INDEX

Explanations

medical terms and drug names

oai_token-act-pair · gpt-3.5-turbo

distinct terms related to chemical substances and medications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 cryst

-0.60

 tremend

-0.58

 skelet

-0.57

 contrasting

-0.57

 behavi

-0.53

 suspic

-0.52

PDATE

-0.51

 crunch

-0.51

 outnumbered

-0.50

 corrid

-0.50

POSITIVE LOGITS

ovich

0.64

aga

0.62

ffe

0.62

usters

0.60

nik

0.59

ious

0.59

zynski

0.58

iaz

0.58

gha

0.58

isse

0.58

Activations Density 0.639%

medical terms and drug names

distinct terms related to chemical substances and medications

No Comments

No Known Activations