INDEX

Explanations

terms related to companies or organizations

oai_token-act-pair · gpt-3.5-turbo

references to specific companies

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

¥µ

-0.78

iste

-0.74

esta

-0.73

lde

-0.71

ature

-0.69

jong

-0.68

 immersion

-0.67

 developmental

-0.64

¸

-0.64

 portable

-0.64

POSITIVE LOGITS

terms

0.72

 Slovenia

0.70

abi

0.70

 Coup

0.69

 Gors

0.66

 Croat

0.64

azel

0.64

CLA

0.63

Ire

0.62

Activations Density 0.000%

terms related to companies or organizations

references to specific companies

No Comments

No Known Activations