Distil-PII-gemma-3-270m-it/training-logs.csv

1		eval_loss	eval_binary	eval_rouge	eval_llm_as_a_judge	eval_runtime	eval_samples_per_second	eval_steps_per_second	epoch	step	loss	grad_norm	learning_rate	train_runtime	train_samples_per_second	train_steps_per_second	total_flos	train_loss
2	0	2.046112060546875	0.0	0.5833271075254629	0.0	28.9848	0.828	0.207	0.0	0
3	1								0.09858044164037855	250	0.5483	12.5625	1.2266009852216749e-05
4	2								0.1971608832807571	500	0.1428	6.03125	2.458128078817734e-05
5	3								0.29574132492113564	750	0.1048	4.5625	3.6896551724137934e-05
6	4								0.3943217665615142	1000	0.0944	3.8125	4.9211822660098524e-05
7	5								0.49290220820189273	1250	0.0833	2.296875	4.939293311887096e-05
8	6								0.5914826498422713	1500	0.0741	3.890625	4.874435739116899e-05
9	7								0.6900630914826499	1750	0.0674	2.3125	4.8095781663467026e-05
10	8								0.7886435331230284	2000	0.0653	2.484375	4.744720593576506e-05
11	9								0.887223974763407	2250	0.0605	2.3125	4.679863020806309e-05
12	10								0.9858044164037855	2500	0.056	1.4609375	4.6150054480361124e-05
13	11	0.12178646773099899	0.16666666666666666	0.9454337879689271	0.25	39.1542	0.613	0.153	1.0	2536
14	12								1.084384858044164	2750	0.0497	1.203125	4.550147875265916e-05
15	13								1.1829652996845426	3000	0.0479	2.46875	4.4852903024957196e-05
16	14								1.2815457413249212	3250	0.0467	1.671875	4.420432729725523e-05
17	15								1.3801261829652998	3500	0.0462	1.8515625	4.355575156955327e-05
18	16								1.4787066246056781	3750	0.0463	1.671875	4.29071758418513e-05
19	17								1.5772870662460567	4000	0.0427	1.1484375	4.2258600114149334e-05
20	18								1.6758675078864353	4250	0.0462	2.03125	4.1610024386447366e-05
21	19								1.774447949526814	4500	0.0411	1.0859375	4.09614486587454e-05
22	20								1.8730283911671926	4750	0.0422	1.5234375	4.031287293104343e-05
23	21								1.971608832807571	5000	0.0405	2.0	3.9664297203341464e-05
24	22	0.12576398253440857	0.16666666666666666	0.9411790557730185	0.25	40.3742	0.594	0.149	2.0	5072
25	23								2.0701892744479493	5250	0.036	1.1953125	3.90157214756395e-05
26	24								2.168769716088328	5500	0.0334	0.91796875	3.836714574793753e-05
27	25								2.2673501577287065	5750	0.035	1.3046875	3.771857002023557e-05
28	26								2.365930599369085	6000	0.0337	1.4609375	3.70699942925336e-05
29	27								2.4645110410094637	6250	0.0334	1.5390625	3.6421418564831635e-05
30	28								2.5630914826498423	6500	0.0322	1.546875	3.577284283712967e-05
31	29								2.661671924290221	6750	0.0344	2.109375	3.51242671094277e-05
32	30								2.7602523659305995	7000	0.0337	1.9609375	3.447569138172573e-05
33	31								2.8588328075709777	7250	0.0321	2.359375	3.3827115654023766e-05
34	32								2.9574132492113563	7500	0.0322	1.078125	3.31785399263218e-05
35	33	0.13283216953277588	0.25	0.9450974688536018	0.25	37.2465	0.644	0.161	3.0	7608
36	34								3.055993690851735	7750	0.0287	0.9765625	3.252996419861983e-05
37	35								3.1545741324921135	8000	0.0265	1.7734375	3.1881388470917864e-05
38	36								3.253154574132492	8250	0.0264	1.0078125	3.12328127432159e-05
39	37								3.3517350157728707	8500	0.028	1.2265625	3.0584237015513936e-05
40	38								3.4503154574132493	8750	0.0281	1.390625	2.993566128781197e-05
41	39								3.548895899053628	9000	0.0276	0.90625	2.928708556011e-05
42	40								3.6474763406940065	9250	0.0258	2.0625	2.8638509832408034e-05
43	41								3.746056782334385	9500	0.0261	0.921875	2.7989934104706067e-05
44	42								3.8446372239747633	9750	0.0263	0.61328125	2.73413583770041e-05
45	43								3.943217665615142	10000	0.027	1.125	2.6692782649302132e-05
46	44	0.14927507936954498	0.25	0.9437395032119867	0.25	36.3204	0.661	0.165	4.0	10144
47	45								4.041798107255521	10250	0.0244	1.5625	2.6044206921600168e-05
48	46								4.140378548895899	10500	0.0217	1.2578125	2.53956311938982e-05
49	47								4.238958990536277	10750	0.0226	1.9765625	2.4747055466196234e-05
50	48								4.337539432176656	11000	0.0215	2.015625	2.4098479738494266e-05
51	49								4.436119873817034	11250	0.023	2.015625	2.34499040107923e-05
52	50								4.534700315457413	11500	0.0225	2.34375	2.2801328283090335e-05
53	51								4.633280757097792	11750	0.021	1.5703125	2.2152752555388368e-05
54	52								4.73186119873817	12000	0.022	1.28125	2.15041768276864e-05
55	53								4.830441640378549	12250	0.021	0.91015625	2.0855601099984433e-05
56	54								4.929022082018927	12500	0.0212	1.84375	2.0207025372282466e-05
57	55	0.16193042695522308	0.2916666666666667	0.9449687757852137	0.3333333333333333	35.8498	0.669	0.167	5.0	12680
58	56								5.027602523659306	12750	0.0208	1.390625	1.9558449644580502e-05
59	57								5.126182965299685	13000	0.0177	1.703125	1.8909873916878535e-05
60	58								5.224763406940063	13250	0.0172	1.6953125	1.8261298189176567e-05
61	59								5.323343848580442	13500	0.0185	1.4375	1.7612722461474604e-05
62	60								5.4219242902208205	13750	0.018	1.1796875	1.6964146733772636e-05
63	61								5.520504731861199	14000	0.0181	2.046875	1.631557100607067e-05
64	62								5.619085173501578	14250	0.0174	3.59375	1.5666995278368705e-05
65	63								5.717665615141955	14500	0.0189	2.96875	1.5018419550666738e-05
66	64								5.816246056782335	14750	0.0183	1.6484375	1.436984382296477e-05
67	65								5.914826498422713	15000	0.0177	1.421875	1.3721268095262805e-05
68	66	0.18538308143615723	0.2916666666666667	0.9420310554561889	0.43478260869565216	36.3402	0.66	0.165	6.0	15216
69	67								6.013406940063091	15250	0.0168	1.84375	1.3072692367560838e-05
70	68								6.11198738170347	15500	0.015	1.703125	1.242411663985887e-05
71	69								6.210567823343848	15750	0.0155	2.484375	1.1775540912156905e-05
72	70								6.309148264984227	16000	0.0154	1.3125	1.1126965184454937e-05
73	71								6.407728706624606	16250	0.0157	1.515625	1.0478389456752972e-05
74	72								6.506309148264984	16500	0.0149	2.71875	9.829813729051005e-06
75	73								6.604889589905363	16750	0.0147	2.328125	9.181238001349037e-06
76	74								6.703470031545741	17000	0.0157	1.8671875	8.532662273647072e-06
77	75								6.80205047318612	17250	0.0163	2.171875	7.884086545945104e-06
78	76								6.900630914826499	17500	0.016	2.421875	7.235510818243138e-06
79	77								6.999211356466877	17750	0.0146	2.375	6.5869350905411715e-06
80	78	0.19330303370952606	0.2916666666666667	0.9420310554561889	0.4583333333333333	41.7416	0.575	0.144	7.0	17752
81	79								7.097791798107256	18000	0.0145	2.421875	5.938359362839206e-06
82	80								7.196372239747634	18250	0.0146	1.5	5.289783635137239e-06
83	81								7.294952681388013	18500	0.0139	1.0703125	4.641207907435272e-06
84	82								7.393533123028391	18750	0.0135	1.890625	3.992632179733306e-06
85	83								7.492113564668769	19000	0.0143	1.9375	3.344056452031339e-06
86	84								7.590694006309148	19250	0.0142	1.7265625	2.695480724329373e-06
87	85								7.6892744479495265	19500	0.0146	1.59375	2.0469049966274064e-06
88	86								7.787854889589905	19750	0.0141	1.1875	1.39832926892544e-06
89	87								7.886435331230284	20000	0.0143	1.71875	7.497535412234734e-07
90	88								7.985015772870662	20250	0.0144	0.96484375	1.0117781352150678e-07
91	89	0.19907398521900177	0.2916666666666667	0.9420310554561889	0.375	41.0489	0.585	0.146	8.0	20288
92	90								8.0	20288				6288.8277	12.903	3.226	6.467727988290509e+16	0.037849801997962625