INDEX

Explanations

terms related to editorial corrections and manuscript review processes

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

Spelling, grammar, and typographical errors

np_acts-logits-general · gemini-2.0-flash

spelling and grammar corrections

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.30.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

AddTagHelper

-0.60

MeasureSpec

-0.48

UnusedPrivate

-0.45

 xoay

-0.44

 informée

-0.41

цездатний

-0.40

dafx

-0.39

 Cord

-0.38

 INTERESAR

-0.38

 externi

-0.37

POSITIVE LOGITS

 grammar

1.47

 grammatical

1.34

 Grammar

1.22

 spelling

1.22

Grammar

1.21

grammar

1.20

 gram

1.20

 gramm

1.06

 typos

1.05

 GRAM

1.03

Activations Density 0.387%

terms related to editorial corrections and manuscript review processes

Spelling, grammar, and typographical errors

spelling and grammar corrections

No Comments

No Known Activations