INDEX

Explanations

mathematical forms or structures presented in a document

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 оригіналу

-0.69

#+#

-0.66

期刊论文

-0.59

usercontent

-0.57

̸

-0.57

."));

-0.57

 مرئيه

-0.56

 raiſ

-0.56

นะนำ

-0.54

 againſt

-0.53

POSITIVE LOGITS

 phrases

0.55

yntaxException

0.54

 <<<<<<<<<<<<<<

0.53

 prayer

0.52

kHz

0.51

UVWXYZ

0.51

Prayer

0.49

pruch

0.49

 phrase

0.48

sivu

0.48

Activations Density 0.742%

mathematical forms or structures presented in a document

No Comments

No Known Activations