INDEX

Explanations

references to body parts and physical features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

crypt

-0.17

lun

-0.16

.scalablytyped

-0.16

-0.15

urat

-0.15

mando

-0.14

 Found

-0.14

ÑĸÐ·

-0.14

scatter

-0.14

ittal

-0.14

POSITIVE LOGITS

 sticking

0.26

 pressed

0.23

ak

0.21

 extended

0.21

 stuck

0.21

 raised

0.21

 cock

0.21

jam

0.20

 partially

0.20

ask

0.19

Activations Density 0.097%

references to body parts and physical features

No Comments

No Known Activations