INDEX

Explanations

prepositions followed by nouns

oai_token-act-pair · gpt-3.5-turbo

legal terminology and concepts related to victimization and responsibility

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

stones

-0.91

tones

-0.87

years

-0.76

olas

-0.75

\.

-0.74

ernels

-0.74

uations

-0.73

 Hispanics

-0.73

 Trees

-0.73

 Bows

-0.73

POSITIVE LOGITS

 collaborator

1.46

 beneficiary

1.41

 buyer

1.38

 follower

1.38

 participant

1.38

 purchaser

1.36

 listener

1.34

 distributor

1.34

 recipient

1.33

 lover

1.33

Activations Density 0.344%

prepositions followed by nouns

legal terminology and concepts related to victimization and responsibility

No Comments

No Known Activations