INDEX

Explanations

details related to a specific object or concept

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

irie

-0.78

idy

-0.70

ãĥīãĥ©

-0.66

mx

-0.64

izen

-0.63

enstein

-0.62

gang

-0.62

ļéĨĴ

-0.62

iple

-0.62

orem

-0.61

POSITIVE LOGITS

 albeit

1.48

 namely

1.38

 although

1.38

 though

1.28

 especially

1.26

 however

1.23

but

1.17

 except

1.10

especially

1.08

 particularly

1.07

Activations Density 1.365%

details related to a specific object or concept

No Comments

No Known Activations