You can deploy this model and start earning money today!

RichardErkhov/
win10_-_phi3.5-pro-10-08-gguf

Name: win10_-_phi3.5-pro-10-08-gguf
Availability: InStock

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

phi3.5-pro-10-08 - GGUF

Model creator: https://huggingface.co/win10/
Original model: https://huggingface.co/win10/phi3.5-pro-10-08/

Name	Quant method	Size
phi3.5-pro-10-08.Q2_K.gguf	Q2_K	1.7GB
phi3.5-pro-10-08.IQ3_XS.gguf	IQ3_XS	1.95GB
phi3.5-pro-10-08.IQ3_S.gguf	IQ3_S	2.02GB
phi3.5-pro-10-08.Q3_K_S.gguf	Q3_K_S	2.02GB
phi3.5-pro-10-08.IQ3_M.gguf	IQ3_M	2.23GB
phi3.5-pro-10-08.Q3_K.gguf	Q3_K	2.35GB
phi3.5-pro-10-08.Q3_K_M.gguf	Q3_K_M	2.35GB
phi3.5-pro-10-08.Q3_K_L.gguf	Q3_K_L	2.52GB
phi3.5-pro-10-08.IQ4_XS.gguf	IQ4_XS	2.5GB
phi3.5-pro-10-08.Q4_0.gguf	Q4_0	2.62GB
phi3.5-pro-10-08.IQ4_NL.gguf	IQ4_NL	2.63GB
phi3.5-pro-10-08.Q4_K_S.gguf	Q4_K_S	2.63GB
phi3.5-pro-10-08.Q4_K.gguf	Q4_K	2.89GB
phi3.5-pro-10-08.Q4_K_M.gguf	Q4_K_M	2.89GB
phi3.5-pro-10-08.Q4_1.gguf	Q4_1	2.9GB
phi3.5-pro-10-08.Q5_0.gguf	Q5_0	3.19GB
phi3.5-pro-10-08.Q5_K_S.gguf	Q5_K_S	3.19GB
phi3.5-pro-10-08.Q5_K.gguf	Q5_K	3.4GB
phi3.5-pro-10-08.Q5_K_M.gguf	Q5_K_M	3.4GB
phi3.5-pro-10-08.Q5_1.gguf	Q5_1	3.47GB
phi3.5-pro-10-08.Q6_K.gguf	Q6_K	3.79GB
phi3.5-pro-10-08.Q8_0.gguf	Q8_0	4.9GB

Original model description:

library_name: transformers license: other tags:

freeze model-index:
name: phi3.5-pro-10-08 results: []

phi3.5-pro-10-08

This model is a fine-tuned version of E:\mergekit\phi3.5-pro on the Magpie_Qwen2_Pro_300k and the Magpie_Llama_3.1_Pro_MT_300k datasets. It achieves the following results on the evaluation set:

Loss: 0.8538

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: 0.0001
train_batch_size: 2
eval_batch_size: 2
seed: 42
gradient_accumulation_steps: 8
total_train_batch_size: 16
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
training_steps: 1000
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.7784	0.0080	100	0.8538
0.8605	0.0160	200	0.8538
0.8386	0.0240	300	0.8538
0.8201	0.0320	400	0.8538
0.8661	0.0400	500	0.8538
0.776	0.0480	600	0.8538
0.8377	0.0561	700	0.8538
0.8541	0.0641	800	0.8538
0.799	0.0721	900	0.8538
0.8146	0.0801	1000	0.8538

Framework versions

Transformers 4.45.0
Pytorch 2.4.0+cu124
Datasets 2.21.0
Tokenizers 0.20.0

RichardErkhov/win10_-_phi3.5-pro-10-08-gguf

Original model description:

phi3.5-pro-10-08

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

RichardErkhov/
win10_-_phi3.5-pro-10-08-gguf