INDEX

Explanations

terms related to cinematography and movie screenings

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

nesses

-0.78

lies

-0.71

lynn

-0.65

REDACTED

-0.64

endif

-0.64

laus

-0.63

le

-0.63

coat

-0.62

lying

-0.61

theless

-0.60

POSITIVE LOGITS

atography

0.95

goers

0.93

ovies

0.93

 theaters

0.87

 cinem

0.85

atic

0.78

 studios

0.78

 theat

0.78

atically

0.76

agraph

0.76

Activations Density 0.042%

terms related to cinematography and movie screenings

No Comments

No Known Activations