INDEX

Explanations

symbols and formatting used in mathematical or technical documentation

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

Big

-0.39

 Cuts

-0.38

Hol

-0.37

Cuts

-0.37

 consolidate

-0.37

 slides

-0.37

 cuts

-0.36

 Kirkpatrick

-0.36

 Opfer

-0.35

Cut

-0.35

POSITIVE LOGITS

 noDo

0.69

Diweddarwch

0.64

parsedMessage

0.61

 gynhyrchwyd

0.57

ImageContext

0.56

 Roskov

0.53

MLLoader

0.52

 الرياضيه

0.52

estacks

0.49

elemField

0.49

Activations Density 0.030%

symbols and formatting used in mathematical or technical documentation

No Comments

No Known Activations