INDEX

Explanations

instances of the concept of being convinced or persuading someone

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

itz

-0.18

dash

-0.16

Trap

-0.15

IES

-0.15

hr

-0.15

fÃ¼

-0.15

 Trap

-0.14

ichel

-0.14

quist

-0.14

POSITIVE LOGITS

ively

0.27

 convinced

0.20

ainty

0.17

iveness

0.17

 convince

0.17

 conviction

0.17

 convincing

0.17

 convinc

0.17

 convin

0.16

 slic

0.16

Activations Density 0.025%

instances of the concept of being convinced or persuading someone

No Comments

No Known Activations