Alignment-handbook Autotrain compatible Base model:datpysci/pythia-1b-... Base model:finetune:datpysci/p... Conversational Dataset:datpysci/spin iter0 Endpoints compatible Generated from trainer Gpt neox Region:us Safetensors
Pythia 1B Self Kto Iter0 Benchmarks
ARC: 30.2 vs 96.7 (so35)-68.8%
HellaSwag: 49.06 vs 95.3 (gpt4)-48.5%
MMLU: 24.11 vs 88.3 (so35)-72.7%
TruthfulQA: 36.35 vs 59 (gpt4)-38.4%
WinoGrande: 53.43 vs 87.5 (gpt4)-38.9%
GSM8K: 1.82 vs 96.4 (so35)-98.1%
LLME Score: 0.24281
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 1B Self Kto Iter0 Parameters and Internals
| LLM Name | Pythia 1B Self Kto Iter0 |
| Repository 🤗 | https://huggingface.co/DatPySci/pythia-1b-self-kto-iter0 |
| Base Model(s) | |
| Model Size | 1b |
| Required VRAM | 2 GB |
| Updated | 2024-09-07 |
| Maintainer | DatPySci |
| Model Type | gpt_neox |
| Model Files | |
| Model Architecture | GPTNeoXForCausalLM |
| License | apache-2.0 |
| Context Length | 2048 |
| Model Max Length | 2048 |
| Transformers Version | 4.38.1 |
| Tokenizer Class | GPTNeoXTokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 50304 |
| Torch Data Type | bfloat16 |
Best Alternatives to Pythia 1B Self Kto Iter0
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Pythia 2.8B Deduped Rp 710M 4K | 4K / 11.7 GB | 9 | 1 |
| Pythia 1.4B Deduped Rp 420M 4K | 4K / 6.1 GB | 11 | 1 |
| Pythia 1.4B Deduped Rp 280M 4K | 4K / 6.1 GB | 11 | 1 |
| ...rAI Pythia 1B Deduped Sft Tldr | 2K / 4 GB | 13661 | 0 |
| Pythia 1B Kto Iter0 | 2K / 2 GB | 6 | 0 |
| Rloo Trial2 | 2K / 2 GB | 8 | 0 |
| Rloo Tldr | 2K / 2 GB | 6 | 0 |
| TinyLM 1B Test2 | 2K / 2.7 GB | 6 | 0 |
| ...hia 1B Deduped Instruct Ckpt 2 | 2K / 2.1 GB | 6 | 0 |
| ...hia 1B Deduped Instruct Ckpt 1 | 2K / 2.1 GB | 6 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than DatPySci/pythia-1b-self-kto-iter0 .
Rank the Pythia 1B Self Kto Iter0 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? 35693 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024072803