INDEX

Explanations

terminology related to processes and actions in technical descriptions

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

performing technical operations

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

########.

-0.49

Bioaccumulative

-0.41

WriteBarrier

-0.41

 JScrollPane

-0.39

Personensuche

-0.38

GraphicsUnit

-0.38

slidesToShow

-0.37

endaft

-0.36

anguera

-0.35

 industria

-0.35

POSITIVE LOGITS

ⓧ

0.54

RegressionTest

0.52

rawDesc

0.50

 UIAlert

0.47

 كومونز

0.45

PyExc

0.43

 snippetHide

0.43

mind

0.42

).__

0.42

 consultato

0.41

Activations Density 0.152%

terminology related to processes and actions in technical descriptions

performing technical operations

No Comments

No Known Activations