INDEX

Explanations

instances of the letter 'a' 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 @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

est

-0.38

que

-0.25

pt

-0.24

bl

-0.24

ffects

-0.24

im

-0.24

esthetic

-0.24

ims

-0.23

äºĽ

-0.23

ffect

-0.23

POSITIVE LOGITS

ustral

0.26

lic

0.22

ustr

0.21

then

0.19

riel

0.18

tras

0.18

eo

0.18

ustralian

0.18

ther

0.18

ero

0.18

Activations Density 0.063%

instances of the letter 'a' in various contexts

No Comments

No Known Activations