Classification Report for Mixtral 8x22B

Invalid Model Output Count

Test	Invalid Count
0%_512	2
0%_1024	1
0%_4096	4
25%_1024	1
25%_4096	1
25%_8192	3
25%_16384	2
25%_32768	2
50%_512	1
50%_4096	1
50%_16384	1
50%_32768	2
75%_256	4
75%_512	1
75%_4096	1

Consolidated Classification Report

Test	Total Samples	True Violations	Predicted Violations	Accuracy	Precision (macro)	Recall (macro)	F1-score (macro)	Precision (weighted)	Recall (weighted)	F1-score (weighted)
base	411	248	286	0.815	0.824	0.787	0.797	0.819	0.815	0.809
0%_64	411	248	287	0.813	0.822	0.784	0.794	0.817	0.813	0.807
0%_128	411	248	291	0.808	0.82	0.777	0.787	0.814	0.808	0.801
0%_256	411	248	283	0.822	0.83	0.796	0.806	0.826	0.822	0.818
0%_512	409	248	286	0.814	0.823	0.785	0.795	0.818	0.814	0.808
0%_1024	410	247	286	0.817	0.828	0.789	0.799	0.822	0.817	0.811
0%_2048	411	248	271	0.837	0.838	0.818	0.825	0.837	0.837	0.834
0%_4096	407	244	247	0.83	0.824	0.822	0.823	0.83	0.83	0.83
0%_8192	411	248	290	0.781	0.787	0.749	0.758	0.784	0.781	0.773
0%_16384	411	248	317	0.754	0.781	0.707	0.714	0.771	0.754	0.736
0%_32768	411	248	311	0.696	0.694	0.649	0.651	0.695	0.696	0.677
25%_64	411	248	297	0.788	0.802	0.753	0.763	0.796	0.788	0.779
25%_128	411	248	306	0.786	0.809	0.746	0.756	0.799	0.786	0.774
25%_256	411	248	305	0.783	0.805	0.744	0.754	0.796	0.783	0.772
25%_512	411	248	322	0.757	0.792	0.707	0.714	0.779	0.757	0.737
25%_1024	410	247	321	0.737	0.763	0.686	0.691	0.753	0.737	0.715
25%_2048	411	248	310	0.757	0.775	0.713	0.721	0.768	0.757	0.742
25%_4096	410	247	282	0.744	0.739	0.714	0.72	0.742	0.744	0.737
25%_8192	408	245	327	0.691	0.705	0.636	0.632	0.701	0.691	0.661
25%_16384	409	247	282	0.68	0.664	0.647	0.65	0.673	0.68	0.671
25%_32768	409	246	276	0.555	0.522	0.52	0.518	0.541	0.555	0.545
50%_64	411	248	308	0.786	0.812	0.745	0.755	0.801	0.786	0.773
50%_128	411	248	306	0.762	0.777	0.72	0.729	0.771	0.762	0.748
50%_256	411	248	310	0.771	0.795	0.728	0.738	0.785	0.771	0.757
50%_512	410	248	326	0.746	0.783	0.693	0.698	0.77	0.746	0.723
50%_1024	411	248	343	0.715	0.765	0.653	0.648	0.749	0.715	0.68
50%_2048	411	248	301	0.764	0.775	0.726	0.734	0.77	0.764	0.752
50%_4096	410	248	282	0.737	0.73	0.706	0.712	0.733	0.737	0.73
50%_8192	411	248	223	0.603	0.595	0.599	0.595	0.615	0.603	0.607
50%_16384	410	247	211	0.59	0.587	0.591	0.585	0.608	0.59	0.595
50%_32768	409	246	266	0.599	0.575	0.571	0.572	0.591	0.599	0.594
75%_64	411	248	316	0.757	0.783	0.71	0.718	0.773	0.757	0.739
75%_128	411	248	311	0.749	0.767	0.705	0.712	0.76	0.749	0.733
75%_256	407	247	322	0.742	0.772	0.688	0.693	0.761	0.742	0.72
75%_512	410	248	328	0.732	0.764	0.677	0.679	0.753	0.732	0.706
75%_1024	411	248	337	0.72	0.761	0.661	0.659	0.748	0.72	0.689
75%_2048	411	248	270	0.723	0.711	0.699	0.702	0.719	0.723	0.718
75%_4096	410	248	239	0.724	0.713	0.717	0.714	0.727	0.724	0.726
75%_8192	411	248	132	0.552	0.602	0.593	0.55	0.631	0.552	0.543
75%_16384	411	248	226	0.586	0.577	0.579	0.576	0.597	0.586	0.59
75%_32768	411	248	214	0.557	0.553	0.555	0.55	0.574	0.557	0.562
100%_64	411	248	315	0.754	0.778	0.708	0.715	0.769	0.754	0.737
100%_128	411	248	326	0.737	0.769	0.684	0.688	0.758	0.737	0.714
100%_256	411	248	331	0.749	0.797	0.695	0.699	0.781	0.749	0.725
100%_512	411	248	325	0.749	0.786	0.698	0.703	0.773	0.749	0.727
100%_1024	411	248	329	0.745	0.785	0.691	0.695	0.771	0.745	0.72
100%_2048	411	248	301	0.73	0.732	0.69	0.696	0.731	0.73	0.717
100%_4096	411	248	231	0.657	0.646	0.651	0.647	0.664	0.657	0.659
100%_8192	411	248	186	0.577	0.587	0.59	0.575	0.61	0.577	0.58
100%_16384	411	248	243	0.637	0.623	0.624	0.623	0.639	0.637	0.638
100%_32768	411	248	301	0.56	0.515	0.512	0.504	0.535	0.56	0.538