INDEX

Explanations

phrases related to conflict, intensity, and negativity

oai_token-act-pair · gpt-3.5-turbo

references to specific groups or individuals involved in negative actions or behaviors

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

 Context

-0.89

significant

-0.83

 Differences

-0.80

irming

-0.76

 Significant

-0.75

 Citation

-0.74

 Factors

-0.74

ECD

-0.74

 Forward

-0.73

 significant

-0.72

POSITIVE LOGITS

 spew

1.45

gob

1.40

 swoop

1.30

wre

1.24

 strut

1.22

 endlessly

1.21

 churn

1.20

 perpetually

1.19

 prow

1.15

 relentlessly

1.13

Activations Density 0.528%

phrases related to conflict, intensity, and negativity

references to specific groups or individuals involved in negative actions or behaviors

No Comments

No Known Activations