INDEX

Explanations

technical jargon and code-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

featureID

-0.89

 Chwiliwch

-0.88

webElementXpaths

-0.86

addCriterion

-0.85

aarrggbb

-0.83

RegistryLite

-0.79

elemField

-0.78

Geplaatst

-0.78

 propOrder

-0.78

WireFormatLite

-0.76

POSITIVE LOGITS

 tils

0.50

 darse

0.49

 след

0.39

dasarkan

0.38

 tidligere

0.38

的她

0.37

ítása

0.36

cir

0.36

맹

0.36

zied

0.35

Activations Density 0.718%

technical jargon and code-related terms

No Comments

No Known Activations