INDEX

Explanations

phrases related to attempting actions or carrying out specific tasks

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Used

-0.69

 Printed

-0.64

faces

-0.63

ITIES

-0.61

 Legs

-0.61

ILY

-0.60

 Spiegel

-0.60

 Houses

-0.60

Flo

-0.60

 Returning

-0.59

POSITIVE LOGITS

 emulate

1.44

 recreate

1.42

 revive

1.38

 convince

1.37

 replicate

1.32

 reconcile

1.32

 minimize

1.31

 imitate

1.30

 capitalize

1.29

 persuade

1.26

Activations Density 0.772%

phrases related to attempting actions or carrying out specific tasks

No Comments

No Known Activations