INDEX

Explanations

references to inventions or innovations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

dfrac

-0.90

bkg

-0.82

ViewFeatures

-0.79

HasForeignKey

-0.79

 Lewandowski

-0.77

emia

-0.77

OneToMany

-0.77

 Barbar

-0.76

 Bergmann

-0.75

intellij

-0.75

POSITIVE LOGITS

 pine

0.77

 fritas

0.74

 amargo

0.72

 Lament

0.71

 Sentry

0.70

 Dage

0.70

िये

0.70

ations

0.69

 Shakspeare

0.69

 Cart

0.69

Activations Density 0.085%

references to inventions or innovations

No Comments

No Known Activations