The answer is yes; for a few nights prior to the impact (assuming they had eyes with a similar sensitivity to our own and could look up!). It could be a bit longer than this if the body was larger than 10 km (it goes up roughly in proportion to the impactor's radius) and could be much longer if the object was a cometary body or had a very high albedo.
Details:
Impacting solar system objects would have relative closing speeds from around 11 to 72 km/s.
We could take the optimal case that the asteroid approaches whilst fully lit by the Sun (which probably precludes the minimum and maximum speed in the range quoted above) and then scale from another similar body - say the asteroid Vesta. This has a diameter of around $a=520$ km, gets as close as $d=1.14$ au from the Earth and has a maximum brightness of about $m=5.2$ apparent magnitude (and is hence just visible to the naked eye) and an observed flux $f = f_0 10^{-0.4m}$, where $f_0$ is a zeropoint for the magnitude scale.
Thus the flux $f_n$ received by a near-Earth asteroid of diameter $a_n$, at a distance $d_n$ from Earth (in au) and with the same reflectivity would be
$$ f_n = f\left(\frac{a_n}{a}\right)^2 \left(\frac{1+d}{1+d_n}\right)^2
\left(\frac{d}{d_n}\right)^2\ .$$
This equation assumes the asteroids are illuminated by the Sun at a distance $(1+d)$ au and then viewed at a distance $d$ au from Earth.
The magnitude of the dinosaur killer would then be
$$m_n = m -2.5\log (f/f_n)$$
To be an at all conspicuous naked eye object, $f_n \geq f$ and the dinosaur-killer becomes a naked-eye object at a distance given by
$$ d_n^2(1+d_n)^2 \leq 0.0022 \left(\frac{a_n}{10{\rm km}}\right)^2\ .$$
The only non-imaginary (i.e. square root of a positive number) solution for $a_n \sim 10$ km is $ d_n \leq 0.047 (a_n/10 {\rm km})$ au or $ \leq 7$ million ($a_n/10 {\rm km})$ km.
Moving at say 30 km/s, then it gets closer by 2.6 million km per day, thus hitting the Earth about 3 days after becoming a naked eye object. Obviously this would be longer for a slower approach speed or for a larger or more reflective asteroid. But shorter for a smaller, faster asteroid or if the asteroid approached from a direction not fully illuminated by the Sun or had a smaller albedo than Vesta.
An interesting trade-off to think about is if what is determined by measurement is the kinetic energy of the impactor. This would be proportional to the product of its mass and the square of its speed. The mass will be proportional to $a^3$. Thus if we fix the kinetic energy and allow the radius to be bigger, then it becomes visible from further away (roughly proportional to $a$) and will also be moving slower. i.e. A more massive, but slower impactor will likely be visible for a longer period of time and vice-versa.
Another possibility is that the object is of a cometary origin with an icy composition. If that were so then it could be much brighter as a result of sublimation, outgassing and having a bright cometary nucleus and tail. The answer would still be yes, but the visibility period could be weeks (comets are rather unpredictably bright).
It thus seems to me that there is a plausible range of parameters and trajectories where a dinosaur-killing asteroid could be observed and then observed to grow brighter over a few nights, but probably not much longer than that unless it was a comet.
Another possibility is of course that the impactor approaches from the sunward side of the Earth. In which case, it would be nearer the Sun, but only partially illuminated as seen from Earth. More importantly, it is likely to be in the daytime sky and would thus remain unobserved until it was, at most, hours away from impact.
