INDEX

Explanations

expressions of hesitation or filler words

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

pit

-0.17

ARRIER

-0.16

pData

-0.15

á»¥n

-0.15

ÑĪÐµ

-0.15

ribbon

-0.15

ideo

-0.14

aney

-0.14

ortex

-0.14

bette

-0.14

POSITIVE LOGITS

bral

0.23

braco

0.20

pper

0.19

eki

0.17

esh

0.17

ESCO

0.15

kehr

0.15

atural

0.15

 skirts

0.15

arked

0.15

Activations Density 0.007%

expressions of hesitation or filler words

No Comments

No Known Activations