INDEX

Explanations

locations or placements within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 auroit

-0.76

 Monfieur

-0.75

 quæ

-0.75

 obstante

-0.72

 fièvre

-0.71

bbene

-0.71

 pouvoit

-0.70

 dezelve

-0.69

 fidé

-0.69

 purpoſe

-0.69

POSITIVE LOGITS

 place

3.60

place

3.29

 Place

3.25

Place

3.19

 PLACE

3.00

PLACE

2.77

 places

2.76

places

2.64

 Places

2.45

Places

2.35

Activations Density 0.101%

locations or placements within the text

No Comments

No Known Activations