INDEX

Explanations

references to animal behavior and human waste

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

robat

-0.18

Ø¬Ø±

-0.16

ubbles

-0.15

neb

-0.14

apus

-0.14

ãĥ«ãĥķ

-0.13

èī¯

-0.13

ummer

-0.13

elig

-0.13

ser

-0.13

POSITIVE LOGITS

fec

0.39

 poop

0.38

poo

0.37

po

0.35

fec

0.34

 stools

0.34

 stool

0.33

ex

0.32

Po

0.31

fa

0.31

Activations Density 0.035%

references to animal behavior and human waste

No Comments

No Known Activations