INDEX

Explanations

acronyms related to technology or organizations, particularly focusing on "RS"

oai_token-act-pair · gpt-3.5-turbo

references to RS (Resource Management) codes or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Contra

-0.77

framework

-0.76

rities

-0.74

 Samoa

-0.72

aneously

-0.72

bernatorial

-0.71

 Antarctic

-0.71

leck

-0.71

enegger

-0.69

berman

-0.69

POSITIVE LOGITS

RS

1.05

VP

1.03

ACK

0.88

acks

0.83

ESSION

0.82

Ãī

0.80

ocial

0.80

heed

0.80

hod

0.80

Activations Density 0.010%

acronyms related to technology or organizations, particularly focusing on "RS"

references to RS (Resource Management) codes or identifiers

No Comments

No Known Activations