sparse_llama_7b_refined_web_50p_2024-03-24 – AI Model by thrunlab | AlphaNeural AI

You can deploy this model and start earning money today!

thrunlab/
sparse_llama_7b_refined_web_50p_2024-03-24

like

0

text-generation

generated_from_trainer

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

autotrain_compatible

Views

No views yet

sparse_llama_7b_refined_web_50p_2024-03-24

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

Loss: 2.1950

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: 1e-05
train_batch_size: 1
eval_batch_size: 1
seed: 0
distributed_type: multi-GPU
num_devices: 4
gradient_accumulation_steps: 8
total_train_batch_size: 32
total_eval_batch_size: 4
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
training_steps: 800

Training results

Training Loss	Epoch	Step	Validation Loss
2.4969	0.01	25	2.7682
2.4532	0.02	50	2.7136
2.4855	0.02	75	2.6372
2.4368	0.03	100	2.6029
2.4952	0.04	125	2.5761
2.3209	0.05	150	2.5665
2.2798	0.06	175	2.5517
2.4447	0.06	200	2.5399
2.4008	0.07	225	2.5317
2.3508	0.08	250	2.5271
2.2851	0.09	275	2.5222
2.3171	0.1	300	2.5151
2.3594	0.1	325	2.5102
2.3233	0.11	350	2.5063
2.2479	0.12	375	2.5039
2.3484	0.13	400	2.5004
2.3252	0.14	425	2.4961
2.2819	0.14	450	2.4951
2.3504	0.15	475	2.4907
2.3745	0.16	500	2.4860
2.2705	0.17	525	2.4860
2.271	0.18	550	2.4836
2.3821	0.18	575	2.4820
2.2663	0.19	600	2.4795
2.2919	0.2	625	2.4764
2.3755	0.21	650	2.4718
2.2654	0.22	675	2.4745
2.2857	0.22	700	2.4723
2.3063	0.23	725	2.4716
2.2062	0.24	750	2.4698
2.2921	0.25	775	2.4664
2.3404	0.26	800	2.4676

Framework versions

Transformers 4.36.2
Pytorch 2.1.1+cu121
Datasets 2.15.0
Tokenizers 0.15.2