Merged Model Autotrain compatible Conversational Endpoints compatible Llama Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/uygarkurt/llama-3-merged-linear
Llama 3 Merged Linear Benchmarks
ARC: 72.78 vs 96.7 (so35)-24.7%
HellaSwag: 88.51 vs 95.3 (gpt4)-7.1%
MMLU: 68.22 vs 88.3 (so35)-22.7%
TruthfulQA: 65.46 vs 59 (gpt4)10.9%
WinoGrande: 79.48 vs 87.5 (gpt4)-9.2%
GSM8K: 69.14 vs 96.4 (so35)-28.3%
LLME Score: 0.26645
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
Llama 3 Merged Linear Parameters and Internals
| LLM Name | Llama 3 Merged Linear |
| Repository 🤗 | https://huggingface.co/uygarkurt/llama-3-merged-linear |
| Merged Model | Yes |
| Model Size | 8b |
| Required VRAM | 16.1 GB |
| Updated | 2024-12-21 |
| Maintainer | uygarkurt |
| Model Type | llama |
| Model Files | |
| Model Architecture | LlamaForCausalLM |
| License | mit |
| Context Length | 8192 |
| Model Max Length | 8192 |
| Transformers Version | 4.40.1 |
| Tokenizer Class | PreTrainedTokenizerFast |
| Vocabulary Size | 128256 |
| Torch Data Type | float16 |
Best Alternatives to Llama 3 Merged Linear
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| ...a 3 8B Instruct Gradient 1048K | 1024K / 16.1 GB | 5528 | 678 |
| Thor V1.4 8B DARK FICTION | 1024K / 16.1 GB | 941 | 2 |
| MrRoboto ProLong 8B V2b | 1024K / 16.1 GB | 177 | 0 |
| MrRoboto ProLong 8B V1a | 1024K / 16.1 GB | 107 | 0 |
| MrRoboto ProLong 8B V2a | 1024K / 16.1 GB | 100 | 0 |
| HEL V0.8 8B LONG DARK | 1024K / 16.1 GB | 184 | 0 |
| MrRoboto ProLong 8B V1n | 1024K / 16.1 GB | 84 | 0 |
| 8B Unaligned BASE V2c | 1024K / 16.1 GB | 128 | 0 |
| MrRoboto ProLong 8B V2f | 1024K / 16.1 GB | 51 | 0 |
| MrRoboto ProLong 8B V1f | 1024K / 16.1 GB | 63 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than uygarkurt/llama-3-merged-linear .
Rank the Llama 3 Merged Linear Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 40013 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241217