INDEX

Explanations

mentions of medical care and treatment

oai_token-act-pair · gpt-3.5-turbo

terms related to healthcare and medical services, particularly in legal and governmental contexts

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

undrum

-0.77

itcher

-0.77

agame

-0.76

rame

-0.72

aiman

-0.71

ndra

-0.71

mire

-0.66

 typo

-0.65

hov

-0.64

auldron

-0.64

POSITIVE LOGITS

 assistance

1.51

 licenses

1.39

 certificates

1.39

 privileges

1.37

 permission

1.37

 permits

1.36

 licences

1.32

 visas

1.31

 authorization

1.27

 permissions

1.25

Activations Density 0.378%

mentions of medical care and treatment

terms related to healthcare and medical services, particularly in legal and governmental contexts

No Comments

No Known Activations