INDEX

Explanations

numerical expressions with positive and negative values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

GEBURTSDATUM

-1.26

帖最后由

-0.94

sizeCache

-0.92

BeginContext

-0.92

 GeoNames

-0.90

HasForeignKey

-0.87

Portail

-0.87

]--;

-0.85

mergeFrom

-0.84

SourceChecksum

-0.83

POSITIVE LOGITS

right

0.62

0.60

0.53

0.52

 right

0.45

ナメント

0.44

há

0.43

iParam

0.41

 colspan

0.41

Activations Density 0.083%

numerical expressions with positive and negative values

No Comments

No Known Activations