Arxiv:2101.00027 Arxiv:2201.07311 Arxiv:2304.01373 Autotrain compatible Dataset:eleutherai/the pile de... En Endpoints compatible Gpt neox Pythia Pytorch Region:us
Pythia 1.4B Deduped Benchmarks
ARC: 32.68 vs 96.7 (so35)-66.2%
HellaSwag: 54.96 vs 95.3 (gpt4)-42.3%
MMLU: 25.56 vs 88.3 (so35)-71.1%
TruthfulQA: 38.66 vs 59 (gpt4)-34.5%
WinoGrande: 57.3 vs 87.5 (gpt4)-34.5%
GSM8K: 0.83 vs 96.4 (so35)-99.1%
LLME Score: 0.18069
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)
Pythia 1.4B Deduped Parameters and Internals
| LLM Name | Pythia 1.4B Deduped |
| Repository 🤗 | https://huggingface.co/EleutherAI/pythia-1.4b-deduped |
| Model Size | 1.4b |
| Required VRAM | 2.9 GB |
| Updated | 2024-09-16 |
| Maintainer | EleutherAI |
| Model Type | gpt_neox |
| Model Files | |
| Supported Languages | en |
| Model Architecture | GPTNeoXForCausalLM |
| License | apache-2.0 |
| Context Length | 2048 |
| Model Max Length | 2048 |
| Transformers Version | 4.24.0 |
| Tokenizer Class | GPTNeoXTokenizer |
| Vocabulary Size | 50304 |
| Torch Data Type | float16 |
Best Alternatives to Pythia 1.4B Deduped
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Pythia 1.4B Deduped 8K Base | 8K / 7.3 GB | 6 | 1 |
| Pythia 1.4B Deduped 4K Base | 4K / 6.1 GB | 5 | 2 |
| Pythia 1.4B Sft Full | 2K / 2.8 GB | 2 | 1 |
| Pythia Delphi Suboptimal | 2K / 5.7 GB | 9 | 0 |
| RCC Ins Reconstruction | 2K / 6.1 GB | 22 | 1 |
| Pythia 1.4B | 2K / 2.9 GB | 121621 | 22 |
| Pythia 1.4b Sft Policy | 2K / 2.9 GB | 914 | 1 |
| Pythia 1.4B Deduped Sharegpt | 2K / 2.8 GB | 717 | 2 |
| Pythia 1.4B Deduped Sharegpt | 2K / 2.8 GB | 711 | 0 |
| ShortKing 1.4B V0.1 | 2K / 2.8 GB | 4304 | 2 |
Note: green Score (e.g. "73.2") means that the model is better than EleutherAI/pythia-1.4b-deduped .
Rank the Pythia 1.4B Deduped 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