INDEX

Explanations

references to barriers or obstacles

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

'))

-1.67

urred

-1.66

hentication

-1.65

'));

-1.57

xiety

-1.52

...](

-1.51

omorphisms

-1.49

agogue

-1.48

acity

-1.47

\\

-1.47

POSITIVE LOGITS

strÃ¶m

2.03

bilt

1.96

ista

1.95

zilla

1.85

gren

1.82

iang

1.79

agem

1.76

istas

1.75

chaft

1.74

iative

1.73

Activations Density 0.050%

references to barriers or obstacles

No Comments

No Known Activations