INDEX

Explanations

verbs indicating actions taken or experiences had

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

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_1m/average_l0_107

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

1,048,576

Data Type

float32

Hook Name

blocks.12.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

IntoConstraints

-0.86

 MainAxisSize

-0.67

 Wiktionnaire

-0.61

phosa

-0.57

 Moslem

-0.57

 hermit

-0.56

InSection

-0.55

BeginContext

-0.55

 populaire

-0.55

sidemargin

-0.55

POSITIVE LOGITS

 Wikimedijinoj

0.52

+#+

0.49

 required

0.48

atier

0.47

 tabPage

0.47

 drew

0.46

SBATCH

0.46

expandindo

0.45

 jedenfalls

0.45

 mostrarán

0.45

Activations Density 0.005%

verbs indicating actions taken or experiences had

No Comments

No Known Activations