INDEX

Explanations

positive terms or emotions

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

loo

-0.82

 Brilliant

-0.76

opsy

-0.72

HAEL

-0.70

stall

-0.68

Leod

-0.68

 Hearts

-0.66

 Recall

-0.66

wine

-0.66

 Mellon

-0.66

POSITIVE LOGITS

itional

1.48

itions

1.26

itivity

1.20

itionally

1.15

idon

1.11

itor

1.07

ited

1.02

icion

1.01

nick

1.00

itiveness

1.00

Activations Density 0.025%

positive terms or emotions

No Comments

No Known Activations