INDEX

Explanations

proper nouns and names

oai_token-act-pair · gpt-3.5-turbo

references to a person named "Sam Raimi" and related content

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

natureconservancy

-0.85

SPONSORED

-0.73

$$$$

-0.73

thia

-0.68

kered

-0.67

hander

-0.64

kees

-0.63

col

-0.63

BOOK

-0.62

picking

-0.62

POSITIVE LOGITS

onica

0.86

ulus

0.84

 Saban

0.83

imum

0.81

aim

0.79

imus

0.78

ati

0.76

eer

0.76

inator

0.75

ments

0.75

Activations Density 0.013%

proper nouns and names

references to a person named "Sam Raimi" and related content

No Comments

No Known Activations