INDEX

Explanations

syntax or formatting structures, particularly with an emphasis on placeholder text or command formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

__':

-0.77

makeText

-0.69

 MainAxisSize

-0.65

getClassLoader

-0.63

__":

-0.62

.},

-0.61

__":

-0.61

发表于

-0.60

 typelib

-0.60

endphp

-0.56

POSITIVE LOGITS

 незавершена

0.54

 mergeFrom

0.52

NameInMap

0.50

multirow

0.49

ilités

0.48

matchCondition

0.48

ỡng

0.46

 subd

0.44

 pomeriggio

0.44

Ks

0.43

Activations Density 0.068%

syntax or formatting structures, particularly with an emphasis on placeholder text or command formats

No Comments

No Known Activations