INDEX

Explanations

terms related to typesetting and text composition processes

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

typesetting and text composition

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

ⓧ

-0.44

 duración

-0.42

 Italijani

-0.42

 estação

-0.40

Extension

-0.39

 duração

-0.38

 possibilidade

-0.36

Espèce

-0.36

AnimationsModule

-0.36

 sentito

-0.36

POSITIVE LOGITS

bufio

0.56

 JTextArea

0.54

TypedDataSet

0.53

 reread

0.53

NewReader

0.53

ContentAsync

0.51

 esternos

0.50

 fonts

0.50

 TextInput

0.50

istoitu

0.49

Activations Density 0.057%

terms related to typesetting and text composition processes

typesetting and text composition

No Comments

No Known Activations