INDEX

Explanations

coding syntax and structure related to logical operations

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

code snippets following `code`

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

ագրություններ

-0.64

 initComponents

-0.60

 Escolar

-0.51

getOutputStream

-0.49

 épisode

-0.46

 архивлан

-0.46

=$?

-0.45

 CallOverrides

-0.45

IsMutable

-0.44

 Biôgrafia

-0.43

POSITIVE LOGITS

 quæ

0.52

<i>

0.44

 nahilalakip

0.44

 avoient

0.44

paramref

0.43

__(/*!

0.41

WEBPACK

0.41

romechanical

0.40

useppe

0.40

 westfalen

0.39

Activations Density 0.034%

coding syntax and structure related to logical operations

code snippets following `code`

No Comments

No Known Activations