INDEX

Explanations

terms related to health, therapy, and familial relationships

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

followed by "-based", "-driven", or "-oriented"

np_acts-logits-general · gemini-2.0-flash

-driven, -based, -oriented concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

\{\\

-0.93

TagMode

-0.83

TagHelper

-0.74

✨:

-0.73

TestingModule

-0.65

 incomplète

-0.64

OGND

-0.63

principalColumn

-0.62

 мәкал

-0.59

antası

-0.59

POSITIVE LOGITS

 based

1.67

 driven

1.56

 oriented

1.49

based

1.40

 related

1.35

 BASED

1.33

 Based

1.32

 inspired

1.30

Based

1.24

 Driven

1.24

Activations Density 1.298%

terms related to health, therapy, and familial relationships

followed by "-based", "-driven", or "-oriented"

-driven, -based, -oriented concepts

No Comments

No Known Activations