INDEX

Explanations

references to hospitals and medical treatment in relation to incidents or accidents

oai_token-act-pair · gpt-3.5-turbo

references to hospitals and medical treatment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

wcsstore

-0.81

artisan

-0.78

BOOK

-0.76

yip

-0.76

 Helpful

-0.72

tail

-0.69

venants

-0.69

 BOOK

-0.68

*/(

-0.68

ezvous

-0.66

POSITIVE LOGITS

ilitating

0.89

ization

0.87

 evacuation

0.82

 outpatient

0.78

 annex

0.78

bed

0.77

 gown

0.77

ysis

0.77

 Ukrain

0.77

itals

0.74

Activations Density 0.023%

references to hospitals and medical treatment in relation to incidents or accidents

references to hospitals and medical treatment

No Comments

No Known Activations