End of training

README.md

@@ -1,7 +1,12 @@
 ---
 base_model: slm-research-vn/Qwen2-7B-Instruct-SPPO-Function-call-v2.1
+datasets:
+- slm-research-vn/dpo-format-function-calling-v2
+- slm-research-vn/dpo-format-glaive-code-assistant-v3-with-mistral-large-slm-iter4
+- argilla/dpo-mix-7k
 library_name: peft
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
@@ -15,17 +20,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Qwen2-7B-Instruct-SPPO-Function-call-v2.4
 
-This model is a fine-tuned version of [slm-research-vn/Qwen2-7B-Instruct-SPPO-Function-call-v2.1](https://huggingface.co/slm-research-vn/Qwen2-7B-Instruct-SPPO-Function-call-v2.1) on the None dataset.
+This model is a fine-tuned version of [slm-research-vn/Qwen2-7B-Instruct-SPPO-Function-call-v2.1](https://huggingface.co/slm-research-vn/Qwen2-7B-Instruct-SPPO-Function-call-v2.1) on the slm-research-vn/dpo-format-function-calling-v2, the slm-research-vn/dpo-format-glaive-code-assistant-v3-with-mistral-large-slm-iter4 and the argilla/dpo-mix-7k datasets.
 It achieves the following results on the evaluation set:
-- Loss: 0.4348
-- Rewards/chosen: 1.2990
-- Rewards/rejected: 0.2819
-- Rewards/accuracies: 0.8222
-- Rewards/margins: 1.0172
-- Logps/rejected: -333.4369
-- Logps/chosen: -261.1283
-- Logits/rejected: -0.7185
-- Logits/chosen: -0.6937
+- Loss: 0.4345
+- Rewards/chosen: 1.3033
+- Rewards/rejected: 0.2776
+- Rewards/accuracies: 0.8185
+- Rewards/margins: 1.0258
+- Logps/rejected: -333.5228
+- Logps/chosen: -261.0424
+- Logits/rejected: -0.7224
+- Logits/chosen: -0.7089
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 0.9996824388694824,
+    "eval_logits/chosen": -0.7089246511459351,
+    "eval_logits/rejected": -0.7224333882331848,
+    "eval_logps/chosen": -261.0423889160156,
+    "eval_logps/rejected": -333.52276611328125,
+    "eval_loss": 0.4344652593135834,
+    "eval_rewards/accuracies": 0.8185185194015503,
+    "eval_rewards/chosen": 1.3033398389816284,
+    "eval_rewards/margins": 1.0257779359817505,
+    "eval_rewards/rejected": 0.27756187319755554,
+    "eval_runtime": 185.0205,
+    "eval_samples": 2155,
+    "eval_samples_per_second": 11.647,
+    "eval_steps_per_second": 1.459,
     "total_flos": 0.0,
     "train_loss": 0.5202696668753327,
     "train_runtime": 5902.1361,

config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "slm-research-vn/Qwen2-7B-Instruct-SPPO-Function-call-v2.1",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "rms_norm_eps": 1e-06,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.44.0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 0.9996824388694824,
+    "eval_logits/chosen": -0.7089246511459351,
+    "eval_logits/rejected": -0.7224333882331848,
+    "eval_logps/chosen": -261.0423889160156,
+    "eval_logps/rejected": -333.52276611328125,
+    "eval_loss": 0.4344652593135834,
+    "eval_rewards/accuracies": 0.8185185194015503,
+    "eval_rewards/chosen": 1.3033398389816284,
+    "eval_rewards/margins": 1.0257779359817505,
+    "eval_rewards/rejected": 0.27756187319755554,
+    "eval_runtime": 185.0205,
+    "eval_samples": 2155,
+    "eval_samples_per_second": 11.647,
+    "eval_steps_per_second": 1.459
+}