INDEX

Explanations

proper nouns related to technology products or concepts

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

Negative Logits

culosis

-0.80

Interstitial

-0.73

istics

-0.71

 prin

-0.63

Tips

-0.62

 Hots

-0.62

éĹĺ

-0.61

TAIN

-0.60

 Halls

-0.60

 tongues

-0.60

POSITIVE LOGITS

opus

1.53

avia

1.08

agon

1.06

av

1.00

agonal

0.99

ogram

0.92

ave

0.92

ogen

0.90

asus

0.86

oc

0.85

Activations Density 0.018%

proper nouns related to technology products or concepts

No Known Activations