INDEX

Explanations

mentions of bans or prohibitions

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

 IMAGES

-0.70

 Generations

-0.67

 Veter

-0.65

mberg

-0.63

eah

-0.63

 Remastered

-0.62

lycer

-0.61

eon

-0.61

prise

-0.61

rious

-0.59

POSITIVE LOGITS

ishment

1.02

zai

0.83

hee

0.82

hammer

0.82

tering

0.81

unal

0.80

ish

0.77

 viol

0.75

ishing

0.74

idding

0.73

Activations Density 0.660%

mentions of bans or prohibitions

No Comments

No Known Activations