INDEX

Explanations

sentences discussing beliefs, opinions, or affirmations

oai_token-act-pair · gpt-3.5-turbo

indications of opinion or belief related to various social and political issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

Pwr

-0.52

cellaneous

-0.49

iasm

-0.49

cour

-0.48

 flurry

-0.47

<-

-0.47

 âĶľâĶĢâĶĢ

-0.47

 Explain

-0.46

 partName

-0.46

:=

-0.45

POSITIVE LOGITS

 unfairly

0.61

 beneficial

0.57

 genuine

0.57

 sufficiently

0.55

 irreversible

0.54

 somehow

0.53

 misunderstood

0.53

 inappropriately

0.53

 undermin

0.52

 undes

0.51

Activations Density 1.701%

sentences discussing beliefs, opinions, or affirmations

indications of opinion or belief related to various social and political issues

No Comments

No Known Activations