INDEX

Explanations

references to names and their associated descriptions

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

parenthetical disambiguation

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

 nakalista

-0.66

AddTagHelper

-0.52

(!__

-0.49

MigrationBuilder

-0.43

Välislingid

-0.43

WireFormat

-0.41

 thither

-0.41

 CreateTagHelper

-0.40

UTELY

-0.40

 lenker

-0.39

POSITIVE LOGITS

 surla

0.53

 وتسجيلات

0.50

ंदीखरीदारी

0.49

thâu

0.46

Portale

0.46

 AssemblyCulture

0.44

 IBOutlet

0.43

ItemBackground

0.42

Portail

0.42

schrank

0.42

Activations Density 0.008%

references to names and their associated descriptions

parenthetical disambiguation

No Comments

No Known Activations