INDEX

Explanations

mentions of the word "Hu" and its variations, indicating a focus on references to a specific entity or type of identity

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

starting with "Hu" or containing "hu"

np_acts-logits-general · gemini-2.0-flash

Hu, hud, hun related to citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

gdx

-0.73

 ProtoMessage

-0.73

RequestMethod

-0.71

Aholisi

-0.69

الحياه

-0.69

OrNil

-0.67

Parcelize

-0.66

SRS

-0.63

Abp

-0.62

 Commercial

-0.61

POSITIVE LOGITS

Hu

0.65

Hust

0.59

Hud

0.58

hu

0.57

Hu

0.52

ipeline

0.50

usermodel

0.48

 Sagan

0.48

bub

0.48

Rotation

0.47

Activations Density 0.079%

mentions of the word "Hu" and its variations, indicating a focus on references to a specific entity or type of identity

starting with "Hu" or containing "hu"

Hu, hud, hun related to citations

No Comments

No Known Activations