INDEX

Explanations

phrases related to rates, measurements, and statistical data.

oai_token-act-pair · gpt-3.5-turbo

phrases or references to rates and measurements related to various phenomena

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

ãĥķãĤ©

-0.89

çīĪ

-0.85

çļ

-0.80

æĦ

-0.78

Õ

-0.75

å·

-0.73

Merit

-0.72

 Aires

-0.71

Īè

-0.70

ENE

-0.70

POSITIVE LOGITS

 occurrence

0.74

 violence

0.71

 parity

0.71

 attendance

0.69

 variance

0.69

 incidence

0.68

 incarceration

0.67

 electricity

0.66

 gunfire

0.66

 interest

0.65

Activations Density 0.063%

**phrases related to rates, measurements, and statistical data.**

phrases or references to rates and measurements related to various phenomena

No Comments

No Known Activations

phrases related to rates, measurements, and statistical data.