INDEX

Explanations

specific locations or landmarks

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

place names and locations

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

 ước

-0.43

 full

-0.40

총

-0.40

-0.39

 gross

-0.39

 Franke

-0.38

真实

-0.38

 Ivanov

-0.37

 druk

-0.37

 Vogt

-0.37

POSITIVE LOGITS

+#+

0.70

 AssemblyCulture

0.67

 disambiguazione

0.66

HasAnnotation

0.66

AddTagHelper

0.63

 Taktlose

0.61

ValueStyle

0.59

MigrationBuilder

0.58

 ExecuteAsync

0.56

MessageOf

0.56

Activations Density 0.027%

specific locations or landmarks

place names and locations

No Comments

No Known Activations