INDEX

Explanations

references to the term "gut" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

eur

-0.18

oding

-0.16

ishing

-0.16

hw

-0.16

oded

-0.16

ql

-0.15

hai

-0.15

hop

-0.15

pole

-0.15

POSITIVE LOGITS

ierrez

0.31

ters

0.29

ted

0.28

gut

0.23

less

0.21

ting

0.20

tsy

0.19

ten

0.19

tae

0.18

achten

0.18

Activations Density 0.006%

references to the term "gut" in various contexts

No Comments

No Known Activations