INDEX

Explanations

expressions of gratitude and acknowledgment from authors

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

"Thank" followed by a person or group

np_acts-logits-general · gemini-2.0-flash

expressions of thanks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

ViewImports

-0.73

tvguidetime

-0.71

 kaarangay

-0.71

 beginnetje

-0.66

parsedMessage

-0.65

AddTagHelper

-0.64

unknownFields

-0.62

InitVars

-0.61

ędziesz

-0.61

 kasarigan

-0.60

POSITIVE LOGITS

 special

1.02

special

0.86

 express

0.79

Special

0.79

 especiales

0.76

 Special

0.76

 spéciale

0.74

 sincere

0.74

 spéciaux

0.73

 especial

0.73

Activations Density 0.115%

expressions of gratitude and acknowledgment from authors

"Thank" followed by a person or group

expressions of thanks

No Comments

No Known Activations