INDEX

Explanations

references to the action of shooting or being shot

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

hir

-0.22

hift

-0.17

loi

-0.17

hill

-0.17

ials

-0.16

hire

-0.16

anders

-0.16

hud

-0.16

ason

-0.16

VAL

-0.16

POSITIVE LOGITS

guns

0.34

gun

0.23

ting

0.21

ì°©

0.17

 hoops

0.17

tps

0.16

shots

0.16

 Fired

0.16

glass

0.15

tober

0.15

Activations Density 0.033%

references to the action of shooting or being shot

No Comments

No Known Activations