INDEX

Explanations

references to particular places or establishments and their significance in the community

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

mares

-0.80

 existed

-0.78

 besides

-0.74

ago

-0.72

each

-0.71

Versions

-0.69

join

-0.68

 relates

-0.68

 exists

-0.67

hm

-0.67

POSITIVE LOGITS

 easiest

1.36

 strongest

1.29

 biggest

1.28

 same

1.28

 largest

1.27

 safest

1.26

 simplest

1.24

 smartest

1.23

 fastest

1.23

 smallest

1.23

Activations Density 0.727%

references to particular places or establishments and their significance in the community

No Comments

No Known Activations