INDEX

Explanations

proper nouns, specifically names and abbreviations related to individuals and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

^(@)

-0.71

 pleaſure

-0.71

__':

-0.70

SupportActionBar

-0.69

وأضاف

-0.69

<?

-0.68

 Preferencias

-0.68

findpost

-0.68

󠁧

-0.67

CrossRef

-0.66

POSITIVE LOGITS

 leads

0.51

 Italijanski

0.50

Bibliograf

0.49

mobileqq

0.49

 atve

0.48

det

0.46

 Werk

0.45

водства

0.43

leads

0.43

ftagPool

0.41

Activations Density 0.762%

proper nouns, specifically names and abbreviations related to individuals and organizations

No Comments

No Known Activations