INDEX

Explanations

phrases related to software and technology

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

 Rowe

-0.63

forms

-0.63

 Publication

-0.59

REDACTED

-0.59

cum

-0.59

personal

-0.59

 Rarity

-0.58

 Watt

-0.57

ãĤ¯

-0.55

 gratification

-0.53

POSITIVE LOGITS

're

1.21

athered

1.20

 ourselves

1.17

've

1.15

asel

1.10

bsite

1.08

ibo

1.07

akening

1.06

aning

1.06

arers

1.05

Activations Density 0.938%

phrases related to software and technology

No Comments

No Known Activations