INDEX

Explanations

instances of the word "Tickets."

oai_token-act-pair · gpt-3.5-turbo

references to tickets for events

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

ths

-0.65

 Galile

-0.65

 invention

-0.65

yan

-0.63

 advers

-0.61

 Forth

-0.61

 competence

-0.59

 Intervention

-0.58

 deception

-0.57

 investigative

-0.57

POSITIVE LOGITS

 Tickets

1.22

Tickets

1.17

 tickets

1.14

ticket

0.91

 Ticket

0.87

 holders

0.82

 ticket

0.81

 rentals

0.77

 purchased

0.76

 prices

0.75

Activations Density 0.035%

instances of the word "Tickets."

references to tickets for events

No Comments

No Known Activations