INDEX

Explanations

terms related to the completion and functionality of processes or materials

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

state of readiness or completion

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

فحة

-0.44

 Autorisations

-0.41

 consideró

-0.40

jetivo

-0.38

ceğine

-0.38

lète

-0.37

ImageLoader

-0.36

 pouvoit

-0.36

 necessária

-0.36

 AssemblyCulture

-0.36

POSITIVE LOGITS

󠁮

0.55

 unread

0.51

 rubber

0.51

GIVEREF

0.49

readable

0.47

rubber

0.47

CGRectMake

0.46

IsMutable

0.46

 juicy

0.45

 unsalted

0.45

Activations Density 0.033%

terms related to the completion and functionality of processes or materials

state of readiness or completion

No Comments

No Known Activations