INDEX

Explanations

keyword terms and references related to information and documentation

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

following "Tech" or "info"

np_acts-logits-general · gemini-2.0-flash

tech info and documentation

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

contentLoaded

-0.77

uxxxx

-0.69

yarnpkg

-0.68

Personendaten

-0.67

UMENT

-0.61



-0.61

 محفوظة

-0.60

otides

-0.59

دانشنامهٔ

-0.59

로운

-0.58

POSITIVE LOGITS

 Neurosci

0.42

Sys

0.39

Sci

0.38

 Toxicol

0.38

Sys

0.38

 Prep

0.37

 meds

0.36

newVal

0.35

 Atty

0.35

Avg

0.34

Activations Density 1.195%

keyword terms and references related to information and documentation

following "Tech" or "info"

tech info and documentation

No Comments

No Known Activations