INDEX

Explanations

references to video game titles and elements related to gaming and reviews

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

Negative Logits

ARGIN

-0.14

ÄĽtÃŃ

-0.14

icism

-0.13

REFIX

-0.13

bic

-0.12

finity

-0.12

rÃ¼n

-0.12

cko

-0.12

ÅĻet

-0.12

obao

-0.11

POSITIVE LOGITS

WEEN

0.16

oppable

0.15

ashtra

0.14

ushima

0.13

rowsable

0.13

perienced

0.13

etheless

0.13

gÄ±n

0.12

nbsp

0.12

Activations Density 1.479%

references to video game titles and elements related to gaming and reviews

No Known Activations