INDEX

Explanations

references to experts and analysts in various fields discussing social and political issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 Civilization

-0.15

ÙĪÙĦØ§

-0.14

Ð±Ð¸ÑĢÐ°

-0.14

ä¼ı

-0.14

Await

-0.13

ÐµÐ·Ð¿ÐµÑĩ

-0.13

ameda

-0.13

Äįem

-0.13

enson

-0.13

.preferences

-0.13

POSITIVE LOGITS

 agree

0.23

 note

0.21

 point

0.21

 alike

0.21

 universally

0.20

 consensus

0.19

 warn

0.19

 agrees

0.19

 such

0.19

 talk

0.18

Activations Density 0.115%

references to experts and analysts in various fields discussing social and political issues

No Comments

No Known Activations