INDEX

Explanations

instances of the word "convenient"

oai_token-act-pair · gpt-3.5-turbo

the concept of convenience

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

kamp

-0.78

master

-0.66

 (Â£

-0.66

stri

-0.63

 twins

-0.62

STAR

-0.61

 Scream

-0.61

 Nile

-0.59

net

-0.58

ric

-0.57

POSITIVE LOGITS

 convenient

3.63

 conveniently

2.17

 inconvenient

2.11

venient

1.67

 advantageous

1.61

 handy

1.59

 useful

1.37

 suitable

1.37

 tempting

1.33

 fashionable

1.32

Activations Density 0.016%

instances of the word "convenient"

the concept of convenience

No Comments

No Known Activations