INDEX

Explanations

phrases related to knowledge or awareness

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

otion

-0.79

isco

-0.75

phrine

-0.75

ItemTracker

-0.74

onding

-0.73

erva

-0.69

ermanent

-0.68

pex

-0.66

acco

-0.65

avored

-0.65

POSITIVE LOGITS

ledged

1.14

ledge

1.11

lege

0.98

 firsthand

0.97

 beforehand

0.96

how

0.92

 instinctively

0.83

nothing

0.83

 nothing

0.79

LED

0.79

Activations Density 0.464%

phrases related to knowledge or awareness

No Comments

No Known Activations