INDEX

Explanations

mathematical constructs and relationships within formal definitions

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

then

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

tagHelperRunner

-0.57

 שוליים

-0.55

RTGC

-0.48

clientY

-0.45

UnusedPrivate

-0.44

 يتيمه

-0.44

 doInBackground

-0.43

CloseOperation

-0.42

WRP

-0.42

 Wikimedijinoj

-0.41

POSITIVE LOGITS

 Roskov

0.47

NSYLVANIA

0.43

Platforms

0.40

 nonUne

0.40

mlı

0.40

 NSCoder

0.40

deauna

0.40

verwijspagina

0.39

película

0.39

Wirtschaft

0.38

Activations Density 0.330%

mathematical constructs and relationships within formal definitions

then

No Comments

No Known Activations