INDEX

Explanations

keywords related to research findings or conclusions

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of research findings

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

mph

-0.69

bid

-0.69

oning

-0.68

tz

-0.66

nas

-0.65

oned

-0.65

 weddings

-0.65

rons

-0.65

yss

-0.64

onement

-0.63

POSITIVE LOGITS

iveness

1.04

DragonMagazine

0.89

 findings

0.88

ĸļ

0.86

ivist

0.82

ãĥĻ

0.79

uggest

0.79

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

0.77

ivity

0.74

inctions

0.70

Activations Density 0.038%

keywords related to research findings or conclusions

repeated mentions of research findings

No Comments

No Known Activations