INDEX

Explanations

information related to technology and services

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

awa

-0.77

 seiz

-0.73

 hump

-0.73

 pardon

-0.72

 washed

-0.72

 saturated

-0.72

 repealing

-0.72

 crop

-0.71

 stroke

-0.70

 eaten

-0.70

POSITIVE LOGITS

LLC

1.33

 Labs

1.30

 Solutions

1.28

âĦ¢

1.28

Ltd

1.27

Â®

1.24

Inc

1.22

 Technologies

1.19

Stud

1.19

 Productions

1.14

Activations Density 2.645%

information related to technology and services

No Comments

No Known Activations