INDEX

Explanations

states in the United States

oai_token-act-pair · gpt-3.5-turbo

mentions of U.S. states and their relevance in the context provided

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

Afee

-0.70

ãĥĻ

-0.69

ufact

-0.68

ãĥĳ

-0.66

LOAD

-0.66

-0.63

Pod

-0.63

BOOK

-0.63

Austral

-0.63

 plur

-0.62

POSITIVE LOGITS

 Rapids

0.89

jri

0.80

achusetts

0.79

burgh

0.79

uyomi

0.72

essee

0.72

sylvania

0.71

;;;;;;;;;;;;

0.71

tu

0.69

 Dept

0.69

Activations Density 0.064%

states in the United States

mentions of U.S. states and their relevance in the context provided

No Comments

No Known Activations