INDEX

Explanations

words related to emergency rooms, medical conditions, and treatments

oai_token-act-pair · gpt-3.5-turbo

references to emergency rooms (ER)

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Finish

-0.78

ucket

-0.74

flake

-0.73

Cro

-0.71

SourceFile

-0.68

Pac

-0.68

utra

-0.66

Activity

-0.66

ãĥķãĤ©

-0.66

orship

-0.66

POSITIVE LOGITS

ER

1.14

 resil

0.92

aser

0.85

 conflic

0.81

ISA

0.80

 satell

0.80

ANGE

0.77

 confir

0.77

senal

0.76

oute

0.76

Activations Density 0.004%

words related to emergency rooms, medical conditions, and treatments

references to emergency rooms (ER)

No Comments

No Known Activations