INDEX

Explanations

instructions related to cooking recipes

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

 derog

-0.73

 retrospective

-0.69

 prostitutes

-0.67

 sponsor

-0.67

 Samar

-0.67

 Canaver

-0.65

 Bans

-0.64

 Odyssey

-0.64

Osw

-0.64

 informants

-0.62

POSITIVE LOGITS

 consistency

0.97

 dried

0.89

 desired

0.88

 crust

0.87

 dough

0.86

 creamy

0.86

 moist

0.84

 finished

0.82

 evenly

0.81

enough

0.81

Activations Density 0.131%

instructions related to cooking recipes

No Comments

No Known Activations