INDEX

Explanations

comments and documentation in code

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

code comments and file headers

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

fromnode

-0.60

 suspected

-0.49

期刊论文

-0.41

 smiles

-0.40

 suspiciously

-0.39

 smiling

-0.39

+:+

-0.39

 Fryer

-0.38

 gauche

-0.37

POSITIVE LOGITS

хьтан

0.75

 ſind

0.60

Jereo

0.57

 Roskov

0.57

WebVitals

0.56

">//

0.56

 queſta

0.54

Cyfeiriadau

0.54

 kasarigan

0.53

Gweler

0.52

Activations Density 0.004%

comments and documentation in code

code comments and file headers

No Comments

No Known Activations