INDEX

Explanations

words related to legal procedures and court opinions

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__mlp-sm_processed/2-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.2.hook_mlp_out

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĻĤ

-1.51

ĸ´

-1.51

 actionable

-1.46

 continued

-1.38

 filed

-1.38

).)

-1.38

 pleaded

-1.38

ĥ½

-1.35

 false

-1.35

 disclaim

-1.33

POSITIVE LOGITS

giving

1.68

 taught

1.64

das

1.52

ienne

1.39

consuming

1.36

dock

1.35

documentclass

1.34

icas

1.34

kÃ¤

1.33

burg

1.32

Activations Density 0.142%

words related to legal procedures and court opinions

No Comments

No Known Activations