INDEX

Explanations

questions or statements inviting feedback or opinions

oai_token-act-pair · gpt-3.5-turbo

expressions inviting feedback or opinions from readers

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

istration

-0.72

alias

-0.68

 Sinai

-0.68

enture

-0.67

orp

-0.65

PHOTOS

-0.65

 Leap

-0.64

sis

-0.61

arcity

-0.60

idad

-0.60

POSITIVE LOGITS

 voices

0.96

 hear

0.88

 Heard

0.83

 hearing

0.81

 Hear

0.80

 heard

0.79

grass

0.75

 aloud

0.75

wered

0.73

cliffe

0.73

Activations Density 0.025%

questions or statements inviting feedback or opinions

expressions inviting feedback or opinions from readers

No Comments

No Known Activations