INDEX

Explanations

concepts related to restrictions or constraints, particularly those that evoke a sense of intensity or pressure

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

 safer

-1.58

 brighter

-1.56

ylum

-1.52

asma

-1.51

mur

-1.48

 wealth

-1.42

osp

-1.40

 clearer

-1.39

 possible

-1.39

 murdered

-1.37

POSITIVE LOGITS

ĻĤ

3.30

·

2.93

į

2.88

ģ

2.80

ľĵ

2.64

Ł

2.43

ī

2.41

ĥ½

2.40

£

2.40

ħ

2.34

Activations Density 0.145%

concepts related to restrictions or constraints, particularly those that evoke a sense of intensity or pressure

No Comments

No Known Activations