That must have been as result of not keeping your column at constant temperature. If not, check the integrity of your solvents, better still use freshly prepared solvent.
Temperature is not the concern here (do you see how the baseline drifts up, then back down starting at 5 minutes into the analysis, not at 0 minutes? Temperature change would be constant).
"Shei S.", You provided no HPLC method or sample information at all. *Based on the chromatogram showing the injection spike (notice the HUGE spike at T0), you may have overloaded the column or injected different solution relative to the one example resulting in a different baseline response (normal). Use the same injection solution for ALL samples. Same volume. Use mobile phase as the injection solution (sample dissolved in mobile phase). Most importantly of all, as you appear to be using HPLC for the first time, with no experience at all. Please contact someone local who is experienced in HPLC to assist you.
To give useful advice, we need to know the solvents used, any solvent modifiers, gradient method, and detection wavelength. One possibility is that the strong solvent absorbs light at the detection wavelength and you are running a gradient causing baseline drift, but we can't be sure without more information.
Positive baseline drift is commonly seen in gradient elution and is often caused by differences in the absorbance properties of the A and B solvents. In the case of reversed-phase HPLC gradient elution, the eluent composition will change from a high percent of A (aqueous) to a high percent of B (organic solvent).