INDEX

Explanations

references to training-related activities and contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

ViewInit

-0.62

SerializedName

-0.54

mbol

-0.53

 Antara

-0.53

bootstrapcdn

-0.52

Твор

-0.52

WriteTagHelper

-0.50

 Heard

-0.50

ents

-0.49

}),

-0.49

POSITIVE LOGITS

wreck

0.74

 ویکی‌پدیا

0.69

 wheels

0.66

wheels

0.64



0.64

DebuggerNonUser

0.64

 propOrder

0.63

RetentionPolicy

0.63

 programmes

0.62

 referenties

0.61

Activations Density 0.069%

references to training-related activities and contexts

No Comments

No Known Activations