INDEX

Explanations

references to various types of schemes or methodologies

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

technical schemes

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

Autowired

-0.51

 fingernails

-0.49

DotNetBar

-0.47

getMinutes

-0.47

 Pitt

-0.44

Butter

-0.43

Fifth

-0.43

last

-0.43

butter

-0.43

 Fifth

-0.42

POSITIVE LOGITS

scheme

1.28

 scheme

1.26

 Scheme

1.25

Scheme

1.24

 SCHEME

1.23

SCHEME

1.19

 schemes

1.16

 Schemes

1.13

 ſche

1.11

cheme

1.10

Activations Density 0.008%

references to various types of schemes or methodologies

technical schemes

No Comments

No Known Activations

references to various types of schemes or methodologies

technical schemes

No Comments

No Known Activations