INDEX

Explanations

terms related to specific qualities or attributes of products and legal disclaimers

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

ioutil

-1.47

 ioutil

-0.43

Hochspringen

-0.39

Dispose

-0.36

ReadAll

-0.34

 puffed

-0.34

 showered

-0.33

嘧

-0.33

due

-0.33

POSITIVE LOGITS

strconv

1.66

ConstraintMaker

1.27

 strconv

1.26

Varint

1.16

Revenir

0.87

LookAnd

0.81

gonic

0.78

 plufieurs

0.74

 ainfi

0.73

 mauva

0.67

Activations Density 0.002%

terms related to specific qualities or attributes of products and legal disclaimers

No Comments

No Known Activations