INDEX

Explanations

actions related to breaking or changing state in various contexts

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

"break" or "breaking"

np_acts-logits-general · gemini-2.0-flash

break

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

 mouseY

-0.57

jubljana

-0.53

ActionCreators

-0.52

 виправивши

-0.51

himovic

-0.50

 بيها

-0.48

Gruß

-0.48

posited

-0.48

Datum

-0.48

 ujednoznacz

-0.47

POSITIVE LOGITS

 breaking

1.58

 break

1.48

 breaks

1.44

Breaking

1.42

 Breaking

1.42

 BREAK

1.40

 broke

1.39

 Break

1.39

 Breaks

1.38

breaking

1.38

Activations Density 0.112%

actions related to breaking or changing state in various contexts

"break" or "breaking"

break

No Comments

No Known Activations