INDEX

Explanations

structures and syntax related to programming and coding constructs

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

technical code snippets

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

setVerticalGroup

-0.62

fromCharCode

-0.36

rrggbb

-0.34

 AssemblyCompany

-0.33

 ویکی‌آمباردا

-0.32

Viitteet

-0.29

 esternos

-0.29

IUrlHelper

-0.29

djangoproject

-0.28

currentColor

-0.28

POSITIVE LOGITS

 infatti

0.64

 '\\;'

0.61

 nahilalakip

0.61

 inderdaad

0.60

੝

0.59

 namelijk

0.58

MemoryWarning

0.58

 pioggia

0.58

 nemlig

0.57

principalColumn

0.57

Activations Density 0.019%

structures and syntax related to programming and coding constructs

technical code snippets

No Comments

No Known Activations