INDEX

Explanations

terms related to mathematical or technical expressions

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Slf

-0.36

 considération

-0.34

Cordialement

-0.33

 المت

-0.33

DIPSETTING

-0.33

RegressionTest

-0.32

 enkelt

-0.31

tre

-0.31

tre

-0.30

hObject

-0.30

POSITIVE LOGITS

 tartalomajánló

0.69

CloseOperation

0.68

httphttps

0.66

 informée

0.65

 transfieras

0.65

 Roskov

0.64

">//

0.63

Билгалдахарш

0.62

ंदीखरीदारी

0.61

الحياه

0.60

Activations Density 0.006%

terms related to mathematical or technical expressions

No Comments

No Known Activations