INDEX

Explanations

phrases related to actions of placing or being placed

oai_token-act-pair · gpt-3.5-turbo

instances of the word "placed" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

externalActionCode

-0.90

esp

-0.69

issance

-0.64

uary

-0.63

eh

-0.60

skin

-0.58

zzy

-0.58

rd

-0.58

ery

-0.57

MED

-0.56

POSITIVE LOGITS

holders

0.94

holder

0.87

arnaev

0.87

 electrodes

0.83

ngth

0.81

sembly

0.79

 bets

0.79

 restraints

0.78

bos

0.75

ronics

0.75

Activations Density 0.036%

phrases related to actions of placing or being placed

instances of the word "placed" and its variations

No Comments

No Known Activations