INDEX

Explanations

references or citations to other content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ness

-0.19

soever

-0.18

land

-0.18

ly

-0.17

nya

-0.17

like

-0.17

wide

-0.17

self

-0.16

most

-0.16

POSITIVE LOGITS

 below

0.29

 also

0.28

-through

0.25

/he

0.23

 Also

0.22

ley

0.22

also

0.21

beck

0.21

LEY

0.21

Also

0.20

Activations Density 0.031%

references or citations to other content

No Comments

No Known Activations