INDEX

Explanations

instances of classes and types related to reading and processing data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

endwhile

-0.67

 muualla

-0.64

DateTimeField

-0.61

 concernant

-0.61

 politiet

-0.60

 autorytatywna

-0.58

triangleq

-0.57

bietet

-0.56

 televisiva

-0.56

namentales

-0.56

POSITIVE LOGITS

Reader

1.06

 Extraction

0.94

 extraction

0.93

 Reader

0.90

alt

0.84

Late

0.84

late

0.83

reader

0.82

Extraction

0.80

extraction

0.78

Activations Density 0.095%

instances of classes and types related to reading and processing data

No Comments

No Known Activations