INDEX

Explanations

phrases that indicate significant features or highlights within descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

vor

-0.17

Eis

-0.15

pan

-0.15

insula

-0.15

Î¼Î±ÏĦÎ¿ÏĤ

-0.14

YST

-0.14

errupt

-0.14

van

-0.14

Creators

-0.14

na

-0.13

POSITIVE LOGITS

 ones

0.23

 Ones

0.18

ones

0.17

 ÙħÙĨÙĩØ§

0.17

tty

0.16

ayette

0.15

ìĭ¬

0.15

 Balt

0.15

addock

0.15

OMEM

0.14

Activations Density 0.287%

phrases that indicate significant features or highlights within descriptions

No Comments

No Known Activations