INDEX

Explanations

Twitter handles and tweets

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 accelerator

-0.84

 covenant

-0.81

 larvae

-0.80

 disciplines

-0.80

 cons

-0.80

 edged

-0.78

 intu

-0.77

 regener

-0.77

 serpent

-0.75

 undert

-0.75

POSITIVE LOGITS

RIP

1.44

Advertisements

1.44

Loading

1.37

Dear

1.36

Looks

1.35

Sorry

1.34

Wow

1.33

HAHAHAHA

1.29

https

1.28

Translation

1.27

Activations Density 0.154%

Twitter handles and tweets

No Comments

No Known Activations