初始化项目，由ModelHub XC社区提供模型

Model: jackf857/llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-q_t-0.43-s_star-0.4 Source: Original Platform
2026-05-26 12:58:23 +08:00
commit 6d0881812d
26 changed files with 13698 additions and 0 deletions
--- a/README.md
+++ b/README.md
@@ -0,0 +1,81 @@
+---
+library_name: transformers
+base_model: W-61/llama-3-8b-base-sft-ultrachat-8xh200
+tags:
+- alignment-handbook
+- new-dpo
+- generated_from_trainer
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
+model-index:
+- name: llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-q_t-0.43-s_star-0.4
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# llama-3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-q_t-0.43-s_star-0.4
+
+This model is a fine-tuned version of [W-61/llama-3-8b-base-sft-ultrachat-8xh200](https://huggingface.co/W-61/llama-3-8b-base-sft-ultrachat-8xh200) on the HuggingFaceH4/ultrafeedback_binarized dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.5983
+- Fcm Dpo/beta: 0.0027
+- Margin Dpo/margin Mean: 105.4011
+- Margin Dpo/margin Std: 182.7466
+- Logps/chosen: -546.8945
+- Logps/rejected: -631.3989
+- Logps/ref Chosen: -287.8268
+- Logps/ref Rejected: -266.9300
+- Kl/chosen Kl Mean: -259.0677
+- Kl/rejected Kl Mean: -364.4689
+- Kl/mean: -311.7683
+- Kl/std: 162.3682
+- Logits/chosen: -0.8539
+- Logits/rejected: -0.8364
+
+## 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: 4
+- eval_batch_size: 2
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 128
+- total_eval_batch_size: 8
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 1
+
+### Training results
+
+| Training Loss | Epoch  | Step | Validation Loss | Fcm Dpo/beta | Margin Dpo/margin Mean | Margin Dpo/margin Std | Logps/chosen | Logps/rejected | Logps/ref Chosen | Logps/ref Rejected | Kl/chosen Kl Mean | Kl/rejected Kl Mean | Kl/mean   | Kl/std   | Logits/chosen | Logits/rejected |
+|:-------------:|:------:|:----:|:---------------:|:------------:|:----------------------:|:---------------------:|:------------:|:--------------:|:----------------:|:------------------:|:-----------------:|:-------------------:|:---------:|:--------:|:-------------:|:---------------:|
+| 4.9173        | 0.4188 | 200  | 0.5938          | 0.0054       | 55.2587                | 96.2979               | -406.3043    | -440.6663      | -287.8268        | -266.9300          | -118.4775         | -173.7362           | -146.1068 | 87.6889  | -0.9034       | -0.8850         |
+| 4.8145        | 0.8377 | 400  | 0.5983          | 0.0027       | 105.4011               | 182.7466              | -546.8945    | -631.3989      | -287.8268        | -266.9300          | -259.0677         | -364.4689           | -311.7683 | 162.3682 | -0.8539       | -0.8364         |
+
+
+### Framework versions
+
+- Transformers 4.51.0
+- Pytorch 2.3.1+cu121
+- Datasets 2.21.0
+- Tokenizers 0.21.4