INDEX

Explanations

references to proximity or closeness to locations or concepts

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.0.hook_mlp_out

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

ORT

-1.60

rap

-1.58

ax

-1.53

pless

-1.48

andra

-1.44

urer

-1.42

kered

-1.37

eps

-1.36

hs

-1.33

 Voor

-1.31

POSITIVE LOGITS

 weaknesses

1.79

¶

1.75

µ

1.71

¼

1.68

 encounters

1.62

ł

1.60

 gaps

1.55

 offence

1.54

Ń

1.53

ķ

1.52

Activations Density 0.413%

references to proximity or closeness to locations or concepts

No Comments

No Known Activations