INDEX

Explanations

words related to negative valuation or reduction in value

oai_token-act-pair · gpt-3.5-turbo

terms related to the devaluation or dishonesty of concepts, entities, or assets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

WORK

-0.72

HEAD

-0.68

ilk

-0.68

 Agency

-0.67

 Sight

-0.67

 Grail

-0.66

 Physics

-0.65

 Parenthood

-0.65

fell

-0.65

Eye

-0.64

POSITIVE LOGITS

 deval

1.39

uation

1.27

uing

1.22

ued

1.21

uations

1.18

ues

0.98

aution

0.96

imar

0.96

pes

0.89

uating

0.88

Activations Density 0.009%

words related to negative valuation or reduction in value

terms related to the devaluation or dishonesty of concepts, entities, or assets

No Comments

No Known Activations