INDEX

Explanations

phrases related to customers

oai_token-act-pair · gpt-3.5-turbo

references to customers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

hesis

-0.73

olon

-0.65

doc

-0.65

 Declaration

-0.62

 Assass

-0.62

ence

-0.61

 Anthropology

-0.60

iral

-0.60

 Movement

-0.60

arian

-0.59

POSITIVE LOGITS

 customers

3.65

 Customers

3.14

 customer

2.49

 clients

2.27

 shoppers

2.18

 consumers

2.18

Customer

2.13

 patrons

2.09

 buyers

2.04

 subscribers

2.01

Activations Density 0.016%

phrases related to customers

references to customers

No Comments

No Known Activations