Llamion 14B Base Benchmarks
ARC: 55.2-42.7%
HellaSwag: 80.57-15.5%
MMLU: 67.89-21.5%
TruthfulQA: 39.79-32.6%
WinoGrande: 73.16-16.4%
GSM8K: 37.98-58.7%
nn.n% — How the model compares to the GPT-4.
Llamion 14B Base Parameters and Internals
| LLM Name | Llamion 14B Base |
| Repository | Open on 🤗 |
| Model Size | 14b |
| Required VRAM | 29 GB |
| Updated | 2024-07-27 |
| Maintainer | vaiv |
| Model Type | llama |
| Model Files | |
| Model Architecture | LlamaForCausalLM |
| License | apache-2.0 |
| Context Length | 4096 |
| Model Max Length | 4096 |
| Transformers Version | 4.37.2 |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | <unk> |
| Vocabulary Size | 84608 |
| Torch Data Type | bfloat16 |
Best Alternatives to Llamion 14B Base
Best Alternatives |
HF Rank |
Context/RAM |
Downloads |
Likes |
|---|---|---|---|---|
| GeM2 Llamion 14B LongChat | 0.3 | 195K / 29 GB | 2484 | 3 |
| 14B | 0.3 | 8K / 28.4 GB | 2170 | 298 |
| Qwen 14B Llamafied | 0.2 | 8K / 28.4 GB | 1412 | 5 |
| Qwen 14B Chat LLaMAfied | 0.2 | 8K / 28.4 GB | 957 | 8 |
| JerseyDevil 14B | 0.2 | 8K / 28.5 GB | 9 | 2 |
| CausalLM Platypus 14B | 0.2 | 8K / 28.4 GB | 927 | 1 |
| SkunkApe 14B | 0.2 | 8K / 28.5 GB | 10 | 3 |
| Qwen 14B Llamafied | 0.2 | 8K / 28.4 GB | 6 | 1 |
| EarlyFailures14B | 0.2 | 8K / 28.4 GB | 18 | 5 |
| POLAR 14B DPO V1.3 | 0.3 | 4K / 28.4 GB | 1971 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than vaiv/llamion-14b-base .
Rank the Llamion 14B Base 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? 34447 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024072501