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: TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
+datasets:
+- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: tinyllama-1.1b-chat-dpo-qlora
   results: []
@@ -16,15 +19,15 @@ should probably proofread and complete it, then remove this comment. -->
 
 # tinyllama-1.1b-chat-dpo-qlora
 
-This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on an unknown dataset.
+This model is a fine-tuned version of [martimfasantos/tinyllama-1.1b-chat-sft-qlora](https://huggingface.co/martimfasantos/tinyllama-1.1b-chat-sft-qlora) on the HuggingFaceH4/ultrafeedback_binarized dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6085
-- Rewards/chosen: -1.0876
-- Rewards/rejected: -1.3914
+- Loss: 0.6084
+- Rewards/chosen: -1.0875
+- Rewards/rejected: -1.3916
 - Rewards/accuracies: 0.6580
-- Rewards/margins: 0.3038
-- Logps/rejected: -490.8211
-- Logps/chosen: -504.9807
+- Rewards/margins: 0.3041
+- Logps/rejected: -490.8393
+- Logps/chosen: -504.9714
 - Logits/rejected: -2.6096
 - Logits/chosen: -2.6425
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.0,
+    "eval_logits/chosen": -2.642491102218628,
+    "eval_logits/rejected": -2.609631299972534,
+    "eval_logps/chosen": -504.97137451171875,
+    "eval_logps/rejected": -490.8393249511719,
+    "eval_loss": 0.6084445714950562,
+    "eval_rewards/accuracies": 0.6579999923706055,
+    "eval_rewards/chosen": -1.0874665975570679,
+    "eval_rewards/margins": 0.3040919303894043,
+    "eval_rewards/rejected": -1.3915585279464722,
+    "eval_runtime": 347.6592,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 5.753,
+    "eval_steps_per_second": 0.719,
     "train_loss": 0.6288731582999011,
     "train_runtime": 37165.2285,
     "train_samples": 61134,

config.json

@@ -0,0 +1,43 @@
+{
+  "_name_or_path": "TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 5632,
+  "max_position_embeddings": 2048,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 22,
+  "num_key_value_heads": 4,
+  "pretraining_tp": 1,
+  "quantization_config": {
+    "_load_in_4bit": true,
+    "_load_in_8bit": false,
+    "bnb_4bit_compute_dtype": "bfloat16",
+    "bnb_4bit_quant_storage": "uint8",
+    "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_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.39.3",
+  "use_cache": true,
+  "vocab_size": 32000
+}

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 1.0,
+    "eval_logits/chosen": -2.642491102218628,
+    "eval_logits/rejected": -2.609631299972534,
+    "eval_logps/chosen": -504.97137451171875,
+    "eval_logps/rejected": -490.8393249511719,
+    "eval_loss": 0.6084445714950562,
+    "eval_rewards/accuracies": 0.6579999923706055,
+    "eval_rewards/chosen": -1.0874665975570679,
+    "eval_rewards/margins": 0.3040919303894043,
+    "eval_rewards/rejected": -1.3915585279464722,
+    "eval_runtime": 347.6592,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 5.753,
+    "eval_steps_per_second": 0.719
+}

runs/Apr24_01-31-15_poseidon/events.out.tfevents.1713959835.poseidon.732971.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:26ee8116844081c9a4472a0ad53547d4f74c6585fc7112bb30015206797f68f0
+size 828