INDEX

Explanations

mention of the word "cancer"

oai_token-act-pair · gpt-3.5-turbo

words related to cancer

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

demand

-0.74

shall

-0.71

oho

-0.68

mediately

-0.66

pled

-0.66

oran

-0.65

ween

-0.64

ppings

-0.62

kept

-0.61

ori

-0.59

POSITIVE LOGITS

ous

0.81

UGH

0.77

ancer

0.76

xual

0.73

NetMessage

0.70

bane

0.69

rics

0.68

llan

0.68

iate

0.68

utics

0.68

Activations Density 0.021%

mention of the word "cancer"

words related to cancer

No Comments

No Known Activations