INDEX

Explanations

sections indicating important contributions or acknowledgments in a text

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

océan

-0.63

-0.61

==

-0.55

 Nilsson

-0.53

to

-0.53

 fièvre

-0.53

ליט

-0.53

et

-0.53

POSITIVE LOGITS

########.

0.87

 ujednoznacz

0.82

$_"

0.82

 unſ

0.80

]='\

0.79

Zy

0.78

ſelves

0.76

NUMX

0.76

 Anſ

0.74

ductory

0.74

Activations Density 0.377%

sections indicating important contributions or acknowledgments in a text

No Comments

No Known Activations