BachelorThesis/LeoMistral-7b_V06_BRONCO_CARDIO_SUMMARY_CATALOG

LeoMistral_V06_BRONCO_CARDIO_SUMMARY_CATALOG

This model is a fine-tuned version of LeoLM/leo-mistral-hessianai-7b on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2802
• Num Input Tokens Seen: 56173093

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: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 50
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Input Tokens Seen
0.7643	0.25	1377	0.7811	2763564
0.6186	0.5	2754	0.6358	5558436
0.5504	0.75	4131	0.5620	8313712
0.5311	1.0	5508	0.5014	11183370
0.3525	1.25	6885	0.4752	13968826
0.4545	1.5	8262	0.4221	16892566
0.4189	1.75	9639	0.3897	19596674
0.297	2.0	11016	0.3561	22472546
0.2219	2.25	12393	0.3447	25171614
0.2962	2.5	13770	0.3171	28070134
0.2244	2.75	15147	0.2974	30841414
0.307	3.0	16524	0.2774	33754586
0.1942	3.25	17901	0.2824	36458478
0.2212	3.5	19278	0.2762	39268846
0.1737	3.75	20655	0.2713	42149206
0.2141	4.0	22032	0.2692	44987298
0.2088	4.25	23409	0.2795	47803982
0.0971	4.5	24786	0.2801	50611334
0.1567	4.75	26163	0.2802	53414782
0.1923	5.0	27540	0.2802	56173093

Framework versions

• Transformers 4.40.2
• Pytorch 2.2.2+cu121
• Datasets 2.19.0
• Tokenizers 0.19.1