INDEX

Explanations

references to online content and communication

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ight

-0.17

ightly

-0.15

ignon

-0.15

ä¸įè¿ĩ

-0.15

arrass

-0.15

opi

-0.14

ONA

-0.14

=~

-0.14

uation

-0.14

omba

-0.14

POSITIVE LOGITS

 elsewhere

0.20

 earlier

0.19

 Crawford

0.18

 previous

0.16

auer

0.15

 liquid

0.14

ãĥĥãĥī

0.14

udiant

0.14

enthal

0.14

Activations Density 0.328%

references to online content and communication

No Comments

No Known Activations