Opt 350M Benchmarks
ARC: 23.55 vs 96.7 (so35)-75.6%
HellaSwag: 36.73 vs 95.3 (gpt4)-61.5%
MMLU: 26.02 vs 88.3 (so35)-70.5%
TruthfulQA: 40.83 vs 59 (gpt4)-30.8%
WinoGrande: 52.64 vs 87.5 (gpt4)-39.8%
GSM8K: 0.3 vs 96.4 (so35)-99.7%
LLME Score: 0.14712
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)
Opt 350M Parameters and Internals
| LLM Name | Opt 350M |
| Repository 🤗 | https://huggingface.co/facebook/opt-350m |
| Model Size | 350m |
| Required VRAM | 0.7 GB |
| Updated | 2024-09-16 |
| Maintainer | |
| Model Type | opt |
| Model Files | |
| Supported Languages | en |
| Model Architecture | OPTForCausalLM |
| License | other |
| Context Length | 2048 |
| Model Max Length | 2048 |
| Transformers Version | 4.20.0.dev0 |
| Beginning of Sentence Token | </s> |
| End of Sentence Token | </s> |
| Unk Token | </s> |
| Vocabulary Size | 50272 |
| Torch Data Type | float16 |
| Activation Function | relu |
| Errors | replace |
Best Alternatives to Opt 350M
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Opt Mini Dataset 0 | 2K / 0.7 GB | 5 | 0 |
| Facebook Opt 350M SFT Korz14 | 2K / 0.7 GB | 5 | 0 |
| Temp Model Sft | 2K / 1.3 GB | 5 | 0 |
| Gpt350 Chat S V0 1 | 2K / 0.7 GB | 10 | 0 |
| Gpt350 Chat S V0 | 2K / 0.7 GB | 6 | 0 |
| Dadjokes Tuned Opt | 2K / 1.3 GB | 4 | 2 |
| Rockyalquimista888 | 2K / 1.3 GB | 7 | 0 |
| Remedycure | 2K / 1.3 GB | 7 | 0 |
| Sunday | 2K / 1.3 GB | 7 | 0 |
| Aira OPT 350M | 2K / 0 GB | 7 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than facebook/opt-350m .
Rank the Opt 350M 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? 35941 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024072803