INDEX

Explanations

adjectives related to comparison and evaluation

oai_token-act-pair · gpt-3.5-turbo

references to emotional states or conditions related to societal issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Selected

-0.71

aval

-0.67

origin

-0.65

agram

-0.62

asus

-0.62

 Variant

-0.62

DNA

-0.61

 Citation

-0.61

jew

-0.61

alion

-0.60

POSITIVE LOGITS

 calmed

1.34

 peaceful

1.34

 calm

1.31

 harmless

1.29

 manageable

1.26

 calming

1.24

 safer

1.22

 toler

1.19

 peace

1.18

 peacefully

1.17

Activations Density 1.298%

adjectives related to comparison and evaluation

references to emotional states or conditions related to societal issues

No Comments

No Known Activations