INDEX

Explanations

patterns of code changes or class/method definitions

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

programming language keywords

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

хьтан

-0.60

posedge

-0.59

 kasarigan

-0.58

原始内容存档于

-0.55

bootstrapcdn

-0.54

WriteBarrier

-0.52

$_"

-0.51

 queſta

-0.51

:✨

-0.50

 betweenstory

-0.49

POSITIVE LOGITS

IContainer

0.49

 initComponents

0.44

">//

0.42

 <<<<<<<<<<<<<<

0.39

gulier

0.37

XmlAccessorType

0.35

RenderAtEndOf

0.35

 EClass

0.35

0.34

oding

0.33

Activations Density 0.005%

patterns of code changes or class/method definitions

programming language keywords

No Comments

No Known Activations

patterns of code changes or class/method definitions

programming language keywords

No Comments

No Known Activations