INDEX

Explanations

technology-related terms, specifically related to Google and its services

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

mble

-0.95

alid

-0.78

philis

-0.71

lain

-0.70

rift

-0.69

quez

-0.68

urst

-0.67

iosis

-0.66

lie

-0.65

rone

-0.65

POSITIVE LOGITS

plex

1.00

 Fiber

1.00

 Maps

0.93

 Doodle

0.93

 Scholar

0.92

geist

0.88

 Chrome

0.86

Maps

0.84

 Glass

0.83

 Lunar

0.82

Activations Density 0.412%

technology-related terms, specifically related to Google and its services

No Comments

No Known Activations