INDEX

Explanations

references to errors and debugging information in code

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

Not in Any Lists

No Comments

Negative Logits

ÅĤo

-0.16

icho

-0.15

lish

-0.14

afone

-0.14

ursal

-0.14

¥Ĳ

-0.14

pson

-0.13

uml

-0.13

agal

-0.13

çŃĴ

-0.13

POSITIVE LOGITS

 line

0.61

 lines

0.56

 Line

0.45

 Lines

0.43

line

0.39

-line

0.37

lines

0.37

-lines

0.36

Lines

0.36

Line

0.36

Activations Density 0.121%

references to errors and debugging information in code

No Comments

No Known Activations