INDEX

Explanations

terms related to capsules and their contexts, particularly in Australia

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

 McDon

-0.92

IBarButtonItem

-0.88

Iq

-0.77

 AppBundle

-0.77

lidean

-0.75

getCmp

-0.74

 Sahib

-0.74

طقة

-0.74

 GoogleFonts

-0.74

ọa

-0.73

POSITIVE LOGITS

 Capsule

0.92

 capsule

0.86

Capsule

0.83

 Lopez

0.81

 halo

0.81

 capsules

0.80

effort

0.78

alo

0.76

Halo

0.76

userInfo

0.75

Activations Density 0.058%

terms related to capsules and their contexts, particularly in Australia

No Comments

No Known Activations