INDEX

Explanations

reference to novels and literary works

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

Non-English programming-related tokens

np_acts-logits-general · gemini-2.0-flash

tiedostot/junar.zip

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

<eos>

-0.51

-0.48

adə

-0.46

bed

-0.44

بي

-0.40

 hoek

-0.40

 folha

-0.40

 ferdig

-0.38

ISupport

-0.38

POSITIVE LOGITS

 '\\;'

0.90

ChildScrollView

0.77

UrlResolution

0.68

InputBorder

0.67

хьтан

0.67

ConstraintMaker

0.66

httphttps

0.65

发表于

0.65

 MotionEvent

0.65

ItemBackground

0.65

Activations Density 0.052%

reference to novels and literary works

Non-English programming-related tokens

tiedostot/junar.zip

No Comments

No Known Activations

reference to novels and literary works

Non-English programming-related tokens

tiedostot/junar.zip

No Comments

No Known Activations