INDEX

Explanations

the letter 'p' followed by a number indicating a specific pattern

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter "p"

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

 compe

-0.90

 Shining

-0.77

ãĤ´ãĥ³

-0.68

roy

-0.67

 diplom

-0.66

 horizont

-0.65

 Rebels

-0.65

adm

-0.63

 Dispatch

-0.62

 Demo

-0.62

POSITIVE LOGITS

orters

1.19

iper

1.19

orter

1.18

ierce

1.17

endant

1.17

uddle

1.16

ixie

1.14

ascal

1.13

uddin

1.11

antom

1.11

Activations Density 0.027%

the letter 'p' followed by a number indicating a specific pattern

occurrences of the letter "p"

No Comments

No Known Activations