INDEX

Explanations

references to components in a technical or formal context

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

ĻĤ

-3.11

Ĥ¬

-2.93

±

-2.81

ĥ½

-2.54

¯

-2.45

-2.43

↵

-2.43

↵

-2.43

↵↵

-2.43

č↵

-2.43

POSITIVE LOGITS

 strengths

2.13

arily

1.78

of

1.75

 thereof

1.74

heses

1.61

hesis

1.55

arity

1.53

tee

1.52

iser

1.52

plates

1.51

Activations Density 0.048%

references to components in a technical or formal context

No Comments

No Known Activations