INDEX

Explanations

sentences indicating the successful completion of a search or discovery process

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

出版年

-0.67

]-->

-0.65

eraard

-0.63

 continú

-0.61

 nécessairement

-0.60

 vždy

-0.59

detectChanges

-0.59

 étoient

-0.58

 Posteriormente

-0.58

 Даль

-0.58

POSITIVE LOGITS

 finally

2.02

finally

1.83

 enfin

1.80

 endlich

1.76

 FINALLY

1.73

 Finally

1.69

 наконец

1.68

Finally

1.65

 finalmente

1.56

reszcie

1.54

Activations Density 0.493%

sentences indicating the successful completion of a search or discovery process

No Comments

No Known Activations