INDEX

Explanations

character names within movie plots

oai_token-act-pair · gpt-3.5-turbo

closing parentheses

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

itory

-0.80

 answ

-0.75

ĸļ

-0.71

ciating

-0.70

ĻĤ

-0.68

 flowering

-0.63

 giveaways

-0.62

 disruptive

-0.62

igmatic

-0.61

criptions

-0.61

POSITIVE LOGITS

 Malf

0.71

 Sabb

0.69

ãĥİ

0.69

 Committees

0.69

 Neph

0.67

 Cursed

0.67

 Syndicate

0.66

Rum

0.65

 Gamb

0.65

 Islands

0.65

Activations Density 0.163%

character names within movie plots

closing parentheses

No Comments

No Known Activations