INDEX

Explanations

references to actions related to opening doors

oai_token-act-pair · gpt-3.5-turbo

references to doors and their actions or states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

rities

-0.84

 Samoa

-0.81

 endorsements

-0.77

nutrition

-0.77

CBO

-0.76

orses

-0.75

 Statistical

-0.75

 Paleo

-0.74

conom

-0.74

stats

-0.74

POSITIVE LOGITS

 doors

1.66

 door

1.58

 unlocked

1.41

door

1.38

 locks

1.35

 Doors

1.33

 entrances

1.32

 latch

1.32

 doorway

1.31

 Door

1.30

Activations Density 0.317%

references to actions related to opening doors

references to doors and their actions or states

No Comments

No Known Activations