INDEX

Explanations

technology-related terms and information, particularly related to software applications and technological advancements

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

inav

-0.76

iates

-0.67

 Moines

-0.64

 Telesc

-0.64

 IMAGES

-0.63

 encounters

-0.63

know

-0.63

 Benefits

-0.62

sburg

-0.61

uld

-0.61

POSITIVE LOGITS

 supposed

1.21

nt

1.13

 currently

1.11

 definitely

1.08

not

1.07

rael

1.07

 still

1.03

 hereby

1.01

 also

1.00

 considered

1.00

Activations Density 1.607%

technology-related terms and information, particularly related to software applications and technological advancements

No Comments

No Known Activations