INDEX

Explanations

instances of the character sequence 'd 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 @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

owitz

-0.17

okrat

-0.17

bers

-0.16

kers

-0.16

antine

-0.16

Ãºi

-0.15

hei

-0.15

Ã§a

-0.15

ÚĺÙĨ

-0.15

yonel

-0.14

POSITIVE LOGITS

esign

0.44

ocument

0.43

isplay

0.42

ownload

0.42

rive

0.41

ouble

0.41

irect

0.40

etails

0.39

irection

0.39

irectory

0.39

Activations Density 0.100%

instances of the character sequence 'd in various contexts

No Comments

No Known Activations