INDEX

Explanations

words related to quick action or instant results

oai_token-act-pair · gpt-3.5-turbo

phrases indicating immediate actions or effects

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

 ANGEL

-0.80

llor

-0.70

ger

-0.68

 setbacks

-0.66

Ã±a

-0.64

 Hann

-0.64

 Millennium

-0.62

may

-0.62

 directives

-0.60

 newsp

-0.59

POSITIVE LOGITS

aneously

1.23

 Leilan

0.81

 transitioned

0.78

 recogn

0.75

 recognizable

0.75

aneous

0.74

immune

0.74

idious

0.72

 identifiable

0.72

ipedia

0.70

Activations Density 0.024%

words related to quick action or instant results

phrases indicating immediate actions or effects

No Comments

No Known Activations