INDEX

Explanations

references to the speaker or first-person perspectives

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

cliffe

-0.19

nya

-0.18

rous

-0.17

stein

-0.17

so

-0.17

-0.16

 themselves

-0.16

lass

-0.16

 itself

-0.16

POSITIVE LOGITS

/us

0.38

SELF

0.23

/her

0.23

adows

0.22

asuring

0.20

zzo

0.19

andering

0.18

zelf

0.18

-mÃªme

0.18

ury

0.17

Activations Density 0.077%

references to the speaker or first-person perspectives

No Comments

No Known Activations