INDEX

Explanations

proper nouns related to various topics

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

STON

-0.74

kefeller

-0.69

Ø©

-0.68

 phosphate

-0.65

 Bravo

-0.63

CLOSE

-0.61

å£«

-0.61

perty

-0.61

UTION

-0.60

conservancy

-0.60

POSITIVE LOGITS

rix

1.34

ypes

1.31

rans

1.26

rogen

1.16

ras

1.14

ronics

1.13

rals

1.13

ral

1.13

rop

1.12

orial

1.12

Activations Density 4.609%

proper nouns related to various topics

No Comments

No Known Activations