INDEX

Explanations

expressions of usefulness and quality in products and services

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

positive qualities

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

 BoxDecoration

-0.45

HTTPResponse

-0.43

setContentView

-0.42

ioutil

-0.41

setBackground

-0.37

 crossorigin

-0.36

randn

-0.36

onActivityResult

-0.36

 चीज़ों

-0.35

ご覧ください

-0.35

POSITIVE LOGITS

.*")]

0.49

 pinulongan

0.49

 transférez

0.48



0.48

///</

0.47

+#+#

0.47

qtype

0.45

 againſt

0.44

advance

0.44

々

0.44

Activations Density 0.062%

expressions of usefulness and quality in products and services

positive qualities

No Comments

No Known Activations