INDEX

Explanations

occurrences of specific programming syntax and commands

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

ols

-1.93

oltz

-1.69

olid

-1.62

lein

-1.62

olver

-1.57

ÃŃf

-1.53

 (âĪĴ

-1.49

bserver

-1.49

ublin

-1.48

antage

-1.47

POSITIVE LOGITS

 hell

1.85

 pity

1.72

 fuck

1.66

thick

1.65

 shit

1.62

 heck

1.60

 liar

1.55

 thicker

1.52

 bitch

1.49

...)

1.46

Activations Density 5.401%

occurrences of specific programming syntax and commands

No Comments

No Known Activations