INDEX

Explanations

numbers related to quantities of people

oai_token-act-pair · gpt-3.5-turbo

references to statistics and the impact on populations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Wong

-0.59

 Ratt

-0.59

ollah

-0.58

 sequels

-0.56

 McKenzie

-0.55

 Nebula

-0.55

 Leigh

-0.55

inventoryQuantity

-0.54

 aggression

-0.54

PO

-0.54

POSITIVE LOGITS

 worldwide

1.31

 annually

1.09

 globally

1.07

 nationwide

1.06

 apiece

1.02

 across

0.86

+.

0.82

 statewide

0.80

 nationally

0.79

 respectively

0.77

Activations Density 0.184%

numbers related to quantities of people

references to statistics and the impact on populations

No Comments

No Known Activations