INDEX

Explanations

the beginning of sentences or paragraphs indicated by specific structural tags

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

 kasarigan

-1.00

bootstrapcdn

-0.97

 ویکی‌پدی

-0.93

StructEnd

-0.89

featureID

-0.88

aarrggbb

-0.84

PropertyChanging

-0.82

fromnode

-0.80

 tartalomajánló

-0.79

NameInMap

-0.77

POSITIVE LOGITS

addGap

0.63

0.58

²(

0.51

ุง

0.45

apy

0.44

 same

0.44

rele

0.43

Endian

0.43

वी

0.42

 great

0.42

Activations Density 0.000%

the beginning of sentences or paragraphs indicated by specific structural tags

No Comments

No Known Activations