INDEX

Explanations

elements related to user data and attributes in a system

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

Formatting and delimiters: ", :, [, ], {, %, etc

np_acts-logits-general · gemini-2.0-flash

user profile attributes

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 defendants

-0.61

ValueStyle

-0.56

 recipients

-0.54

 plaintiffs

-0.53

 umane

-0.52

Authors

-0.52

 authors

-0.52

 culprits

-0.52

 appellants

-0.50

選手

-0.50

POSITIVE LOGITS

 roles

0.76

 role

0.75

 hobbies

0.74

 profile

0.72

 Profile

0.72

 Roles

0.70

 preferences

0.69

 avatar

0.67

 gender

0.66

 Role

0.66

Activations Density 0.208%

elements related to user data and attributes in a system

Formatting and delimiters: ", :, [, ], {, %, etc

user profile attributes

No Comments

No Known Activations