Model save

README.md

@@ -7,6 +7,7 @@ license: llama3.1
 tags:
 - trl
 - sft
+- alignment-handbook
 - generated_from_trainer
 model-index:
 - name: llama-3-1-sft-qlora-test
@@ -19,8 +20,6 @@ should probably proofread and complete it, then remove this comment. -->
 # llama-3-1-sft-qlora-test
 
 This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on the generator dataset.
-It achieves the following results on the evaluation set:
-- Loss: 1.1645
 
 ## Model description
 
@@ -53,9 +52,6 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss |
-|:-------------:|:------:|:----:|:---------------:|
-| 1.2385        | 0.9934 | 75   | 1.1645          |
 
 
 ### Framework versions

adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "v_proj",
     "gate_proj",
     "k_proj",
-    "o_proj",
     "q_proj",
-    "down_proj",
-    "up_proj"
+    "o_proj",
+    "v_proj",
+    "up_proj",
+    "down_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

all_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 0.9933774834437086,
     "total_flos": 5.564154814608179e+16,
-    "train_loss": 1.2146572240193685,
-    "train_runtime": 3027.7749,
+    "train_loss": 0.0,
+    "train_runtime": 0.9326,
     "train_samples": 1039,
-    "train_samples_per_second": 0.199,
-    "train_steps_per_second": 0.025
+    "train_samples_per_second": 645.502,
+    "train_steps_per_second": 80.42
 }

config.json

@@ -0,0 +1,51 @@
+{
+  "_name_or_path": "meta-llama/Llama-3.1-8B",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 128000,
+  "eos_token_id": 128001,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 131072,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "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": {
+    "factor": 8.0,
+    "high_freq_factor": 4.0,
+    "low_freq_factor": 1.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "llama3"
+  },
+  "rope_theta": 500000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.45.1",
+  "use_cache": true,
+  "vocab_size": 128256
+}

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 0.9933774834437086,
     "total_flos": 5.564154814608179e+16,
-    "train_loss": 1.2146572240193685,
-    "train_runtime": 3027.7749,
+    "train_loss": 0.0,
+    "train_runtime": 0.9326,
     "train_samples": 1039,
-    "train_samples_per_second": 0.199,
-    "train_steps_per_second": 0.025
+    "train_samples_per_second": 645.502,
+    "train_steps_per_second": 80.42
 }

trainer_state.json

@@ -120,22 +120,14 @@
       "loss": 1.2385,
       "step": 75
     },
-    {
-      "epoch": 0.9933774834437086,
-      "eval_loss": 1.1645171642303467,
-      "eval_runtime": 2610.6612,
-      "eval_samples_per_second": 5.155,
-      "eval_steps_per_second": 0.645,
-      "step": 75
-    },
     {
       "epoch": 0.9933774834437086,
       "step": 75,
       "total_flos": 5.564154814608179e+16,
-      "train_loss": 1.2146572240193685,
-      "train_runtime": 3027.7749,
-      "train_samples_per_second": 0.199,
-      "train_steps_per_second": 0.025
+      "train_loss": 0.0,
+      "train_runtime": 0.9326,
+      "train_samples_per_second": 645.502,
+      "train_steps_per_second": 80.42
     }
   ],
   "logging_steps": 5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be928b88e5a6832a2c69dba0813b990375959c7861a41535e5947c8446b5c82a
+oid sha256:ea6564d1fa58a3ec4d8ab7340d84be9f250343f22726715020d63cc1621efd91
 size 6264