INDEX

Explanations

quantitative references related to datasets and statistical features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 autorytatywna

-1.11

StoryboardSegue

-0.92

 Roskov

-0.88

 BeautifulSoup

-0.84

InjectAttribute

-0.82

Bigr

-0.80

 esternos

-0.79

 >=",

-0.79

 groupBox

-0.78

RTEE

-0.78

POSITIVE LOGITS

 overall

0.61

 totalt

0.52

 total

0.47

per

0.47

 totale

0.47

 Overall

0.46

 comples

0.44

overall

0.43

Overall

0.42

on

0.42

Activations Density 0.693%

quantitative references related to datasets and statistical features

No Comments

No Known Activations