INDEX

Explanations

references to leadership and political identity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

å°ĸ

-0.14

zier

-0.14

eka

-0.14

ÑĥÑī

-0.14

ãĤ¤ãĤº

-0.14

 anarchists

-0.14

itung

-0.14

å¢ĥ

-0.14

mploy

-0.14

POSITIVE LOGITS

 states

0.23

 figure

0.21

 moderate

0.21

 cent

0.19

 outsider

0.19

states

0.19

 cerebral

0.19

 inexperienced

0.18

 States

0.18

/pop

0.17

Activations Density 0.162%

references to leadership and political identity

No Comments

No Known Activations