INDEX

Explanations

instances of the word "small."

oai_token-act-pair · gpt-3.5-turbo

references to the concept of 'smallness' or small entities

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

alde

-0.70

 anthem

-0.67

arov

-0.67

endas

-0.66

fet

-0.66

 partying

-0.65

otle

-0.64

oÄŁ

-0.63

 allegiance

-0.62

fal

-0.62

POSITIVE LOGITS

 Small

3.43

Small

2.79

small

2.11

 Large

1.96

Large

1.61

 Tiny

1.56

 small

1.53

 Medium

1.51

 Minor

1.39

 Huge

1.33

Activations Density 0.009%

instances of the word "small."

references to the concept of 'smallness' or small entities

No Comments

No Known Activations