INDEX

Explanations

references to improvement or enhancement

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

skosten

-0.69

 fiscale

-0.64

Vorte

-0.62

 Skyscanner

-0.61

lectricité

-0.60

 Steinberg

-0.60

Kos

-0.60

Jackie

-0.60

geology

-0.60

 jLabel

-0.59

POSITIVE LOGITS

 BETTER

1.45

Better

1.44

better

1.40

 Better

1.40

 better

1.40

 bedre

0.99

 beter

0.98

 besseren

0.93

 mieux

0.93

 betterment

0.92

Activations Density 0.060%

references to improvement or enhancement

No Comments

No Known Activations