INDEX

Explanations

mathematical operations and evaluations

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 AssemblyCulture

-0.88

ValueStyle

-0.67

AddTagHelper

-0.66

########.

-0.63

 nakalista

-0.58

المناصب

-0.52

 correctes

-0.52

 nahilalakip

-0.51

AndEndTag

-0.47

RTGC

-0.47

POSITIVE LOGITS

 otomatig

0.52

 informée

0.50

featureID

0.47

MessageTagHelper

0.46

Rhestr

0.45

 defaultstate

0.44

PreInfinity

0.43

𝒯

0.41

 ComVisible

0.41

 JScrollPane

0.39

Activations Density 10.660%

mathematical operations and evaluations

No Comments

No Known Activations