INDEX

Explanations

words related to medication or medical treatment

oai_token-act-pair · gpt-3.5-turbo

references to medication or medical terms

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

 forwarding

-0.70

 shedding

-0.64

 newsp

-0.63

CRIP

-0.60

RP

-0.60

 Static

-0.59

 outgoing

-0.58

 attracting

-0.58

 torch

-0.58

 donkey

-0.57

POSITIVE LOGITS

med

4.67

ming

2.30

MED

2.17

Med

1.77

mer

1.70

ms

1.52

mers

1.46

ned

1.44

sed

1.38

med

1.37

Activations Density 0.009%

words related to medication or medical treatment

references to medication or medical terms

No Comments

No Known Activations