INDEX

Explanations

specific instances of the word "use"

oai_token-act-pair · gpt-3.5-turbo

references to rules, regulations, or guidelines related to usage

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ylum

-0.83

arious

-0.80

 Barth

-0.78

 marqu

-0.76

 Jean

-0.70

 Ralph

-0.69

aro

-0.69

 Reson

-0.68

aram

-0.67

illo

-0.67

POSITIVE LOGITS

use

1.87

use

1.72

 uses

1.71

USE

1.70

Use

1.63

Use

1.61

 usage

1.55

 Uses

1.55

USE

1.48

using

1.48

Activations Density 0.245%

specific instances of the word "use"

references to rules, regulations, or guidelines related to usage

No Comments

No Known Activations