INDEX

Explanations

time-related phrases and references

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

-1.83

naire

-1.54

Ļ

-1.52

 anchor

-1.45

 oxidase

-1.45

gered

-1.44

OTO

-1.43

ħ

-1.43

%%

-1.39

 indicators

-1.36

POSITIVE LOGITS

 vicinity

1.65

 Sessions

1.59

cliffe

1.45

hers

1.42

field

1.39

oft

1.37

combe

1.36

dale

1.34

ford

1.33

>'

1.33

Activations Density 0.284%

time-related phrases and references

No Comments

No Known Activations