INDEX

Explanations

terms related to love and affection

oai_token-act-pair · gpt-3.5-turbo

repetitions of the word "love."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

•layer 1 - love

No Comments

Negative Logits

 reconc

-0.67

 removable

-0.64

 misleading

-0.63

 rise

-0.63

 unnatural

-0.63

externalActionCode

-0.62

İĭ

-0.60

 decline

-0.59

 Incarnation

-0.59

 unrecogn

-0.58

POSITIVE LOGITS

ove

1.19

tsky

1.08

lette

0.97

ments

0.93

anu

0.90

pedia

0.85

cot

0.84

rette

0.82

itsch

0.82

ices

0.81

Activations Density 0.008%

terms related to love and affection

repetitions of the word "love."

No Comments

No Known Activations