INDEX

Explanations

formatting or structural elements in code

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

noscript

-0.63

MLLoader

-0.57

 plus

-0.52

entee

-0.50

QUADS

-0.49

ácio

-0.49

plus

-0.47

 Plus

-0.47

Plus

-0.47

encodeWith

-0.47

POSITIVE LOGITS

 purpoſe

0.77

 للاسماء

0.69

 houſe

0.67

 ſtate

0.65

 Majefty

0.65

OGND

0.63

Personendaten

0.63

 themſelves

0.62

TypedDataSet

0.62

enumi

0.61

Activations Density 0.018%

formatting or structural elements in code

No Comments

No Known Activations