Autotrain compatible Base model:finetune:tinyllama/... Base model:tinyllama/tinyllama... Conversational Dataset:generator Endpoints compatible Generated from trainer Llama Region:us Safetensors Sft Tensorboard Trl
Tukan 1.1B Chat Reasoning Sft COLA Benchmarks
ARC: 34.13-64.6%
HellaSwag: 59.78-37.3%
MMLU: 24.86-71.3%
TruthfulQA: 38.25-35.2%
WinoGrande: 60.77-30.5%
GSM8K: 1.36-98.5%
nn.n% — How the model compares to the GPT-4.
Tukan 1.1B Chat Reasoning Sft COLA Parameters and Internals
| LLM Name | Tukan 1.1B Chat Reasoning Sft COLA |
| Repository | Open on 🤗 |
| Base Model(s) | |
| Model Size | 1.1b |
| Required VRAM | 4.4 GB |
| Updated | 2024-07-27 |
| Maintainer | alexredna |
| Model Type | llama |
| Model Files | |
| Model Architecture | LlamaForCausalLM |
| License | apache-2.0 |
| Context Length | 2048 |
| Model Max Length | 2048 |
| Transformers Version | 4.36.2 |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | </s> |
| Vocabulary Size | 32000 |
| Torch Data Type | float32 |
Best Alternatives to Tukan 1.1B Chat Reasoning Sft COLA
Best Alternatives |
HF Rank |
Context/RAM |
Downloads |
Likes |
|---|---|---|---|---|
| Coven Tiny 1.1b 32k Orpo Alpha | 0.3 | 32K / 2.2 GB | 281 | 2 |
| Test Mix 01 | 0.2 | 32K / 2.2 GB | 6 | 0 |
| Palmer Merge Test 5 | 0.2 | 32K / 2.2 GB | 7 | 0 |
| ...llama 1.1B 16K Instructions V4 | 0.2 | 32K / 2.2 GB | 363 | 0 |
| TinyLlama 1.1B 32K Instruct | 0.2 | 32K / 2.2 GB | 977 | 13 |
| TinyLlama 1.1B 32K | 0.2 | 32K / 2.2 GB | 414 | 29 |
| ...lama 1.1B 16K Instructions Rag | 0.2 | 32K / 2.2 GB | 73 | 0 |
| Palmer 002 32K | 0.2 | 32K / 2.2 GB | 14 | 0 |
| TinyJ.O.S.I.E. 1.1B 32K Base | 0.2 | 32K / 2.2 GB | 6 | 1 |
| Tinyllama History Chat V1.1 | 0.2 | 32K / 2.2 GB | 9 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than alexredna/Tukan-1.1B-Chat-reasoning-sft-COLA .
Rank the Tukan 1.1B Chat Reasoning Sft COLA 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? 34446 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v2024072501