INDEX

Explanations

punctuation marks and structural elements in writing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

kus

-0.18

Ã©is

-0.15

azon

-0.14

ovnÄĽ

-0.14

ews

-0.14

krom

-0.14

Ø§Ø¦Ùģ

-0.14

edla

-0.14

VML

-0.14

 Profession

-0.14

POSITIVE LOGITS

–

0.18

stdarg

0.15

noDB

0.15

aton

0.15

undle

0.15

 redirects

0.15

0.14

appa

0.14

heck

0.14

ĥĿ

0.14

Activations Density 0.000%

punctuation marks and structural elements in writing

No Comments

No Known Activations