INDEX

Explanations

names of researchers or scientists

oai_token-act-pair · gpt-3.5-turbo

parts of proper nouns or names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ĸļ

-0.83

EStream

-0.72

dfx

-0.71

 psychiat

-0.69

ĻĤ

-0.68

orate

-0.68

PDATE

-0.66

 fashionable

-0.66

ÃĥÃĤ

-0.64

 undermin

-0.62

POSITIVE LOGITS

 Farms

1.00

 Technologies

0.99

 Enterprises

0.96

 Productions

0.94

 Industries

0.92

 Associates

0.89

 County

0.89

 Labs

0.86

 Publishing

0.85

 Manor

0.84

Activations Density 0.329%

names of researchers or scientists

parts of proper nouns or names

No Comments

No Known Activations