INDEX

Explanations

references to data processing and analysis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.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

 however

-0.94

 للاسماء

-0.80

however

-0.75

 però

-0.72

 tuttavia

-0.70

However

-0.69

RegressionTest

-0.67

 entanto

-0.66

прочем

-0.66

 However

-0.65

POSITIVE LOGITS

 przecież

0.74

畢竟

0.70

毕竟

0.61

 ohnehin

0.57

 already

0.57

 inherently

0.52

 involves

0.50

 involve

0.49

already

0.48

 Schließlich

0.47

Activations Density 0.919%

references to data processing and analysis

No Comments

No Known Activations