INDEX

Explanations

elements related to mathematical or statistical operations and notations

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

specific references to sources, especially related to interviews or citations

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

written by, calculated

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

—

-0.81

-0.69

;-)

-0.65

новниш

-0.64

RRect

-0.64

Rüyada

-0.62

—

-0.61

^(@)

-0.60

 Wikispecies

-0.60

 Offisielt

-0.59

POSITIVE LOGITS

||

0.76

0.66

0.61

 Aesthetic

0.61

setPreferred

0.60

♡

0.59

edits

0.58

 aesthetic

0.58

 NSCoder

0.57

Activations Density 0.020%

elements related to mathematical or statistical operations and notations

specific references to sources, especially related to interviews or citations

written by, calculated

No Comments

No Known Activations