INDEX

Explanations

structured data representations, such as JSON format elements

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

<eos>

-0.66

-0.63

↵↵

-0.56

-0.54

-0.51

.</

-0.51

,”

-0.50

…

-0.49

</h2>

-0.48

-0.46

POSITIVE LOGITS

 myſelf

1.01

 CreateTagHelper

1.00

 Monfieur

0.91

NameInMap

0.87

CloseOperation

0.87

 itſelf

0.85

kloped

0.84

цездатний

0.83

 ſeveral

0.82

 himſelf

0.82

Activations Density 0.103%

structured data representations, such as JSON format elements

No Comments

No Known Activations