INDEX

Explanations

references to food and nutrition-related products and companies

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Pearce

-0.15

reno

-0.15

stice

-0.14

á»įt

-0.14

 tá»¥

-0.14

à¸Ļà¸²à¸¢

-0.14

 ÐºÐ¾Ð¼Ñĥ

-0.14

rott

-0.14

 prostituer

-0.14

_PROVID

-0.13

POSITIVE LOGITS

 Nest

0.35

 ingredient

0.27

 packaging

0.26

 processors

0.25

 packaged

0.24

 nest

0.24

 Kraft

0.24

 food

0.24

 brands

0.23

 processor

0.23

Activations Density 0.197%

references to food and nutrition-related products and companies

No Comments

No Known Activations