INDEX

Explanations

references to statistical or mathematical modeling concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_22/width_16k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_mlp_out

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

Ha

-0.46

va

-0.44

Ha

-0.43

 рее

-0.43

 kaynağından

-0.43

 PyTuple

-0.43

@[+][

-0.42

 Horne

-0.42

 Numerology

-0.41

celotti

-0.41

POSITIVE LOGITS

 الحره

0.70

PYX

0.69

mybatisplus

0.68

AddTagHelper

0.66

ValueStyle

0.59

 nakalista

0.59

 ویکی‌پدیا

0.58

 autorytatywna

0.57

 Wicidata

0.57

Jeografia

0.56

Activations Density 4.933%

references to statistical or mathematical modeling concepts

No Comments

No Known Activations