INDEX

Explanations

physical descriptions or processes related to medical conditions

oai_token-act-pair · gpt-3.5-turbo

phrases indicating potential dangers or serious consequences

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

 Priority

-0.80

 Equity

-0.79

 Kahn

-0.75

 Ranking

-0.74

 Rural

-0.71

abouts

-0.70

 reimbursement

-0.68

 Loan

-0.68

Offline

-0.67

 Ranked

-0.67

POSITIVE LOGITS

 liqu

1.37

 disinteg

1.25

 evapor

1.24

 oxid

1.21

 vapor

1.21

 expel

1.17

 dissolve

1.14

 explode

1.14

 melts

1.13

 swell

1.12

Activations Density 0.250%

physical descriptions or processes related to medical conditions

phrases indicating potential dangers or serious consequences

No Comments

No Known Activations