You can deploy this model and start earning money today!

RichardErkhov/
tanliboy_-_zephyr-qwen2-7b-dpo-gguf

Name: tanliboy_-_zephyr-qwen2-7b-dpo-gguf
Availability: InStock

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

zephyr-qwen2-7b-dpo - GGUF

Model creator: https://huggingface.co/tanliboy/
Original model: https://huggingface.co/tanliboy/zephyr-qwen2-7b-dpo/

Name	Quant method	Size
zephyr-qwen2-7b-dpo.Q2_K.gguf	Q2_K	2.81GB
zephyr-qwen2-7b-dpo.IQ3_XS.gguf	IQ3_XS	3.12GB
zephyr-qwen2-7b-dpo.IQ3_S.gguf	IQ3_S	3.26GB
zephyr-qwen2-7b-dpo.Q3_K_S.gguf	Q3_K_S	3.25GB
zephyr-qwen2-7b-dpo.IQ3_M.gguf	IQ3_M	3.33GB
zephyr-qwen2-7b-dpo.Q3_K.gguf	Q3_K	3.55GB
zephyr-qwen2-7b-dpo.Q3_K_M.gguf	Q3_K_M	3.55GB
zephyr-qwen2-7b-dpo.Q3_K_L.gguf	Q3_K_L	3.81GB
zephyr-qwen2-7b-dpo.IQ4_XS.gguf	IQ4_XS	3.96GB
zephyr-qwen2-7b-dpo.Q4_0.gguf	Q4_0	4.13GB
zephyr-qwen2-7b-dpo.IQ4_NL.gguf	IQ4_NL	4.16GB
zephyr-qwen2-7b-dpo.Q4_K_S.gguf	Q4_K_S	4.15GB
zephyr-qwen2-7b-dpo.Q4_K.gguf	Q4_K	4.36GB
zephyr-qwen2-7b-dpo.Q4_K_M.gguf	Q4_K_M	4.36GB
zephyr-qwen2-7b-dpo.Q4_1.gguf	Q4_1	4.54GB
zephyr-qwen2-7b-dpo.Q5_0.gguf	Q5_0	4.95GB
zephyr-qwen2-7b-dpo.Q5_K_S.gguf	Q5_K_S	4.95GB
zephyr-qwen2-7b-dpo.Q5_K.gguf	Q5_K	5.07GB
zephyr-qwen2-7b-dpo.Q5_K_M.gguf	Q5_K_M	5.07GB
zephyr-qwen2-7b-dpo.Q5_1.gguf	Q5_1	5.36GB
zephyr-qwen2-7b-dpo.Q6_K.gguf	Q6_K	5.82GB
zephyr-qwen2-7b-dpo.Q8_0.gguf	Q8_0	7.54GB

Original model description:

license: apache-2.0 base_model: tanliboy/zephyr-qwen2-7b-sft tags:

alignment-handbook
trl
dpo
generated_from_trainer
trl
dpo
generated_from_trainer datasets:
HuggingFaceH4/ultrafeedback_binarized model-index:
name: zephyr-qwen2-7b-dpo results: []

zephyr-qwen2-7b-dpo

This model is a fine-tuned version of tanliboy/zephyr-qwen2-7b-sft on the HuggingFaceH4/ultrafeedback_binarized dataset. It achieves the following results on the evaluation set:

Loss: 0.5492
Rewards/chosen: -0.3377
Rewards/rejected: -0.7347
Rewards/accuracies: 0.7120
Rewards/margins: 0.3970
Logps/rejected: -339.1164
Logps/chosen: -334.2959
Logits/rejected: 0.0096
Logits/chosen: 0.1430

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-07
train_batch_size: 2
eval_batch_size: 2
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 8
total_train_batch_size: 128
total_eval_batch_size: 16
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.6439	0.2094	100	0.6410	-0.0729	-0.1774	0.6280	0.1045	-283.3870	-307.8225	-0.0798	0.0658
0.5947	0.4187	200	0.5743	-0.3018	-0.6030	0.6800	0.3013	-325.9524	-330.7078	-0.0418	0.1044
0.5465	0.6281	300	0.5582	-0.2623	-0.6201	0.7080	0.3578	-327.6540	-326.7564	0.0043	0.1374
0.5513	0.8375	400	0.5497	-0.3456	-0.7435	0.7040	0.3979	-340.0029	-335.0901	0.0097	0.1434

Framework versions

Transformers 4.40.2
Pytorch 2.3.0+cu121
Datasets 2.19.1
Tokenizers 0.19.1

RichardErkhov/tanliboy_-_zephyr-qwen2-7b-dpo-gguf

Original model description:

zephyr-qwen2-7b-dpo

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

RichardErkhov/
tanliboy_-_zephyr-qwen2-7b-dpo-gguf