INDEX

Explanations

words related to computer programming and commands

oai_token-act-pair · gpt-3.5-turbo

strings of characters or specific letter sequences

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

 Thrones

-0.67

 undert

-0.67

 Nost

-0.65

 Founding

-0.63

 Stones

-0.63

 Demons

-0.62

 Electricity

-0.61

 Chatt

-0.61

 Furious

-0.61

 Union

-0.60

POSITIVE LOGITS

td

1.22

tf

1.21

dm

1.20

tg

1.17

tp

1.17

sf

1.15

sd

1.13

sb

1.12

sr

1.12

tm

1.11

Activations Density 0.109%

words related to computer programming and commands

strings of characters or specific letter sequences

No Comments

No Known Activations