INDEX

Explanations

terms related to robustness or stability in various contexts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 AssemblyCulture

-0.85

\{\\

-0.73

DockStyle

-0.68

Билгалдахарш

-0.68

 Numerade

-0.68

 للمعارف

-0.67

 esternos

-0.66

󠁴

-0.64

RegressionTest

-0.63

UserScript

-0.63

POSITIVE LOGITS

人

0.91

日

0.67

人

0.56

他

0.55

月

0.54

인

0.50

 obviously

0.48

 Obviously

0.47

일

0.46

Obviously

0.45

Activations Density 0.101%

terms related to robustness or stability in various contexts

No Known Activations