INDEX

Explanations

dimensions and measurements of objects

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

length and width measurements

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

########.

-0.65

LookAnd

-0.63



-0.61

AsUp

-0.60

Personendaten

-0.57

 skimage

-0.52

QMetaType

-0.50

 Infórmanos

-0.50

 Shades

-0.48

casio

-0.47

POSITIVE LOGITS

 ciasc

0.46

UpInside

0.45

 lenker

0.43

alive

0.42

splitlines

0.41

 yaşında

0.40

 FormGroup

0.39

 laikā

0.39

 edible

0.38

 arşivlendi

0.38

Activations Density 0.012%

dimensions and measurements of objects

length and width measurements

No Comments

No Known Activations