tiny-llama-lora

xshubhamx/tiny-llama-lora

tiny-llama-lora is a machine learning model from xshubhamx. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.

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

Downloads · 30 days

Access

Public

Updated Apr 19, 2024

Repo size

1.8 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.pt1.1 GB · 62%

At a glance

License: apache-2.0
Access: Public
Created: Apr 16, 2024
Updated: Apr 19, 2024
SHA: a84e55a2

Base models

TinyLlama/TinyLlama-1.1B-Chat-v1.0

Model card

From the Hugging Face model README

tiny-llama

This model is a fine-tuned version of TinyLlama/TinyLlama-1.1B-Chat-v1.0 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.6966
Accuracy: 0.8195
Precision: 0.8222
Recall: 0.8195
Precision Macro: 0.7955
Recall Macro: 0.7536
Macro Fpr: 0.0148
Weighted Fpr: 0.0141
Weighted Specificity: 0.9765
Macro Specificity: 0.9873
Weighted Sensitivity: 0.8327
Macro Sensitivity: 0.7536
F1 Micro: 0.8327
F1 Macro: 0.7609
F1 Weighted: 0.8291

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: 2
eval_batch_size: 2
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 8
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 10
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	Precision Macro	Recall Macro	Macro Fpr	Weighted Fpr	Weighted Specificity	Macro Specificity	Weighted Sensitivity	Macro Sensitivity	F1 Micro	F1 Macro	F1 Weighted
1.0444	1.0	642	0.5968	0.8056	0.8050	0.8056	0.7122	0.6995	0.0175	0.0169	0.9730	0.9852	0.8056	0.6995	0.8056	0.6986	0.8014
0.4788	2.0	1284	0.6966	0.8195	0.8222	0.8195	0.8092	0.7825	0.0161	0.0155	0.9755	0.9863	0.8195	0.7825	0.8195	0.7849	0.8172
0.3354	3.0	1926	0.8046	0.8327	0.8276	0.8327	0.8058	0.7582	0.0148	0.0141	0.9758	0.9872	0.8327	0.7582	0.8327	0.7742	0.8282
0.0571	4.0	2569	1.1143	0.8265	0.8312	0.8265	0.7904	0.7763	0.0152	0.0148	0.9772	0.9869	0.8265	0.7763	0.8265	0.7690	0.8262
0.0187	5.0	3211	1.1104	0.8319	0.8316	0.8319	0.7745	0.7724	0.0149	0.0142	0.9770	0.9873	0.8319	0.7724	0.8319	0.7638	0.8303
0.0071	6.0	3853	1.1445	0.8242	0.8210	0.8242	0.7684	0.7384	0.0157	0.0150	0.9755	0.9866	0.8242	0.7384	0.8242	0.7451	0.8209
0.0002	7.0	4495	1.2032	0.8327	0.8302	0.8327	0.7985	0.7529	0.0148	0.0141	0.9765	0.9873	0.8327	0.7529	0.8327	0.7617	0.8293
0.0028	8.0	5138	1.1918	0.8257	0.8226	0.8257	0.7738	0.7493	0.0155	0.0149	0.9756	0.9868	0.8257	0.7493	0.8257	0.7552	0.8229
0.0	9.0	5780	1.2181	0.8311	0.8286	0.8311	0.7935	0.7522	0.0150	0.0143	0.9764	0.9872	0.8311	0.7522	0.8311	0.7592	0.8276
0.0018	10.0	6420	1.2265	0.8327	0.8301	0.8327	0.7955	0.7536	0.0148	0.0141	0.9765	0.9873	0.8327	0.7536	0.8327	0.7609	0.8291

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu121
Datasets 2.18.0
Tokenizers 0.15.1

License: apache-2.0
Created: Apr 16, 2024
Updated: Apr 19, 2024