INDEX

Explanations

technical terms and references in scientific or programming contexts

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

allowing retrieval methods

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



-0.66

RegistryLite

-0.63

 CURIAM

-0.53

 Bioaccumulative

-0.52

 समीक्षक

-0.50

VolleyError

-0.50

recated

-0.47

MessageTagHelper

-0.46

NUMX

-0.46

 useNavigate

-0.46

POSITIVE LOGITS

rentiel

0.42

 allowing

0.40

openqa

0.40

 позволит

0.38

allows

0.37

allowing

0.37

 autorytatywna

0.36

bo

0.35

ker

0.35

 permettant

0.35

Activations Density 0.180%

technical terms and references in scientific or programming contexts

allowing retrieval methods

No Comments

No Known Activations