INDEX

Explanations

professional titles and names in a document

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

Ļ½

-0.88

ãĥ¥

-0.64

İĭ

-0.64

¾

-0.64

 answ

-0.61

ãĥĦ

-0.59

 artif

-0.58

Ń·

-0.58

cdn

-0.57

ãĥ´ãĤ¡

-0.56

POSITIVE LOGITS

/"

0.69

ãĥİ

0.65

TPS

0.64

*.

0.64

eous

0.59

0.58

 constitu

0.56

uve

0.56

arthed

0.54

Activations Density 1.984%

professional titles and names in a document

No Comments

No Known Activations