INDEX

Explanations

references to specific metrics or indicators, often in a table format

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/1-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.1.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

ies

-1.72

heads

-1.67

iles

-1.60

Ã©s

-1.55

ids

-1.53

astics

-1.52

edly

-1.51

oning

-1.51

works

-1.50

anes

-1.50

POSITIVE LOGITS

leans

1.54

 grep

1.45

jour

1.41

 dated

1.40

 prescribe

1.39

 date

1.36

inel

1.35

agra

1.35

Wire

1.35

½

1.33

Activations Density 0.062%

references to specific metrics or indicators, often in a table format

No Comments

No Known Activations