INDEX

Explanations

specific numbers

oai_token-act-pair · gpt-3.5-turbo

words related to action and movement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

atility

-0.73

 âĢº

-0.71

ULE

-0.65

alogue

-0.64

 Coliseum

-0.64

UCT

-0.63

;;;;;;;;;;;;

-0.62

ega

-0.62

unda

-0.61

 Thumbnails

-0.61

POSITIVE LOGITS

tif

0.71

 resemb

0.71

 foreskin

0.69

 repr

0.69

whe

0.65

minist

0.65

ly

0.63

unc

0.62

xx

0.61

rex

0.60

Activations Density 0.000%

specific numbers

words related to action and movement

No Comments

No Known Activations