INDEX

Explanations

contractions

oai_token-act-pair · gpt-3.5-turbo

negative contractions and phrases indicating a state of being or condition

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

ESA

-0.83

jri

-0.82

anon

-0.70

udos

-0.65

Jackson

-0.65

lez

-0.64

âĢİ

-0.64

xit

-0.64

lash

-0.63

iffin

-0.63

POSITIVE LOGITS

 outnumbered

0.79

 cheap

0.72

 deduct

0.72

 cheaper

0.71

 outwe

0.67

 already

0.66

 messed

0.66

 collateral

0.65

 poorly

0.65

 pree

0.65

Activations Density 0.392%

contractions

negative contractions and phrases indicating a state of being or condition

No Comments

No Known Activations