INDEX

Explanations

phrases related to cognitive biases and logical fallacies

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

 Canaver

-0.73

outhern

-0.67

 Torrent

-0.65

undrum

-0.64

mobi

-0.64

cephal

-0.61

 Cosponsors

-0.61

 Charity

-0.60

çīĪ

-0.60

 Luxem

-0.60

POSITIVE LOGITS

 oneself

0.98

)).

0.98

 predetermined

0.93

?".

0.89

 theirs

0.88

etc

0.83

 themselves

0.83

 others

0.82

them

0.81

 somebody

0.80

Activations Density 7.723%

phrases related to cognitive biases and logical fallacies

No Comments

No Known Activations