mlabonne_-_chesspythia-70m-gguf

RichardErkhov/mlabonne_-_chesspythia-70m-gguf

mlabonne_-_chesspythia-70m-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.

Downloads · 30 days

1% of all-time downloads

All-time downloads

3.6K

Public

Repo size

943 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.gguf1.1 GB · 100%

At a glance

Access: Public
Created: Jul 2, 2024
Updated: Jul 2, 2024
SHA: 2fc2079a

Model card

From the Hugging Face model README

Quantization made by Richard Erkhov.

Github

Discord

Request more models

chesspythia-70m - GGUF

Model creator: https://huggingface.co/mlabonne/
Original model: https://huggingface.co/mlabonne/chesspythia-70m/

Name	Quant method	Size
chesspythia-70m.Q2_K.gguf	Q2_K	0.04GB
chesspythia-70m.IQ3_XS.gguf	IQ3_XS	0.04GB
chesspythia-70m.IQ3_S.gguf	IQ3_S	0.04GB
chesspythia-70m.Q3_K_S.gguf	Q3_K_S	0.04GB
chesspythia-70m.IQ3_M.gguf	IQ3_M	0.04GB
chesspythia-70m.Q3_K.gguf	Q3_K	0.04GB
chesspythia-70m.Q3_K_M.gguf	Q3_K_M	0.04GB
chesspythia-70m.Q3_K_L.gguf	Q3_K_L	0.04GB
chesspythia-70m.IQ4_XS.gguf	IQ4_XS	0.04GB
chesspythia-70m.Q4_0.gguf	Q4_0	0.04GB
chesspythia-70m.IQ4_NL.gguf	IQ4_NL	0.04GB
chesspythia-70m.Q4_K_S.gguf	Q4_K_S	0.04GB
chesspythia-70m.Q4_K.gguf	Q4_K	0.05GB
chesspythia-70m.Q4_K_M.gguf	Q4_K_M	0.05GB
chesspythia-70m.Q4_1.gguf	Q4_1	0.05GB
chesspythia-70m.Q5_0.gguf	Q5_0	0.05GB
chesspythia-70m.Q5_K_S.gguf	Q5_K_S	0.05GB
chesspythia-70m.Q5_K.gguf	Q5_K	0.05GB
chesspythia-70m.Q5_K_M.gguf	Q5_K_M	0.05GB
chesspythia-70m.Q5_1.gguf	Q5_1	0.05GB
chesspythia-70m.Q6_K.gguf	Q6_K	0.06GB
chesspythia-70m.Q8_0.gguf	Q8_0	0.07GB

Original model description:

license: apache-2.0 base_model: EleutherAI/pythia-70m-deduped tags:

generated_from_trainer model-index:
name: results results: []

results

This model is a fine-tuned version of EleutherAI/pythia-70m-deduped on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 1.2691

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: 100
eval_batch_size: 100
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss
2.852	0.1	1	3.1074
3.0923	0.2	2	2.3879
2.3371	0.3	3	2.1025
2.1166	0.4	4	1.9761
2.0538	0.5	5	1.8446
1.8972	0.6	6	1.7470
1.8356	0.7	7	1.6615
1.702	0.8	8	1.6187
1.6907	0.9	9	1.6626
1.5877	1.0	10	1.6192
1.6332	1.1	11	1.5464
1.4906	1.2	12	1.5091
1.5267	1.3	13	1.4850
1.4857	1.4	14	1.4572
1.4247	1.5	15	1.4319
1.4815	1.6	16	1.4207
1.3584	1.7	17	1.4092
1.4812	1.8	18	1.4196
1.4381	1.9	19	1.4021
1.453	2.0	20	1.4013
1.3468	2.1	21	1.3781
1.3327	2.2	22	1.3598
1.3623	2.3	23	1.3516
1.2876	2.4	24	1.3384
1.374	2.5	25	1.3366
1.3863	2.6	26	1.3265
1.3327	2.7	27	1.3186
1.2886	2.8	28	1.3130
1.3842	2.9	29	1.3024
1.3105	3.0	30	1.2986
1.2331	3.1	31	1.2966
1.3227	3.2	32	1.2954
1.2923	3.3	33	1.2928
1.2976	3.4	34	1.2901
1.3207	3.5	35	1.2879
1.2455	3.6	36	1.2834
1.2546	3.7	37	1.2779
1.2999	3.8	38	1.2744
1.2484	3.9	39	1.2723
1.281	4.0	40	1.2720
1.2134	4.1	41	1.2722
1.214	4.2	42	1.2721
1.3031	4.3	43	1.2715
1.2174	4.4	44	1.2708
1.2359	4.5	45	1.2703
1.2578	4.6	46	1.2699
1.2815	4.7	47	1.2695
1.2866	4.8	48	1.2693
1.2878	4.9	49	1.2691
1.2214	5.0	50	1.2691

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu121
Datasets 2.16.1
Tokenizers 0.15.0

Created: Jul 2, 2024
Updated: Jul 2, 2024