INDEX

Explanations

terms and structures related to templates in programming or documentation

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

code templates

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

 duele

-0.49

 čin

-0.47

getNumber

-0.46

 Bioaccumulative

-0.44

 humaine

-0.44

 bordada

-0.44

 siento

-0.44

rijven

-0.42

 Jurí

-0.42

 tertinggi

-0.41

POSITIVE LOGITS

 template

1.29

 Template

1.23

Template

1.18

template

1.14

 templates

1.07

 Templates

1.04

 TEMPLATE

1.02

TEMPLATE

0.95

模板

0.95

Templates

0.90

Activations Density 0.008%

terms and structures related to templates in programming or documentation

code templates

No Comments

No Known Activations