INDEX

Explanations

technical terms related to technology and legislation

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

$.

-0.76

}.

-0.73

.</

-0.70

.).

-0.69

.[

-0.67

 thereof

-0.67

+.

-0.66

]."

-0.65

().

-0.64

>.

-0.63

POSITIVE LOGITS

 Canaver

0.56

resa

0.50

estern

0.46

 dilemma

0.42

 partName

0.42

 FANTASY

0.41

 Announce

0.41

undrum

0.40

odore

0.40

 whirlwind

0.40

Activations Density 13.715%

technical terms related to technology and legislation

No Comments

No Known Activations