INDEX

Explanations

occurrences of control characters or formatting symbols

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

writeFieldEnd

-0.66

AddTagHelper

-0.56

 propOrder

-0.51

LabelTagHelper

-0.47

archiviato

-0.46

nonatomic

-0.45

 Wikiseite

-0.45

ReusableCell

-0.45

IUrlHelper

-0.44

MemoryWarning

-0.42

POSITIVE LOGITS

+=

0.58

0.54

()+

0.53

&+

0.53

]+=

0.53

0.52

0.50

}+\

0.48

+\

0.47

Activations Density 0.011%

occurrences of control characters or formatting symbols

No Comments

No Known Activations