INDEX

Explanations

instances of the letters "ri" and its variations in different words

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

an

-0.24

ãĥ³

-0.21

Ñı

-0.19

ãĥ¼

-0.18

ÙĨ

-0.18

ãĥ¼ãĥ³

-0.17

°

-0.17

cec

-0.17

Ð½

-0.16

alogy

-0.16

POSITIVE LOGITS

0.27

kk

0.26

eh

0.24

pped

0.23

eff

0.23

ev

0.22

ech

0.21

ever

0.21

kt

0.20

ppy

0.20

Activations Density 0.023%

instances of the letters "ri" and its variations in different words

No Comments

No Known Activations