INDEX

Explanations

qualities or states related to simplicity or visual attributes

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

precise descriptions

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

 Majefty

-0.56

 betweenstory

-0.51

 becauſe

-0.47

⟬

-0.46

 pleaſure

-0.44

VIAF

-0.43

 cantons

-0.43

balleur

-0.43

 Reſ

-0.42

nastics

-0.42

POSITIVE LOGITS

msgTypes

0.43

green

0.36

 למש

0.35

 green

0.35

 الاطلاع

0.34

 prí

0.34

 Navarro

0.34

GRANTED

0.34

rungsseite

0.33

 pří

0.32

Activations Density 0.165%

qualities or states related to simplicity or visual attributes

precise descriptions

No Comments

No Known Activations