INDEX

Explanations

technical terms related to electrical signals and interrupt handling

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

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

dex

-0.57

Dex

-0.55

dex

-0.54

ρεί

-0.54

DEX

-0.53

<?

-0.50

spher

-0.50

lexikon

-0.49

AppCompat

-0.49

 Alvar

-0.49

POSITIVE LOGITS

 pins

0.84

pin

0.79

Pin

0.73

 Pins

0.69

 signals

0.69

Pin

0.67

pins

0.67

pin

0.62

 pull

0.62

 rising

0.62

Activations Density 0.465%

technical terms related to electrical signals and interrupt handling

No Comments

No Known Activations