INDEX

Explanations

time references within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

principalTable

-0.83

ftagPool

-0.75

WithIOException

-0.73

 pinulongan

-0.72

 betweenstory

-0.70

 ujednoznacz

-0.67

Демографія

-0.67

Personendaten

-0.66

MigrationBuilder

-0.66

 kaarangay

-0.65

POSITIVE LOGITS

 writing

1.65

writing

1.40

 Writing

1.24

 write

1.22

Writing

1.22

 WRITING

1.15

write

1.09

WRITING

1.08

 writes

1.00

 escribir

0.99

Activations Density 0.154%

time references within the text

No Comments

No Known Activations