INDEX

Explanations

references to "car" and its variations in context

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.2.hook_resid_post

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

ĥ½

-2.49

Ĥ¬

-2.28

£

-2.09

ĥ

-1.83

CDATA

-1.77

Ļ

-1.75

ĨĴ

-1.70

ĩ

-1.67

ense

-1.61

rapeut

-1.59

POSITIVE LOGITS

ousel

2.60

avan

2.35

illon

2.04

riages

1.95

ving

1.94

riers

1.93

rier

1.87

pool

1.76

ved

1.69

gos

1.68

Activations Density 0.112%

references to "car" and its variations in context

No Comments

No Known Activations