INDEX

Explanations

mentions of files and file-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 reaſon

-0.87

 pleaſure

-0.86

 Monfieur

-0.85

 Chriftian

-0.83

 greateſt

-0.79

 Pingback

-0.78

 uſed

-0.75

 ſeveral

-0.71

 itſelf

-0.71

 Theſe

-0.71

POSITIVE LOGITS

 file

4.36

file

3.63

 File

3.46

File

3.24

 FILE

2.90

 files

2.88

文件

2.52

FILE

2.51

 Files

2.29

ファイル

2.26

Activations Density 0.093%

mentions of files and file-related terms

No Comments

No Known Activations