INDEX

Explanations

instances of the word "fight" and its variations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ities

-0.17

eward

-0.15

xit

-0.15

ATE

-0.15

WARD

-0.15

ulers

-0.15

heit

-0.14

icit

-0.14

ITY

-0.14

ieties

-0.14

POSITIVE LOGITS

club

0.23

back

0.23

 against

0.19

against

0.17

 Against

0.17

dirty

0.17

 tooth

0.17

-hard

0.16

axe

0.16

 club

0.16

Activations Density 0.024%

instances of the word "fight" and its variations in different contexts

No Comments

No Known Activations