INDEX

Explanations

references to virulence in various contexts

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

 different

-1.64

our

-1.48

 same

-1.40

 much

-1.37

son

-1.33

growth

-1.32

scribe

-1.31

 rapid

-1.31

cer

-1.29

 pure

-1.28

POSITIVE LOGITS

ariate

2.20

ulent

2.10

ĻĤ

2.04

ģ

1.98

gins

1.87

idian

1.85

ulus

1.82

ulence

1.75

imes

1.68

iously

1.67

Activations Density 0.010%

references to virulence in various contexts

No Comments

No Known Activations