INDEX

Explanations

commands or instructional information

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

 paperback

-0.70

 document

-0.66

 Airbnb

-0.65

 monopoly

-0.61

 guest

-0.60

 coax

-0.60

 charter

-0.60

 Garmin

-0.60

 horizont

-0.59

 primaries

-0.59

POSITIVE LOGITS

âĢ

1.78

Â»

1.31

Â

1.31

1.30

Â«

1.29

"""

1.26

1.25

Â´

1.24

**

1.24

"(

1.23

Activations Density 2.145%

commands or instructional information

No Comments

No Known Activations