INDEX

Explanations

programming-related terms and annotations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_12/width_16k/average_l0_130

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 JADX

-0.55

 infinitely

-0.52

likan

-0.51

 Photocase

-0.51

MeasureSpec

-0.51

Victory

-0.50

 infinite

-0.49

 victory

-0.49

 Victory

-0.47

 MotionEvent

-0.47

POSITIVE LOGITS

stereotype

2.44

 stereotype

0.85

Stere

0.69

 Stere

0.68

 stereotypes

0.67

stere

0.64

 stereotyp

0.63

RestController

0.62

 stere

0.52

Stateless

0.49

Activations Density 0.028%

programming-related terms and annotations

No Comments

No Known Activations

programming-related terms and annotations

No Comments

No Known Activations