INDEX

Explanations

dialogues and conversational exchanges

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

Beginning of an answer or response

np_acts-logits-general · gemini-2.0-flash

affirmative interjections and responses

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

twimg

-0.91

 kasarigan

-0.88

bootstrapcdn

-0.77

findpost

-0.76

 EconPapers

-0.72

soort

-0.67

saraba

-0.65

 ModelExpression

-0.64

aarrggbb

-0.64

ⓧ

-0.64

POSITIVE LOGITS

 absolutely

0.76

 Absolutely

0.71

 listen

0.70

Absolutely

0.70

absolutely

0.66

 Yeah

0.64

 yeah

0.63

Yeah

0.61

Listen

0.59

 Listen

0.59

Activations Density 0.055%

dialogues and conversational exchanges

Beginning of an answer or response

affirmative interjections and responses

No Comments

No Known Activations