INDEX

Explanations

mentions of political figures and governmental actions

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

lections

-0.68

boro

-0.66

utical

-0.66

 endeav

-0.63

minist

-0.63

heit

-0.60

 endeavour

-0.59

 everyday

-0.57

 equip

-0.56

 offline

-0.56

POSITIVE LOGITS

PHOTOS

0.96

Scroll

0.94

Asked

0.89

However

0.84

 Related

0.82

Bus

0.81

Authorities

0.80

Meanwhile

0.79

SEE

0.79

Earlier

0.78

Activations Density 1.337%

mentions of political figures and governmental actions

No Comments

No Known Activations