INDEX

Explanations

programming code elements related to data manipulation and visualization in Python

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

pandas df.loc indexing

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

-0.45

༘

-0.39

-0.36

 laajenta

-0.36

GraphicsUnit

-0.35

機場

-0.34

Ifc

-0.33

BaseAdapter

-0.33

 will

-0.33

 achieved

-0.32

POSITIVE LOGITS

 autorytatywna

0.65

 kaarangay

0.63

getColumnIndex

0.59

AndEndTag

0.57

principalColumn

0.54

GOTREF

0.51

usercontent

0.50

__(/*!

0.50

kháu

0.49

 themſelves

0.49

Activations Density 0.015%

programming code elements related to data manipulation and visualization in Python

pandas df.loc indexing

No Comments

No Known Activations