INDEX

Explanations

identifiers or codes related to software components and their properties

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

script tags and URLs

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

 الحره

-0.49

 pericolo

-0.43

 الرياضيه

-0.42

big

-0.41

 Thatcher

-0.41

 kän

-0.41

ressional

-0.40

 perigo

-0.40

 soutenu

-0.40

 oiled

-0.40

POSITIVE LOGITS

хьтан

0.59

 autorytatywna

0.54

 snippetHide

0.51

 kasarigan

0.49

 useAppContext

0.46

ù

0.43

 ब्रेकडाउन

0.40

 invokingState

0.40

UrlResolution

0.39

 jdbcTemplate

0.39

Activations Density 0.010%

identifiers or codes related to software components and their properties

script tags and URLs

No Comments

No Known Activations