Merged Models
Collection
These are models created by merging existing models that are already fine tuned or even merged themselves. • 5 items • Updated • 1
Transformers
Safetensors
GGUF
llama
Merge
mergekit
lazymergekit
theprint/Code-Llama-Bagel-8B
ajibawa-2023/Code-Llama-3-8B
jondurbin/bagel-8b-v1.0
Eval Results (legacy)
text-generation-inference
conversational
How to use from
llama.cppInstall from WinGet (Windows)
winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf theprint/Code-Llama-Bagel-8B:# Run inference directly in the terminal:
llama-cli -hf theprint/Code-Llama-Bagel-8B:Use pre-built binary
# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf theprint/Code-Llama-Bagel-8B:# Run inference directly in the terminal:
./llama-cli -hf theprint/Code-Llama-Bagel-8B:Build from source code
git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf theprint/Code-Llama-Bagel-8B:# Run inference directly in the terminal:
./build/bin/llama-cli -hf theprint/Code-Llama-Bagel-8B:Use Docker
docker model run hf.co/theprint/Code-Llama-Bagel-8B:Quick Links
Code-Llama-Bagel-8B
Code-Llama-Bagel-8B is a merge of the following models using LazyMergekit:
🧩 Configuration
slices:
- sources:
- model: ajibawa-2023/Code-Llama-3-8B
layer_range: [0, 32]
- model: jondurbin/bagel-8b-v1.0
layer_range: [0, 32]
merge_method: slerp
base_model: ajibawa-2023/Code-Llama-3-8B
parameters:
t:
- filter: self_attn
value: [0, 0.5, 0.3, 0.7, 1]
- filter: mlp
value: [1, 0.5, 0.7, 0.3, 0]
- value: 0.5
dtype: bfloat16
💻 Usage
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "theprint/Code-Llama-Bagel-8B"
messages = [{"role": "user", "content": "What is a large language model?"}]
tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
"text-generation",
model=model,
torch_dtype=torch.float16,
device_map="auto",
)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 14.48 |
| IFEval (0-Shot) | 25.30 |
| BBH (3-Shot) | 25.34 |
| MATH Lvl 5 (4-Shot) | 4.98 |
| GPQA (0-shot) | 3.47 |
| MuSR (0-shot) | 7.53 |
| MMLU-PRO (5-shot) | 20.24 |
- Downloads last month
- 193
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for theprint/Code-Llama-Bagel-8B
Collection including theprint/Code-Llama-Bagel-8B
Evaluation results
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard25.300
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard25.340
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard4.980
- acc_norm on GPQA (0-shot)Open LLM Leaderboard3.470
- acc_norm on MuSR (0-shot)Open LLM Leaderboard7.530
- accuracy on MMLU-PRO (5-shot)test set Open LLM Leaderboard20.240
Install from brew
# Start a local OpenAI-compatible server with a web UI: llama-server -hf theprint/Code-Llama-Bagel-8B:# Run inference directly in the terminal: llama-cli -hf theprint/Code-Llama-Bagel-8B: