INDEX

Explanations

discussions around conditions and qualifiers, often related to logic or comparisons

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-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

incón

-0.50

劲

-0.48

cipta

-0.48

 gyhoeddwyd

-0.47

-----

-0.47

 tricot

-0.47

gelopen

-0.46

ard

-0.46

potranspiration

-0.45

 nemlig

-0.45

POSITIVE LOGITS

 otherwise

0.85

 OTHERWISE

0.84

otherwise

0.78

 else

0.77

Otherwise

0.76

 alternatively

0.73

 zumindest

0.71

 Otherwise

0.70

CreateModel

0.69

 surla

0.67

Activations Density 0.252%

discussions around conditions and qualifiers, often related to logic or comparisons

No Comments

No Known Activations