INDEX

Explanations

phrases that indicate time periods and durations

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

Time period/duration

np_acts-logits-general · gemini-2.0-flash

short time durations

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

 surla

-0.74

 FetchType

-0.73

 }}"></

-0.69

MLLoader

-0.67

*/;

-0.66

ondissement

-0.65

原始内容存档于

-0.64

complexContent

-0.62

providedIn

-0.62

 mips

-0.61

POSITIVE LOGITS

 only

1.72

 just

1.72

 less

1.63

just

1.47

only

1.46

 slechts

1.46

わずか

1.43

 apenas

1.41

Only

1.41

 mere

1.35

Activations Density 0.443%

phrases that indicate time periods and durations

Time period/duration

short time durations

No Comments

No Known Activations