INDEX

Explanations

overall performance metrics in sports or competitive contexts

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

Preceding "high", "low", or "best"

np_acts-logits-general · gemini-2.0-flash

high or best following time/career

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

 ivelany

-0.71

 виправивши

-0.66

OGND

-0.63

 eventdata

-0.63

Демографія

-0.61

 Wikimedijinoj

-0.61

providedIn

-0.60

QMetaType

-0.59

LookAnd

-0.59

 ujednoznacz

-0.58

POSITIVE LOGITS

 high

3.84

high

3.47

High

3.09

 High

3.02

 HIGH

2.84

HIGH

2.81

low

2.72

高

2.64

 higher

2.58

 tinggi

2.40

Activations Density 1.172%

overall performance metrics in sports or competitive contexts

Preceding "high", "low", or "best"

high or best following time/career

No Comments

No Known Activations