INDEX

Explanations

references to specialized areas of expertise or products within a broader category

oai_token-act-pair · gpt-3.5-turbo

terms related to specialties and unique categories of expertise

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iven

-0.86

 Babel

-0.81

obyl

-0.71

ils

-0.70

-0.69

ilib

-0.69

 Migration

-0.67

IL

-0.66

ILS

-0.65

construct

-0.65

POSITIVE LOGITS

 specialty

2.47

 vintage

1.79

 antique

1.30

 prescription

1.29

wax

1.06

 persuasion

1.04

 novelty

1.03

 Schwartz

1.01

 prescriptions

1.00

 bent

0.96

Activations Density 0.036%

references to specialized areas of expertise or products within a broader category

terms related to specialties and unique categories of expertise

No Comments

No Known Activations