INDEX

Explanations

mathematical labels or annotations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

 Reſ

-0.49

 ſta

-0.48

 Inſ

-0.43

vů

-0.43

 sweter

-0.42

 ſal

-0.42

 Anſ

-0.41

 ſche

-0.39

 eſſ

-0.39

 extranjera

-0.38

POSITIVE LOGITS

TabIndex

5.00

adaptiveStyles

1.13

awtextra

0.99

 kasarigan

0.98

 AssemblyTitle

0.98

UseVisualStyle

0.97

 resourceCulture

0.93

 <<<<<<<<<<<<<<

0.91

ValueStyle

0.88

TypedDataSet

0.85

Activations Density 0.017%

mathematical labels or annotations

No Comments

No Known Activations