INDEX

Explanations

instances of the word "use" in various forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 unforgettable

-0.78

 Macaulay

-0.77

([\

-0.77

 jumbo

-0.77

McM

-0.77

 MacGregor

-0.76

 GIPHY

-0.73

HttpHeaders

-0.72

 McFar

-0.72

Šalt

-0.71

POSITIVE LOGITS

use

1.93

use

1.68

Use

1.64

Use

1.59

USE

1.51

USE

1.40

uso

1.34

 uses

1.11

Uso

1.11

 usage

1.05

Activations Density 0.146%

instances of the word "use" in various forms

No Comments

No Known Activations

instances of the word "use" in various forms

No Comments

No Known Activations