INDEX

Explanations

references to research findings or study results

oai_token-act-pair · gpt-3.5-turbo

references to research findings or results

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

bid

-0.68

 leisure

-0.64

pload

-0.64

mph

-0.63

 residence

-0.63

 monop

-0.63

yss

-0.63

 goose

-0.62

oning

-0.61

yne

-0.61

POSITIVE LOGITS

iveness

1.04

 findings

0.91

ĸļ

0.89

uggest

0.87

DragonMagazine

0.82

ãĥĻ

0.79

ivist

0.78

èĥ

0.76

~~~~~~~~~~~~~~~~

0.74

ãĤ¦ãĤ¹

0.74

Activations Density 0.028%

references to research findings or study results

references to research findings or results

No Comments

No Known Activations