INDEX

Explanations

information related to data analysis, including calculations, measurements, and statistics

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

zech

-0.79

uve

-0.68

 charism

-0.67

 tant

-0.67

killing

-0.67

mare

-0.67

Ãª

-0.66

 evangel

-0.66

ivas

-0.66

 pleas

-0.65

POSITIVE LOGITS

 statistical

1.46

 graphs

1.41

 probabilities

1.36

verages

1.34

 dataset

1.31

 averages

1.30

 spreadsheet

1.29

 estimating

1.22

 estimates

1.22

 metrics

1.22

Activations Density 2.775%

information related to data analysis, including calculations, measurements, and statistics

No Comments

No Known Activations