INDEX

Explanations

references to geographical regions, specifically those labeled as "Western" or "Eastern."

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

empl

-0.16

Sad

-0.15

ige

-0.15

agn

-0.15

opher

-0.15

ward

-0.14

beh

-0.14

chin

-0.14

éĴ

-0.14

uar

-0.14

POSITIVE LOGITS

most

0.30

ization

0.24

ers

0.24

ized

0.23

-most

0.21

 blot

0.20

izes

0.19

isation

0.19

izing

0.18

izer

0.18

Activations Density 0.018%

references to geographical regions, specifically those labeled as "Western" or "Eastern."

No Comments

No Known Activations