INDEX

Explanations

information related to political events, protests, and demonstrations

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

catentry

-0.77

 hindsight

-0.75

 inherited

-0.75

 Decay

-0.71

 Acqu

-0.70

CLASSIFIED

-0.69

 Gloss

-0.69

 inher

-0.68

 causation

-0.67

ahime

-0.67

POSITIVE LOGITS

 chanting

1.53

 protest

1.41

 protesting

1.39

 Protest

1.31

 chanted

1.27

 rally

1.26

 protestors

1.25

 cheering

1.24

 boycott

1.21

 demonstrators

1.21

Activations Density 2.880%

information related to political events, protests, and demonstrations

No Comments

No Known Activations