End of training

README.md

@@ -2,10 +2,16 @@
 license: apache-2.0
 library_name: peft
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
 base_model: mistralai/Mistral-7B-Instruct-v0.2
+datasets:
+- nthakur/multilingual-ultrafeedback-binarized-dpo-v0.1
+- nthakur/multilingual-distilabel-intel-orca-dpo-pairs-v0.1
+- nthakur/multilingual-truthy-dpo-pairs-v0.1
+- nthakur/GSM8KInstruct-Parallel-instruct-dpo-v0.1
 model-index:
 - name: Mistral-7B-Instruct-v0.2-multilingual-dpo-v1.0-v2
   results: []
@@ -16,17 +22,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Mistral-7B-Instruct-v0.2-multilingual-dpo-v1.0-v2
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the None dataset.
+This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the nthakur/multilingual-ultrafeedback-binarized-dpo-v0.1, the nthakur/multilingual-distilabel-intel-orca-dpo-pairs-v0.1, the nthakur/multilingual-truthy-dpo-pairs-v0.1 and the nthakur/GSM8KInstruct-Parallel-instruct-dpo-v0.1 datasets.
 It achieves the following results on the evaluation set:
-- Loss: 0.1328
-- Rewards/chosen: -2.7014
-- Rewards/rejected: -12.3976
-- Rewards/accuracies: 0.9383
-- Rewards/margins: 9.6962
-- Logps/rejected: -1531.6849
-- Logps/chosen: -609.8344
-- Logits/rejected: 0.5002
-- Logits/chosen: 0.3026
+- Loss: 0.1324
+- Rewards/chosen: -2.6738
+- Rewards/rejected: -12.2394
+- Rewards/accuracies: 0.9377
+- Rewards/margins: 9.5656
+- Logps/rejected: -1515.8665
+- Logps/chosen: -607.0774
+- Logits/rejected: 0.4952
+- Logits/chosen: 0.3030
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 0.9998638529611981,
+    "eval_logits/chosen": 0.303020715713501,
+    "eval_logits/rejected": 0.4951964318752289,
+    "eval_logps/chosen": -607.077392578125,
+    "eval_logps/rejected": -1515.866455078125,
+    "eval_loss": 0.13239584863185883,
+    "eval_rewards/accuracies": 0.9376528263092041,
+    "eval_rewards/chosen": -2.673825979232788,
+    "eval_rewards/margins": 9.565610885620117,
+    "eval_rewards/rejected": -12.239436149597168,
+    "eval_runtime": 3761.8245,
+    "eval_samples": 4903,
+    "eval_samples_per_second": 1.303,
+    "eval_steps_per_second": 0.109,
     "total_flos": 0.0,
     "train_loss": 0.32278433931516665,
     "train_runtime": 141933.118,

config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "mistralai/Mistral-7B-Instruct-v0.2",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 32768,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.41.2",
+  "use_cache": true,
+  "vocab_size": 32000
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 0.9998638529611981,
+    "eval_logits/chosen": 0.303020715713501,
+    "eval_logits/rejected": 0.4951964318752289,
+    "eval_logps/chosen": -607.077392578125,
+    "eval_logps/rejected": -1515.866455078125,
+    "eval_loss": 0.13239584863185883,
+    "eval_rewards/accuracies": 0.9376528263092041,
+    "eval_rewards/chosen": -2.673825979232788,
+    "eval_rewards/margins": 9.565610885620117,
+    "eval_rewards/rejected": -12.239436149597168,
+    "eval_runtime": 3761.8245,
+    "eval_samples": 4903,
+    "eval_samples_per_second": 1.303,
+    "eval_steps_per_second": 0.109
+}