INDEX

Explanations

numerical values or references to specific values in data contexts

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

hexadecimal constants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

archiviato

-0.50

帖最后由

-0.49

 propOrder

-0.49

 ویکی‌آمباردا

-0.47

principalColumn

-0.46

 विश्वसनीयता

-0.44

 Sommers

-0.43

Glej

-0.41

Datuak

-0.41

Rüyada

-0.39

POSITIVE LOGITS

########.

0.59

0.56

 nitro

0.52

 Nitro

0.51

Kyoto

0.48

ſelves

0.48

HttpGet

0.48

NgModule

0.46

 Trolley

0.46

 tfsi

0.46

Activations Density 0.002%

numerical values or references to specific values in data contexts

hexadecimal constants

No Comments

No Known Activations