INDEX

Explanations

phrases that indicate critique and improvement in academic writing

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

authors acknowledging corrections

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

AddTagHelper

-0.65

 pinulongan

-0.60

Tembelea

-0.58

WebVitals

-0.57

 ContentValues

-0.56

XmlAccessType

-0.56

iſen

-0.55

 PeEnEo

-0.54

 Вікі

-0.53

curities

-0.53

POSITIVE LOGITS

帖最后由

0.33

fjspx

0.33

 previously

0.31

 added

0.30

 conceded

0.29

خر

0.29

uvwxyz

0.29

 genel

0.29

 increased

0.28

侧

0.27

Activations Density 0.078%

phrases that indicate critique and improvement in academic writing

authors acknowledging corrections

No Comments

No Known Activations