INDEX

Explanations

references to physical actions and bodily sensations, especially in tense situations

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ocument

-0.93

thood

-0.88

å¥³

-0.88

ovo

-0.88

utical

-0.87

ablishment

-0.86

-0.85

ocumented

-0.85

Chall

-0.85

ourced

-0.84

POSITIVE LOGITS

own

1.40

 lips

1.26

 chest

1.24

 cheeks

1.23

 knees

1.21

 ears

1.20

 forehead

1.19

 gaze

1.18

 clipboard

1.14

 fingers

1.13

Activations Density 8.539%

references to physical actions and bodily sensations, especially in tense situations

No Comments

No Known Activations