INDEX

Explanations

mentions of chips and processors in a technical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_mlp_out

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

 chip

-2.17

 chips

-1.95

chip

-1.91

Chip

-1.75

 Chip

-1.74

 Chips

-1.57

chips

-1.52

Chips

-1.51

 CHIP

-1.21

CHIP

-1.20

POSITIVE LOGITS

 BoxDecoration

0.79

__*/

0.77

 становника

0.73

GOTREF

0.71

WebElementEntity

0.67

 חיצוניים

0.66

hnia

0.65

########.

0.65

abestanden

0.64

 kaarangay

0.62

Activations Density 0.299%

mentions of chips and processors in a technical context

No Comments

No Known Activations