End of training

README.md

@@ -1,8 +1,11 @@
 ---
 base_model: alignment-handbook/zephyr-7b-sft-full
+datasets:
+- generation/GPT4
 library_name: peft
 license: apache-2.0
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
@@ -16,20 +19,20 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-dpo-qlora-gpt4-5e-6
 
-This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the None dataset.
+This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the generation/GPT4 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.9371
-- Rewards/chosen: -6.0411
-- Rewards/rejected: -6.9811
-- Rewards/accuracies: 0.6450
-- Rewards/margins: 0.9400
-- Rewards/margins Max: 4.7437
-- Rewards/margins Min: -3.0368
-- Rewards/margins Std: 2.6550
-- Logps/rejected: -956.6877
-- Logps/chosen: -888.7039
-- Logits/rejected: -1.4320
-- Logits/chosen: -1.4909
+- Loss: 0.9318
+- Rewards/chosen: -6.0570
+- Rewards/rejected: -7.0161
+- Rewards/accuracies: 0.6430
+- Rewards/margins: 0.9590
+- Rewards/margins Max: 4.7709
+- Rewards/margins Min: -3.0068
+- Rewards/margins Std: 2.6570
+- Logps/rejected: -960.1845
+- Logps/chosen: -890.2942
+- Logits/rejected: -1.4367
+- Logits/chosen: -1.4956
 
 ## Model description
 

all_results.json

@@ -1,5 +1,21 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -1.4956016540527344,
+    "eval_logits/rejected": -1.4366858005523682,
+    "eval_logps/chosen": -890.294189453125,
+    "eval_logps/rejected": -960.1845092773438,
+    "eval_loss": 0.9318017959594727,
+    "eval_rewards/accuracies": 0.6430000066757202,
+    "eval_rewards/chosen": -6.057007789611816,
+    "eval_rewards/margins": 0.9590479731559753,
+    "eval_rewards/margins_max": 4.770871162414551,
+    "eval_rewards/margins_min": -3.0068037509918213,
+    "eval_rewards/margins_std": 2.65700101852417,
+    "eval_rewards/rejected": -7.016055107116699,
+    "eval_runtime": 428.551,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 4.667,
+    "eval_steps_per_second": 0.292,
     "train_loss": 0.24729210323011372,
     "train_runtime": 4030.3896,
     "train_samples": 5678,

config.json

@@ -0,0 +1,40 @@
+{
+  "_name_or_path": "alignment-handbook/zephyr-7b-sft-full",
+  "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,
+  "quantization_config": {
+    "_load_in_4bit": true,
+    "_load_in_8bit": false,
+    "bnb_4bit_compute_dtype": "bfloat16",
+    "bnb_4bit_quant_type": "nf4",
+    "bnb_4bit_use_double_quant": false,
+    "llm_int8_enable_fp32_cpu_offload": false,
+    "llm_int8_has_fp16_weight": false,
+    "llm_int8_skip_modules": null,
+    "llm_int8_threshold": 6.0,
+    "load_in_4bit": true,
+    "load_in_8bit": false,
+    "quant_method": "bitsandbytes"
+  },
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.39.0.dev0",
+  "use_cache": true,
+  "vocab_size": 32000
+}

eval_results.json

@@ -0,0 +1,19 @@
+{
+    "epoch": 1.0,
+    "eval_logits/chosen": -1.4956016540527344,
+    "eval_logits/rejected": -1.4366858005523682,
+    "eval_logps/chosen": -890.294189453125,
+    "eval_logps/rejected": -960.1845092773438,
+    "eval_loss": 0.9318017959594727,
+    "eval_rewards/accuracies": 0.6430000066757202,
+    "eval_rewards/chosen": -6.057007789611816,
+    "eval_rewards/margins": 0.9590479731559753,
+    "eval_rewards/margins_max": 4.770871162414551,
+    "eval_rewards/margins_min": -3.0068037509918213,
+    "eval_rewards/margins_std": 2.65700101852417,
+    "eval_rewards/rejected": -7.016055107116699,
+    "eval_runtime": 428.551,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 4.667,
+    "eval_steps_per_second": 0.292
+}

runs/Jul28_07-09-30_notebook-deployment-48-7d9b6c99-khd85/events.out.tfevents.1722155175.notebook-deployment-48-7d9b6c99-khd85.1186377.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7927582e389e6e28d580165e70a4f5cf4a71bdfae296b5ca13624e6271a5f579
+size 1017