INDEX

Explanations

code structures and syntax related to programming and data processing

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

surveys adapter

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

Personendaten

-0.52

ftagPool

-0.52

outSlope

-0.52

⤹

-0.50

 ModelExpression

-0.49

Seeder

-0.45

MessageTagHelper

-0.42

 ComVisible

-0.42

SequentialGroup

-0.42

setVerticalGroup

-0.41

POSITIVE LOGITS

forhold

0.40

InputTagHelper

0.39

当たり

0.38

ंदीखरीदारी

0.37

 Ursprung

0.36

 omkring

0.36

Betty

0.36

 artig

0.36

OpenHelper

0.35

ValueStyle

0.35

Activations Density 0.029%

code structures and syntax related to programming and data processing

surveys adapter

No Comments

No Known Activations