INDEX

Explanations

references to English-speaking populations 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 @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

rieve

-0.15

tep

-0.15

ESH

-0.14

Å¡il

-0.14

ghi

-0.14

wright

-0.14

etes

-0.14

ocha

-0.14

ialis

-0.13

 INTERRUPTION

-0.13

POSITIVE LOGITS

phone

0.24

-American

0.21

 Norm

0.20

Norm

0.20

 Anglo

0.19

phones

0.19

axon

0.18

-Israel

0.17

-phone

0.17

uhn

0.17

Activations Density 0.013%

references to English-speaking populations or communities

No Comments

No Known Activations