INDEX

Explanations

expressions related to complexity and richness in human experiences

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

 even

-0.34

ÓS

-0.33

 form

-0.33

 OMITBAD

-0.32

 shape

-0.31

 acer

-0.31

 appear

-0.31

 stronger

-0.30

cal

-0.30

 Shape

-0.30

POSITIVE LOGITS

PreInfinity

0.67

WriteBarrier

0.65

Климаты

0.65

SourceChecksum

0.64

 للمعارف

0.61

ftagPool

0.59

visející

0.59

setVerticalGroup

0.59

niſſe

0.58

neſs

0.57

Activations Density 0.067%

expressions related to complexity and richness in human experiences

No Comments

No Known Activations