INDEX

Explanations

references to artistic works and their creators, particularly focusing on books, poems, films, and music

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

atif

-0.15

-0.14

ls

-0.14

czy

-0.14

oring

-0.14

clud

-0.13

egers

-0.13

urer

-0.13

ander

-0.13

POSITIVE LOGITS

 worth

0.25

 unto

0.21

 nobody

0.19

 Worth

0.19

 that

0.18

 unlike

0.18

 everyone

0.18

 everybody

0.17

that

0.16

 whose

0.15

Activations Density 0.212%

references to artistic works and their creators, particularly focusing on books, poems, films, and music

No Comments

No Known Activations