INDEX

Explanations

references to historical time periods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

an

-0.16

CO

-0.15

par

-0.15

 elsewhere

-0.14

Th

-0.14

 Fact

-0.14

 actually

-0.14

Ã¡li

-0.14

POSITIVE LOGITS

esen

0.17

ITTER

0.15

unifu

0.15

wcs

0.15

ãĥĭãĤ¢

0.14

ÙĨØ²

0.14

ighter

0.14

usp

0.14

olon

0.14

áº

0.14

Activations Density 0.009%

references to historical time periods

No Comments

No Known Activations