INDEX

Explanations

phrases related to politics and policy discussions

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

 diluted

-0.68

Spread

-0.67

Ü

-0.65

 unlaw

-0.64

 bleed

-0.63

Christmas

-0.63

 coupon

-0.62

 gust

-0.61

 stale

-0.61

 stray

-0.61

POSITIVE LOGITS

 fame

1.01

 overseeing

0.80

 Edwin

0.74

 told

0.74

 specializing

0.70

who

0.66

 Nanto

0.66

 Connie

0.65

ignty

0.65

ells

0.64

Activations Density 2.194%

phrases related to politics and policy discussions

No Comments

No Known Activations