Are all iterations saved in resudiuals and jacobian matrix with lsqnonlin?

Hello I estimate a model with [k_optim, resnorm, residual, exitflag, output, lambda, jacobian] = lsqnonlin(f,k0,lb,ub,options)

My exitflag is 3. Is this the reason why I my residuals and jacobian has three times more rows as observations, respectively as neccessary? I guess its quite obvious, but I just want to make sure.