INDEX

Explanations

negative sentiment or criticism in text

oai_token-act-pair · gpt-3.5-turbo

negative sentiments or criticisms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Tiny

-0.80

 Feder

-0.80

 Dynam

-0.78

 Ballistic

-0.74

 Helic

-0.73

FIN

-0.72

 Yellow

-0.71

 Mystic

-0.71

 Gilbert

-0.71

UC

-0.70

POSITIVE LOGITS

direction

1.30

based

1.29

family

1.23

repeat

1.21

bearing

1.21

level

1.20

axis

1.20

related

1.19

backer

1.18

induced

1.18

Activations Density 0.121%

negative sentiment or criticism in text

negative sentiments or criticisms

No Comments

No Known Activations