INDEX

Explanations

the word "prior" in various contexts, indicating a focus on temporal references or previous events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

chin

-0.16

kre

-0.16

iber

-0.15

apor

-0.15

bru

-0.15

sterol

-0.14

pool

-0.14

au

-0.14

else

-0.14

 hrom

-0.14

POSITIVE LOGITS

/current

0.20

lush

0.15

lus

0.15

ê¹

0.15

-last

0.15

imes

0.15

wner

0.15

 ÄĳÃ¢y

0.14

/original

0.14

avad

0.14

Activations Density 0.013%

the word "prior" in various contexts, indicating a focus on temporal references or previous events

No Comments

No Known Activations