INDEX

Explanations

occurrences of rules in a programming or procedural context

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

 electrons

-1.86

---|---

-1.85

#{

-1.81

 shame

-1.73

 suppose

-1.70

?'

-1.67

?"

-1.64

¿½

-1.62

 welcome

-1.59

ĥ½

-1.59

POSITIVE LOGITS

eria

2.18

ista

1.92

zilla

1.85

naire

1.84

etable

1.83

ariate

1.79

aggio

1.78

aho

1.78

maker

1.75

rooms

1.74

Activations Density 0.010%

occurrences of rules in a programming or procedural context

No Comments

No Known Activations