INDEX

Explanations

mentions of event locations

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

catentry

-0.78

aceutical

-0.77

employment

-0.76

phabet

-0.73

gans

-0.69

itivity

-0.69

chrome

-0.67

piracy

-0.67

VIDEO

-0.66

INAL

-0.65

POSITIVE LOGITS

 Wembley

1.11

rium

1.06

 noon

0.92

 Fort

0.90

 Disneyland

0.88

 halftime

0.87

 Dull

0.86

 Arlington

0.86

 Marriott

0.86

onement

0.86

Activations Density 0.110%

mentions of event locations

No Comments

No Known Activations