INDEX

Explanations

references to awards and nominations related to performance in film or television

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

"appeared" or "as" indicating roles/names

np_acts-logits-general · gemini-2.0-flash

guest role appearance

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 initComponents

-0.85

出版年

-0.75

 Wicidata

-0.70

 Houſe

-0.69

 greateſt

-0.68

 wireType

-0.67

enumi

-0.67

 pleaſure

-0.67

شوف

-0.66

 يتيمه

-0.66

POSITIVE LOGITS

 cameo

0.69

 guest

0.67

guest

0.65

 alongside

0.52

 appeared

0.50

 reprise

0.50

 appearing

0.50

 role

0.49

Guest

0.48

 featured

0.48

Activations Density 0.107%

references to awards and nominations related to performance in film or television

"appeared" or "as" indicating roles/names

guest role appearance

No Comments

No Known Activations