INDEX

Explanations

specific technical or scientific terms and their relationships within detailed descriptions

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

technical terms or names

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

loff

-0.46

fum

-0.46

hum

-0.43

urre

-0.41

itra

-0.40

dom

-0.38

Exo

-0.36

 boarding

-0.36

amen

-0.35

bea

-0.35

POSITIVE LOGITS

rungsseite

0.65

 незавершена

0.60

MemoryWarning

0.59

 utafitiHapana

0.58

 المعيارى

0.56

 OFDb

0.55

parsedMessage

0.53

 AssemblyTitle

0.52

GraphicsUnit

0.50

 ObjectParameter

0.50

Activations Density 0.068%

specific technical or scientific terms and their relationships within detailed descriptions

technical terms or names

No Comments

No Known Activations