INDEX

Explanations

programming terminology, particularly related to utility functions and coding structure

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

archaic reflexive pronouns

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

-0.69

UnusedPrivate

-0.67

 autorytatywna

-0.59

DockStyle

-0.59

 Infórmanos

-0.58

новниш

-0.57

Autoritní

-0.57

 initComponents

-0.55

CloseOperation

-0.53

 propOrder

-0.53

POSITIVE LOGITS

 itſelf

0.40

 myſelf

0.37

 vizsg

0.37

 themſelves

0.35

 himſelf

0.33

 ſche

0.32

 gastron

0.31

 loam

0.30

 juſ

0.30

 muñ

0.29

Activations Density 0.001%

programming terminology, particularly related to utility functions and coding structure

archaic reflexive pronouns

No Comments

No Known Activations