INDEX

Explanations

hints or suggestions within a given context

oai_token-act-pair · gpt-3.5-turbo

phrases or words suggesting anticipation or indication of forthcoming information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

nea

-0.79

fare

-0.73

frey

-0.72

portion

-0.70

lux

-0.69

rament

-0.69

rior

-0.69

artney

-0.67

vict

-0.67

bred

-0.65

POSITIVE LOGITS

 hint

1.51

 hints

1.31

 clue

1.02

 hinted

0.93

 clues

0.87

 glimps

0.82

 wink

0.81

 intim

0.79

 warning

0.73

 signs

0.71

Activations Density 0.035%

hints or suggestions within a given context

phrases or words suggesting anticipation or indication of forthcoming information

No Comments

No Known Activations