Modelpublic

ferrazzipietro/crfTask-unsup-Llama-3.2-1B-Instruct-datav2-60k

sourceHugging Facellama3.2updated 7mo agoView on Hugging Face

0likes8downloads

Model Card

crfTask-unsup-Llama-3.2-1B-Instruct-datav2-60k

This model is a fine-tuned version of ferrazzipietro/unsup-Llama-3.2-1B-Instruct-datav2 on an unknown dataset. It achieves the following results on the evaluation set:

—Loss: 0.8659
—F1 Micro: 0.8339
—F1 Macro: 0.2019
—F1 Weighted: 0.8335

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

—learning_rate: 0.0003
—trainbatchsize: 64
—evalbatchsize: 128
—seed: 42
—distributed_type: multi-GPU
—gradientaccumulationsteps: 2
—totaltrainbatch_size: 128
—optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-07 and optimizerargs=No additional optimizer arguments
—lrschedulertype: cosine
—lrschedulerwarmup_ratio: 0.1
—num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Micro	F1 Macro	F1 Weighted
4.7656	0.0047	2	2.8021	0.0	0.0	0.0
4.7656	0.0095	4	2.8021	0.0	0.0	0.0
4.8281	0.0142	6	2.8021	0.0	0.0	0.0
4.8281	0.0189	8	2.7969	0.0	0.0	0.0
4.6375	0.0236	10	2.7760	0.0	0.0	0.0
4.6375	0.0284	12	2.7552	0.0	0.0	0.0
4.6375	0.0331	14	2.7240	0.0	0.0	0.0
4.5438	0.0378	16	2.6458	0.0	0.0	0.0
4.5438	0.0426	18	2.5677	0.0	0.0	0.0
4.3563	0.0473	20	2.4740	0.0	0.0	0.0
4.3563	0.0520	22	2.375	0.0	0.0	0.0
4.3563	0.0567	24	2.2604	0.0	0.0	0.0
4.1453	0.0615	26	2.1510	0.0	0.0	0.0
4.1453	0.0662	28	2.0417	0.0133	0.0012	0.0256
3.6313	0.0709	30	1.9375	0.0930	0.0074	0.1392
3.6313	0.0757	32	1.8333	0.1960	0.0146	0.1768
3.6313	0.0804	34	1.7266	0.2757	0.0192	0.1673
3.45	0.0851	36	1.6172	0.3056	0.0186	0.1682
3.45	0.0898	38	1.5286	0.2990	0.0222	0.1637
2.9641	0.0946	40	1.4557	0.2924	0.0288	0.1645
2.9641	0.0993	42	1.3984	0.3322	0.0397	0.1730
2.9641	0.1040	44	1.3464	0.3522	0.0361	0.2297
2.6031	0.1087	46	1.3047	0.5482	0.0629	0.4901
2.6031	0.1135	48	1.2656	0.3754	0.0525	0.2672
2.4203	0.1182	50	1.2370	0.3987	0.0543	0.3106
2.4203	0.1229	52	1.2083	0.5847	0.0711	0.4941
2.4203	0.1277	54	1.1901	0.4352	0.0774	0.3745
2.4328	0.1324	56	1.1745	0.5449	0.1157	0.5619
2.4328	0.1371	58	1.1589	0.6246	0.1092	0.5399
2.3781	0.1418	60	1.1432	0.4618	0.1190	0.4001
2.3781	0.1466	62	1.1328	0.6146	0.1394	0.4923
2.3781	0.1513	64	1.1146	0.4319	0.1614	0.3802
2.1828	0.1560	66	1.0990	0.7209	0.2174	0.7057
2.1828	0.1608	68	1.0833	0.6744	0.1650	0.6233
2.1469	0.1655	70	1.0729	0.5017	0.1536	0.4851
2.1469	0.1702	72	1.0625	0.6213	0.1219	0.5163
2.1469	0.1749	74	1.0521	0.5150	0.1628	0.4699
2.15	0.1797	76	1.0417	0.6346	0.1487	0.5324
2.15	0.1844	78	1.0312	0.6645	0.2108	0.6729
2.0406	0.1891	80	1.0221	0.6611	0.1956	0.6909
2.0406	0.1939	82	1.0143	0.7475	0.1863	0.7257
2.0406	0.1986	84	1.0065	0.6811	0.1798	0.6720
1.9727	0.2033	86	1.0013	0.7176	0.1633	0.6775
1.9727	0.2080	88	0.9974	0.7043	0.1702	0.6956
1.9203	0.2128	90	0.9896	0.6877	0.1797	0.6660
1.9203	0.2175	92	0.9857	0.6844	0.1863	0.7027
1.9203	0.2222	94	0.9831	0.7608	0.1917	0.7485
1.9281	0.2270	96	0.9753	0.7841	0.1913	0.7685
1.9281	0.2317	98	0.9714	0.7542	0.1975	0.7587
2.0344	0.2364	100	0.9635	0.7907	0.2053	0.7843
2.0344	0.2411	102	0.9609	0.7409	0.1722	0.7391
2.0344	0.2459	104	0.9609	0.7076	0.1429	0.7070
1.9953	0.2506	106	0.9531	0.7176	0.1682	0.6865
1.9953	0.2553	108	0.9531	0.7243	0.2074	0.7315
1.918	0.2600	110	0.9505	0.6412	0.1729	0.5437
1.918	0.2648	112	0.9453	0.7641	0.1928	0.7530
1.918	0.2695	114	0.9440	0.7276	0.1945	0.7539
1.8633	0.2742	116	0.9414	0.7608	0.1736	0.7332
1.8633	0.2790	118	0.9414	0.7741	0.1810	0.7644
1.8508	0.2837	120	0.9401	0.8073	0.2483	0.7988
1.8508	0.2884	122	0.9349	0.8140	0.2762	0.8104
1.8508	0.2931	124	0.9349	0.7243	0.2022	0.7658
1.9297	0.2979	126	0.9323	0.7641	0.1864	0.7521
1.9297	0.3026	128	0.9284	0.8106	0.2141	0.7991
1.8219	0.3073	130	0.9271	0.7043	0.1595	0.6983
1.8219	0.3121	132	0.9258	0.7409	0.2419	0.7453
1.8219	0.3168	134	0.9232	0.7276	0.2425	0.7542
1.7641	0.3215	136	0.9219	0.8339	0.2032	0.8159
1.7641	0.3262	138	0.9219	0.6877	0.1810	0.6787
1.8227	0.3310	140	0.9193	0.7907	0.2049	0.7705
1.8227	0.3357	142	0.9180	0.7409	0.2216	0.7688
1.8227	0.3404	144	0.9167	0.7475	0.1874	0.7331
1.8313	0.3452	146	0.9167	0.6611	0.2095	0.6644
1.8313	0.3499	148	0.9141	0.7475	0.1730	0.7195
1.818	0.3546	150	0.9115	0.7708	0.2092	0.7727
1.818	0.3593	152	0.9115	0.8239	0.2236	0.8146
1.818	0.3641	154	0.9115	0.6977	0.2253	0.7036
1.8836	0.3688	156	0.9115	0.7708	0.1847	0.7493
1.8836	0.3735	158	0.9115	0.6611	0.2410	0.6604
1.9234	0.3783	160	0.9089	0.7342	0.2011	0.7095
1.9234	0.3830	162	0.9089	0.7774	0.2684	0.7923
1.9234	0.3877	164	0.9062	0.8505	0.2094	0.8467
1.8266	0.3924	166	0.9062	0.8140	0.2397	0.8178
1.8266	0.3972	168	0.9062	0.8272	0.2630	0.8264
1.918	0.4019	170	0.9049	0.8140	0.2638	0.8142
1.918	0.4066	172	0.9023	0.8206	0.2550	0.8185
1.918	0.4113	174	0.9023	0.8306	0.2432	0.8324
1.7836	0.4161	176	0.9023	0.7874	0.2670	0.8001
1.7836	0.4208	178	0.9023	0.8239	0.2116	0.8231
1.8461	0.4255	180	0.9010	0.8106	0.2753	0.8215
1.8461	0.4303	182	0.8984	0.8472	0.2050	0.8416
1.8461	0.4350	184	0.8984	0.8272	0.2052	0.8219
1.7688	0.4397	186	0.8971	0.8306	0.2623	0.8340
1.7688	0.4444	188	0.8971	0.7940	0.3258	0.8052
1.8984	0.4492	190	0.8971	0.8272	0.3293	0.8361
1.8984	0.4539	192	0.8971	0.7874	0.3204	0.8026
1.8984	0.4586	194	0.8958	0.8140	0.1930	0.8057
1.8672	0.4634	196	0.8958	0.7309	0.3147	0.7396
1.8672	0.4681	198	0.8958	0.7807	0.2747	0.7714
1.8461	0.4728	200	0.8958	0.7508	0.3229	0.7601
1.8461	0.4775	202	0.8958	0.7176	0.1764	0.6778
1.8461	0.4823	204	0.8958	0.6678	0.2080	0.6749
1.7781	0.4870	206	0.8906	0.8472	0.2630	0.8489
1.7781	0.4917	208	0.8906	0.8339	0.2076	0.8329
1.8289	0.4965	210	0.8906	0.8239	0.2090	0.8255
1.8289	0.5012	212	0.8906	0.8306	0.2102	0.8255
1.8289	0.5059	214	0.8906	0.7907	0.1939	0.8001
1.7492	0.5106	216	0.8906	0.8306	0.3139	0.8284
1.7492	0.5154	218	0.8880	0.8272	0.3175	0.8198
1.75	0.5201	220	0.8867	0.8040	0.2962	0.8032
1.75	0.5248	222	0.8867	0.8140	0.2841	0.8235
1.75	0.5296	224	0.8867	0.8239	0.2055	0.8203
1.7859	0.5343	226	0.8867	0.7973	0.2196	0.7952
1.7859	0.5390	228	0.8867	0.7243	0.1676	0.6905
1.7891	0.5437	230	0.8854	0.7940	0.2421	0.8017
1.7891	0.5485	232	0.8854	0.8106	0.2491	0.8182
1.7891	0.5532	234	0.8841	0.8173	0.1971	0.8083
1.7406	0.5579	236	0.8841	0.8206	0.2344	0.8256
1.7406	0.5626	238	0.8841	0.8272	0.2805	0.8350
1.7883	0.5674	240	0.8828	0.8140	0.3173	0.8141
1.7883	0.5721	242	0.8828	0.8007	0.3061	0.8087
1.7883	0.5768	244	0.8828	0.8272	0.2776	0.8295
1.8352	0.5816	246	0.8828	0.8372	0.2243	0.8288
1.8352	0.5863	248	0.8815	0.8272	0.2678	0.8309
1.7063	0.5910	250	0.8815	0.7874	0.2411	0.7956
1.7063	0.5957	252	0.8815	0.7841	0.2251	0.7740
1.7063	0.6005	254	0.8815	0.8272	0.2469	0.8258
1.7063	0.6052	256	0.8802	0.7973	0.2859	0.8101
1.7063	0.6099	258	0.8789	0.7774	0.2543	0.7717
1.7094	0.6147	260	0.8789	0.8239	0.3612	0.8306
1.7094	0.6194	262	0.8789	0.7807	0.3228	0.7872
1.7094	0.6241	264	0.8789	0.8073	0.2908	0.7950
1.6531	0.6288	266	0.8776	0.8272	0.3392	0.8287
1.6531	0.6336	268	0.8776	0.8339	0.3553	0.8322
1.7484	0.6383	270	0.8776	0.8306	0.3882	0.8328
1.7484	0.6430	272	0.8776	0.8272	0.3509	0.8250
1.7484	0.6478	274	0.8776	0.8206	0.2882	0.8202
1.8133	0.6525	276	0.8776	0.8106	0.3657	0.8220
1.8133	0.6572	278	0.8776	0.8272	0.3844	0.8388
1.7383	0.6619	280	0.8737	0.8073	0.1887	0.8048
1.7383	0.6667	282	0.8737	0.8372	0.2043	0.8369
1.7383	0.6714	284	0.8737	0.7940	0.2127	0.7961
1.7859	0.6761	286	0.8737	0.8372	0.2231	0.8343
1.7859	0.6809	288	0.8724	0.8173	0.2188	0.8106
1.7586	0.6856	290	0.8724	0.8439	0.2351	0.8434
1.7586	0.6903	292	0.8724	0.8405	0.2176	0.8376
1.7586	0.6950	294	0.8724	0.8472	0.2063	0.8436
1.7695	0.6998	296	0.8724	0.8439	0.2969	0.8403
1.7695	0.7045	298	0.8724	0.8605	0.3687	0.8603
1.7461	0.7092	300	0.8724	0.8538	0.3926	0.8580
1.7461	0.7139	302	0.8724	0.8339	0.3943	0.8375
1.7461	0.7187	304	0.8724	0.8272	0.4357	0.8299
1.7414	0.7234	306	0.8698	0.8372	0.4128	0.8452
1.7414	0.7281	308	0.8698	0.8306	0.3726	0.8394
1.8297	0.7329	310	0.8698	0.8505	0.3630	0.8513
1.8297	0.7376	312	0.8685	0.8538	0.2990	0.8499
1.8297	0.7423	314	0.8685	0.8439	0.2734	0.8390
1.7344	0.7470	316	0.8685	0.8306	0.2472	0.8283
1.7344	0.7518	318	0.8685	0.8272	0.2499	0.8273
1.7586	0.7565	320	0.8685	0.8239	0.2247	0.8305
1.7586	0.7612	322	0.8685	0.8239	0.2258	0.8301
1.7586	0.7660	324	0.8685	0.8206	0.2897	0.8257
1.7367	0.7707	326	0.8685	0.8306	0.2860	0.8387
1.7367	0.7754	328	0.8685	0.8007	0.2764	0.8103
1.8266	0.7801	330	0.8685	0.8206	0.2425	0.8287
1.8266	0.7849	332	0.8685	0.8272	0.2666	0.8296
1.8266	0.7896	334	0.8685	0.8173	0.1903	0.8151
1.8016	0.7943	336	0.8685	0.8173	0.1986	0.8144
1.8016	0.7991	338	0.8685	0.8472	0.2062	0.8470
1.718	0.8038	340	0.8685	0.8372	0.2047	0.8377
1.718	0.8085	342	0.8685	0.8472	0.2500	0.8479
1.718	0.8132	344	0.8685	0.8439	0.2492	0.8433
1.7641	0.8180	346	0.8685	0.8206	0.2409	0.8181
1.7641	0.8227	348	0.8685	0.8339	0.2015	0.8322
1.6781	0.8274	350	0.8685	0.8306	0.1981	0.8316
1.6781	0.8322	352	0.8685	0.8372	0.3619	0.8404
1.6781	0.8369	354	0.8685	0.8372	0.3619	0.8404
1.8313	0.8416	356	0.8685	0.8405	0.3609	0.8446
1.8313	0.8463	358	0.8685	0.8405	0.3109	0.8472
1.6625	0.8511	360	0.8685	0.8439	0.3967	0.8497
1.6625	0.8558	362	0.8685	0.8372	0.3956	0.8430
1.6625	0.8605	364	0.8685	0.8439	0.3966	0.8491
1.7258	0.8652	366	0.8672	0.8306	0.3959	0.8345
1.7258	0.8700	368	0.8672	0.8306	0.3059	0.8341
1.707	0.8747	370	0.8672	0.8439	0.3600	0.8451
1.707	0.8794	372	0.8659	0.8405	0.3577	0.8428
1.707	0.8842	374	0.8659	0.8405	0.3546	0.8443
1.6922	0.8889	376	0.8659	0.8439	0.3615	0.8481
1.6922	0.8936	378	0.8659	0.8339	0.2696	0.8378
1.7188	0.8983	380	0.8659	0.8405	0.3609	0.8446
1.7188	0.9031	382	0.8659	0.8306	0.3575	0.8350
1.7188	0.9078	384	0.8659	0.8339	0.3582	0.8385
1.7023	0.9125	386	0.8659	0.8372	0.2701	0.8413
1.7023	0.9173	388	0.8659	0.8405	0.3609	0.8446
1.7406	0.9220	390	0.8659	0.8372	0.2733	0.8397
1.7406	0.9267	392	0.8659	0.8372	0.2670	0.8390
1.7406	0.9314	394	0.8659	0.8439	0.3599	0.8451
1.6633	0.9362	396	0.8659	0.8439	0.2698	0.8452
1.6633	0.9409	398	0.8659	0.8405	0.2693	0.8417
1.8211	0.9456	400	0.8659	0.8405	0.2693	0.8417
1.8211	0.9504	402	0.8659	0.8372	0.3588	0.8382
1.8211	0.9551	404	0.8659	0.8339	0.2682	0.8349
1.7023	0.9598	406	0.8659	0.8339	0.2665	0.8355
1.7023	0.9645	408	0.8659	0.8339	0.2019	0.8335
1.7844	0.9693	410	0.8659	0.8372	0.3569	0.8389
1.7844	0.9740	412	0.8659	0.8339	0.2665	0.8355
1.7844	0.9787	414	0.8659	0.8272	0.1991	0.8273
1.6906	0.9835	416	0.8659	0.8372	0.2688	0.8383
1.6906	0.9882	418	0.8659	0.8339	0.2002	0.8342
1.7188	0.9929	420	0.8659	0.8339	0.2019	0.8335
1.7188	0.9976	422	0.8659	0.8339	0.2019	0.8335

Framework versions

—PEFT 0.18.1
—Transformers 4.51.0
—Pytorch 2.8.0+cu128
—Datasets 3.6.0
—Tokenizers 0.21.0