INDEX

Explanations

descriptions related to robotics and artificial intelligence

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

 resignation

-0.77

-0.76

jured

-0.75

Sexual

-0.72

 Lann

-0.71

 Deaths

-0.71

 Prayer

-0.70

 solemn

-0.70

 sorrow

-0.70

 Sergeant

-0.69

POSITIVE LOGITS

 technologies

1.54

 scalable

1.42

 technology

1.41

 architectures

1.38

 interoper

1.37

 automation

1.26

technology

1.26

 innovations

1.26

 prototypes

1.25

 capabilities

1.24

Activations Density 4.024%

descriptions related to robotics and artificial intelligence

No Comments

No Known Activations