INDEX

Explanations

pieces of information related to housing affordability and geopolitical news, particularly around navy cruisers and military budgets

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

 seiz

-0.70

awa

-0.70

 hump

-0.70

 arsen

-0.66

 disapp

-0.65

 pardon

-0.63

akespe

-0.62

 typical

-0.61

 federally

-0.61

 saturated

-0.61

POSITIVE LOGITS

LLC

1.29

Ltd

1.24

 Solutions

1.21

Inc

1.19

 Labs

1.13

Magazine

1.08

Â®

1.08

 Magazine

1.07

âĦ¢

1.06

 Technologies

1.04

Activations Density 1.488%

pieces of information related to housing affordability and geopolitical news, particularly around navy cruisers and military budgets

No Comments

No Known Activations