INDEX

Explanations

dates and numerical references in scientific or medical contexts

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

Adverbs ending in "-ly"

np_acts-logits-general · gemini-2.0-flash

adverb describing state followed by adjective

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

worthiness

-0.68

DockStyle

-0.67

 fraîche

-0.67

 professionalism

-0.64

 portability

-0.63

 acceptability

-0.63

compatibility

-0.63

mindedness

-0.62

SequentialGroup

-0.61

 appropriateness

-0.61

POSITIVE LOGITS

aarrggbb

0.71

 thermally

0.67

styleable

0.62

urally

0.57

 heavily

0.56

 emotionally

0.55

 musically

0.54

 highly

0.53

وردار

0.53

 decorated

0.52

Activations Density 0.936%

dates and numerical references in scientific or medical contexts

Adverbs ending in "-ly"

adverb describing state followed by adjective

No Comments

No Known Activations