INDEX

Explanations

references to indicators and parameters used in scientific measurement or analysis

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

the beginning of documents or inputs, activating strongly on the initial token.

oai_token-act-pair · claude-3-7-sonnet-20250219 Triggered by @neilrathi

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

Prompts (Dashboard)

24,576 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

ConstraintMaker

-0.67

MLLoader

-0.65

KommentareTeilen

-0.64

 ब्रेकडाउन

-0.63

 queſta

-0.62

NUMX

-0.61

IsContent

-0.60

IntoConstraints

-0.60

 فريبيس

-0.59

OGND

-0.59

POSITIVE LOGITS

<bos>

0.56

The

0.39

<strong>

0.35

0.33

0.32

0.31

Activations Density 50.614%

references to indicators and parameters used in scientific measurement or analysis

the beginning of documents or inputs, activating strongly on the initial token.

No Comments

No Known Activations