INDEX

Explanations

profession or occupation titles

oai_token-act-pair · gpt-3.5-turbo

terms related to various occupations and professions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

edIn

-0.78

ecause

-0.74

Ø§ÙĦ

-0.73

 Members

-0.69

fw

-0.68

ourcing

-0.68

atch

-0.68

equality

-0.67

require

-0.67

english

-0.67

POSITIVE LOGITS

osphere

0.97

 himself

0.88

 laureate

0.84

 Gabriel

0.82

 Eugene

0.80

 Ernest

0.78

 whom

0.78

 extraord

0.78

liest

0.77

who

0.77

Activations Density 0.234%

profession or occupation titles

terms related to various occupations and professions

No Comments

No Known Activations