INDEX

Explanations

phrases or expressions related to personal experiences or reflections

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

hips

-0.67

idth

-0.60

 Passenger

-0.58

ILE

-0.58

 Friend

-0.57

 Pirates

-0.57

 Frontier

-0.56

 Balloon

-0.56

 Petroleum

-0.54

 Wanted

-0.54

POSITIVE LOGITS

alian

1.39

self

1.24

chy

1.22

unes

1.10

iner

0.96

atic

0.88

selves

0.87

asca

0.87

ueller

0.83

MpServer

0.82

Activations Density 0.593%

phrases or expressions related to personal experiences or reflections

No Comments

No Known Activations