INDEX

Explanations

positive descriptions or evaluations, particularly related to written works or products

oai_token-act-pair · gpt-3.5-turbo

positive evaluations or recommendations for books and stories

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

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

ogenesis

-0.83

foreseen

-0.80

gres

-0.79

venants

-0.78

undreds

-0.76

ousands

-0.74

soDeliveryDate

-0.74

Phase

-0.73

zees

-0.72

iances

-0.70

POSITIVE LOGITS

 nonetheless

0.95

 piece

0.95

 reminder

0.94

 specimen

0.94

 overall

0.93

 combination

0.93

 example

0.90

 addition

0.89

 underrated

0.89

 overview

0.87

Activations Density 0.323%

positive descriptions or evaluations, particularly related to written works or products

positive evaluations or recommendations for books and stories

No Comments

No Known Activations