INDEX

Explanations

numeric values and their associated variables or identifiers

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

probability calculations with letters

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 foncé

-0.41

enumi

-0.41

 bäst

-0.41

 पद

-0.40

Kop

-0.40

mainwindow

-0.40

 Residential

-0.40

Residential

-0.40

 sauvages

-0.40

poison

-0.39

POSITIVE LOGITS

ьаж

0.50

AccessorTable

0.45

 snippetHide

0.45



0.41

 propOrder

0.40

ništ

0.39

Portale

0.39

ganu

0.37

)))),

0.37

めでとう

0.36

Activations Density 0.251%

numeric values and their associated variables or identifiers

probability calculations with letters

No Comments

No Known Activations