INDEX

Explanations

numerical values indicating intensity or significance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 25-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.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

 seven

-0.71

 Seven

-0.69

 SEVEN

-0.68

七

-0.68

seven

-0.67

VII

-0.66

 seventh

-0.65

Personensuche

-0.64

七

-0.63

 Bourgoin

-0.62

POSITIVE LOGITS

0.62

 ivelany

0.57

ConstraintMaker

0.54

４

0.53

 čty

0.52

0.51

annica

0.49

 čtyř

0.48

 keempat

0.48

Activations Density 0.057%

numerical values indicating intensity or significance

No Comments

No Known Activations