INDEX

Explanations

occasional mentions of the word "PRO" 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

him

-1.84

iginally

-1.83

 temper

-1.79

Him

-1.67

 humor

-1.65

 humour

-1.62

his

-1.62

 kindness

-1.59

 calm

-1.58

 others

-1.56

POSITIVE LOGITS

FILE

2.12

GRAM

2.11

JECT

1.95

 legend

1.78

ium

1.77

idium

1.73

fit

1.64

blems

1.61

leaf

1.60

pad

1.58

Activations Density 0.007%

occasional mentions of the word "PRO" in various contexts

No Comments

No Known Activations