INDEX

Explanations

individuals and information related to their professions or roles

oai_token-act-pair · gpt-3.5-turbo

phrases that describe professions or roles related to various activities and contributions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SPONSORED

-0.84

 sooner

-0.77

eworthy

-0.66

truth

-0.65

 unforeseen

-0.65

ozo

-0.63

evaluate

-0.62

itized

-0.62

 precon

-0.62

zo

-0.61

POSITIVE LOGITS

byss

0.68

 Crescent

0.67

 nearby

0.67

é¾

0.64

 Devon

0.64

ometimes

0.63

 Fenrir

0.63

 northeast

0.62

Ĥª

0.61

 affiliate

0.61

Activations Density 0.262%

individuals and information related to their professions or roles

phrases that describe professions or roles related to various activities and contributions

No Comments

No Known Activations