INDEX

Explanations

medical acronyms related to a specific medical condition, specifically "SS" with high activation values

oai_token-act-pair · gpt-3.5-turbo

the acronym "SS" which pertains to relevant contexts and discussions

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

 Flore

-0.77

 Fiat

-0.72

ney

-0.69

 disputes

-0.69

 Wayne

-0.69

 coun

-0.68

 Painter

-0.65

 Bloomberg

-0.64

 Fare

-0.61

 Calgary

-0.61

POSITIVE LOGITS

SS

4.19

SS

2.09

ss

1.74

SSL

1.56

SR

1.48

SD

1.47

RS

1.44

DD

1.39

LS

1.35

SL

1.31

Activations Density 0.008%

medical acronyms related to a specific medical condition, specifically "SS" with high activation values

the acronym "SS" which pertains to relevant contexts and discussions

No Comments

No Known Activations