INDEX

Explanations

references to size and sizing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

})));

-0.70

CrossRef

-0.69

Diwedd

-0.65

 depoz

-0.64

 yoksa

-0.63

ängel

-0.63

多い

-0.61

 aldus

-0.61

[],

-0.61

{},

-0.61

POSITIVE LOGITS

 size

1.09

 SIZE

0.96

 Size

0.88

siz

0.87

 sizes

0.83

Enllaços

0.83

 SMALL

0.81

 sized

0.77

 tamaño

0.77

Демографія

0.75

Activations Density 0.026%

references to size and sizing

No Comments

No Known Activations