INDEX

Explanations

descriptions of actions or situations that are considered unacceptable

oai_token-act-pair · gpt-3.5-turbo

terms related to unacceptability and moral judgment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Insight

-0.70

craft

-0.66

 Fortune

-0.65

Mov

-0.65

mus

-0.63

ier

-0.63

stone

-0.60

 Born

-0.60

 Heal

-0.59

 Speed

-0.59

POSITIVE LOGITS

 unacceptable

3.36

 intolerable

2.29

acceptable

1.98

 undesirable

1.84

 unsustainable

1.72

 inappropriate

1.66

 objectionable

1.65

 appalling

1.61

 unbearable

1.61

 acceptable

1.57

Activations Density 0.021%

descriptions of actions or situations that are considered unacceptable

terms related to unacceptability and moral judgment

No Comments

No Known Activations