INDEX

Explanations

terms related to specific types of neurons and their properties

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

neurons and neurotransmitters

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

 happening

-0.44

(=)

-0.36

queryInterface

-0.34

 coroa

-0.32

 dropping

-0.31

 crown

-0.31

<b>

-0.31

 canina

-0.31

 geda

-0.31

brand

-0.31

POSITIVE LOGITS

 informée

0.61

 Chwiliwch

0.58

 tartalomajánló

0.57

tagHelperRunner

0.57

 שוליים

0.56

Tikang

0.56

 neurons

0.52

hoeddwyd

0.52

 CanadaChoose

0.51

 CreateTagHelper

0.51

Activations Density 0.238%

terms related to specific types of neurons and their properties

neurons and neurotransmitters

No Comments

No Known Activations