INDEX

Explanations

sequences of underscores or spacing patterns in a structured format, possibly for code or formatting purposes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 unknownFields

-0.62

TagHelpers

-0.58

<eos>

-0.56

 late

-0.51

Late

-0.48

');?>

-0.47

RegistryLite

-0.47

Vidu

-0.47

Vidite

-0.46

RequestMapping

-0.46

POSITIVE LOGITS

enumi

0.80

 pleaſure

0.80

 greateſt

0.76

amaño

0.72

 purpoſe

0.71

 peines

0.69

 uſe

0.68

 Bayard

0.68

sự

0.67

 houſe

0.66

Activations Density 0.545%

sequences of underscores or spacing patterns in a structured format, possibly for code or formatting purposes

No Comments

No Known Activations