INDEX

Explanations

the concept of matching or compatibility in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ties

-0.21

isset

-0.16

ated

-0.16

alfa

-0.16

idget

-0.16

oted

-0.16

İ

-0.15

elfare

-0.15

zos

-0.15

zzo

-0.15

POSITIVE LOGITS

(es

0.26

maker

0.26

stick

0.24

-ups

0.21

ett

0.20

makers

0.20

sticks

0.20

DispatchToProps

0.20

ings

0.19

enko

0.19

Activations Density 0.043%

the concept of matching or compatibility in various contexts

No Comments

No Known Activations