INDEX

Explanations

passive voice constructions, specifically focusing on the verb "to be" followed by a past participle

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Immunity

-0.73

 fray

-0.71

vas

-0.68

 Stain

-0.66

 Sina

-0.65

 Moose

-0.63

icago

-0.63

 Newspaper

-0.62

 Corpus

-0.60

 Dresden

-0.60

POSITIVE LOGITS

 reckoned

1.04

 seen

0.95

 explored

0.90

 avoided

0.88

 replaced

0.88

 judged

0.87

ige

0.85

 taken

0.85

 found

0.84

 evaluated

0.83

Activations Density 0.090%

passive voice constructions, specifically focusing on the verb "to be" followed by a past participle

No Comments

No Known Activations