INDEX

Explanations

keywords related to products, services, features, and attributes in text

oai_token-act-pair · gpt-3.5-turbo

terms associated with various categories of products, services, and roles

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

 avenues

-0.79

 corridors

-0.79

 Dragons

-0.73

 mechanisms

-0.70

 collaborations

-0.69

 constants

-0.69

 waterways

-0.69

 sources

-0.68

 fronts

-0.67

 priorities

-0.67

POSITIVE LOGITS

or

0.81

ogram

0.80

alyst

0.74

robe

0.72

hole

0.71

ribe

0.70

 âľ

0.69

artist

0.68

igmatic

0.67

ictionary

0.67

Activations Density 0.725%

keywords related to products, services, features, and attributes in text

terms associated with various categories of products, services, and roles

No Comments

No Known Activations