INDEX

Explanations

phrases related to taking action or performing specific tasks

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

pmwiki

-0.73

 wont

-0.66

ItemTracker

-0.61

æ©Ł

-0.59

To

-0.58

ãĤ´ãĥ³

-0.58

 marked

-0.55

Downloadha

-0.55

To

-0.54

to

-0.54

POSITIVE LOGITS

uate

1.14

enance

0.98

 oneself

0.94

igate

0.87

dress

0.80

ISE

0.77

 them

0.75

ezvous

0.74

ulate

0.74

him

0.71

Activations Density 2.298%

phrases related to taking action or performing specific tasks

No Comments

No Known Activations