End of training

README.md

@@ -1,7 +1,12 @@
 ---
 base_model: slm-research-vn/Qwen2-7B-Instruct-SPPO-Function-call-v2.8
+datasets:
+- slm-research-vn/dpo-format-function-calling-v4
+- 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-v2.12
 
-This model is a fine-tuned version of [slm-research-vn/Qwen2-7B-Instruct-SPPO-Function-call-v2.8](https://huggingface.co/slm-research-vn/Qwen2-7B-Instruct-SPPO-Function-call-v2.8) on the None dataset.
+This model is a fine-tuned version of [slm-research-vn/Qwen2-7B-Instruct-SPPO-Function-call-v2.8](https://huggingface.co/slm-research-vn/Qwen2-7B-Instruct-SPPO-Function-call-v2.8) on the slm-research-vn/dpo-format-function-calling-v4, 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: 0.3328
-- Rewards/chosen: 0.5530
-- Rewards/rejected: -0.6974
-- Rewards/accuracies: 0.9075
-- Rewards/margins: 1.2504
-- Logps/rejected: -278.6715
-- Logps/chosen: -129.0585
-- Logits/rejected: -0.5998
-- Logits/chosen: -0.7754
+- Loss: 0.3322
+- Rewards/chosen: 0.5523
+- Rewards/rejected: -0.7005
+- Rewards/accuracies: 0.9017
+- Rewards/margins: 1.2528
+- Logps/rejected: -278.7327
+- Logps/chosen: -129.0717
+- Logits/rejected: -0.5984
+- Logits/chosen: -0.7738
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 0.9995419147961521,
+    "eval_logits/chosen": -0.7738187313079834,
+    "eval_logits/rejected": -0.5984290242195129,
+    "eval_logps/chosen": -129.0717315673828,
+    "eval_logps/rejected": -278.73272705078125,
+    "eval_loss": 0.33219456672668457,
+    "eval_rewards/accuracies": 0.9017341136932373,
+    "eval_rewards/chosen": 0.552308976650238,
+    "eval_rewards/margins": 1.252788782119751,
+    "eval_rewards/rejected": -0.7004798650741577,
+    "eval_runtime": 251.2644,
+    "eval_samples": 2763,
+    "eval_samples_per_second": 10.996,
+    "eval_steps_per_second": 1.377,
     "total_flos": 0.0,
     "train_loss": 0.4306906612229719,
     "train_runtime": 8689.0202,

config.json

@@ -0,0 +1,27 @@
+{
+  "_name_or_path": "slm-research-vn/Qwen2-7B-Instruct-SPPO-Function-call-v2.8",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "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": 0.9995419147961521,
+    "eval_logits/chosen": -0.7738187313079834,
+    "eval_logits/rejected": -0.5984290242195129,
+    "eval_logps/chosen": -129.0717315673828,
+    "eval_logps/rejected": -278.73272705078125,
+    "eval_loss": 0.33219456672668457,
+    "eval_rewards/accuracies": 0.9017341136932373,
+    "eval_rewards/chosen": 0.552308976650238,
+    "eval_rewards/margins": 1.252788782119751,
+    "eval_rewards/rejected": -0.7004798650741577,
+    "eval_runtime": 251.2644,
+    "eval_samples": 2763,
+    "eval_samples_per_second": 10.996,
+    "eval_steps_per_second": 1.377
+}