INDEX

Explanations

words related to medical procedures or treatments

oai_token-act-pair · gpt-3.5-turbo

words related to being subjected to oppression or restrictions

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

 Express

-0.63

 Apex

-0.59

 Pell

-0.59

 head

-0.59

 Head

-0.58

 Prime

-0.57

 formulation

-0.57

 Treasurer

-0.56

 prime

-0.56

Shape

-0.56

POSITIVE LOGITS

ied

4.38

ying

2.56

IED

2.55

ies

2.15

IES

1.44

ieth

1.42

iable

1.42

ief

1.42

iest

1.36

iers

1.34

Activations Density 0.014%

words related to medical procedures or treatments

words related to being subjected to oppression or restrictions

No Comments

No Known Activations