INDEX

Explanations

words associated with implications or consequences

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

Following "imp" or "sp" or "st"

np_acts-logits-general · gemini-2.0-flash

implying or impacting

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

IFrame

Link

Negative Logits

PerformLayout

-0.74

styrene

-0.73

betical

-0.73

propylene

-0.72

AndroidJUnit

-0.71

AndView

-0.70

visiae

-0.69

adecimal

-0.68

 diſt

-0.67

imuth

-0.67

POSITIVE LOGITS

 doInBackground

0.64

DockStyle

0.54

 deberes

0.52

BindingResult

0.52

 skydd

0.51

NewUrlParser

0.51

 Nare

0.49

Slf

0.48

thew

0.48

道を

0.48

Activations Density 1.030%

words associated with implications or consequences

Following "imp" or "sp" or "st"

implying or impacting

No Known Activations