INDEX

Explanations

terms related to physical harm or damage

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

ubert

-1.66

asters

-1.61

usterity

-1.57

áĢ¯

-1.47

---|---|---

-1.44

dtd

-1.43

atern

-1.43

summary

-1.41

omorphisms

-1.41

astery

-1.40

POSITIVE LOGITS

uous

1.81

ilated

1.73

eted

1.57

acted

1.52

stre

1.52

 offence

1.50

IBLE

1.49

 vest

1.49

fulness

1.47

ense

1.46

Activations Density 0.026%

terms related to physical harm or damage

No Comments

No Known Activations