End of training

README.md

@@ -2,10 +2,13 @@
 license: apache-2.0
 library_name: peft
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
 base_model: alignment-handbook/zephyr-7b-sft-full
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: zephyr-7b-dpo-uffull-qlora-5e-7
   results: []
@@ -16,20 +19,20 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-dpo-uffull-qlora-5e-7
 
-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 HuggingFaceH4/ultrafeedback_binarized dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5925
-- Rewards/chosen: -0.2517
-- Rewards/rejected: -0.6019
-- Rewards/accuracies: 0.7341
-- Rewards/margins: 0.3502
-- Rewards/margins Max: 1.2316
-- Rewards/margins Min: -0.5546
+- Loss: 0.5924
+- Rewards/chosen: -0.2516
+- Rewards/rejected: -0.6013
+- Rewards/accuracies: 0.7321
+- Rewards/margins: 0.3497
+- Rewards/margins Max: 1.2300
+- Rewards/margins Min: -0.5547
 - Rewards/margins Std: 0.6038
-- Logps/rejected: -322.3433
-- Logps/chosen: -309.6684
-- Logits/rejected: -2.6801
-- Logits/chosen: -2.7126
+- Logps/rejected: -322.2831
+- Logps/chosen: -309.6581
+- Logits/rejected: -2.6832
+- Logits/chosen: -2.7155
 
 ## Model description
 

all_results.json

@@ -1,5 +1,21 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -2.715510368347168,
+    "eval_logits/rejected": -2.6831843852996826,
+    "eval_logps/chosen": -309.6580505371094,
+    "eval_logps/rejected": -322.2830810546875,
+    "eval_loss": 0.5924258232116699,
+    "eval_rewards/accuracies": 0.7321428656578064,
+    "eval_rewards/chosen": -0.2516312599182129,
+    "eval_rewards/margins": 0.3497060239315033,
+    "eval_rewards/margins_max": 1.2300177812576294,
+    "eval_rewards/margins_min": -0.5546701550483704,
+    "eval_rewards/margins_std": 0.6038431525230408,
+    "eval_rewards/rejected": -0.6013372540473938,
+    "eval_runtime": 424.8015,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 4.708,
+    "eval_steps_per_second": 0.148,
     "train_loss": 0.619671463092813,
     "train_runtime": 44477.573,
     "train_samples": 61134,

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": -2.715510368347168,
+    "eval_logits/rejected": -2.6831843852996826,
+    "eval_logps/chosen": -309.6580505371094,
+    "eval_logps/rejected": -322.2830810546875,
+    "eval_loss": 0.5924258232116699,
+    "eval_rewards/accuracies": 0.7321428656578064,
+    "eval_rewards/chosen": -0.2516312599182129,
+    "eval_rewards/margins": 0.3497060239315033,
+    "eval_rewards/margins_max": 1.2300177812576294,
+    "eval_rewards/margins_min": -0.5546701550483704,
+    "eval_rewards/margins_std": 0.6038431525230408,
+    "eval_rewards/rejected": -0.6013372540473938,
+    "eval_runtime": 424.8015,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 4.708,
+    "eval_steps_per_second": 0.148
+}

runs/Jul22_23-14-44_node26/events.out.tfevents.1721703042.node26.3249241.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5923c5a556d7f3c6ad13accf57059669986d82813f65d311bb2a2b8d10bf524e
+size 1017