INDEX

Explanations

mathematical constructs and notation related to processes and causal relationships

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

stopping time and temporal order

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

Not in Any Lists

No Comments

Negative Logits

colony

-0.42

 Masking

-0.41

 myſelf

-0.41

immel

-0.41

ValueStyle

-0.41

 Cocina

-0.40

rawdę

-0.40

fanart

-0.40

-------

-0.40

glue

-0.40

POSITIVE LOGITS

yntaxException

0.59

 BoxFit

0.52

writeFieldEnd

0.50

OGND

0.50

 Italijanski

0.47

Autoritní

0.45

EndContext

0.44

+:+

0.42

SequentialGroup

0.42

angliski

0.41

Activations Density 0.001%

mathematical constructs and notation related to processes and causal relationships

stopping time and temporal order

No Comments

No Known Activations