INDEX

Explanations

geographical locations and organizational affiliations

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

names and affiliations

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

 AssemblyCulture

-0.84

hyrchwyd

-0.48

 snippetHide

-0.45

tagHelperRunner

-0.42

 itself

-0.41

endphp

-0.41

 transfieras

-0.40

MigrationBuilder

-0.40

 Exacts

-0.39

 Infórmanos

-0.38

POSITIVE LOGITS

who

0.54

astéroïdes

0.50

 whose

0.50

gilan

0.48

whose

0.47

smoking

0.44

fielder

0.44

 borealis

0.43

Бил

0.42

ditor

0.42

Activations Density 0.028%

geographical locations and organizational affiliations

names and affiliations

No Comments

No Known Activations