End of training

README.md

@@ -3,9 +3,15 @@ library_name: transformers
 license: llama3.2
 base_model: meta-llama/Llama-3.2-1B
 tags:
+- alignment-handbook
 - trl
 - sft
 - generated_from_trainer
+- trl
+- sft
+- generated_from_trainer
+datasets:
+- JunxiongWang/sftdataset
 model-index:
 - name: 1b_distill_width_prune
   results: []
@@ -16,9 +22,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # 1b_distill_width_prune
 
-This model is a fine-tuned version of [meta-llama/Llama-3.2-1B](https://huggingface.co/meta-llama/Llama-3.2-1B) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Llama-3.2-1B](https://huggingface.co/meta-llama/Llama-3.2-1B) on the JunxiongWang/sftdataset dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.5713
+- Loss: 1.5712
 
 ## Model description
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 0.08267797093601252,
+    "eval_loss": 1.5711677074432373,
+    "eval_runtime": 394.1141,
+    "eval_samples": 4096,
+    "eval_samples_per_second": 10.393,
+    "eval_steps_per_second": 2.598,
     "total_flos": 0.0,
     "train_loss": 6.551032936291113e-05,
     "train_runtime": 86.8457,

config.json

@@ -0,0 +1,40 @@
+{
+  "_attn_implementation_autoset": true,
+  "_name_or_path": "meta-llama/Llama-3.2-1B",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 128000,
+  "eos_token_id": 128001,
+  "head_dim": 64,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "kv_dim": 512,
+  "max_position_embeddings": 131072,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "n_layer": 16,
+  "num_attention_heads": 32,
+  "num_hidden_layers": 16,
+  "num_key_value_heads": 8,
+  "num_kv_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": {
+    "factor": 32.0,
+    "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": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.47.0.dev0",
+  "use_cache": true,
+  "vocab_size": 128256
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.08267797093601252,
+    "eval_loss": 1.5711677074432373,
+    "eval_runtime": 394.1141,
+    "eval_samples": 4096,
+    "eval_samples_per_second": 10.393,
+    "eval_steps_per_second": 2.598
+}