INDEX

Explanations

specific structural elements or commands within coding or programming documentation

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

 Paglinawan

-1.06

StoryboardSegue

-0.80

 myſelf

-0.78

الدراسه

-0.76

 ſtate

-0.73

 Shakspeare

-0.69

 Jefus

-0.69

 itſelf

-0.69

 Monfieur

-0.68

Efq

-0.68

POSITIVE LOGITS

<eos>

0.69

↵↵↵

0.57

{~

0.53

 rowspan

0.53

 mathvariant

0.52

0.51

...

0.51

nonatomic

0.50

Activations Density 0.218%

specific structural elements or commands within coding or programming documentation

No Comments

No Known Activations