INDEX

Explanations

government or legal offices in specific locations

oai_token-act-pair · gpt-3.5-turbo

references to legal or governmental offices

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

isers

-0.83

ickets

-0.82

pread

-0.76

ervative

-0.74

iser

-0.74

ãģĦ

-0.74

izers

-0.67

isting

-0.66

é¾įå¥ĳå£«

-0.66

hemer

-0.66

POSITIVE LOGITS

 prosec

0.83

 raided

0.72

 Department

0.71

 branch

0.71

 alleges

0.67

 Dept

0.67

 Cyrus

0.65

 Office

0.65

 Cheong

0.64

 Authority

0.63

Activations Density 0.023%

government or legal offices in specific locations

references to legal or governmental offices

No Comments

No Known Activations