INDEX

Explanations

references to large-scale systems or studies

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

Preceding "scale" or "large"

np_acts-logits-general · gemini-2.0-flash

large-scale

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

demic

-0.75

<?,

-0.73

yntaxException

-0.72

̓

-0.70

UnitTesting

-0.68

︎

-0.65

ActionTypes

-0.64

theless

-0.63

 Gables

-0.62

жидан

-0.59

POSITIVE LOGITS

 scale

1.19

scale

1.03

 Scale

0.89

Scale

0.85

mouth

0.82

 SCALE

0.80

cale

0.79

SCALE

0.75

 enough

0.72

 amounts

0.71

Activations Density 0.094%

references to large-scale systems or studies

Preceding "scale" or "large"

large-scale

No Comments

No Known Activations