INDEX

Explanations

references to prototypes and prototyping-related terminology

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

prototype and prototypes

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

Hochspringen

-0.58

Skocz

-0.57

AxisAlignment

-0.56

 שוליים

-0.55

vistar

-0.52

 تعدى

-0.51

<h2>

-0.49

ìm

-0.49

 bancaire

-0.48

setList

-0.47

POSITIVE LOGITS

 prototype

1.25

 prototypes

1.09

 Prototype

1.08

Prototype

1.05

prototype

0.96

 prototyping

0.93

 proto

0.83

 Proto

0.79

 PROTO

0.73

Proto

0.72

Activations Density 0.006%

references to prototypes and prototyping-related terminology

prototype and prototypes

No Comments

No Known Activations