INDEX

Explanations

phrases related to training or educational programs

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

"Training" token

np_acts-logits-general · gemini-2.0-flash

training process and context

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bootstrapcdn

-0.56

cemia

-0.53

ysty

-0.52

SerializedName

-0.51

addFlags

-0.49

internalType

-0.49

зировать

-0.49

__(

-0.49

 Antara

-0.49

 laterale

-0.49

POSITIVE LOGITS

wreck

0.81

 Roskov

0.77

 ویکی‌پدیا

0.67

 propOrder

0.66



0.62

 AttributeSet

0.60

 Waray

0.59

ifflin

0.58

 FetchType

0.58

 montage

0.56

Activations Density 0.074%

phrases related to training or educational programs

"Training" token

training process and context

No Comments

No Known Activations