INDEX

Explanations

political and governmental terms and contexts

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

-,

-0.49

 gener

-0.48

 Dise

-0.48

 hindsight

-0.48

avia

-0.47

Loading

-0.47

ffen

-0.47

 Oliv

-0.46

 bounce

-0.45

 complication

-0.44

POSITIVE LOGITS

etheless

0.98

 respectively

0.76

 âĵĺ

0.58

 hereby

0.57

iannopoulos

0.57

NetMessage

0.56

rael

0.52

collect

0.52

usalem

0.51

stellar

0.51

Activations Density 7.945%

political and governmental terms and contexts

No Comments

No Known Activations