INDEX

Explanations

phrases related to job positions and roles

oai_token-act-pair · gpt-3.5-turbo

references to roles or stances held by individuals or groups

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

go

-0.73

vine

-0.72

amaz

-0.69

 THANK

-0.65

jee

-0.65

bey

-0.64

WARNING

-0.62

agic

-0.61

bane

-0.61

Limited

-0.60

POSITIVE LOGITS

 positions

3.96

 position

2.66

 stances

2.09

 Position

2.08

Position

1.86

 roles

1.76

Pos

1.67

 spots

1.51

 positioning

1.47

 heights

1.41

Activations Density 0.009%

phrases related to job positions and roles

references to roles or stances held by individuals or groups

No Comments

No Known Activations