INDEX

Explanations

phrases related to quotations and unidentified referenced terms

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

Ļ½

-0.92

stant

-0.75

ĻĤ

-0.74

¾

-0.69

ãĥ´ãĤ¡

-0.68

¸

-0.66

¿

-0.65

icular

-0.62

ushi

-0.62

paren

-0.62

POSITIVE LOGITS

/"

1.31

 moniker

0.72

 motto

0.66

 mantra

0.65

aneers

0.63

 mentality

0.63

aka

0.60

 designation

0.60

 trademark

0.60

ãģ®ç

0.59

Activations Density 0.936%

phrases related to quotations and unidentified referenced terms

No Comments

No Known Activations