INDEX

Explanations

mentions of medical care units or intensive care related terms

oai_token-act-pair · gpt-3.5-turbo

terms related to intensive care and labor-intensive processes

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

omsky

-0.74

opp

-0.69

NPR

-0.68

udder

-0.67

aws

-0.67

izons

-0.66

uid

-0.66

ulet

-0.66

obyl

-0.65

warm

-0.65

POSITIVE LOGITS

ly

1.12

mble

1.01

 intensive

0.98

lly

0.79

urally

0.78

icult

0.77

intensive

0.76

IPM

0.76

itures

0.75

entimes

0.74

Activations Density 0.016%

mentions of medical care units or intensive care related terms

terms related to intensive care and labor-intensive processes

No Comments

No Known Activations