iris-llama-3.2-3b-combined-multihop-k10-ep4

daryaZare/iris-llama-3.2-3b-combined-multihop-k10-ep4

iris-llama-3.2-3b-combined-multihop-k10-ep4 is a text generation model from daryaZare. Use it when you need the model to write or continue text. It is set up for peft. The card lists the license as llama3.2.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

31% of all-time downloads

All-time downloads

112

Public

Repo size

1.2 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors876 MB · 61%

At a glance

Task: Text Generation
Library: peft
License: llama3.2
Access: Public
Created: Sep 24, 2026
Updated: Sep 25, 2026
SHA: a4425bc9

Try a prompt

Base models

meta-llama/Llama-3.2-3B-Instruct

Model card

From the Hugging Face model README

iris-llama-3.2-3b-combined-multihop-k10-ep4

This model is a fine-tuned version of meta-llama/Llama-3.2-3B-Instruct on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.6651
Model Preparation Time: 0.0275
Soft Mae: 0.0631
Soft Brier: 0.0125
Student Prelevantmean: 0.2882
Teacher Prelevantmean: 0.2965
Bin F1: 0.8634
Cal Ece: 0.0989
Cal Brier: 0.0823
Cal Auroc: 0.9771
Info Ndcg@p8: 0.9770
Info Pairwiseacc: 0.8708
Num Questions: 110
Num Pairs: 1071

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: 5e-05
train_batch_size: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 16
total_train_batch_size: 16
optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: constant
lr_scheduler_warmup_steps: 0.03
num_epochs: 4.0

Training results

Training Loss	Epoch	Step	Bin F1	Cal Auroc	Cal Brier	Cal Ece	Info Pairwiseacc	Info Ndcg@p8	Soft Brier	Soft Mae	Student Prelevantmean	Teacher Prelevantmean	Validation Loss	Model Preparation Time	Num Pairs	Num Questions
1.1158	0.0410	25	0.8066	0.9386	0.1038	0.0623	0.8124	0.9518	0.0316	0.1162	0.3301	0.2965	1.0422	0.0255	1071	110
0.8121	0.0820	50	0.8103	0.9446	0.1098	0.0893	0.8309	0.9608	0.0272	0.1068	0.2859	0.2965	0.9690	0.0255	1071	110
0.9291	0.1231	75	0.7465	0.9450	0.1244	0.1357	0.8210	0.9626	0.0317	0.1092	0.2322	0.2965	1.0204	0.0255	1071	110
0.8330	0.1641	100	0.7982	0.9564	0.1083	0.1175	0.8290	0.9638	0.0263	0.1020	0.2576	0.2965	0.9339	0.0255	1071	110
0.9089	0.2051	125	0.8040	0.9577	0.1015	0.1041	0.8294	0.9683	0.0237	0.0923	0.2638	0.2965	0.9071	0.0255	1071	110
0.8424	0.2461	150	0.8158	0.9618	0.0938	0.1010	0.8338	0.9713	0.0252	0.0911	0.2669	0.2965	0.9577	0.0255	1071	110
0.8013	0.2872	175	0.8642	0.9634	0.0968	0.1149	0.8448	0.9746	0.0230	0.0982	0.3392	0.2965	0.9574	0.0255	1071	110
0.8167	0.3282	200	0.8556	0.9647	0.0868	0.0821	0.8467	0.9761	0.0195	0.0822	0.2904	0.2965	0.8031	0.0255	1071	110
0.8984	0.3692	225	0.8371	0.9606	0.1022	0.1149	0.8388	0.9739	0.0209	0.0958	0.3004	0.2965	0.8112	0.0255	1071	110
0.7115	0.4102	250	0.8213	0.9644	0.0990	0.1094	0.8542	0.9729	0.0192	0.0850	0.2710	0.2965	0.7559	0.0255	1071	110
0.7933	0.4512	275	0.8653	0.9698	0.0984	0.1283	0.8558	0.9769	0.0182	0.0912	0.3002	0.2965	0.8103	0.0255	1071	110
0.8110	0.4923	300	0.8324	0.9665	0.0982	0.1126	0.8502	0.9748	0.0191	0.0811	0.2609	0.2965	0.7523	0.0255	1071	110
0.7711	0.5333	325	0.8118	0.9724	0.0946	0.1313	0.8540	0.9790	0.0174	0.0879	0.2828	0.2965	0.7463	0.0255	1071	110
0.8096	0.5743	350	0.8119	0.9700	0.0963	0.1245	0.8461	0.9739	0.0177	0.0876	0.2759	0.2965	0.8110	0.0255	1071	110
0.7820	0.6153	375	0.8677	0.9667	0.0936	0.1075	0.8525	0.9741	0.0168	0.0822	0.2987	0.2965	0.7484	0.0255	1071	110
0.8306	0.6563	400	0.8095	0.9679	0.1020	0.1161	0.8550	0.9754	0.0185	0.0840	0.2554	0.2965	0.7778	0.0255	1071	110
0.7680	0.6974	425	0.8094	0.9707	0.0921	0.1110	0.8631	0.9778	0.0161	0.0792	0.2822	0.2965	0.7185	0.0255	1071	110
0.6715	0.7384	450	0.8383	0.9721	0.0970	0.1262	0.8541	0.9745	0.0194	0.0836	0.2478	0.2965	0.8001	0.0255	1071	110
0.8800	0.7794	475	0.8520	0.9743	0.0874	0.1181	0.8532	0.9759	0.0137	0.0752	0.2945	0.2965	0.7046	0.0255	1071	110
0.7691	0.8204	500	0.7512	0.9724	0.1079	0.1408	0.8530	0.9746	0.0222	0.0904	0.2289	0.2965	0.8321	0.0255	1071	110
0.7112	0.8615	525	0.7711	0.9678	0.1083	0.1317	0.8500	0.9751	0.0211	0.0866	0.2382	0.2965	0.9112	0.0255	1071	110
0.6609	0.9025	550	0.8662	0.9686	0.0790	0.0735	0.8612	0.9758	0.0155	0.0756	0.3189	0.2965	0.7716	0.0255	1071	110
0.6082	0.9435	575	0.8752	0.9717	0.0982	0.1355	0.8596	0.9787	0.0170	0.0879	0.2959	0.2965	0.8216	0.0255	1071	110
0.6978	0.9845	600	0.7982	0.9779	0.0902	0.1201	0.8591	0.9802	0.0164	0.0769	0.2610	0.2965	0.7159	0.0255	1071	110
0.5699	1.0246	625	0.8491	0.9737	0.0897	0.1046	0.8642	0.9794	0.0141	0.0689	0.2786	0.2965	0.6843	0.0255	1071	110
0.4579	1.0656	650	0.8633	0.9738	0.0836	0.0955	0.8645	0.9799	0.0142	0.0673	0.2831	0.2965	0.6650	0.0255	1071	110
0.4530	1.1067	675	0.8624	0.9687	0.0903	0.0983	0.8628	0.9773	0.0149	0.0731	0.2917	0.2965	0.7401	0.0255	1071	110
0.4755	1.1477	700	0.8842	0.9768	0.0870	0.0990	0.8606	0.9790	0.0135	0.0660	0.2740	0.2965	0.7140	0.0255	1071	110
0.4278	1.1887	725	0.8619	0.9779	0.0888	0.1130	0.8580	0.9791	0.0156	0.0699	0.2583	0.2965	0.7183	0.0255	1071	110
0.5246	1.2297	750	0.8827	0.9763	0.0880	0.1192	0.8699	0.9799	0.0130	0.0700	0.2938	0.2965	0.6801	0.0255	1071	110
0.5448	1.2707	775	0.8194	0.9728	0.0999	0.1291	0.8586	0.9753	0.0174	0.0802	0.2537	0.2965	0.7505	0.0255	1071	110
0.4517	1.3118	800	0.8501	0.9704	0.0866	0.0885	0.8638	0.9758	0.0157	0.0705	0.2794	0.2965	0.6859	0.0255	1071	110
0.5358	1.3528	825	0.8802	0.9773	0.0815	0.1050	0.8615	0.9767	0.0127	0.0637	0.2906	0.2965	0.6172	0.0255	1071	110
0.5138	1.3938	850	0.8816	0.9773	0.0830	0.1072	0.8588	0.9766	0.0125	0.0641	0.2868	0.2965	0.6169	0.0255	1071	110
0.5337	1.4348	875	0.8421	0.9743	0.0887	0.1047	0.8591	0.9778	0.0159	0.0705	0.2680	0.2965	0.6726	0.0255	1071	110
0.4772	1.4758	900	0.8733	0.9763	0.0863	0.1003	0.8583	0.9772	0.0145	0.0675	0.2730	0.2965	0.6778	0.0255	1071	110
0.3949	1.5169	925	0.8110	0.0275	0.0713	0.0148	0.2795	0.2965	0.8877	0.1086	0.0896	0.9725	0.9795	0.8594	110	1071
0.6050	1.5579	950	0.6587	0.0275	0.0701	0.0150	0.2942	0.2965	0.8698	0.0932	0.0839	0.9715	0.9769	0.8626	110	1071
0.6209	1.5989	975	0.6753	0.0275	0.0697	0.0161	0.2667	0.2965	0.8559	0.1058	0.0838	0.9758	0.9785	0.8618	110	1071
0.4565	1.6399	1000	0.6817	0.0275	0.0690	0.0152	0.2743	0.2965	0.8693	0.0954	0.0821	0.9761	0.9792	0.8629	110	1071
0.4698	1.6810	1025	0.7246	0.0275	0.0681	0.0141	0.2683	0.2965	0.8615	0.1040	0.0831	0.9792	0.9825	0.8611	110	1071
0.5193	1.7220	1050	0.6538	0.0275	0.0689	0.0141	0.2699	0.2965	0.8627	0.1081	0.0864	0.9768	0.9811	0.8663	110	1071
0.5434	1.7630	1075	0.6772	0.0275	0.0743	0.0140	0.3146	0.2965	0.8921	0.1236	0.0862	0.9754	0.9792	0.8624	110	1071
0.4621	1.8040	1100	0.8179	0.0275	0.0782	0.0182	0.2539	0.2965	0.8083	0.1226	0.0960	0.9741	0.9759	0.8616	110	1071
0.4878	1.8450	1125	0.7413	0.0275	0.0681	0.0161	0.2603	0.2965	0.8506	0.1076	0.0860	0.9771	0.9798	0.8615	110	1071
0.4514	1.8861	1150	0.6913	0.0275	0.0662	0.0139	0.2803	0.2965	0.8770	0.1033	0.0824	0.9759	0.9799	0.8667	110	1071
0.5688	1.9271	1175	0.6661	0.0275	0.0645	0.0140	0.2757	0.2965	0.8655	0.0933	0.0832	0.9754	0.9790	0.8620	110	1071
0.5586	1.9681	1200	0.6523	0.0275	0.0641	0.0132	0.2865	0.2965	0.8836	0.0944	0.0778	0.9763	0.9767	0.8620	110	1071
0.4451	2.0082	1225	0.7545	0.0275	0.0671	0.0150	0.2680	0.2965	0.8623	0.1013	0.0852	0.9749	0.9789	0.8608	110	1071
0.4141	2.0492	1250	0.6660	0.0275	0.0630	0.0127	0.2779	0.2965	0.8862	0.1053	0.0810	0.9793	0.9802	0.8596	110	1071
0.4891	2.0902	1275	0.7079	0.0275	0.0658	0.0146	0.2613	0.2965	0.8631	0.1080	0.0857	0.9771	0.9781	0.8505	110	1071
0.3872	2.1313	1300	0.7531	0.0275	0.0649	0.0145	0.2781	0.2965	0.8674	0.0979	0.0846	0.9744	0.9761	0.8508	110	1071
0.3783	2.1723	1325	0.7723	0.0275	0.0718	0.0167	0.2563	0.2965	0.8355	0.1130	0.0925	0.9749	0.9798	0.8598	110	1071
0.3632	2.2133	1350	0.7172	0.0275	0.0697	0.0133	0.2738	0.2965	0.8547	0.1267	0.0882	0.9783	0.9811	0.8644	110	1071
0.3853	2.2543	1375	0.8668	0.0275	0.0737	0.0164	0.2499	0.2965	0.8457	0.1237	0.0926	0.9780	0.9818	0.8566	110	1071
0.4235	2.2954	1400	0.6654	0.0275	0.0667	0.0123	0.3110	0.2965	0.8975	0.1183	0.0814	0.9792	0.9805	0.8594	110	1071
0.4499	2.3364	1425	0.6385	0.0275	0.0613	0.0117	0.2796	0.2965	0.8841	0.1048	0.0833	0.9787	0.9823	0.8591	110	1071
0.3881	2.3774	1450	0.8063	0.0275	0.0689	0.0168	0.2626	0.2965	0.8544	0.1055	0.0840	0.9770	0.9766	0.8416	110	1071
0.4321	2.4184	1475	0.7250	0.0275	0.0704	0.0142	0.2778	0.2965	0.8660	0.1092	0.0888	0.9734	0.9784	0.8236	110	1071
0.3849	2.4594	1500	0.8029	0.0275	0.0739	0.0195	0.2530	0.2965	0.8379	0.1149	0.0895	0.9740	0.9767	0.8607	110	1071
0.3415	2.5005	1525	0.7318	0.0275	0.0659	0.0141	0.2699	0.2965	0.8623	0.1042	0.0851	0.9768	0.9795	0.8623	110	1071
0.3188	2.5415	1550	0.8381	0.0275	0.0744	0.0181	0.2462	0.2965	0.8523	0.1217	0.0932	0.9742	0.9789	0.8628	110	1071
0.3321	2.5825	1575	0.8261	0.0275	0.0710	0.0180	0.2663	0.2965	0.8359	0.1016	0.0894	0.9706	0.9793	0.8647	110	1071
0.3439	2.6235	1600	0.6591	0.0275	0.0633	0.0119	0.2927	0.2965	0.8834	0.1035	0.0818	0.9782	0.9806	0.8655	110	1071
0.3909	2.6645	1625	0.7562	0.0275	0.0699	0.0174	0.2586	0.2965	0.8442	0.1093	0.0898	0.9736	0.9770	0.8609	110	1071
0.4346	2.7056	1650	0.6191	0.0275	0.0664	0.0126	0.2904	0.2965	0.8698	0.1040	0.0845	0.9755	0.9792	0.8666	110	1071
0.3684	2.7466	1675	0.6813	0.0275	0.0620	0.0130	0.2863	0.2965	0.8684	0.0926	0.0859	0.9725	0.9788	0.8577	110	1071
0.3645	2.7876	1700	0.6715	0.0275	0.0646	0.0144	0.2745	0.2965	0.8560	0.1018	0.0823	0.9773	0.9803	0.8633	110	1071
0.3377	2.8286	1725	0.6751	0.0275	0.0611	0.0127	0.2828	0.2965	0.8747	0.0947	0.0795	0.9776	0.9801	0.8651	110	1071
0.3617	2.8697	1750	0.8021	0.0275	0.0736	0.0141	0.2640	0.2965	0.8331	0.1255	0.0951	0.9779	0.9792	0.8688	110	1071
0.3862	2.9107	1775	0.7453	0.0275	0.0639	0.0145	0.2642	0.2965	0.8642	0.1049	0.0845	0.9776	0.9789	0.8638	110	1071
0.3718	2.9517	1800	0.6553	0.0275	0.0618	0.0128	0.2818	0.2965	0.8602	0.1001	0.0815	0.9767	0.9796	0.8671	110	1071
0.4129	2.9927	1825	0.7096	0.0275	0.0670	0.0145	0.2675	0.2965	0.8388	0.1056	0.0866	0.9764	0.9789	0.8664	110	1071
0.2616	3.0328	1850	0.7601	0.0275	0.0667	0.0121	0.2906	0.2965	0.8694	0.1120	0.0861	0.9763	0.9810	0.8673	110	1071
0.3494	3.0738	1875	0.7839	0.0275	0.0727	0.0149	0.2615	0.2965	0.8406	0.1195	0.0948	0.9767	0.9810	0.8668	110	1071
0.3499	3.1149	1900	0.7452	0.0275	0.0658	0.0123	0.2805	0.2965	0.875	0.1225	0.0874	0.9794	0.9788	0.8698	110	1071
0.3771	3.1559	1925	0.6704	0.0275	0.0618	0.0113	0.2894	0.2965	0.8745	0.1131	0.0834	0.9794	0.9819	0.8724	110	1071
0.3829	3.1969	1950	0.6378	0.0275	0.0602	0.0118	0.2869	0.2965	0.8735	0.0968	0.0798	0.9779	0.9798	0.8688	110	1071
0.2628	3.2379	1975	0.6874	0.0275	0.0583	0.0121	0.2736	0.2965	0.8678	0.1058	0.0814	0.9791	0.9806	0.8710	110	1071
0.3333	3.2789	2000	0.6680	0.0275	0.0624	0.0128	0.2724	0.2965	0.8579	0.1098	0.0854	0.9776	0.9793	0.8709	110	1071
0.3014	3.3200	2025	0.7296	0.0275	0.0577	0.0118	0.2834	0.2965	0.8856	0.1055	0.0802	0.9790	0.9790	0.8741	110	1071
0.2716	3.3610	2050	0.7458	0.0275	0.0616	0.0133	0.2682	0.2965	0.8740	0.1060	0.0888	0.9769	0.9801	0.8730	110	1071
0.2615	3.4020	2075	0.7351	0.0275	0.0613	0.0138	0.2845	0.2965	0.8769	0.0912	0.0801	0.9754	0.9801	0.8783	110	1071
0.2882	3.4430	2100	0.6625	0.0275	0.0640	0.0136	0.2816	0.2965	0.8692	0.1005	0.0829	0.9739	0.9788	0.8694	110	1071
0.2544	3.4841	2125	0.7259	0.0275	0.0595	0.0129	0.2792	0.2965	0.8883	0.0991	0.0828	0.9772	0.9811	0.8649	110	1071
0.3302	3.5251	2150	0.6858	0.0275	0.0614	0.0130	0.2836	0.2965	0.8796	0.1056	0.0837	0.9761	0.9794	0.8651	110	1071
0.2858	3.5661	2175	0.7181	0.0275	0.0620	0.0135	0.3016	0.2965	0.8863	0.0813	0.0754	0.9755	0.9796	0.8712	110	1071
0.2946	3.6071	2200	0.6455	0.0275	0.0593	0.0123	0.2851	0.2965	0.8788	0.1000	0.0790	0.9784	0.9779	0.8654	110	1071
0.2712	3.6481	2225	0.6763	0.0275	0.0630	0.0130	0.2744	0.2965	0.8662	0.1122	0.0824	0.9790	0.9808	0.8677	110	1071
0.3423	3.6892	2250	0.6691	0.0275	0.0583	0.0117	0.2755	0.2965	0.8677	0.1076	0.0791	0.9813	0.9805	0.8603	110	1071
0.2788	3.7302	2275	0.7675	0.0275	0.0654	0.0141	0.2656	0.2965	0.8611	0.1154	0.0884	0.9778	0.9783	0.8657	110	1071
0.3149	3.7712	2300	0.7049	0.0275	0.0653	0.0135	0.2932	0.2965	0.8698	0.0920	0.0785	0.9751	0.9804	0.8689	110	1071
0.2989	3.8122	2325	0.9831	0.0275	0.0788	0.0204	0.2416	0.2965	0.8171	0.1282	0.0931	0.9810	0.9791	0.8678	110	1071
0.2357	3.8532	2350	0.7080	0.0275	0.0607	0.0124	0.2820	0.2965	0.8785	0.1013	0.0819	0.9785	0.9782	0.8669	110	1071
0.3350	3.8943	2375	0.7374	0.0275	0.0673	0.0148	0.2601	0.2965	0.8531	0.1184	0.0887	0.9804	0.9800	0.8740	110	1071
0.2727	3.9353	2400	0.7246	0.0275	0.0603	0.0136	0.2745	0.2965	0.8820	0.0978	0.0821	0.9773	0.9787	0.8586	110	1071
0.3047	3.9763	2425	0.7737	0.0275	0.0699	0.0165	0.2585	0.2965	0.8225	0.1095	0.0906	0.9753	0.9790	0.8618	110	1071
0.3031	4.0	2440	0.6651	0.0275	0.0631	0.0125	0.2882	0.2965	0.8634	0.0989	0.0823	0.9771	0.9770	0.8708	110	1071

Framework versions

PEFT 0.19.1
Transformers 5.7.0
Pytorch 2.11.0+cu130
Datasets 4.8.5
Tokenizers 0.22.2

Task: Text Generation
Library: peft
License: llama3.2
Created: Sep 24, 2026
Updated: Sep 25, 2026