INDEX

Explanations

details related to testimony and legal proceedings

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

followed by prepositions or punctuation

np_acts-logits-general · gemini-2.0-flash

progress and responsibilities

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

ustainable

-0.82

-0.81

Lastly

-0.70

 incentiv

-0.67

 Thankfully

-0.66

 Whilst

-0.65

Thankfully

-0.63

Whilst

-0.63

 Lastly

-0.62

 Jazeera

-0.61

POSITIVE LOGITS

 mußten

0.76

 Moslem

0.67

SourceChecksum

0.66

 Negroes

0.66

 employes

0.65

 iprot

0.65

 FontFamily

0.65

 Wikimedijinoj

0.64

TagMode

0.63

 referenties

0.63

Activations Density 0.439%

details related to testimony and legal proceedings

followed by prepositions or punctuation

progress and responsibilities

No Comments

No Known Activations