INDEX

Explanations

specific programming languages or technology-related terms

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

ftagPool

-0.44

 nakalista

-0.42

Aholisi

-0.40

новниш

-0.39

 autorytatywna

-0.37

 présidentielle

-0.36

WebElementEntity

-0.35

 corporativa

-0.33

 respectivamente

-0.33

 oprot

-0.33

POSITIVE LOGITS

0.54

########.

0.53

 المعيارى

0.51

rief

0.51

esModule

0.50

MLLoader

0.49

:+:

0.48

PerformLayout

0.47

."]

0.47

oka

0.46

Activations Density 0.003%

specific programming languages or technology-related terms

No Comments

No Known Activations