INDEX

Explanations

words and phrases indicating redundancy or excessive repetition

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

chwitz

-0.18

 denen

-0.17

ono

-0.16

 erotische

-0.15

hev

-0.15

clock

-0.14

Ã¨ve

-0.14

ierge

-0.14

amik

-0.14

 clock

-0.14

POSITIVE LOGITS

 respective

0.18

oric

0.18

orie

0.17

 hereby

0.16

ese

0.16

tas

0.15

 successors

0.15

UTO

0.14

 subsid

0.14

imers

0.14

Activations Density 0.128%

words and phrases indicating redundancy or excessive repetition

No Comments

No Known Activations