INDEX

Explanations

locations or directions

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

 Likes

-0.80

vous

-0.80

 Items

-0.79

 Parenthood

-0.78

 Meter

-0.78

 Experiment

-0.76

 Pieces

-0.73

Testing

-0.73

 Inspect

-0.72

 Numbers

-0.72

POSITIVE LOGITS

ward

1.25

side

1.08

 coast

1.07

wards

0.99

bound

0.96

central

0.88

 side

0.87

 hemisphere

0.85

 coasts

0.84

 wing

0.82

Activations Density 0.022%

locations or directions

No Comments

No Known Activations