INDEX

Explanations

references to crime-related details and evidence

oai_token-act-pair · gpt-3.5-turbo

phrases indicating evidence and investigation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Indie

-0.71

usterity

-0.71

profits

-0.70

 bandwagon

-0.70

oliberal

-0.70

 startups

-0.70

 indie

-0.69

 Capitalism

-0.69

reprene

-0.68

 Esports

-0.68

POSITIVE LOGITS

 fingerprints

1.17

 markings

1.06

 incrim

0.99

 protr

0.99

 belonged

0.97

 charred

0.96

 fingerprint

0.95

 resemblance

0.92

 resembled

0.92

DNA

0.90

Activations Density 0.855%

references to crime-related details and evidence

phrases indicating evidence and investigation

No Comments

No Known Activations