INDEX

Explanations

phrases indicating something or someone is under a particular authority or regulation

oai_token-act-pair · gpt-3.5-turbo

phrases concerning regulatory compliance or conditions requiring oversight

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

kr

-0.61

 Americas

-0.60

Zur

-0.60

 cane

-0.58

 plat

-0.58

 latt

-0.58

Tib

-0.56

 Zimmer

-0.56

 Sutherland

-0.56

 Euras

-0.55

POSITIVE LOGITS

ively

1.23

ivity

1.12

ioned

0.96

ivist

0.94

ivities

0.94

ive

0.93

ivating

0.93

itatively

0.92

Reviewer

0.89

iving

0.87

Activations Density 0.024%

phrases indicating something or someone is under a particular authority or regulation

phrases concerning regulatory compliance or conditions requiring oversight

No Comments

No Known Activations