INDEX

Explanations

mentions of frequency or repetition

oai_token-act-pair · gpt-3.5-turbo

the word "each" to highlight instances of collective or individual consideration

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hin

-0.78

dad

-0.64

soType

-0.60

Sport

-0.60

news

-0.57

ensen

-0.57

 Prediction

-0.56

audi

-0.56

ota

-0.54

sic

-0.54

POSITIVE LOGITS

 each

3.34

each

2.67

 Each

2.27

Each

2.26

 apiece

1.97

 every

1.71

every

1.23

 annually

1.21

 respectively

1.19

 EVERY

1.16

Activations Density 0.032%

mentions of frequency or repetition

the word "each" to highlight instances of collective or individual consideration

No Comments

No Known Activations