INDEX

Explanations

references to individuals being native to specific places or communities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

native

-0.20

 native

-0.18

Ùī

-0.17

edList

-0.17

Native

-0.17

_native

-0.17

rij

-0.17

rooms

-0.16

_NATIVE

-0.16

nice

-0.16

POSITIVE LOGITS

-born

0.31

born

0.26

 born

0.23

 tongue

0.23

-spe

0.21

 speaker

0.20

/original

0.19

son

0.19

/local

0.19

 Born

0.19

Activations Density 0.015%

references to individuals being native to specific places or communities

No Comments

No Known Activations