INDEX

Explanations

references to software tools and their functionalities

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

áĢ¯

-2.18

áĢº

-2.16

ureus

-1.80

vow

-1.61

áĢ¬

-1.55

ãģ¾ãģĽ

-1.54

woke

-1.51

itÃ¤t

-1.49

 births

-1.45

ahoma

-1.45

POSITIVE LOGITS

tip

2.56

kit

2.46

set

2.35

maker

2.35

bars

2.14

makers

2.01

sets

1.99

chain

1.97

box

1.94

nos

1.91

Activations Density 0.160%

references to software tools and their functionalities

No Comments

No Known Activations