llama-3-tulu-2-70b

Home/Models/allenai/llama-3-tulu-2-70b

llama-3-tulu-2-70b

allenai/llama-3-tulu-2-70b

kind: original
quant: bf16 (16-bit)
family: llama
modality: text-generation
root model: meta-llama/Meta-Llama-3-70B
provider: allenai
license: apache-2.0
language: en
context: —
library: —
vRAM (GB): —
size on disk: —
downloads: 52
likes: 0
created: —
last modified: 2024-08-05T16:59:47+00:00 (2024)
huggingface: https://huggingface.co/allenai/llama-3-tulu-2-70b
tags: transformerssafetensorsllamatext-generationconversationalendataset:allenai/tulu-v2-sft-mixturearxiv:2311.10702base_model:meta-llama/Meta-Llama-3-70Bbase_model:finetune:meta-llama/Meta-Llama-3-70Blicense:apache-2.0text-generation-inferenceendpoints_compatibleregion:us

Benchmark scores (8)

benchmark	category	score	context	conf	evidence
AlpacaEval	chat	94.65	GSM8k 8-shot cot \| BBH 3-shot cot	low	`\| [Llama 3 8b instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) \| 0.626 \| 0.770 \| 0.606 \| 59.04 \| 0.799 \| 94.65 \| 23.12 \| 0.682 \| 0.741 \| 0.028 \| 0.115 \| 70.36`
TydiQA	knowledge	66.8	GSM8k 8-shot cot \| BBH 3-shot cot	low	`\| [Llama 3 8b base](https://huggingface.co/meta-llama/Meta-Llama-3-8B) \| 0.649 \| 0.565 \| 0.653 \| 66.80 \| 0.664 \| - \| - \| 0.299 \| 0.146 \| 0.200 \| 0.390 \| 54.36 \|`
HumanEval	coding	66.4	GSM8k 8-shot cot \| BBH 3-shot cot	low	`\| [Llama 3 8b base](https://huggingface.co/meta-llama/Meta-Llama-3-8B) \| 0.649 \| 0.565 \| 0.653 \| 66.80 \| 0.664 \| - \| - \| 0.299 \| 0.146 \| 0.200 \| 0.390 \| 54.36 \|`
BBH	reasoning	65.3	GSM8k 8-shot cot \| BBH 3-shot cot	low	`\| [Llama 3 8b base](https://huggingface.co/meta-llama/Meta-Llama-3-8B) \| 0.649 \| 0.565 \| 0.653 \| 66.80 \| 0.664 \| - \| - \| 0.299 \| 0.146 \| 0.200 \| 0.390 \| 54.36 \|`
MMLU	knowledge	64.9	GSM8k 8-shot cot \| BBH 3-shot cot	low	`\| [Llama 3 8b base](https://huggingface.co/meta-llama/Meta-Llama-3-8B) \| 0.649 \| 0.565 \| 0.653 \| 66.80 \| 0.664 \| - \| - \| 0.299 \| 0.146 \| 0.200 \| 0.390 \| 54.36 \|`
GSM8K	math	56.5	GSM8k 8-shot cot \| BBH 3-shot cot	low	`\| [Llama 3 8b base](https://huggingface.co/meta-llama/Meta-Llama-3-8B) \| 0.649 \| 0.565 \| 0.653 \| 66.80 \| 0.664 \| - \| - \| 0.299 \| 0.146 \| 0.200 \| 0.390 \| 54.36 \|`
TruthfulQA	safety	29.9	GSM8k 8-shot cot \| BBH 3-shot cot	low	`\| [Llama 3 8b base](https://huggingface.co/meta-llama/Meta-Llama-3-8B) \| 0.649 \| 0.565 \| 0.653 \| 66.80 \| 0.664 \| - \| - \| 0.299 \| 0.146 \| 0.200 \| 0.390 \| 54.36 \|`
IFEval	instruction-following	14.6	GSM8k 8-shot cot \| BBH 3-shot cot	low	`\| [Llama 3 8b base](https://huggingface.co/meta-llama/Meta-Llama-3-8B) \| 0.649 \| 0.565 \| 0.653 \| 66.80 \| 0.664 \| - \| - \| 0.299 \| 0.146 \| 0.200 \| 0.390 \| 54.36 \|`

Discussions (0)

none scraped.