INDEX

Explanations

phrases indicating high levels of quality or recommendation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.82

-0.79

tableFuture

-0.68

 Вікіпе

-0.66

 scrubs

-0.65

onel

-0.65

-0.63

<b>

-0.63

ubation

-0.62

 Mabel

-0.61

POSITIVE LOGITS

angliski

0.95

highly

0.95

%")

0.92

 Highly

0.91

Highly

0.91

%"),

0.91

ientras

0.88

 متحده

0.88

 highly

0.86

 muſt

0.85

Activations Density 0.139%

phrases indicating high levels of quality or recommendation

No Comments

No Known Activations