INDEX

Explanations

phrases related to comparing quantities by using multiples or ratios

oai_token-act-pair · gpt-3.5-turbo

repeated phrases or patterns related to comparative ratios and multiples

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

PLA

-0.90

 Leilan

-0.76

ADRA

-0.68

RAW

-0.67

NING

-0.65

Chart

-0.65

 Blaze

-0.63

azel

-0.62

NULL

-0.62

amac

-0.61

POSITIVE LOGITS

 amount

1.06

 usual

0.98

 extent

0.92

 average

0.91

 size

0.91

 thickness

0.90

 pace

0.79

 fastest

0.79

 diameter

0.78

 magnitude

0.77

Activations Density 0.064%

phrases related to comparing quantities by using multiples or ratios

repeated phrases or patterns related to comparative ratios and multiples

No Comments

No Known Activations