INDEX

Explanations

numerical values associated with coding constructs

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

testing framework assertions

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

 defaultstate

-0.61

ItemBackground

-0.50

 Italijani

-0.49

bewerken

-0.43

 ModelExpression

-0.43

awaiter

-0.43

OrNil

-0.42

TableAdapter

-0.42

 Signalez

-0.42

 мәкал

-0.41

POSITIVE LOGITS

 nahilalakip

0.58

endphp

0.52

 ब्रेकडाउन

0.52

 AssemblyTitle

0.47

tagHelperRunner

0.46

 povezave

0.43

 Wicidata

0.43

 iconLine

0.42

UnusedPrivate

0.42

BeginContext

0.42

Activations Density 0.046%

numerical values associated with coding constructs

testing framework assertions

No Comments

No Known Activations