INDEX

Explanations

questions and phrases related to the concept of "how."

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

hesis

-1.68

 whose

-1.67

achus

-1.61

burg

-1.58

ylvania

-1.57

(\#

-1.55

 (“

-1.52

 whom

-1.51

aho

-1.44

 footnote

-1.42

POSITIVE LOGITS

ĥ½

2.52

¾

2.35

↵↵

2.24

<|outofrange|>

2.24

č↵

2.24

↵

2.24

<|outofrange|>

2.24

↵

2.24

<|outofrange|>

2.24

Activations Density 0.108%

questions and phrases related to the concept of "how."

No Comments

No Known Activations