INDEX

Explanations

structural elements and properties related to code and data formats

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

references to figures or text

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

RegressionTest

-0.65



-0.56

SequentialGroup

-0.56

AccessorTable

-0.54

 betweenstory

-0.53

AndEndTag

-0.49

TagMode

-0.48

 zawsze

-0.46

AnimationsModule

-0.46

awtextra

-0.46

POSITIVE LOGITS

extAlignment

0.60

بوابة

0.50

 transfieras

0.49

DOCTYPE

0.49

PyExc

0.47

 MessageBoxIcon

0.46

glGen

0.45

UIControlState

0.45

setDo

0.43

Портал

0.42

Activations Density 0.082%

structural elements and properties related to code and data formats

references to figures or text

No Comments

No Known Activations