INDEX

Explanations

comparisons of things or people to indicate superiority or predominance

oai_token-act-pair · gpt-3.5-turbo

phrases implying comparison or superiority

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

istries

-0.79

 Christy

-0.74

CBC

-0.67

mary

-0.63

kay

-0.62

itent

-0.61

dal

-0.61

 Brock

-0.59

cu

-0.59

hn

-0.58

POSITIVE LOGITS

THING

0.82

 predecessors

0.79

ç¥ŀ

0.73

ł

0.71

 conceivable

0.71

 predecessor

0.70

 predec

0.69

¿½

0.67

 peers

0.67

 competitors

0.65

Activations Density 0.081%

comparisons of things or people to indicate superiority or predominance

phrases implying comparison or superiority

No Comments

No Known Activations