INDEX

Explanations

words related to various categories or types of items

oai_token-act-pair · gpt-3.5-turbo

categories and classifications related to various topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

bda

-0.85

stice

-0.75

ãĤ¦ãĤ¹

-0.71

mosp

-0.70

 ç¥ŀ

-0.67

}}}

-0.65

rawdownloadcloneembedreportprint

-0.64

 Flavoring

-0.63

onz

-0.63

\\\\\\\\

-0.62

POSITIVE LOGITS

 considered

0.85

 touted

0.84

we

0.76

 covered

0.76

 hardest

0.74

 referenced

0.74

igmat

0.73

 eligible

0.73

 featured

0.72

 favored

0.72

Activations Density 0.403%

words related to various categories or types of items

categories and classifications related to various topics

No Comments

No Known Activations