INDEX

Explanations

references to conspiracy theories and their proponents

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

bourg

-0.88

pex

-0.84

emouth

-0.81

atal

-0.76

amen

-0.73

emale

-0.72

furt

-0.71

artney

-0.70

Delivery

-0.69

ophon

-0.68

POSITIVE LOGITS

 theories

1.14

 theorists

0.92

 theorist

0.91

 debunk

0.90

 concoct

0.86

 conspiracy

0.84

 theor

0.84

 abound

0.81

 explanations

0.81

 twist

0.80

Activations Density 0.021%

references to conspiracy theories and their proponents

No Comments

No Known Activations