INDEX

Explanations

words related to self-control or moderation

oai_token-act-pair · gpt-3.5-turbo

concepts related to restraint and decency

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

hops

-0.84

reau

-0.77

 Selected

-0.71

alez

-0.70

 SOLD

-0.67

ales

-0.67

~/

-0.67

wang

-0.66

valued

-0.64

asca

-0.63

POSITIVE LOGITS

 restraint

2.69

 decency

2.13

 hypocrisy

2.09

 intolerance

1.56

 deprivation

1.48

 coward

1.40

 negligence

1.39

 intimidation

1.37

 modesty

1.30

 insecurity

1.25

Activations Density 0.030%

words related to self-control or moderation

concepts related to restraint and decency

No Comments

No Known Activations