INDEX

Explanations

locations and geographical names associated with specific areas or landmarks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

bootstrapcdn

-0.84

ForValue

-0.67

 LUMP

-0.64

rawDesc

-0.61

Rptr

-0.60

 CascadeType

-0.59

 Kauai

-0.57

 Dalai

-0.57

 Alaskan

-0.56

PrototypeOf

-0.56

POSITIVE LOGITS

 Paglinawan

0.74

 posta

0.63

########.

0.57

例句

0.56

north

0.55

 poch

0.53

 Concord

0.52

tuck

0.50

 Савезне

0.50

 divertimento

0.50

Activations Density 0.303%

locations and geographical names associated with specific areas or landmarks

No Comments

No Known Activations