INDEX

Explanations

instances of "new" in the context of objects or features being created or added

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

new account, new game, new tab

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

 domestiques

-0.36

UDAD

-0.35

miyorum

-0.35

 Olivenöl

-0.35

 solidaridad

-0.34

 zís

-0.34

 dolayı

-0.34

 organes

-0.33

 DOMIN

-0.33

เต็ม

-0.33

POSITIVE LOGITS

تقاوى

0.66

DockStyle

0.65

MigrationBuilder

0.60

 препратки

0.57

httphttps

0.56

addCriterion

0.56

 GenerationType

0.56

SharedCtor

0.56

 otomatig

0.56

CreateModel

0.56

Activations Density 0.089%

instances of "new" in the context of objects or features being created or added

new account, new game, new tab

No Comments

No Known Activations