INDEX

Explanations

terms related to health and medical topics

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

ĭ

-1.90

ĨĴ

-1.87

Ļª

-1.81

¯

-1.79

ı

-1.72

´

-1.72

ľ

-1.70

µ

-1.66

deal

-1.60

Į

-1.59

POSITIVE LOGITS

care

1.78

grass

1.66

iterranean

1.58

lichen

1.53

ocs

1.53

birds

1.53

table

1.52

ubot

1.52

idelines

1.45

assic

1.40

Activations Density 0.006%

terms related to health and medical topics

No Comments

No Known Activations