INDEX

Explanations

references to educational or professional roles and group identities

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

‘s

-0.20

sWith

-0.19

ezi

-0.19

Â´s

-0.18

`s

-0.17

’s

-0.17

?s

-0.16

swire

-0.16

Ã©ric

-0.16

ï¿½s

-0.15

POSITIVE LOGITS

0.71

’

0.62

who

0.56

who

0.46

 whom

0.44

hips

0.39

'/

0.37

'-

0.36

hip

0.36

 whose

0.35

Activations Density 0.888%

references to educational or professional roles and group identities

No Comments

No Known Activations