INDEX

Explanations

terms related to scientific terminology and discourse

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

eyer

-0.16

ationally

-0.16

sut

-0.15

tyard

-0.15

utenberg

-0.15

undry

-0.15

adies

-0.15

antine

-0.15

reau

-0.15

ILT

-0.15

POSITIVE LOGITS

ific

0.42

IFIC

0.33

ÃŃf

0.32

ifique

0.29

ifi

0.27

ifica

0.26

ifik

0.26

ifice

0.26

ometrics

0.24

fic

0.24

Activations Density 0.005%

terms related to scientific terminology and discourse

No Comments

No Known Activations