INDEX

Explanations

instances of the word "strong" and its variations, indicating strength or power

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

ized

-0.18

bian

-0.17

ë¡ľ

-0.17

jur

-0.16

fulness

-0.15

umas

-0.15

aled

-0.15

ION

-0.15

otros

-0.15

cha

-0.14

POSITIVE LOGITS

holds

0.36

 máº½

0.25

bow

0.24

(er

0.23

hold

0.22

-arm

0.21

-strong

0.21

/we

0.20

 bonds

0.19

 enough

0.19

Activations Density 0.054%

instances of the word "strong" and its variations, indicating strength or power

No Comments

No Known Activations