INDEX

Explanations

words related to medical conditions and terms

oai_token-act-pair · gpt-3.5-turbo

mentions of specific medical conditions or treatments

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

PRES

-0.68

 Chancellor

-0.63

 count

-0.63

Madison

-0.63

 chancellor

-0.61

 blatant

-0.60

LA

-0.60

PDATE

-0.60

eve

-0.60

 Wass

-0.60

POSITIVE LOGITS

atform

1.29

asts

1.14

asso

1.12

asma

1.08

asms

1.06

enty

1.01

ifting

0.98

utonium

0.97

asm

0.95

acer

0.95

Activations Density 0.014%

words related to medical conditions and terms

mentions of specific medical conditions or treatments

No Comments

No Known Activations