INDEX

Explanations

references to specific scientific terms and concepts, particularly those related to quantum mechanics

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

 Lichten

-0.76

board

-0.73

 Martine

-0.72

 Dumont

-0.72

 Lockwood

-0.71

mtable

-0.71

 Fonten

-0.69

 Croydon

-0.69

 Berkeley

-0.69

Georges

-0.68

POSITIVE LOGITS

Sif

0.90

Sif

0.84

 MACH

0.83

 Peel

0.76

 Argo

0.75

 ujednoznacz

0.73

 Wade

0.73

 Mach

0.73

 Ramb

0.72

 resultMap

0.71

Activations Density 0.055%

references to specific scientific terms and concepts, particularly those related to quantum mechanics

No Comments

No Known Activations