INDEX

Explanations

the presence of the word "ver" and its derivatives related to perception and vision

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ert

-0.19

 Feinstein

-0.17

mite

-0.15

Poh

-0.15

 pros

-0.15

åħĲ

-0.14

ery

-0.14

 pert

-0.14

à¸ļ

-0.14

èĲ

-0.14

POSITIVE LOGITS

ded

0.26

arming

0.20

ander

0.19

der

0.19

bindings

0.19

onica

0.18

andering

0.18

Ded

0.17

adem

0.17

dere

0.16

Activations Density 0.005%

the presence of the word "ver" and its derivatives related to perception and vision

No Comments

No Known Activations