INDEX

Explanations

SQL-related terms and structures in code snippets

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

user queries about users

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

✭✭

-0.43

 camera

-0.33

 uintptr

-0.32

dyna

-0.32

measurement

-0.31

nav

-0.31

 desp

-0.31

 Nagel

-0.31

 transi

-0.30

 camer

-0.30

POSITIVE LOGITS

rungsseite

0.74

PMailer

0.66

 betweenstory

0.59

enderror

0.54

 ***!

0.52

 CanadaChoose

0.52

 Grüsse

0.52

 AssemblyCompany

0.50

 particulières

0.50

ⓧ

0.49

Activations Density 0.005%

SQL-related terms and structures in code snippets

user queries about users

No Comments

No Known Activations