INDEX

Explanations

phrases indicating contrast or change in direction

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

acus

-0.16

avig

-0.15

ãĤ¤ãĥ¤

-0.14

uele

-0.14

aje

-0.14

elps

-0.13

bih

-0.13

 ongoing

-0.13

lat

-0.13

POSITIVE LOGITS

/fwlink

0.23

ober

0.18

VERN

0.18

ÅĽcie

0.16

 tangent

0.16

 route

0.15

SSIP

0.15

 Ð´ÐµÐ»Ð¾

0.15

itant

0.15

 broke

0.15

Activations Density 0.101%

phrases indicating contrast or change in direction

No Comments

No Known Activations