INDEX

Explanations

verbs followed by prepositions that indicate action or consequence

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

nexus

-0.60

 simultane

-0.58

members

-0.56

 hijacked

-0.56

 logs

-0.55

 Advertisement

-0.54

 vows

-0.54

 manuscripts

-0.54

esthes

-0.52

 elev

-0.52

POSITIVE LOGITS

 contend

1.03

ggles

0.96

 satisfy

0.94

 prove

0.93

wered

0.88

 explore

0.87

 settle

0.86

 choose

0.86

 celebrate

0.85

 learn

0.83

Activations Density 0.849%

verbs followed by prepositions that indicate action or consequence

No Comments

No Known Activations