INDEX

Explanations

phrases related to political and social issues

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

 whenever

-0.68

 accordingly

-0.68

uate

-0.67

iffe

-0.66

deals

-0.65

 aloud

-0.65

 besides

-0.65

 elsewhere

-0.64

 suppose

-0.62

 with

-0.62

POSITIVE LOGITS

 same

1.54

 slightest

1.50

 requisite

1.34

 utmost

1.30

 highest

1.26

same

1.22

 widest

1.22

 greatest

1.21

 smallest

1.20

 brunt

1.19

Activations Density 1.357%

phrases related to political and social issues

No Comments

No Known Activations