fine-tune-radia-v3

joedonino/fine-tune-radia-v3

fine-tune-radia-v3 is a machine learning model from joedonino. Use it for the machine learning task on the model card, and read the license before you ship it in a product.

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

Downloads · 30 days

Access

Public

Updated Nov 16, 2023

Repo size

130 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors129 MB · 98%

At a glance

Access: Public
Created: Nov 16, 2023
Updated: Nov 16, 2023
SHA: 63359a0a

Base models

meta-llama/Llama-2-7b-hf

Model card

From the Hugging Face model README

fine-tune-radia-v3

This model is a fine-tuned version of meta-llama/Llama-2-7b-hf on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.4142

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: 0.0002
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.3
num_epochs: 6

Training results

Training Loss	Epoch	Step	Validation Loss
1.1521	0.09	5	1.1761
1.1482	0.17	10	1.1711
1.2242	0.26	15	1.1610
1.1821	0.34	20	1.1427
1.1905	0.43	25	1.1125
1.108	0.52	30	1.0602
1.0857	0.6	35	0.9833
0.9924	0.69	40	0.8858
0.8337	0.78	45	0.8109
0.8399	0.86	50	0.7848
0.7903	0.95	55	0.7641
0.8023	1.03	60	0.7455
0.7848	1.12	65	0.7305
0.7655	1.21	70	0.7155
0.7169	1.29	75	0.6981
0.7095	1.38	80	0.6774
0.6762	1.47	85	0.6524
0.7198	1.55	90	0.6331
0.6491	1.64	95	0.6082
0.6619	1.72	100	0.5881
0.645	1.81	105	0.5722
0.5815	1.9	110	0.5592
0.5906	1.98	115	0.5439
0.6282	2.07	120	0.5348
0.5758	2.16	125	0.5222
0.5252	2.24	130	0.5097
0.5269	2.33	135	0.5015
0.4826	2.41	140	0.4897
0.5443	2.5	145	0.4823
0.4485	2.59	150	0.4757
0.4363	2.67	155	0.4718
0.4674	2.76	160	0.4649
0.5294	2.84	165	0.4581
0.4529	2.93	170	0.4536
0.4642	3.02	175	0.4488
0.4224	3.1	180	0.4464
0.3956	3.19	185	0.4477
0.4081	3.28	190	0.4424
0.4678	3.36	195	0.4414
0.4652	3.45	200	0.4375
0.4233	3.53	205	0.4360
0.3717	3.62	210	0.4320
0.4267	3.71	215	0.4262
0.3676	3.79	220	0.4263
0.411	3.88	225	0.4206
0.4197	3.97	230	0.4215
0.373	4.05	235	0.4239
0.3347	4.14	240	0.4224
0.3834	4.22	245	0.4192
0.3055	4.31	250	0.4219
0.359	4.4	255	0.4182
0.3374	4.48	260	0.4219
0.3173	4.57	265	0.4207
0.3619	4.66	270	0.4153
0.3975	4.74	275	0.4141
0.4167	4.83	280	0.4135
0.42	4.91	285	0.4111
0.3459	5.0	290	0.4137
0.2955	5.09	295	0.4173
0.3668	5.17	300	0.4185
0.324	5.26	305	0.4152
0.3058	5.34	310	0.4162
0.3568	5.43	315	0.4155
0.3144	5.52	320	0.4151
0.2713	5.6	325	0.4153
0.3373	5.69	330	0.4143
0.3762	5.78	335	0.4140
0.3217	5.86	340	0.4140
0.3507	5.95	345	0.4142

Framework versions

Transformers 4.36.0.dev0
Pytorch 2.1.0+cu118
Datasets 2.14.7
Tokenizers 0.15.0

Created: Nov 16, 2023
Updated: Nov 16, 2023