INDEX

Explanations

comparative and evaluative questions regarding numerical values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

-0.80

-0.65

-0.63

-0.60

-0.59

–

-0.59

and

-0.58

-0.57

POSITIVE LOGITS

 незавершена

1.28

 AssemblyCulture

1.28

featureID

1.25

 ModelExpression

1.25



1.20

AndEndTag

1.19

 kaarangay

1.18

 Мексичка

1.17

PreferredItem

1.16

 myſelf

1.15

Activations Density 2.266%

comparative and evaluative questions regarding numerical values

No Comments

No Known Activations