INDEX

Explanations

references to cookies and chocolate-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 الحره

-0.76

Cort

-0.70

 Castilla

-0.67

 reft

-0.65

Vim

-0.63

 Griechen

-0.62

Mir

-0.61

Vy

-0.61

<.

-0.60

Conway

-0.58

POSITIVE LOGITS

 chocolate

0.96

 Chocolate

0.94

bru

0.93

chocolate

0.91

Chocolate

0.91

OCOLATE

0.90

🍫

0.87

cookie

0.86

 chocol

0.85

 brown

0.85

Activations Density 0.346%

references to cookies and chocolate-related terms

No Comments

No Known Activations

references to cookies and chocolate-related terms

No Comments

No Known Activations