INDEX

Explanations

references to actions or events happening at someone's home

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "home."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

andum

-0.75

uality

-0.71

yz

-0.69

rencies

-0.68

emonic

-0.67

immers

-0.65

ãģĨ

-0.64

OTUS

-0.64

utical

-0.63

acebook

-0.63

POSITIVE LOGITS

chool

1.00

coming

0.91

nikov

0.86

opath

0.84

stead

0.83

boy

0.81

owner

0.78

buy

0.78

pun

0.77

keeper

0.77

Activations Density 0.051%

references to actions or events happening at someone's home

mentions of the word "home."

No Comments

No Known Activations