INDEX

Explanations

the word "suggestions"

oai_token-act-pair · gpt-3.5-turbo

expressions related to suggestions and recommendations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ipel

-0.77

ophon

-0.72

mberg

-0.68

tails

-0.68

IGH

-0.67

 Strait

-0.64

inals

-0.64

estine

-0.63

seconds

-0.63

otype

-0.63

POSITIVE LOGITS

 sugg

0.96

 suggestions

0.92

uggest

0.88

 suggestion

0.83

ļéĨĴ

0.78

 Suggest

0.77

uyomi

0.77

 hint

0.77

 pane

0.76

 recommendations

0.71

Activations Density 0.023%

the word "suggestions"

expressions related to suggestions and recommendations

No Comments

No Known Activations