INDEX

Explanations

phrases indicating visual perception or interaction with people

oai_token-act-pair · gpt-3.5-turbo

personal pronouns and references to individual relationships

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

ente

-0.71

ģĸ

-0.69

ĵ

-0.68

isma

-0.67

ĨĴ

-0.63

icion

-0.62

©¶æ¥µ

-0.62

idium

-0.61

İ

-0.59

creation

-0.58

POSITIVE LOGITS

 unfold

0.85

 alive

0.80

 perform

0.78

 interact

0.77

 smiling

0.76

 closely

0.74

 naked

0.72

 slumped

0.71

 silhou

0.71

 firsthand

0.69

Activations Density 0.190%

phrases indicating visual perception or interaction with people

personal pronouns and references to individual relationships

No Comments

No Known Activations