INDEX

Explanations

references to the word "Per" or "Peru"

oai_token-act-pair · gpt-3.5-turbo

the repeated mention of the word "Per," likely indicating a focus on a specific entity or keyword

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

DERR

-0.80

enegger

-0.79

INESS

-0.70

 pneum

-0.67

 Raleigh

-0.66

 Carbuncle

-0.66

orld

-0.64

 Lazarus

-0.63

ATIONAL

-0.63

Ic

-0.62

POSITIVE LOGITS

ceptions

1.24

 capita

1.22

secut

1.17

ipher

1.14

uvian

1.14

iod

1.14

due

1.14

ceived

1.11

missible

1.09

secution

1.09

Activations Density 0.019%

references to the word "Per" or "Peru"

the repeated mention of the word "Per," likely indicating a focus on a specific entity or keyword

No Comments

No Known Activations