INDEX

Explanations

references to the term "user" and its variations in context

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

user

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

 cherchés

-0.48

 ARRANG

-0.45

siguiente

-0.45

Билгалдахарш

-0.41

IGraphics

-0.40

arrêter

-0.40

setVerticalGroup

-0.40

 ComVisible

-0.39

 propOrder

-0.39

itself

-0.38

POSITIVE LOGITS

 user

0.75

 User

0.69

 USER

0.63

 users

0.62

user

0.61

 person

0.59

USER

0.57

 getUser

0.57

 Users

0.56

User

0.55

Activations Density 0.018%

references to the term "user" and its variations in context

user

No Comments

No Known Activations