INDEX

Explanations

references related to the term "Per" with different numerical values assigned to the term

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "Per" in various contexts

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

enegger

-0.79

INESS

-0.77

å¸

-0.75

 Osiris

-0.68

 Lazarus

-0.68

 Tycoon

-0.67

ä

-0.66

DERR

-0.65

 Carbuncle

-0.65

 Raleigh

-0.65

POSITIVE LOGITS

due

1.17

secut

1.17

iod

1.17

 capita

1.15

ipher

1.10

ceptions

1.08

secution

1.06

forming

1.04

pet

1.03

missible

1.01

Activations Density 0.016%

references related to the term "Per" with different numerical values assigned to the term

occurrences of the term "Per" in various contexts

No Comments

No Known Activations