INDEX

Explanations

historical and statistical data related to growth rates and rankings

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate rankings or distinctions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Kits

-0.72

 Debor

-0.71

 Bullets

-0.69

 Wilson

-0.65

 Shaw

-0.64

 Borders

-0.64

Cox

-0.64

 Spock

-0.62

 Indigo

-0.62

etooth

-0.61

POSITIVE LOGITS

ranked

1.26

ever

1.21

ranking

1.19

worst

1.15

selling

1.14

performing

1.12

generation

1.10

tier

1.09

favorite

1.05

funded

1.03

Activations Density 0.049%

historical and statistical data related to growth rates and rankings

phrases that indicate rankings or distinctions

No Comments

No Known Activations