INDEX

Explanations

words related to medical conditions or treatments

oai_token-act-pair · gpt-3.5-turbo

variations of the word "atheism" or related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 millisec

-0.72

 overtime

-0.71

 Domestic

-0.70

 Yards

-0.70

 bloss

-0.66

 advertisements

-0.66

([

-0.65

 masks

-0.64

 Schultz

-0.64

 domestic

-0.64

POSITIVE LOGITS

athe

4.90

athed

2.35

athing

1.95

aths

1.78

ath

1.69

 Athe

1.29

othe

1.25

ATH

1.17

 athe

1.15

alde

1.08

Activations Density 0.023%

words related to medical conditions or treatments

variations of the word "atheism" or related terms

No Comments

No Known Activations