INDEX

Explanations

specific details or items in a list

oai_token-act-pair · gpt-3.5-turbo

specific nouns and terms related to details, structures, and systems

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

atorium

-0.71

 Amen

-0.70

 Dragonbound

-0.68

asca

-0.66

levels

-0.64

ebin

-0.64

ODY

-0.64

rius

-0.63

kus

-0.63

 Flavoring

-0.62

POSITIVE LOGITS

 thrown

1.18

 tossed

1.18

 popping

1.16

 disappearing

1.15

 swapped

1.11

 piled

1.10

 piling

1.10

 slipping

1.05

 discarded

1.05

 shoved

1.03

Activations Density 0.399%

specific details or items in a list

specific nouns and terms related to details, structures, and systems

No Comments

No Known Activations