K_M.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:77ae4fb2401d828c31f8f788b7a133f3dfdb651503ef4ad2778218b27dd2280d +size 4783256576 diff --git a/CodeLlama-7B-KStack.Q5_K_S.gguf b/CodeLlama-7B-KStack.Q5_K_S.gguf new file mode 100644 index 0000000..b38016c --- /dev/null +++ b/CodeLlama-7B-KStack.Q5_K_S.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f9b927ca37a72f3d2a2eefb303600e395a6263a03d800a584ffb8b0e610241df +size 4651791360 diff --git a/CodeLlama-7B-KStack.Q6_K.gguf b/CodeLlama-7B-KStack.Q6_K.gguf new file mode 100644 index 0000000..1531a2b --- /dev/null +++ b/CodeLlama-7B-KStack.Q6_K.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:67b906a1734e2c6fc7b2b8ec4fa404f1fa2e6d9c7127d00d6180519b869f7e0f +size 5529302528 diff --git a/CodeLlama-7B-KStack.Q8_0.gguf b/CodeLlama-7B-KStack.Q8_0.gguf new file mode 100644 index 0000000..d9e7e8a --- /dev/null +++ b/CodeLlama-7B-KStack.Q8_0.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e836d8ffd145a930fb18f7b797ca58b9f6d170b820767447a6f2fd2d60889b14 +size 7161229824 diff --git a/README.md b/README.md new file mode 100644 index 0000000..b047d98 --- /dev/null +++ b/README.md @@ -0,0 +1,78 @@ +--- +license: apache-2.0 +datasets: +- JetBrains/KStack +results: +- task: + type: text-generation + dataset: + name: MultiPL-HumanEval (Kotlin) + type: openai_humaneval + metrics: + - name: pass@1 + type: pass@1 + value: 29.19 +tags: +- code +library_name: transformers +pipeline_tag: text-generation +base_model: JetBrains/CodeLlama-7B-KStack +--- + +# CodeLlama-7B-KStack-GGUF +This is quantized version of [JetBrains/CodeLlama-7B-KStack](https://huggingface.co/JetBrains/CodeLlama-7B-KStack) created using llama.cpp +# Model description + +This is a repository for the **CodeLlama-7b** model fine-tuned on the [KStack](https://huggingface.co/datasets/JetBrains/KStack) dataset with rule-based filtering, in the *Hugging Face Transformers* format. KStack is the largest collection of permissively licensed Kotlin code, and so the model is fine-tuned to work better with Kotlin code. + +# How to use + +As with the base model, we can use FIM. To do this, the following format must be used: +``` +'
 ' + prefix + '  ' + suffix + ' '
+```
+
+# Training setup
+
+The model was trained on one A100 GPU with following hyperparameters:
+
+|         **Hyperparameter**           |             **Value**              |
+|:---------------------------:|:----------------------------------------:|
+|           `warmup`            |           5%            |
+|        `max_lr`        |          1e-6          |
+|        `num_epochs`        |          1          |
+|        'attention_dropout' |          0.1        |
+|        `scheduler`        |          cosine          |
+|        `total_batch_size`        |          128 (~65K tokens per step)          |
+|        `num_epochs`        |          1          |
+
+More details about fine-tuning can be found in the technical report (coming soon!).
+
+# Fine-tuning data
+
+For tuning the model, we used the [KStack](https://huggingface.co/datasets/JetBrains/KStack) dataset, the largest collection of permissively licensed Kotlin code. To increase the quality of the dataset and filter out outliers, such as homework assignments, we filter out the dataset entries according to the following rules:
+* We filter out files, which belong to low-popular repos (the sum of stars and forks is less than 6)
+* Next, we filter out files, which belong to repos with less than 5 Kotlin files
+* Finally, we remove files which have fewer than 20 SLOC
+
+We clean the content of the remaining dataset entries according to the following rules:
+* We remove all non-ASCII entries
+* We remove all package lines, such as _package kotlinx.coroutines.channels_
+* We remove half of the import lines
+
+We removed half of the imports to avoid potential hallucinations by the model, where it might attempt to import unnecessary libraries. Additionally, packages were removed because this information is only useful at the project level and may introduce additional noise during the learning process.
+
+# Evaluation 
+
+For evaluation, we used the [Kotlin HumanEval](https://huggingface.co/datasets/JetBrains/Kotlin_HumanEval) dataset, which contains all 161 tasks from HumanEval translated into Kotlin by human experts. You can find more details about the pre-processing necessary to obtain our results, including the code for running, on the [datasets's page](https://huggingface.co/datasets/JetBrains/Kotlin_HumanEval).
+
+Here are the results of our evaluation:
+
+|         **Model name**           |             **Kotlin HumanEval Pass Rate**              |
+|:---------------------------:|:----------------------------------------:|
+|           `CodeLlama-7B`            |           26.09            |
+|        `CodeLlama-7B-KStack`        |          **29.19**         |
+
+# Ethical Considerations and Limitations
+
+CodeLlama-7B-KStack is a new technology that carries risks with use. The testing conducted to date has not covered, nor could it cover all scenarios. For these reasons, as with all LLMs, CodeLlama-7B-KStack's potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate or objectionable responses to user prompts. The model was fine-tuned on a specific data format (Kotlin tasks), and deviation from this format can also lead to inaccurate or undesirable responses to user queries. Therefore, before deploying any applications of CodeLlama-7B-KStack, developers should perform safety testing and tuning tailored to their specific applications of the model.
\ No newline at end of file
diff --git a/configuration.json b/configuration.json
new file mode 100644
index 0000000..bbeeda1
--- /dev/null
+++ b/configuration.json
@@ -0,0 +1 @@
+{"framework": "pytorch", "task": "text-generation", "allow_remote": true}
\ No newline at end of file