INDEX

Explanations

code-related terms and identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

Personendaten

-1.15

IsContent

-0.79

PreferredItem

-0.77

BindingResult

-0.76

featureID

-0.76

GeneratedCode

-0.73

XmlAccessType

-0.72

 Wikimedijinoj

-0.72

 propOrder

-0.71

CodedInputStream

-0.69

POSITIVE LOGITS

</tr>

0.52

tyard

0.50

 Learning

0.49

0.48

digheden

0.48

szy

0.48

 tenderness

0.47

etype

0.47

 Lear

0.47

 Wali

0.47

Activations Density 0.032%

code-related terms and identifiers

No Comments

No Known Activations