INDEX

Explanations

variations of the word "len" or related morphological forms

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/1-res-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.1.hook_resid_post

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 poor

-1.61

LES

-1.49

Donnell

-1.48

 suspected

-1.46

 reason

-1.46

 myself

-1.44

 dwarf

-1.43

 meant

-1.42

 large

-1.41

 slowing

-1.40

POSITIVE LOGITS

baum

2.21

esis

2.13

heimer

2.10

vironment

2.09

heim

2.05

emy

1.83

unci

1.79

ev

1.78

hower

1.75

burgh

1.73

Activations Density 0.199%

variations of the word "len" or related morphological forms

No Comments

No Known Activations