INDEX

Explanations

terms related to spectral analysis

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĥ½

-2.71

½

-2.10

↵

-2.10

↵

-2.10

<|outofrange|>

-2.10

↵

-2.10

↵

-2.10

↵

-2.10

POSITIVE LOGITS

finals

1.67

 ordinance

1.59

iences

1.54

ised

1.48

 room

1.47

idad

1.47

weather

1.45

charts

1.44

 amounts

1.41

idades

1.39

Activations Density 0.006%

terms related to spectral analysis

No Comments

No Known Activations