INDEX

Explanations

words related to positive experiences and emotions

oai_token-act-pair · gpt-3.5-turbo

descriptors related to enjoyment and positive experiences

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.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

Sources

-0.87

odox

-0.83

æ©Ł

-0.83

osponsors

-0.82

bane

-0.79

reditary

-0.79

 actionGroup

-0.78

brids

-0.78

Requires

-0.78

vae

-0.77

POSITIVE LOGITS

 experience

1.51

 moment

1.37

 trip

1.34

 weekend

1.33

 feeling

1.30

 evening

1.29

 night

1.26

 meal

1.25

day

1.23

 occasion

1.22

Activations Density 0.375%

words related to positive experiences and emotions

descriptors related to enjoyment and positive experiences

No Comments

No Known Activations