INDEX

Explanations

references to specific individuals and their political affiliations or actions

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

NYT

-0.15

 Strip

-0.15

 strip

-0.14

rej

-0.14

 kidn

-0.14

æī«

-0.14

hist

-0.14

NPR

-0.14

aes

-0.13

vale

-0.13

POSITIVE LOGITS

WND

0.24

itbart

0.23

 Breitbart

0.21

CNS

0.19

 Epoch

0.19

Fox

0.17

wnd

0.17

rieb

0.16

 liberty

0.16

 Judicial

0.16

Activations Density 0.145%

references to specific individuals and their political affiliations or actions

No Comments

No Known Activations