INDEX

Explanations

statements or questions related to reviews or assessments

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

idth

-0.63

 sling

-0.61

ãĥ³ãĤ¸

-0.60

srf

-0.58

dim

-0.56

 Pione

-0.56

 quarrel

-0.56

ĪĴ

-0.55

Bo

-0.54

 Crush

-0.53

POSITIVE LOGITS

 anonymously

0.96

 myself

0.95

 ourselves

0.94

 firsthand

0.93

 yourself

0.93

 manually

0.91

 herself

0.87

 anyway

0.86

 aloud

0.86

.</

0.85

Activations Density 1.940%

statements or questions related to reviews or assessments

No Comments

No Known Activations