INDEX

Explanations

references to companionship or association

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

ÙĪÙĨØ¯

-0.17

kins

-0.15

uet

-0.15

ãĥ³ãĥĲ

-0.15

erosis

-0.15

antro

-0.14

(++

-0.14

nid

-0.14

 Ð½ÑĸÐ¶

-0.14

iy

-0.14

POSITIVE LOGITS

 gÃ³

0.17

_go

0.16

 Flem

0.15

 accompanies

0.14

uron

0.14

éĻĦ

0.14

Conv

0.14

oday

0.13

ypsum

0.13

Activations Density 0.198%

references to companionship or association

No Comments

No Known Activations