INDEX

Explanations

phrases related to different potentials or possibilities

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

cipline

-0.88

ulu

-0.87

bane

-0.77

car

-0.75

zee

-0.73

creen

-0.73

men

-0.72

gar

-0.72

toe

-0.71

uart

-0.70

POSITIVE LOGITS

 pitfalls

1.01

ities

0.89

 dangers

0.88

 future

0.87

 hazards

0.87

 usefulness

0.86

 threats

0.85

 successors

0.85

 solutions

0.85

 futures

0.82

Activations Density 0.585%

phrases related to different potentials or possibilities

No Comments

No Known Activations