Upload folder using huggingface_hub

added_tokens.json

@@ -0,0 +1,13 @@
+{
+  "<|assistant|>": 32001,
+  "<|endoftext|>": 32000,
+  "<|end|>": 32007,
+  "<|placeholder1|>": 32002,
+  "<|placeholder2|>": 32003,
+  "<|placeholder3|>": 32004,
+  "<|placeholder4|>": 32005,
+  "<|placeholder5|>": 32008,
+  "<|placeholder6|>": 32009,
+  "<|system|>": 32006,
+  "<|user|>": 32010
+}

all_results.json

@@ -0,0 +1,24 @@
+{
+    "epoch": 0.9880609304240429,
+    "eval_logits/chosen": 9.119073867797852,
+    "eval_logits/rejected": 9.919014930725098,
+    "eval_logps/chosen": -109.64173889160156,
+    "eval_logps/ref_chosen": -106.16099548339844,
+    "eval_logps/ref_rejected": -119.78473663330078,
+    "eval_logps/rejected": -123.23009490966797,
+    "eval_loss": 0.6934764981269836,
+    "eval_rewards/accuracies": 0.5018518567085266,
+    "eval_rewards/chosen": -0.034807514399290085,
+    "eval_rewards/margins": -0.0003539324679877609,
+    "eval_rewards/rejected": -0.034453582018613815,
+    "eval_runtime": 161.8044,
+    "eval_samples": 1080,
+    "eval_samples_per_second": 6.675,
+    "eval_steps_per_second": 1.669,
+    "total_flos": 0.0,
+    "train_loss": 0.6918170555432638,
+    "train_runtime": 7488.2587,
+    "train_samples": 9715,
+    "train_samples_per_second": 1.297,
+    "train_steps_per_second": 0.01
+}

checkpoint-12/added_tokens.json

@@ -0,0 +1,13 @@
+{
+  "<|assistant|>": 32001,
+  "<|endoftext|>": 32000,
+  "<|end|>": 32007,
+  "<|placeholder1|>": 32002,
+  "<|placeholder2|>": 32003,
+  "<|placeholder3|>": 32004,
+  "<|placeholder4|>": 32005,
+  "<|placeholder5|>": 32008,
+  "<|placeholder6|>": 32009,
+  "<|system|>": 32006,
+  "<|user|>": 32010
+}

checkpoint-12/config.json

@@ -0,0 +1,36 @@
+{
+  "_name_or_path": "/home/v-liuzichuan/huggingface/Phi-3-mini-4k-instruct",
+  "architectures": [
+    "Phi3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_phi3.Phi3Config",
+    "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
+  },
+  "bos_token_id": 1,
+  "embd_pdrop": 0.0,
+  "eos_token_id": 32000,
+  "hidden_act": "silu",
+  "hidden_size": 3072,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 4096,
+  "model_type": "phi3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "original_max_position_embeddings": 4096,
+  "pad_token_id": 32000,
+  "resid_pdrop": 0.0,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "sliding_window": 2047,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.45.2",
+  "use_cache": false,
+  "vocab_size": 32064
+}

checkpoint-12/generation_config.json

@@ -0,0 +1,11 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": [
+    32000,
+    32001,
+    32007
+  ],
+  "pad_token_id": 32000,
+  "transformers_version": "4.45.2"
+}

checkpoint-12/model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a2bcfea8eeb2555724d32128ab082d0623e2df052df8e10e3025bc37235e43c
+size 4972489328

checkpoint-12/model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:180b1c92903401c30e82eb9864f2d1b8217db1665fa1da1bc7b2bfe948456e82
+size 2669692552

checkpoint-12/model.safetensors.index.json

@@ -0,0 +1,202 @@
+{
+  "metadata": {
+    "total_size": 7642159104
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}

checkpoint-12/special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-12/tokenizer_config.json

@@ -0,0 +1,131 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "32000": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32002": {
+      "content": "<|placeholder1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32003": {
+      "content": "<|placeholder2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32004": {
+      "content": "<|placeholder3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32005": {
+      "content": "<|placeholder4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32006": {
+      "content": "<|system|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32007": {
+      "content": "<|end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32008": {
+      "content": "<|placeholder5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32009": {
+      "content": "<|placeholder6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32010": {
+      "content": "<|user|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% for message in messages %}{% if message['role'] == 'system' %}{{'<|system|>\n' + message['content'] + '<|end|>\n'}}{% elif message['role'] == 'user' %}{{'<|user|>\n' + message['content'] + '<|end|>\n'}}{% elif message['role'] == 'assistant' %}{{'<|assistant|>\n' + message['content'] + '<|end|>\n'}}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>\n' }}{% else %}{{ eos_token }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "legacy": false,
+  "model_max_length": 4096,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

checkpoint-12/trainer_state.json

@@ -0,0 +1,237 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.15808974886784685,
+  "eval_steps": 500,
+  "global_step": 12,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.013174145738987238,
+      "grad_norm": 0.48070791363716125,
+      "learning_rate": 6.25e-08,
+      "logits/chosen": 10.32492733001709,
+      "logits/rejected": 10.282785415649414,
+      "logps/chosen": -176.09544372558594,
+      "logps/ref_chosen": -176.09544372558594,
+      "logps/ref_rejected": -181.75552368164062,
+      "logps/rejected": -181.75552368164062,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.0,
+      "rewards/chosen": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/rejected": 0.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.026348291477974475,
+      "grad_norm": 0.4185059070587158,
+      "learning_rate": 1.25e-07,
+      "logits/chosen": 10.530074119567871,
+      "logits/rejected": 10.672085762023926,
+      "logps/chosen": -173.2084503173828,
+      "logps/ref_chosen": -173.2084503173828,
+      "logps/ref_rejected": -187.02206420898438,
+      "logps/rejected": -187.02206420898438,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.0,
+      "rewards/chosen": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/rejected": 0.0,
+      "step": 2
+    },
+    {
+      "epoch": 0.03952243721696171,
+      "grad_norm": 0.3848375082015991,
+      "learning_rate": 1.875e-07,
+      "logits/chosen": 10.340475082397461,
+      "logits/rejected": 10.486526489257812,
+      "logps/chosen": -169.6623992919922,
+      "logps/ref_chosen": -169.080810546875,
+      "logps/ref_rejected": -182.52792358398438,
+      "logps/rejected": -182.80642700195312,
+      "loss": 0.6947,
+      "rewards/accuracies": 0.3828125,
+      "rewards/chosen": -0.005815833806991577,
+      "rewards/margins": -0.0030306153930723667,
+      "rewards/rejected": -0.002785218646749854,
+      "step": 3
+    },
+    {
+      "epoch": 0.05269658295594895,
+      "grad_norm": 0.7261964678764343,
+      "learning_rate": 2.5e-07,
+      "logits/chosen": 10.3652925491333,
+      "logits/rejected": 10.412069320678711,
+      "logps/chosen": -177.03428649902344,
+      "logps/ref_chosen": -176.78369140625,
+      "logps/ref_rejected": -180.0931396484375,
+      "logps/rejected": -180.21463012695312,
+      "loss": 0.6938,
+      "rewards/accuracies": 0.4765625,
+      "rewards/chosen": -0.0025058696046471596,
+      "rewards/margins": -0.0012909012148156762,
+      "rewards/rejected": -0.0012149682734161615,
+      "step": 4
+    },
+    {
+      "epoch": 0.06587072869493618,
+      "grad_norm": 0.3572319746017456,
+      "learning_rate": 3.1249999999999997e-07,
+      "logits/chosen": 10.757452011108398,
+      "logits/rejected": 10.791389465332031,
+      "logps/chosen": -169.12631225585938,
+      "logps/ref_chosen": -168.522216796875,
+      "logps/ref_rejected": -176.08982849121094,
+      "logps/rejected": -176.59901428222656,
+      "loss": 0.6936,
+      "rewards/accuracies": 0.4609375,
+      "rewards/chosen": -0.0060411859303712845,
+      "rewards/margins": -0.0009493756806477904,
+      "rewards/rejected": -0.0050918105989694595,
+      "step": 5
+    },
+    {
+      "epoch": 0.07904487443392343,
+      "grad_norm": 0.4071093201637268,
+      "learning_rate": 3.75e-07,
+      "logits/chosen": 11.003137588500977,
+      "logits/rejected": 11.056475639343262,
+      "logps/chosen": -174.662353515625,
+      "logps/ref_chosen": -174.5843048095703,
+      "logps/ref_rejected": -181.65040588378906,
+      "logps/rejected": -181.72401428222656,
+      "loss": 0.6932,
+      "rewards/accuracies": 0.5078125,
+      "rewards/chosen": -0.0007804610067978501,
+      "rewards/margins": -4.43047538283281e-05,
+      "rewards/rejected": -0.0007361561874859035,
+      "step": 6
+    },
+    {
+      "epoch": 0.09221902017291066,
+      "grad_norm": 0.7951124906539917,
+      "learning_rate": 4.375e-07,
+      "logits/chosen": 10.003807067871094,
+      "logits/rejected": 10.067156791687012,
+      "logps/chosen": -171.4049530029297,
+      "logps/ref_chosen": -170.4031219482422,
+      "logps/ref_rejected": -181.00917053222656,
+      "logps/rejected": -181.8280792236328,
+      "loss": 0.6941,
+      "rewards/accuracies": 0.4453125,
+      "rewards/chosen": -0.010018287226557732,
+      "rewards/margins": -0.0018291514134034514,
+      "rewards/rejected": -0.008189136162400246,
+      "step": 7
+    },
+    {
+      "epoch": 0.1053931659118979,
+      "grad_norm": 0.5591660737991333,
+      "learning_rate": 5e-07,
+      "logits/chosen": 10.874868392944336,
+      "logits/rejected": 10.79828929901123,
+      "logps/chosen": -179.36700439453125,
+      "logps/ref_chosen": -178.47369384765625,
+      "logps/ref_rejected": -184.52243041992188,
+      "logps/rejected": -185.44284057617188,
+      "loss": 0.693,
+      "rewards/accuracies": 0.5078125,
+      "rewards/chosen": -0.008933190256357193,
+      "rewards/margins": 0.0002710025873966515,
+      "rewards/rejected": -0.009204192087054253,
+      "step": 8
+    },
+    {
+      "epoch": 0.11856731165088513,
+      "grad_norm": 0.45744186639785767,
+      "learning_rate": 4.997252228714278e-07,
+      "logits/chosen": 10.472944259643555,
+      "logits/rejected": 10.640350341796875,
+      "logps/chosen": -175.36553955078125,
+      "logps/ref_chosen": -174.38418579101562,
+      "logps/ref_rejected": -182.0985565185547,
+      "logps/rejected": -183.1609649658203,
+      "loss": 0.6928,
+      "rewards/accuracies": 0.5234375,
+      "rewards/chosen": -0.009813489392399788,
+      "rewards/margins": 0.0008107352769002318,
+      "rewards/rejected": -0.010624224320054054,
+      "step": 9
+    },
+    {
+      "epoch": 0.13174145738987236,
+      "grad_norm": 0.3784727156162262,
+      "learning_rate": 4.989014955054745e-07,
+      "logits/chosen": 10.471797943115234,
+      "logits/rejected": 10.57437515258789,
+      "logps/chosen": -173.78768920898438,
+      "logps/ref_chosen": -172.48837280273438,
+      "logps/ref_rejected": -179.625,
+      "logps/rejected": -180.91595458984375,
+      "loss": 0.6932,
+      "rewards/accuracies": 0.4765625,
+      "rewards/chosen": -0.012993087992072105,
+      "rewards/margins": -8.335959864780307e-05,
+      "rewards/rejected": -0.012909728102385998,
+      "step": 10
+    },
+    {
+      "epoch": 0.14491560312885962,
+      "grad_norm": 0.4041476547718048,
+      "learning_rate": 4.975306286336627e-07,
+      "logits/chosen": 10.539186477661133,
+      "logits/rejected": 10.477883338928223,
+      "logps/chosen": -172.62416076660156,
+      "logps/ref_chosen": -169.97216796875,
+      "logps/ref_rejected": -180.00955200195312,
+      "logps/rejected": -182.72525024414062,
+      "loss": 0.6929,
+      "rewards/accuracies": 0.546875,
+      "rewards/chosen": -0.026519589126110077,
+      "rewards/margins": 0.0006374535150825977,
+      "rewards/rejected": -0.027157040312886238,
+      "step": 11
+    },
+    {
+      "epoch": 0.15808974886784685,
+      "grad_norm": 0.5964677333831787,
+      "learning_rate": 4.956156357188939e-07,
+      "logits/chosen": 10.28010082244873,
+      "logits/rejected": 10.401304244995117,
+      "logps/chosen": -178.43714904785156,
+      "logps/ref_chosen": -175.83639526367188,
+      "logps/ref_rejected": -181.73045349121094,
+      "logps/rejected": -184.20223999023438,
+      "loss": 0.6938,
+      "rewards/accuracies": 0.4375,
+      "rewards/chosen": -0.026007818058133125,
+      "rewards/margins": -0.0012897354317829013,
+      "rewards/rejected": -0.024718083441257477,
+      "step": 12
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 75,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 12,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-24/added_tokens.json

@@ -0,0 +1,13 @@
+{
+  "<|assistant|>": 32001,
+  "<|endoftext|>": 32000,
+  "<|end|>": 32007,
+  "<|placeholder1|>": 32002,
+  "<|placeholder2|>": 32003,
+  "<|placeholder3|>": 32004,
+  "<|placeholder4|>": 32005,
+  "<|placeholder5|>": 32008,
+  "<|placeholder6|>": 32009,
+  "<|system|>": 32006,
+  "<|user|>": 32010
+}

checkpoint-24/config.json

@@ -0,0 +1,36 @@
+{
+  "_name_or_path": "/home/v-liuzichuan/huggingface/Phi-3-mini-4k-instruct",
+  "architectures": [
+    "Phi3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_phi3.Phi3Config",
+    "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
+  },
+  "bos_token_id": 1,
+  "embd_pdrop": 0.0,
+  "eos_token_id": 32000,
+  "hidden_act": "silu",
+  "hidden_size": 3072,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 4096,
+  "model_type": "phi3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "original_max_position_embeddings": 4096,
+  "pad_token_id": 32000,
+  "resid_pdrop": 0.0,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "sliding_window": 2047,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.45.2",
+  "use_cache": false,
+  "vocab_size": 32064
+}

checkpoint-24/generation_config.json

@@ -0,0 +1,11 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": [
+    32000,
+    32001,
+    32007
+  ],
+  "pad_token_id": 32000,
+  "transformers_version": "4.45.2"
+}

checkpoint-24/model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9903cca5d41647e253d986bff1db308641492f23b4d3ae1a3d95d410cf069ba
+size 4972489328

checkpoint-24/model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f2ba7b65cbd6188d832c4b7aad8966474aa357323d506e993ac5fa062e61489
+size 2669692552

checkpoint-24/model.safetensors.index.json

@@ -0,0 +1,202 @@
+{
+  "metadata": {
+    "total_size": 7642159104
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}

checkpoint-24/special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-24/tokenizer_config.json

@@ -0,0 +1,131 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "32000": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32002": {
+      "content": "<|placeholder1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32003": {
+      "content": "<|placeholder2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32004": {
+      "content": "<|placeholder3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32005": {
+      "content": "<|placeholder4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32006": {
+      "content": "<|system|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32007": {
+      "content": "<|end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32008": {
+      "content": "<|placeholder5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32009": {
+      "content": "<|placeholder6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32010": {
+      "content": "<|user|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% for message in messages %}{% if message['role'] == 'system' %}{{'<|system|>\n' + message['content'] + '<|end|>\n'}}{% elif message['role'] == 'user' %}{{'<|user|>\n' + message['content'] + '<|end|>\n'}}{% elif message['role'] == 'assistant' %}{{'<|assistant|>\n' + message['content'] + '<|end|>\n'}}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>\n' }}{% else %}{{ eos_token }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "legacy": false,
+  "model_max_length": 4096,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

checkpoint-24/trainer_state.json

@@ -0,0 +1,441 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.3161794977356937,
+  "eval_steps": 500,
+  "global_step": 24,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.013174145738987238,
+      "grad_norm": 0.48070791363716125,
+      "learning_rate": 6.25e-08,
+      "logits/chosen": 10.32492733001709,
+      "logits/rejected": 10.282785415649414,
+      "logps/chosen": -176.09544372558594,
+      "logps/ref_chosen": -176.09544372558594,
+      "logps/ref_rejected": -181.75552368164062,
+      "logps/rejected": -181.75552368164062,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.0,
+      "rewards/chosen": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/rejected": 0.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.026348291477974475,
+      "grad_norm": 0.4185059070587158,
+      "learning_rate": 1.25e-07,
+      "logits/chosen": 10.530074119567871,
+      "logits/rejected": 10.672085762023926,
+      "logps/chosen": -173.2084503173828,
+      "logps/ref_chosen": -173.2084503173828,
+      "logps/ref_rejected": -187.02206420898438,
+      "logps/rejected": -187.02206420898438,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.0,
+      "rewards/chosen": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/rejected": 0.0,
+      "step": 2
+    },
+    {
+      "epoch": 0.03952243721696171,
+      "grad_norm": 0.3848375082015991,
+      "learning_rate": 1.875e-07,
+      "logits/chosen": 10.340475082397461,
+      "logits/rejected": 10.486526489257812,
+      "logps/chosen": -169.6623992919922,
+      "logps/ref_chosen": -169.080810546875,
+      "logps/ref_rejected": -182.52792358398438,
+      "logps/rejected": -182.80642700195312,
+      "loss": 0.6947,
+      "rewards/accuracies": 0.3828125,
+      "rewards/chosen": -0.005815833806991577,
+      "rewards/margins": -0.0030306153930723667,
+      "rewards/rejected": -0.002785218646749854,
+      "step": 3
+    },
+    {
+      "epoch": 0.05269658295594895,
+      "grad_norm": 0.7261964678764343,
+      "learning_rate": 2.5e-07,
+      "logits/chosen": 10.3652925491333,
+      "logits/rejected": 10.412069320678711,
+      "logps/chosen": -177.03428649902344,
+      "logps/ref_chosen": -176.78369140625,
+      "logps/ref_rejected": -180.0931396484375,
+      "logps/rejected": -180.21463012695312,
+      "loss": 0.6938,
+      "rewards/accuracies": 0.4765625,
+      "rewards/chosen": -0.0025058696046471596,
+      "rewards/margins": -0.0012909012148156762,
+      "rewards/rejected": -0.0012149682734161615,
+      "step": 4
+    },
+    {
+      "epoch": 0.06587072869493618,
+      "grad_norm": 0.3572319746017456,
+      "learning_rate": 3.1249999999999997e-07,
+      "logits/chosen": 10.757452011108398,
+      "logits/rejected": 10.791389465332031,
+      "logps/chosen": -169.12631225585938,
+      "logps/ref_chosen": -168.522216796875,
+      "logps/ref_rejected": -176.08982849121094,
+      "logps/rejected": -176.59901428222656,
+      "loss": 0.6936,
+      "rewards/accuracies": 0.4609375,
+      "rewards/chosen": -0.0060411859303712845,
+      "rewards/margins": -0.0009493756806477904,
+      "rewards/rejected": -0.0050918105989694595,
+      "step": 5
+    },
+    {
+      "epoch": 0.07904487443392343,
+      "grad_norm": 0.4071093201637268,
+      "learning_rate": 3.75e-07,
+      "logits/chosen": 11.003137588500977,
+      "logits/rejected": 11.056475639343262,
+      "logps/chosen": -174.662353515625,
+      "logps/ref_chosen": -174.5843048095703,
+      "logps/ref_rejected": -181.65040588378906,
+      "logps/rejected": -181.72401428222656,
+      "loss": 0.6932,
+      "rewards/accuracies": 0.5078125,
+      "rewards/chosen": -0.0007804610067978501,
+      "rewards/margins": -4.43047538283281e-05,
+      "rewards/rejected": -0.0007361561874859035,
+      "step": 6
+    },
+    {
+      "epoch": 0.09221902017291066,
+      "grad_norm": 0.7951124906539917,
+      "learning_rate": 4.375e-07,
+      "logits/chosen": 10.003807067871094,
+      "logits/rejected": 10.067156791687012,
+      "logps/chosen": -171.4049530029297,
+      "logps/ref_chosen": -170.4031219482422,
+      "logps/ref_rejected": -181.00917053222656,
+      "logps/rejected": -181.8280792236328,
+      "loss": 0.6941,
+      "rewards/accuracies": 0.4453125,
+      "rewards/chosen": -0.010018287226557732,
+      "rewards/margins": -0.0018291514134034514,
+      "rewards/rejected": -0.008189136162400246,
+      "step": 7
+    },
+    {
+      "epoch": 0.1053931659118979,
+      "grad_norm": 0.5591660737991333,
+      "learning_rate": 5e-07,
+      "logits/chosen": 10.874868392944336,
+      "logits/rejected": 10.79828929901123,
+      "logps/chosen": -179.36700439453125,
+      "logps/ref_chosen": -178.47369384765625,
+      "logps/ref_rejected": -184.52243041992188,
+      "logps/rejected": -185.44284057617188,
+      "loss": 0.693,
+      "rewards/accuracies": 0.5078125,
+      "rewards/chosen": -0.008933190256357193,
+      "rewards/margins": 0.0002710025873966515,
+      "rewards/rejected": -0.009204192087054253,
+      "step": 8
+    },
+    {
+      "epoch": 0.11856731165088513,
+      "grad_norm": 0.45744186639785767,
+      "learning_rate": 4.997252228714278e-07,
+      "logits/chosen": 10.472944259643555,
+      "logits/rejected": 10.640350341796875,
+      "logps/chosen": -175.36553955078125,
+      "logps/ref_chosen": -174.38418579101562,
+      "logps/ref_rejected": -182.0985565185547,
+      "logps/rejected": -183.1609649658203,
+      "loss": 0.6928,
+      "rewards/accuracies": 0.5234375,
+      "rewards/chosen": -0.009813489392399788,
+      "rewards/margins": 0.0008107352769002318,
+      "rewards/rejected": -0.010624224320054054,
+      "step": 9
+    },
+    {
+      "epoch": 0.13174145738987236,
+      "grad_norm": 0.3784727156162262,
+      "learning_rate": 4.989014955054745e-07,
+      "logits/chosen": 10.471797943115234,
+      "logits/rejected": 10.57437515258789,
+      "logps/chosen": -173.78768920898438,
+      "logps/ref_chosen": -172.48837280273438,
+      "logps/ref_rejected": -179.625,
+      "logps/rejected": -180.91595458984375,
+      "loss": 0.6932,
+      "rewards/accuracies": 0.4765625,
+      "rewards/chosen": -0.012993087992072105,
+      "rewards/margins": -8.335959864780307e-05,
+      "rewards/rejected": -0.012909728102385998,
+      "step": 10
+    },
+    {
+      "epoch": 0.14491560312885962,
+      "grad_norm": 0.4041476547718048,
+      "learning_rate": 4.975306286336627e-07,
+      "logits/chosen": 10.539186477661133,
+      "logits/rejected": 10.477883338928223,
+      "logps/chosen": -172.62416076660156,
+      "logps/ref_chosen": -169.97216796875,
+      "logps/ref_rejected": -180.00955200195312,
+      "logps/rejected": -182.72525024414062,
+      "loss": 0.6929,
+      "rewards/accuracies": 0.546875,
+      "rewards/chosen": -0.026519589126110077,
+      "rewards/margins": 0.0006374535150825977,
+      "rewards/rejected": -0.027157040312886238,
+      "step": 11
+    },
+    {
+      "epoch": 0.15808974886784685,
+      "grad_norm": 0.5964677333831787,
+      "learning_rate": 4.956156357188939e-07,
+      "logits/chosen": 10.28010082244873,
+      "logits/rejected": 10.401304244995117,
+      "logps/chosen": -178.43714904785156,
+      "logps/ref_chosen": -175.83639526367188,
+      "logps/ref_rejected": -181.73045349121094,
+      "logps/rejected": -184.20223999023438,
+      "loss": 0.6938,
+      "rewards/accuracies": 0.4375,
+      "rewards/chosen": -0.026007818058133125,
+      "rewards/margins": -0.0012897354317829013,
+      "rewards/rejected": -0.024718083441257477,
+      "step": 12
+    },
+    {
+      "epoch": 0.17126389460683408,
+      "grad_norm": 0.488361120223999,
+      "learning_rate": 4.931607263312032e-07,
+      "logits/chosen": 10.292243957519531,
+      "logits/rejected": 10.386707305908203,
+      "logps/chosen": -173.99058532714844,
+      "logps/ref_chosen": -170.8806610107422,
+      "logps/ref_rejected": -180.6468963623047,
+      "logps/rejected": -184.01853942871094,
+      "loss": 0.6919,
+      "rewards/accuracies": 0.5234375,
+      "rewards/chosen": -0.03109920769929886,
+      "rewards/margins": 0.002617327030748129,
+      "rewards/rejected": -0.033716537058353424,
+      "step": 13
+    },
+    {
+      "epoch": 0.1844380403458213,
+      "grad_norm": 0.45636966824531555,
+      "learning_rate": 4.9017129689421e-07,
+      "logits/chosen": 10.729025840759277,
+      "logits/rejected": 10.921786308288574,
+      "logps/chosen": -174.65345764160156,
+      "logps/ref_chosen": -171.15963745117188,
+      "logps/ref_rejected": -180.73471069335938,
+      "logps/rejected": -184.30807495117188,
+      "loss": 0.6928,
+      "rewards/accuracies": 0.515625,
+      "rewards/chosen": -0.03493800014257431,
+      "rewards/margins": 0.0007955244509503245,
+      "rewards/rejected": -0.03573352098464966,
+      "step": 14
+    },
+    {
+      "epoch": 0.19761218608480857,
+      "grad_norm": 0.3838670551776886,
+      "learning_rate": 4.866539188226085e-07,
+      "logits/chosen": 10.265353202819824,
+      "logits/rejected": 10.378411293029785,
+      "logps/chosen": -177.72991943359375,
+      "logps/ref_chosen": -173.84222412109375,
+      "logps/ref_rejected": -182.6996612548828,
+      "logps/rejected": -186.89825439453125,
+      "loss": 0.6916,
+      "rewards/accuracies": 0.515625,
+      "rewards/chosen": -0.03887684643268585,
+      "rewards/margins": 0.003108714707195759,
+      "rewards/rejected": -0.04198555648326874,
+      "step": 15
+    },
+    {
+      "epoch": 0.2107863318237958,
+      "grad_norm": 0.3698700964450836,
+      "learning_rate": 4.826163240767716e-07,
+      "logits/chosen": 10.751184463500977,
+      "logits/rejected": 10.945128440856934,
+      "logps/chosen": -181.60301208496094,
+      "logps/ref_chosen": -178.638671875,
+      "logps/ref_rejected": -185.8243408203125,
+      "logps/rejected": -188.93057250976562,
+      "loss": 0.6925,
+      "rewards/accuracies": 0.5234375,
+      "rewards/chosen": -0.029643364250659943,
+      "rewards/margins": 0.0014189946232363582,
+      "rewards/rejected": -0.03106236271560192,
+      "step": 16
+    },
+    {
+      "epoch": 0.22396047756278303,
+      "grad_norm": 0.5419145822525024,
+      "learning_rate": 4.780673881662242e-07,
+      "logits/chosen": 10.279337882995605,
+      "logits/rejected": 10.265542984008789,
+      "logps/chosen": -176.72821044921875,
+      "logps/ref_chosen": -172.87332153320312,
+      "logps/ref_rejected": -177.30355834960938,
+      "logps/rejected": -181.23831176757812,
+      "loss": 0.6928,
+      "rewards/accuracies": 0.484375,
+      "rewards/chosen": -0.038549020886421204,
+      "rewards/margins": 0.0007985997945070267,
+      "rewards/rejected": -0.03934762626886368,
+      "step": 17
+    },
+    {
+      "epoch": 0.23713462330177026,
+      "grad_norm": 0.5288915038108826,
+      "learning_rate": 4.730171106393466e-07,
+      "logits/chosen": 10.508716583251953,
+      "logits/rejected": 10.601346969604492,
+      "logps/chosen": -176.69735717773438,
+      "logps/ref_chosen": -172.82644653320312,
+      "logps/ref_rejected": -180.93768310546875,
+      "logps/rejected": -184.82861328125,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.5234375,
+      "rewards/chosen": -0.038708824664354324,
+      "rewards/margins": 0.00020039017545059323,
+      "rewards/rejected": -0.03890921175479889,
+      "step": 18
+    },
+    {
+      "epoch": 0.2503087690407575,
+      "grad_norm": 0.45863935351371765,
+      "learning_rate": 4.6747659310219757e-07,
+      "logits/chosen": 10.509140014648438,
+      "logits/rejected": 10.403264999389648,
+      "logps/chosen": -183.62411499023438,
+      "logps/ref_chosen": -179.5069122314453,
+      "logps/ref_rejected": -185.01487731933594,
+      "logps/rejected": -189.2952880859375,
+      "loss": 0.6924,
+      "rewards/accuracies": 0.5078125,
+      "rewards/chosen": -0.041171781718730927,
+      "rewards/margins": 0.0016320659779012203,
+      "rewards/rejected": -0.04280385002493858,
+      "step": 19
+    },
+    {
+      "epoch": 0.2634829147797447,
+      "grad_norm": 0.6024115085601807,
+      "learning_rate": 4.6145801481477433e-07,
+      "logits/chosen": 10.452649116516113,
+      "logits/rejected": 10.525249481201172,
+      "logps/chosen": -174.3675994873047,
+      "logps/ref_chosen": -170.23623657226562,
+      "logps/ref_rejected": -177.55618286132812,
+      "logps/rejected": -181.85946655273438,
+      "loss": 0.6923,
+      "rewards/accuracies": 0.5078125,
+      "rewards/chosen": -0.04131368175148964,
+      "rewards/margins": 0.0017191548831760883,
+      "rewards/rejected": -0.04303283616900444,
+      "step": 20
+    },
+    {
+      "epoch": 0.276657060518732,
+      "grad_norm": 0.4595341384410858,
+      "learning_rate": 4.549746059183561e-07,
+      "logits/chosen": 10.422194480895996,
+      "logits/rejected": 10.483356475830078,
+      "logps/chosen": -180.44906616210938,
+      "logps/ref_chosen": -176.3700408935547,
+      "logps/ref_rejected": -182.80032348632812,
+      "logps/rejected": -187.17276000976562,
+      "loss": 0.6917,
+      "rewards/accuracies": 0.5703125,
+      "rewards/chosen": -0.04079030081629753,
+      "rewards/margins": 0.0029342519119381905,
+      "rewards/rejected": -0.04372455179691315,
+      "step": 21
+    },
+    {
+      "epoch": 0.28983120625771924,
+      "grad_norm": 0.3111652731895447,
+      "learning_rate": 4.480406183527823e-07,
+      "logits/chosen": 10.481276512145996,
+      "logits/rejected": 10.559721946716309,
+      "logps/chosen": -177.63865661621094,
+      "logps/ref_chosen": -173.29742431640625,
+      "logps/ref_rejected": -177.54916381835938,
+      "logps/rejected": -182.2364044189453,
+      "loss": 0.6915,
+      "rewards/accuracies": 0.6015625,
+      "rewards/chosen": -0.043412309139966965,
+      "rewards/margins": 0.003460067557170987,
+      "rewards/rejected": -0.04687237739562988,
+      "step": 22
+    },
+    {
+      "epoch": 0.3030053519967065,
+      "grad_norm": 0.520114541053772,
+      "learning_rate": 4.4067129452759546e-07,
+      "logits/chosen": 10.153634071350098,
+      "logits/rejected": 10.433453559875488,
+      "logps/chosen": -174.9644012451172,
+      "logps/ref_chosen": -170.41177368164062,
+      "logps/ref_rejected": -182.7073974609375,
+      "logps/rejected": -187.5763397216797,
+      "loss": 0.6916,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.04552610218524933,
+      "rewards/margins": 0.0031630881130695343,
+      "rewards/rejected": -0.04868919029831886,
+      "step": 23
+    },
+    {
+      "epoch": 0.3161794977356937,
+      "grad_norm": 0.38157743215560913,
+      "learning_rate": 4.3288283381591725e-07,
+      "logits/chosen": 10.212328910827637,
+      "logits/rejected": 10.149028778076172,
+      "logps/chosen": -178.75759887695312,
+      "logps/ref_chosen": -174.15003967285156,
+      "logps/ref_rejected": -181.65594482421875,
+      "logps/rejected": -186.58934020996094,
+      "loss": 0.6916,
+      "rewards/accuracies": 0.5703125,
+      "rewards/chosen": -0.046075694262981415,
+      "rewards/margins": 0.003258442971855402,
+      "rewards/rejected": -0.04933413863182068,
+      "step": 24
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 75,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 12,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-36/added_tokens.json

@@ -0,0 +1,13 @@
+{
+  "<|assistant|>": 32001,
+  "<|endoftext|>": 32000,
+  "<|end|>": 32007,
+  "<|placeholder1|>": 32002,
+  "<|placeholder2|>": 32003,
+  "<|placeholder3|>": 32004,
+  "<|placeholder4|>": 32005,
+  "<|placeholder5|>": 32008,
+  "<|placeholder6|>": 32009,
+  "<|system|>": 32006,
+  "<|user|>": 32010
+}

checkpoint-36/config.json

@@ -0,0 +1,36 @@
+{
+  "_name_or_path": "/home/v-liuzichuan/huggingface/Phi-3-mini-4k-instruct",
+  "architectures": [
+    "Phi3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_phi3.Phi3Config",
+    "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
+  },
+  "bos_token_id": 1,
+  "embd_pdrop": 0.0,
+  "eos_token_id": 32000,
+  "hidden_act": "silu",
+  "hidden_size": 3072,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 4096,
+  "model_type": "phi3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "original_max_position_embeddings": 4096,
+  "pad_token_id": 32000,
+  "resid_pdrop": 0.0,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "sliding_window": 2047,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.45.2",
+  "use_cache": false,
+  "vocab_size": 32064
+}

checkpoint-36/generation_config.json

@@ -0,0 +1,11 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": [
+    32000,
+    32001,
+    32007
+  ],
+  "pad_token_id": 32000,
+  "transformers_version": "4.45.2"
+}

checkpoint-36/model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6d1d77f4dcadb7404c9b80f486411ac3418085f1e47c3f4ead051a3b729ccaf
+size 4972489328

checkpoint-36/model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0caafaeb430d9a079eec8c0574f306fa1687482e6c6da3fcf7514eb7546355f8
+size 2669692552

checkpoint-36/model.safetensors.index.json

@@ -0,0 +1,202 @@
+{
+  "metadata": {
+    "total_size": 7642159104
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}

checkpoint-36/special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-36/tokenizer_config.json

@@ -0,0 +1,131 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "32000": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32002": {
+      "content": "<|placeholder1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32003": {
+      "content": "<|placeholder2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32004": {
+      "content": "<|placeholder3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32005": {
+      "content": "<|placeholder4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32006": {
+      "content": "<|system|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32007": {
+      "content": "<|end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32008": {
+      "content": "<|placeholder5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32009": {
+      "content": "<|placeholder6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32010": {
+      "content": "<|user|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% for message in messages %}{% if message['role'] == 'system' %}{{'<|system|>\n' + message['content'] + '<|end|>\n'}}{% elif message['role'] == 'user' %}{{'<|user|>\n' + message['content'] + '<|end|>\n'}}{% elif message['role'] == 'assistant' %}{{'<|assistant|>\n' + message['content'] + '<|end|>\n'}}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>\n' }}{% else %}{{ eos_token }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "legacy": false,
+  "model_max_length": 4096,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

checkpoint-36/trainer_state.json

@@ -0,0 +1,645 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.47426924660354053,
+  "eval_steps": 500,
+  "global_step": 36,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.013174145738987238,
+      "grad_norm": 0.48070791363716125,
+      "learning_rate": 6.25e-08,
+      "logits/chosen": 10.32492733001709,
+      "logits/rejected": 10.282785415649414,
+      "logps/chosen": -176.09544372558594,
+      "logps/ref_chosen": -176.09544372558594,
+      "logps/ref_rejected": -181.75552368164062,
+      "logps/rejected": -181.75552368164062,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.0,
+      "rewards/chosen": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/rejected": 0.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.026348291477974475,
+      "grad_norm": 0.4185059070587158,
+      "learning_rate": 1.25e-07,
+      "logits/chosen": 10.530074119567871,
+      "logits/rejected": 10.672085762023926,
+      "logps/chosen": -173.2084503173828,
+      "logps/ref_chosen": -173.2084503173828,
+      "logps/ref_rejected": -187.02206420898438,
+      "logps/rejected": -187.02206420898438,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.0,
+      "rewards/chosen": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/rejected": 0.0,
+      "step": 2
+    },
+    {
+      "epoch": 0.03952243721696171,
+      "grad_norm": 0.3848375082015991,
+      "learning_rate": 1.875e-07,
+      "logits/chosen": 10.340475082397461,
+      "logits/rejected": 10.486526489257812,
+      "logps/chosen": -169.6623992919922,
+      "logps/ref_chosen": -169.080810546875,
+      "logps/ref_rejected": -182.52792358398438,
+      "logps/rejected": -182.80642700195312,
+      "loss": 0.6947,
+      "rewards/accuracies": 0.3828125,
+      "rewards/chosen": -0.005815833806991577,
+      "rewards/margins": -0.0030306153930723667,
+      "rewards/rejected": -0.002785218646749854,
+      "step": 3
+    },
+    {
+      "epoch": 0.05269658295594895,
+      "grad_norm": 0.7261964678764343,
+      "learning_rate": 2.5e-07,
+      "logits/chosen": 10.3652925491333,
+      "logits/rejected": 10.412069320678711,
+      "logps/chosen": -177.03428649902344,
+      "logps/ref_chosen": -176.78369140625,
+      "logps/ref_rejected": -180.0931396484375,
+      "logps/rejected": -180.21463012695312,
+      "loss": 0.6938,
+      "rewards/accuracies": 0.4765625,
+      "rewards/chosen": -0.0025058696046471596,
+      "rewards/margins": -0.0012909012148156762,
+      "rewards/rejected": -0.0012149682734161615,
+      "step": 4
+    },
+    {
+      "epoch": 0.06587072869493618,
+      "grad_norm": 0.3572319746017456,
+      "learning_rate": 3.1249999999999997e-07,
+      "logits/chosen": 10.757452011108398,
+      "logits/rejected": 10.791389465332031,
+      "logps/chosen": -169.12631225585938,
+      "logps/ref_chosen": -168.522216796875,
+      "logps/ref_rejected": -176.08982849121094,
+      "logps/rejected": -176.59901428222656,
+      "loss": 0.6936,
+      "rewards/accuracies": 0.4609375,
+      "rewards/chosen": -0.0060411859303712845,
+      "rewards/margins": -0.0009493756806477904,
+      "rewards/rejected": -0.0050918105989694595,
+      "step": 5
+    },
+    {
+      "epoch": 0.07904487443392343,
+      "grad_norm": 0.4071093201637268,
+      "learning_rate": 3.75e-07,
+      "logits/chosen": 11.003137588500977,
+      "logits/rejected": 11.056475639343262,
+      "logps/chosen": -174.662353515625,
+      "logps/ref_chosen": -174.5843048095703,
+      "logps/ref_rejected": -181.65040588378906,
+      "logps/rejected": -181.72401428222656,
+      "loss": 0.6932,
+      "rewards/accuracies": 0.5078125,
+      "rewards/chosen": -0.0007804610067978501,
+      "rewards/margins": -4.43047538283281e-05,
+      "rewards/rejected": -0.0007361561874859035,
+      "step": 6
+    },
+    {
+      "epoch": 0.09221902017291066,
+      "grad_norm": 0.7951124906539917,
+      "learning_rate": 4.375e-07,
+      "logits/chosen": 10.003807067871094,
+      "logits/rejected": 10.067156791687012,
+      "logps/chosen": -171.4049530029297,
+      "logps/ref_chosen": -170.4031219482422,
+      "logps/ref_rejected": -181.00917053222656,
+      "logps/rejected": -181.8280792236328,
+      "loss": 0.6941,
+      "rewards/accuracies": 0.4453125,
+      "rewards/chosen": -0.010018287226557732,
+      "rewards/margins": -0.0018291514134034514,
+      "rewards/rejected": -0.008189136162400246,
+      "step": 7
+    },
+    {
+      "epoch": 0.1053931659118979,
+      "grad_norm": 0.5591660737991333,
+      "learning_rate": 5e-07,
+      "logits/chosen": 10.874868392944336,
+      "logits/rejected": 10.79828929901123,
+      "logps/chosen": -179.36700439453125,
+      "logps/ref_chosen": -178.47369384765625,
+      "logps/ref_rejected": -184.52243041992188,
+      "logps/rejected": -185.44284057617188,
+      "loss": 0.693,
+      "rewards/accuracies": 0.5078125,
+      "rewards/chosen": -0.008933190256357193,
+      "rewards/margins": 0.0002710025873966515,
+      "rewards/rejected": -0.009204192087054253,
+      "step": 8
+    },
+    {
+      "epoch": 0.11856731165088513,
+      "grad_norm": 0.45744186639785767,
+      "learning_rate": 4.997252228714278e-07,
+      "logits/chosen": 10.472944259643555,
+      "logits/rejected": 10.640350341796875,
+      "logps/chosen": -175.36553955078125,
+      "logps/ref_chosen": -174.38418579101562,
+      "logps/ref_rejected": -182.0985565185547,
+      "logps/rejected": -183.1609649658203,
+      "loss": 0.6928,
+      "rewards/accuracies": 0.5234375,
+      "rewards/chosen": -0.009813489392399788,
+      "rewards/margins": 0.0008107352769002318,
+      "rewards/rejected": -0.010624224320054054,
+      "step": 9
+    },
+    {
+      "epoch": 0.13174145738987236,
+      "grad_norm": 0.3784727156162262,
+      "learning_rate": 4.989014955054745e-07,
+      "logits/chosen": 10.471797943115234,
+      "logits/rejected": 10.57437515258789,
+      "logps/chosen": -173.78768920898438,
+      "logps/ref_chosen": -172.48837280273438,
+      "logps/ref_rejected": -179.625,
+      "logps/rejected": -180.91595458984375,
+      "loss": 0.6932,
+      "rewards/accuracies": 0.4765625,
+      "rewards/chosen": -0.012993087992072105,
+      "rewards/margins": -8.335959864780307e-05,
+      "rewards/rejected": -0.012909728102385998,
+      "step": 10
+    },
+    {
+      "epoch": 0.14491560312885962,
+      "grad_norm": 0.4041476547718048,
+      "learning_rate": 4.975306286336627e-07,
+      "logits/chosen": 10.539186477661133,
+      "logits/rejected": 10.477883338928223,
+      "logps/chosen": -172.62416076660156,
+      "logps/ref_chosen": -169.97216796875,
+      "logps/ref_rejected": -180.00955200195312,
+      "logps/rejected": -182.72525024414062,
+      "loss": 0.6929,
+      "rewards/accuracies": 0.546875,
+      "rewards/chosen": -0.026519589126110077,
+      "rewards/margins": 0.0006374535150825977,
+      "rewards/rejected": -0.027157040312886238,
+      "step": 11
+    },
+    {
+      "epoch": 0.15808974886784685,
+      "grad_norm": 0.5964677333831787,
+      "learning_rate": 4.956156357188939e-07,
+      "logits/chosen": 10.28010082244873,
+      "logits/rejected": 10.401304244995117,
+      "logps/chosen": -178.43714904785156,
+      "logps/ref_chosen": -175.83639526367188,
+      "logps/ref_rejected": -181.73045349121094,
+      "logps/rejected": -184.20223999023438,
+      "loss": 0.6938,
+      "rewards/accuracies": 0.4375,
+      "rewards/chosen": -0.026007818058133125,
+      "rewards/margins": -0.0012897354317829013,
+      "rewards/rejected": -0.024718083441257477,
+      "step": 12
+    },
+    {
+      "epoch": 0.17126389460683408,
+      "grad_norm": 0.488361120223999,
+      "learning_rate": 4.931607263312032e-07,
+      "logits/chosen": 10.292243957519531,
+      "logits/rejected": 10.386707305908203,
+      "logps/chosen": -173.99058532714844,
+      "logps/ref_chosen": -170.8806610107422,
+      "logps/ref_rejected": -180.6468963623047,
+      "logps/rejected": -184.01853942871094,
+      "loss": 0.6919,
+      "rewards/accuracies": 0.5234375,
+      "rewards/chosen": -0.03109920769929886,
+      "rewards/margins": 0.002617327030748129,
+      "rewards/rejected": -0.033716537058353424,
+      "step": 13
+    },
+    {
+      "epoch": 0.1844380403458213,
+      "grad_norm": 0.45636966824531555,
+      "learning_rate": 4.9017129689421e-07,
+      "logits/chosen": 10.729025840759277,
+      "logits/rejected": 10.921786308288574,
+      "logps/chosen": -174.65345764160156,
+      "logps/ref_chosen": -171.15963745117188,
+      "logps/ref_rejected": -180.73471069335938,
+      "logps/rejected": -184.30807495117188,
+      "loss": 0.6928,
+      "rewards/accuracies": 0.515625,
+      "rewards/chosen": -0.03493800014257431,
+      "rewards/margins": 0.0007955244509503245,
+      "rewards/rejected": -0.03573352098464966,
+      "step": 14
+    },
+    {
+      "epoch": 0.19761218608480857,
+      "grad_norm": 0.3838670551776886,
+      "learning_rate": 4.866539188226085e-07,
+      "logits/chosen": 10.265353202819824,
+      "logits/rejected": 10.378411293029785,
+      "logps/chosen": -177.72991943359375,
+      "logps/ref_chosen": -173.84222412109375,
+      "logps/ref_rejected": -182.6996612548828,
+      "logps/rejected": -186.89825439453125,
+      "loss": 0.6916,
+      "rewards/accuracies": 0.515625,
+      "rewards/chosen": -0.03887684643268585,
+      "rewards/margins": 0.003108714707195759,
+      "rewards/rejected": -0.04198555648326874,
+      "step": 15
+    },
+    {
+      "epoch": 0.2107863318237958,
+      "grad_norm": 0.3698700964450836,
+      "learning_rate": 4.826163240767716e-07,
+      "logits/chosen": 10.751184463500977,
+      "logits/rejected": 10.945128440856934,
+      "logps/chosen": -181.60301208496094,
+      "logps/ref_chosen": -178.638671875,
+      "logps/ref_rejected": -185.8243408203125,
+      "logps/rejected": -188.93057250976562,
+      "loss": 0.6925,
+      "rewards/accuracies": 0.5234375,
+      "rewards/chosen": -0.029643364250659943,
+      "rewards/margins": 0.0014189946232363582,
+      "rewards/rejected": -0.03106236271560192,
+      "step": 16
+    },
+    {
+      "epoch": 0.22396047756278303,
+      "grad_norm": 0.5419145822525024,
+      "learning_rate": 4.780673881662242e-07,
+      "logits/chosen": 10.279337882995605,
+      "logits/rejected": 10.265542984008789,
+      "logps/chosen": -176.72821044921875,
+      "logps/ref_chosen": -172.87332153320312,
+      "logps/ref_rejected": -177.30355834960938,
+      "logps/rejected": -181.23831176757812,
+      "loss": 0.6928,
+      "rewards/accuracies": 0.484375,
+      "rewards/chosen": -0.038549020886421204,
+      "rewards/margins": 0.0007985997945070267,
+      "rewards/rejected": -0.03934762626886368,
+      "step": 17
+    },
+    {
+      "epoch": 0.23713462330177026,
+      "grad_norm": 0.5288915038108826,
+      "learning_rate": 4.730171106393466e-07,
+      "logits/chosen": 10.508716583251953,
+      "logits/rejected": 10.601346969604492,
+      "logps/chosen": -176.69735717773438,
+      "logps/ref_chosen": -172.82644653320312,
+      "logps/ref_rejected": -180.93768310546875,
+      "logps/rejected": -184.82861328125,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.5234375,
+      "rewards/chosen": -0.038708824664354324,
+      "rewards/margins": 0.00020039017545059323,
+      "rewards/rejected": -0.03890921175479889,
+      "step": 18
+    },
+    {
+      "epoch": 0.2503087690407575,
+      "grad_norm": 0.45863935351371765,
+      "learning_rate": 4.6747659310219757e-07,
+      "logits/chosen": 10.509140014648438,
+      "logits/rejected": 10.403264999389648,
+      "logps/chosen": -183.62411499023438,
+      "logps/ref_chosen": -179.5069122314453,
+      "logps/ref_rejected": -185.01487731933594,
+      "logps/rejected": -189.2952880859375,
+      "loss": 0.6924,
+      "rewards/accuracies": 0.5078125,
+      "rewards/chosen": -0.041171781718730927,
+      "rewards/margins": 0.0016320659779012203,
+      "rewards/rejected": -0.04280385002493858,
+      "step": 19
+    },
+    {
+      "epoch": 0.2634829147797447,
+      "grad_norm": 0.6024115085601807,
+      "learning_rate": 4.6145801481477433e-07,
+      "logits/chosen": 10.452649116516113,
+      "logits/rejected": 10.525249481201172,
+      "logps/chosen": -174.3675994873047,
+      "logps/ref_chosen": -170.23623657226562,
+      "logps/ref_rejected": -177.55618286132812,
+      "logps/rejected": -181.85946655273438,
+      "loss": 0.6923,
+      "rewards/accuracies": 0.5078125,
+      "rewards/chosen": -0.04131368175148964,
+      "rewards/margins": 0.0017191548831760883,
+      "rewards/rejected": -0.04303283616900444,
+      "step": 20
+    },
+    {
+      "epoch": 0.276657060518732,
+      "grad_norm": 0.4595341384410858,
+      "learning_rate": 4.549746059183561e-07,
+      "logits/chosen": 10.422194480895996,
+      "logits/rejected": 10.483356475830078,
+      "logps/chosen": -180.44906616210938,
+      "logps/ref_chosen": -176.3700408935547,
+      "logps/ref_rejected": -182.80032348632812,
+      "logps/rejected": -187.17276000976562,
+      "loss": 0.6917,
+      "rewards/accuracies": 0.5703125,
+      "rewards/chosen": -0.04079030081629753,
+      "rewards/margins": 0.0029342519119381905,
+      "rewards/rejected": -0.04372455179691315,
+      "step": 21
+    },
+    {
+      "epoch": 0.28983120625771924,
+      "grad_norm": 0.3111652731895447,
+      "learning_rate": 4.480406183527823e-07,
+      "logits/chosen": 10.481276512145996,
+      "logits/rejected": 10.559721946716309,
+      "logps/chosen": -177.63865661621094,
+      "logps/ref_chosen": -173.29742431640625,
+      "logps/ref_rejected": -177.54916381835938,
+      "logps/rejected": -182.2364044189453,
+      "loss": 0.6915,
+      "rewards/accuracies": 0.6015625,
+      "rewards/chosen": -0.043412309139966965,
+      "rewards/margins": 0.003460067557170987,
+      "rewards/rejected": -0.04687237739562988,
+      "step": 22
+    },
+    {
+      "epoch": 0.3030053519967065,
+      "grad_norm": 0.520114541053772,
+      "learning_rate": 4.4067129452759546e-07,
+      "logits/chosen": 10.153634071350098,
+      "logits/rejected": 10.433453559875488,
+      "logps/chosen": -174.9644012451172,
+      "logps/ref_chosen": -170.41177368164062,
+      "logps/ref_rejected": -182.7073974609375,
+      "logps/rejected": -187.5763397216797,
+      "loss": 0.6916,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.04552610218524933,
+      "rewards/margins": 0.0031630881130695343,
+      "rewards/rejected": -0.04868919029831886,
+      "step": 23
+    },
+    {
+      "epoch": 0.3161794977356937,
+      "grad_norm": 0.38157743215560913,
+      "learning_rate": 4.3288283381591725e-07,
+      "logits/chosen": 10.212328910827637,
+      "logits/rejected": 10.149028778076172,
+      "logps/chosen": -178.75759887695312,
+      "logps/ref_chosen": -174.15003967285156,
+      "logps/ref_rejected": -181.65594482421875,
+      "logps/rejected": -186.58934020996094,
+      "loss": 0.6916,
+      "rewards/accuracies": 0.5703125,
+      "rewards/chosen": -0.046075694262981415,
+      "rewards/margins": 0.003258442971855402,
+      "rewards/rejected": -0.04933413863182068,
+      "step": 24
+    },
+    {
+      "epoch": 0.32935364347468093,
+      "grad_norm": 0.5030648112297058,
+      "learning_rate": 4.246923569447104e-07,
+      "logits/chosen": 10.425580978393555,
+      "logits/rejected": 10.567748069763184,
+      "logps/chosen": -181.23281860351562,
+      "logps/ref_chosen": -177.0232391357422,
+      "logps/ref_rejected": -184.0133056640625,
+      "logps/rejected": -188.3357696533203,
+      "loss": 0.6926,
+      "rewards/accuracies": 0.5234375,
+      "rewards/chosen": -0.04209586977958679,
+      "rewards/margins": 0.0011287340894341469,
+      "rewards/rejected": -0.04322460666298866,
+      "step": 25
+    },
+    {
+      "epoch": 0.34252778921366817,
+      "grad_norm": 0.3347052335739136,
+      "learning_rate": 4.161178683597054e-07,
+      "logits/chosen": 10.395452499389648,
+      "logits/rejected": 10.5776948928833,
+      "logps/chosen": -177.6590576171875,
+      "logps/ref_chosen": -172.2715606689453,
+      "logps/ref_rejected": -180.46673583984375,
+      "logps/rejected": -186.09349060058594,
+      "loss": 0.692,
+      "rewards/accuracies": 0.5703125,
+      "rewards/chosen": -0.05387478694319725,
+      "rewards/margins": 0.002392976311966777,
+      "rewards/rejected": -0.056267764419317245,
+      "step": 26
+    },
+    {
+      "epoch": 0.3557019349526554,
+      "grad_norm": 0.40187934041023254,
+      "learning_rate": 4.0717821664772124e-07,
+      "logits/chosen": 10.416847229003906,
+      "logits/rejected": 10.693917274475098,
+      "logps/chosen": -180.66729736328125,
+      "logps/ref_chosen": -174.87091064453125,
+      "logps/ref_rejected": -183.6770477294922,
+      "logps/rejected": -189.53399658203125,
+      "loss": 0.6929,
+      "rewards/accuracies": 0.4921875,
+      "rewards/chosen": -0.057963769882917404,
+      "rewards/margins": 0.0006057576974853873,
+      "rewards/rejected": -0.0585695244371891,
+      "step": 27
+    },
+    {
+      "epoch": 0.3688760806916426,
+      "grad_norm": 0.537835419178009,
+      "learning_rate": 3.978930531033806e-07,
+      "logits/chosen": 10.315240859985352,
+      "logits/rejected": 10.52540111541748,
+      "logps/chosen": -178.289794921875,
+      "logps/ref_chosen": -172.75225830078125,
+      "logps/ref_rejected": -180.3443145751953,
+      "logps/rejected": -186.29811096191406,
+      "loss": 0.6911,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.05537542328238487,
+      "rewards/margins": 0.004162484314292669,
+      "rewards/rejected": -0.05953790992498398,
+      "step": 28
+    },
+    {
+      "epoch": 0.3820502264306299,
+      "grad_norm": 0.9168035984039307,
+      "learning_rate": 3.882827885312998e-07,
+      "logits/chosen": 10.197809219360352,
+      "logits/rejected": 10.199935913085938,
+      "logps/chosen": -181.69606018066406,
+      "logps/ref_chosen": -176.17613220214844,
+      "logps/ref_rejected": -180.6341552734375,
+      "logps/rejected": -186.46124267578125,
+      "loss": 0.6917,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.05519918352365494,
+      "rewards/margins": 0.0030717107001692057,
+      "rewards/rejected": -0.0582708939909935,
+      "step": 29
+    },
+    {
+      "epoch": 0.39522437216961714,
+      "grad_norm": 0.3577282428741455,
+      "learning_rate": 3.7836854837871044e-07,
+      "logits/chosen": 10.497368812561035,
+      "logits/rejected": 10.725122451782227,
+      "logps/chosen": -184.75241088867188,
+      "logps/ref_chosen": -178.8612060546875,
+      "logps/ref_rejected": -187.1546173095703,
+      "logps/rejected": -193.5741424560547,
+      "loss": 0.6906,
+      "rewards/accuracies": 0.5859375,
+      "rewards/chosen": -0.058911994099617004,
+      "rewards/margins": 0.0052833897061645985,
+      "rewards/rejected": -0.06419539451599121,
+      "step": 30
+    },
+    {
+      "epoch": 0.4083985179086044,
+      "grad_norm": 0.46136385202407837,
+      "learning_rate": 3.681721262971413e-07,
+      "logits/chosen": 10.34453010559082,
+      "logits/rejected": 10.35668659210205,
+      "logps/chosen": -178.43930053710938,
+      "logps/ref_chosen": -172.60848999023438,
+      "logps/ref_rejected": -179.45013427734375,
+      "logps/rejected": -185.967529296875,
+      "loss": 0.6898,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.0583079531788826,
+      "rewards/margins": 0.006865999195724726,
+      "rewards/rejected": -0.06517395377159119,
+      "step": 31
+    },
+    {
+      "epoch": 0.4215726636475916,
+      "grad_norm": 0.3592880368232727,
+      "learning_rate": 3.577159362352426e-07,
+      "logits/chosen": 10.357152938842773,
+      "logits/rejected": 10.40371322631836,
+      "logps/chosen": -181.8780975341797,
+      "logps/ref_chosen": -175.29119873046875,
+      "logps/ref_rejected": -182.2250213623047,
+      "logps/rejected": -189.01063537597656,
+      "loss": 0.6923,
+      "rewards/accuracies": 0.5390625,
+      "rewards/chosen": -0.06586895883083344,
+      "rewards/margins": 0.001987436320632696,
+      "rewards/rejected": -0.06785639375448227,
+      "step": 32
+    },
+    {
+      "epoch": 0.43474680938657884,
+      "grad_norm": 0.4338357746601105,
+      "learning_rate": 3.470229631680624e-07,
+      "logits/chosen": 10.255874633789062,
+      "logits/rejected": 10.296948432922363,
+      "logps/chosen": -183.95484924316406,
+      "logps/ref_chosen": -177.42674255371094,
+      "logps/ref_rejected": -181.18801879882812,
+      "logps/rejected": -188.07952880859375,
+      "loss": 0.6914,
+      "rewards/accuracies": 0.5859375,
+      "rewards/chosen": -0.065280981361866,
+      "rewards/margins": 0.003634275868535042,
+      "rewards/rejected": -0.06891525536775589,
+      "step": 33
+    },
+    {
+      "epoch": 0.44792095512556607,
+      "grad_norm": 0.4542549252510071,
+      "learning_rate": 3.361167125710832e-07,
+      "logits/chosen": 10.056829452514648,
+      "logits/rejected": 10.121377944946289,
+      "logps/chosen": -181.4767303466797,
+      "logps/ref_chosen": -173.72467041015625,
+      "logps/ref_rejected": -180.5912322998047,
+      "logps/rejected": -188.53164672851562,
+      "loss": 0.6923,
+      "rewards/accuracies": 0.4921875,
+      "rewards/chosen": -0.07752064615488052,
+      "rewards/margins": 0.0018832057248800993,
+      "rewards/rejected": -0.0794038400053978,
+      "step": 34
+    },
+    {
+      "epoch": 0.4610951008645533,
+      "grad_norm": 0.4729287028312683,
+      "learning_rate": 3.2502115875008516e-07,
+      "logits/chosen": 10.46947193145752,
+      "logits/rejected": 10.667445182800293,
+      "logps/chosen": -182.939208984375,
+      "logps/ref_chosen": -175.0980987548828,
+      "logps/ref_rejected": -181.65821838378906,
+      "logps/rejected": -189.89309692382812,
+      "loss": 0.6913,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.07841099798679352,
+      "rewards/margins": 0.00393773103132844,
+      "rewards/rejected": -0.08234872668981552,
+      "step": 35
+    },
+    {
+      "epoch": 0.47426924660354053,
+      "grad_norm": 0.8570416569709778,
+      "learning_rate": 3.137606921404191e-07,
+      "logits/chosen": 10.280503273010254,
+      "logits/rejected": 10.623340606689453,
+      "logps/chosen": -182.2921142578125,
+      "logps/ref_chosen": -174.0312042236328,
+      "logps/ref_rejected": -181.84324645996094,
+      "logps/rejected": -190.1889190673828,
+      "loss": 0.6928,
+      "rewards/accuracies": 0.484375,
+      "rewards/chosen": -0.08260920643806458,
+      "rewards/margins": 0.0008472882909700274,
+      "rewards/rejected": -0.08345648646354675,
+      "step": 36
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 75,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 12,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-48/added_tokens.json

@@ -0,0 +1,13 @@
+{
+  "<|assistant|>": 32001,
+  "<|endoftext|>": 32000,
+  "<|end|>": 32007,
+  "<|placeholder1|>": 32002,
+  "<|placeholder2|>": 32003,
+  "<|placeholder3|>": 32004,
+  "<|placeholder4|>": 32005,
+  "<|placeholder5|>": 32008,
+  "<|placeholder6|>": 32009,
+  "<|system|>": 32006,
+  "<|user|>": 32010
+}

checkpoint-48/config.json

@@ -0,0 +1,36 @@
+{
+  "_name_or_path": "/home/v-liuzichuan/huggingface/Phi-3-mini-4k-instruct",
+  "architectures": [
+    "Phi3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_phi3.Phi3Config",
+    "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
+  },
+  "bos_token_id": 1,
+  "embd_pdrop": 0.0,
+  "eos_token_id": 32000,
+  "hidden_act": "silu",
+  "hidden_size": 3072,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 4096,
+  "model_type": "phi3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "original_max_position_embeddings": 4096,
+  "pad_token_id": 32000,
+  "resid_pdrop": 0.0,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "sliding_window": 2047,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.45.2",
+  "use_cache": false,
+  "vocab_size": 32064
+}

checkpoint-48/generation_config.json

@@ -0,0 +1,11 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": [
+    32000,
+    32001,
+    32007
+  ],
+  "pad_token_id": 32000,
+  "transformers_version": "4.45.2"
+}

checkpoint-48/model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e990237fb825f6313636d6cc7722fc91e4d062413088db68c3dde02274edda4
+size 4972489328

checkpoint-48/model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64832e83d10c15b0a46b0b3c790f66ea215f71e167df40ae6f84b2afdf911376
+size 2669692552

checkpoint-48/model.safetensors.index.json

@@ -0,0 +1,202 @@
+{
+  "metadata": {
+    "total_size": 7642159104
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}

checkpoint-48/special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-48/tokenizer_config.json

@@ -0,0 +1,131 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "32000": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32002": {
+      "content": "<|placeholder1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32003": {
+      "content": "<|placeholder2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32004": {
+      "content": "<|placeholder3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32005": {
+      "content": "<|placeholder4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32006": {
+      "content": "<|system|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32007": {
+      "content": "<|end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32008": {
+      "content": "<|placeholder5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32009": {
+      "content": "<|placeholder6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32010": {
+      "content": "<|user|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% for message in messages %}{% if message['role'] == 'system' %}{{'<|system|>\n' + message['content'] + '<|end|>\n'}}{% elif message['role'] == 'user' %}{{'<|user|>\n' + message['content'] + '<|end|>\n'}}{% elif message['role'] == 'assistant' %}{{'<|assistant|>\n' + message['content'] + '<|end|>\n'}}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>\n' }}{% else %}{{ eos_token }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "legacy": false,
+  "model_max_length": 4096,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

checkpoint-48/trainer_state.json

@@ -0,0 +1,849 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.6323589954713874,
+  "eval_steps": 500,
+  "global_step": 48,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.013174145738987238,
+      "grad_norm": 0.48070791363716125,
+      "learning_rate": 6.25e-08,
+      "logits/chosen": 10.32492733001709,
+      "logits/rejected": 10.282785415649414,
+      "logps/chosen": -176.09544372558594,
+      "logps/ref_chosen": -176.09544372558594,
+      "logps/ref_rejected": -181.75552368164062,
+      "logps/rejected": -181.75552368164062,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.0,
+      "rewards/chosen": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/rejected": 0.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.026348291477974475,
+      "grad_norm": 0.4185059070587158,
+      "learning_rate": 1.25e-07,
+      "logits/chosen": 10.530074119567871,
+      "logits/rejected": 10.672085762023926,
+      "logps/chosen": -173.2084503173828,
+      "logps/ref_chosen": -173.2084503173828,
+      "logps/ref_rejected": -187.02206420898438,
+      "logps/rejected": -187.02206420898438,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.0,
+      "rewards/chosen": 0.0,
+      "rewards/margins": 0.0,
+      "rewards/rejected": 0.0,
+      "step": 2
+    },
+    {
+      "epoch": 0.03952243721696171,
+      "grad_norm": 0.3848375082015991,
+      "learning_rate": 1.875e-07,
+      "logits/chosen": 10.340475082397461,
+      "logits/rejected": 10.486526489257812,
+      "logps/chosen": -169.6623992919922,
+      "logps/ref_chosen": -169.080810546875,
+      "logps/ref_rejected": -182.52792358398438,
+      "logps/rejected": -182.80642700195312,
+      "loss": 0.6947,
+      "rewards/accuracies": 0.3828125,
+      "rewards/chosen": -0.005815833806991577,
+      "rewards/margins": -0.0030306153930723667,
+      "rewards/rejected": -0.002785218646749854,
+      "step": 3
+    },
+    {
+      "epoch": 0.05269658295594895,
+      "grad_norm": 0.7261964678764343,
+      "learning_rate": 2.5e-07,
+      "logits/chosen": 10.3652925491333,
+      "logits/rejected": 10.412069320678711,
+      "logps/chosen": -177.03428649902344,
+      "logps/ref_chosen": -176.78369140625,
+      "logps/ref_rejected": -180.0931396484375,
+      "logps/rejected": -180.21463012695312,
+      "loss": 0.6938,
+      "rewards/accuracies": 0.4765625,
+      "rewards/chosen": -0.0025058696046471596,
+      "rewards/margins": -0.0012909012148156762,
+      "rewards/rejected": -0.0012149682734161615,
+      "step": 4
+    },
+    {
+      "epoch": 0.06587072869493618,
+      "grad_norm": 0.3572319746017456,
+      "learning_rate": 3.1249999999999997e-07,
+      "logits/chosen": 10.757452011108398,
+      "logits/rejected": 10.791389465332031,
+      "logps/chosen": -169.12631225585938,
+      "logps/ref_chosen": -168.522216796875,
+      "logps/ref_rejected": -176.08982849121094,
+      "logps/rejected": -176.59901428222656,
+      "loss": 0.6936,
+      "rewards/accuracies": 0.4609375,
+      "rewards/chosen": -0.0060411859303712845,
+      "rewards/margins": -0.0009493756806477904,
+      "rewards/rejected": -0.0050918105989694595,
+      "step": 5
+    },
+    {
+      "epoch": 0.07904487443392343,
+      "grad_norm": 0.4071093201637268,
+      "learning_rate": 3.75e-07,
+      "logits/chosen": 11.003137588500977,
+      "logits/rejected": 11.056475639343262,
+      "logps/chosen": -174.662353515625,
+      "logps/ref_chosen": -174.5843048095703,
+      "logps/ref_rejected": -181.65040588378906,
+      "logps/rejected": -181.72401428222656,
+      "loss": 0.6932,
+      "rewards/accuracies": 0.5078125,
+      "rewards/chosen": -0.0007804610067978501,
+      "rewards/margins": -4.43047538283281e-05,
+      "rewards/rejected": -0.0007361561874859035,
+      "step": 6
+    },
+    {
+      "epoch": 0.09221902017291066,
+      "grad_norm": 0.7951124906539917,
+      "learning_rate": 4.375e-07,
+      "logits/chosen": 10.003807067871094,
+      "logits/rejected": 10.067156791687012,
+      "logps/chosen": -171.4049530029297,
+      "logps/ref_chosen": -170.4031219482422,
+      "logps/ref_rejected": -181.00917053222656,
+      "logps/rejected": -181.8280792236328,
+      "loss": 0.6941,
+      "rewards/accuracies": 0.4453125,
+      "rewards/chosen": -0.010018287226557732,
+      "rewards/margins": -0.0018291514134034514,
+      "rewards/rejected": -0.008189136162400246,
+      "step": 7
+    },
+    {
+      "epoch": 0.1053931659118979,
+      "grad_norm": 0.5591660737991333,
+      "learning_rate": 5e-07,
+      "logits/chosen": 10.874868392944336,
+      "logits/rejected": 10.79828929901123,
+      "logps/chosen": -179.36700439453125,
+      "logps/ref_chosen": -178.47369384765625,
+      "logps/ref_rejected": -184.52243041992188,
+      "logps/rejected": -185.44284057617188,
+      "loss": 0.693,
+      "rewards/accuracies": 0.5078125,
+      "rewards/chosen": -0.008933190256357193,
+      "rewards/margins": 0.0002710025873966515,
+      "rewards/rejected": -0.009204192087054253,
+      "step": 8
+    },
+    {
+      "epoch": 0.11856731165088513,
+      "grad_norm": 0.45744186639785767,
+      "learning_rate": 4.997252228714278e-07,
+      "logits/chosen": 10.472944259643555,
+      "logits/rejected": 10.640350341796875,
+      "logps/chosen": -175.36553955078125,
+      "logps/ref_chosen": -174.38418579101562,
+      "logps/ref_rejected": -182.0985565185547,
+      "logps/rejected": -183.1609649658203,
+      "loss": 0.6928,
+      "rewards/accuracies": 0.5234375,
+      "rewards/chosen": -0.009813489392399788,
+      "rewards/margins": 0.0008107352769002318,
+      "rewards/rejected": -0.010624224320054054,
+      "step": 9
+    },
+    {
+      "epoch": 0.13174145738987236,
+      "grad_norm": 0.3784727156162262,
+      "learning_rate": 4.989014955054745e-07,
+      "logits/chosen": 10.471797943115234,
+      "logits/rejected": 10.57437515258789,
+      "logps/chosen": -173.78768920898438,
+      "logps/ref_chosen": -172.48837280273438,
+      "logps/ref_rejected": -179.625,
+      "logps/rejected": -180.91595458984375,
+      "loss": 0.6932,
+      "rewards/accuracies": 0.4765625,
+      "rewards/chosen": -0.012993087992072105,
+      "rewards/margins": -8.335959864780307e-05,
+      "rewards/rejected": -0.012909728102385998,
+      "step": 10
+    },
+    {
+      "epoch": 0.14491560312885962,
+      "grad_norm": 0.4041476547718048,
+      "learning_rate": 4.975306286336627e-07,
+      "logits/chosen": 10.539186477661133,
+      "logits/rejected": 10.477883338928223,
+      "logps/chosen": -172.62416076660156,
+      "logps/ref_chosen": -169.97216796875,
+      "logps/ref_rejected": -180.00955200195312,
+      "logps/rejected": -182.72525024414062,
+      "loss": 0.6929,
+      "rewards/accuracies": 0.546875,
+      "rewards/chosen": -0.026519589126110077,
+      "rewards/margins": 0.0006374535150825977,
+      "rewards/rejected": -0.027157040312886238,
+      "step": 11
+    },
+    {
+      "epoch": 0.15808974886784685,
+      "grad_norm": 0.5964677333831787,
+      "learning_rate": 4.956156357188939e-07,
+      "logits/chosen": 10.28010082244873,
+      "logits/rejected": 10.401304244995117,
+      "logps/chosen": -178.43714904785156,
+      "logps/ref_chosen": -175.83639526367188,
+      "logps/ref_rejected": -181.73045349121094,
+      "logps/rejected": -184.20223999023438,
+      "loss": 0.6938,
+      "rewards/accuracies": 0.4375,
+      "rewards/chosen": -0.026007818058133125,
+      "rewards/margins": -0.0012897354317829013,
+      "rewards/rejected": -0.024718083441257477,
+      "step": 12
+    },
+    {
+      "epoch": 0.17126389460683408,
+      "grad_norm": 0.488361120223999,
+      "learning_rate": 4.931607263312032e-07,
+      "logits/chosen": 10.292243957519531,
+      "logits/rejected": 10.386707305908203,
+      "logps/chosen": -173.99058532714844,
+      "logps/ref_chosen": -170.8806610107422,
+      "logps/ref_rejected": -180.6468963623047,
+      "logps/rejected": -184.01853942871094,
+      "loss": 0.6919,
+      "rewards/accuracies": 0.5234375,
+      "rewards/chosen": -0.03109920769929886,
+      "rewards/margins": 0.002617327030748129,
+      "rewards/rejected": -0.033716537058353424,
+      "step": 13
+    },
+    {
+      "epoch": 0.1844380403458213,
+      "grad_norm": 0.45636966824531555,
+      "learning_rate": 4.9017129689421e-07,
+      "logits/chosen": 10.729025840759277,
+      "logits/rejected": 10.921786308288574,
+      "logps/chosen": -174.65345764160156,
+      "logps/ref_chosen": -171.15963745117188,
+      "logps/ref_rejected": -180.73471069335938,
+      "logps/rejected": -184.30807495117188,
+      "loss": 0.6928,
+      "rewards/accuracies": 0.515625,
+      "rewards/chosen": -0.03493800014257431,
+      "rewards/margins": 0.0007955244509503245,
+      "rewards/rejected": -0.03573352098464966,
+      "step": 14
+    },
+    {
+      "epoch": 0.19761218608480857,
+      "grad_norm": 0.3838670551776886,
+      "learning_rate": 4.866539188226085e-07,
+      "logits/chosen": 10.265353202819824,
+      "logits/rejected": 10.378411293029785,
+      "logps/chosen": -177.72991943359375,
+      "logps/ref_chosen": -173.84222412109375,
+      "logps/ref_rejected": -182.6996612548828,
+      "logps/rejected": -186.89825439453125,
+      "loss": 0.6916,
+      "rewards/accuracies": 0.515625,
+      "rewards/chosen": -0.03887684643268585,
+      "rewards/margins": 0.003108714707195759,
+      "rewards/rejected": -0.04198555648326874,
+      "step": 15
+    },
+    {
+      "epoch": 0.2107863318237958,
+      "grad_norm": 0.3698700964450836,
+      "learning_rate": 4.826163240767716e-07,
+      "logits/chosen": 10.751184463500977,
+      "logits/rejected": 10.945128440856934,
+      "logps/chosen": -181.60301208496094,
+      "logps/ref_chosen": -178.638671875,
+      "logps/ref_rejected": -185.8243408203125,
+      "logps/rejected": -188.93057250976562,
+      "loss": 0.6925,
+      "rewards/accuracies": 0.5234375,
+      "rewards/chosen": -0.029643364250659943,
+      "rewards/margins": 0.0014189946232363582,
+      "rewards/rejected": -0.03106236271560192,
+      "step": 16
+    },
+    {
+      "epoch": 0.22396047756278303,
+      "grad_norm": 0.5419145822525024,
+      "learning_rate": 4.780673881662242e-07,
+      "logits/chosen": 10.279337882995605,
+      "logits/rejected": 10.265542984008789,
+      "logps/chosen": -176.72821044921875,
+      "logps/ref_chosen": -172.87332153320312,
+      "logps/ref_rejected": -177.30355834960938,
+      "logps/rejected": -181.23831176757812,
+      "loss": 0.6928,
+      "rewards/accuracies": 0.484375,
+      "rewards/chosen": -0.038549020886421204,
+      "rewards/margins": 0.0007985997945070267,
+      "rewards/rejected": -0.03934762626886368,
+      "step": 17
+    },
+    {
+      "epoch": 0.23713462330177026,
+      "grad_norm": 0.5288915038108826,
+      "learning_rate": 4.730171106393466e-07,
+      "logits/chosen": 10.508716583251953,
+      "logits/rejected": 10.601346969604492,
+      "logps/chosen": -176.69735717773438,
+      "logps/ref_chosen": -172.82644653320312,
+      "logps/ref_rejected": -180.93768310546875,
+      "logps/rejected": -184.82861328125,
+      "loss": 0.6931,
+      "rewards/accuracies": 0.5234375,
+      "rewards/chosen": -0.038708824664354324,
+      "rewards/margins": 0.00020039017545059323,
+      "rewards/rejected": -0.03890921175479889,
+      "step": 18
+    },
+    {
+      "epoch": 0.2503087690407575,
+      "grad_norm": 0.45863935351371765,
+      "learning_rate": 4.6747659310219757e-07,
+      "logits/chosen": 10.509140014648438,
+      "logits/rejected": 10.403264999389648,
+      "logps/chosen": -183.62411499023438,
+      "logps/ref_chosen": -179.5069122314453,
+      "logps/ref_rejected": -185.01487731933594,
+      "logps/rejected": -189.2952880859375,
+      "loss": 0.6924,
+      "rewards/accuracies": 0.5078125,
+      "rewards/chosen": -0.041171781718730927,
+      "rewards/margins": 0.0016320659779012203,
+      "rewards/rejected": -0.04280385002493858,
+      "step": 19
+    },
+    {
+      "epoch": 0.2634829147797447,
+      "grad_norm": 0.6024115085601807,
+      "learning_rate": 4.6145801481477433e-07,
+      "logits/chosen": 10.452649116516113,
+      "logits/rejected": 10.525249481201172,
+      "logps/chosen": -174.3675994873047,
+      "logps/ref_chosen": -170.23623657226562,
+      "logps/ref_rejected": -177.55618286132812,
+      "logps/rejected": -181.85946655273438,
+      "loss": 0.6923,
+      "rewards/accuracies": 0.5078125,
+      "rewards/chosen": -0.04131368175148964,
+      "rewards/margins": 0.0017191548831760883,
+      "rewards/rejected": -0.04303283616900444,
+      "step": 20
+    },
+    {
+      "epoch": 0.276657060518732,
+      "grad_norm": 0.4595341384410858,
+      "learning_rate": 4.549746059183561e-07,
+      "logits/chosen": 10.422194480895996,
+      "logits/rejected": 10.483356475830078,
+      "logps/chosen": -180.44906616210938,
+      "logps/ref_chosen": -176.3700408935547,
+      "logps/ref_rejected": -182.80032348632812,
+      "logps/rejected": -187.17276000976562,
+      "loss": 0.6917,
+      "rewards/accuracies": 0.5703125,
+      "rewards/chosen": -0.04079030081629753,
+      "rewards/margins": 0.0029342519119381905,
+      "rewards/rejected": -0.04372455179691315,
+      "step": 21
+    },
+    {
+      "epoch": 0.28983120625771924,
+      "grad_norm": 0.3111652731895447,
+      "learning_rate": 4.480406183527823e-07,
+      "logits/chosen": 10.481276512145996,
+      "logits/rejected": 10.559721946716309,
+      "logps/chosen": -177.63865661621094,
+      "logps/ref_chosen": -173.29742431640625,
+      "logps/ref_rejected": -177.54916381835938,
+      "logps/rejected": -182.2364044189453,
+      "loss": 0.6915,
+      "rewards/accuracies": 0.6015625,
+      "rewards/chosen": -0.043412309139966965,
+      "rewards/margins": 0.003460067557170987,
+      "rewards/rejected": -0.04687237739562988,
+      "step": 22
+    },
+    {
+      "epoch": 0.3030053519967065,
+      "grad_norm": 0.520114541053772,
+      "learning_rate": 4.4067129452759546e-07,
+      "logits/chosen": 10.153634071350098,
+      "logits/rejected": 10.433453559875488,
+      "logps/chosen": -174.9644012451172,
+      "logps/ref_chosen": -170.41177368164062,
+      "logps/ref_rejected": -182.7073974609375,
+      "logps/rejected": -187.5763397216797,
+      "loss": 0.6916,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.04552610218524933,
+      "rewards/margins": 0.0031630881130695343,
+      "rewards/rejected": -0.04868919029831886,
+      "step": 23
+    },
+    {
+      "epoch": 0.3161794977356937,
+      "grad_norm": 0.38157743215560913,
+      "learning_rate": 4.3288283381591725e-07,
+      "logits/chosen": 10.212328910827637,
+      "logits/rejected": 10.149028778076172,
+      "logps/chosen": -178.75759887695312,
+      "logps/ref_chosen": -174.15003967285156,
+      "logps/ref_rejected": -181.65594482421875,
+      "logps/rejected": -186.58934020996094,
+      "loss": 0.6916,
+      "rewards/accuracies": 0.5703125,
+      "rewards/chosen": -0.046075694262981415,
+      "rewards/margins": 0.003258442971855402,
+      "rewards/rejected": -0.04933413863182068,
+      "step": 24
+    },
+    {
+      "epoch": 0.32935364347468093,
+      "grad_norm": 0.5030648112297058,
+      "learning_rate": 4.246923569447104e-07,
+      "logits/chosen": 10.425580978393555,
+      "logits/rejected": 10.567748069763184,
+      "logps/chosen": -181.23281860351562,
+      "logps/ref_chosen": -177.0232391357422,
+      "logps/ref_rejected": -184.0133056640625,
+      "logps/rejected": -188.3357696533203,
+      "loss": 0.6926,
+      "rewards/accuracies": 0.5234375,
+      "rewards/chosen": -0.04209586977958679,
+      "rewards/margins": 0.0011287340894341469,
+      "rewards/rejected": -0.04322460666298866,
+      "step": 25
+    },
+    {
+      "epoch": 0.34252778921366817,
+      "grad_norm": 0.3347052335739136,
+      "learning_rate": 4.161178683597054e-07,
+      "logits/chosen": 10.395452499389648,
+      "logits/rejected": 10.5776948928833,
+      "logps/chosen": -177.6590576171875,
+      "logps/ref_chosen": -172.2715606689453,
+      "logps/ref_rejected": -180.46673583984375,
+      "logps/rejected": -186.09349060058594,
+      "loss": 0.692,
+      "rewards/accuracies": 0.5703125,
+      "rewards/chosen": -0.05387478694319725,
+      "rewards/margins": 0.002392976311966777,
+      "rewards/rejected": -0.056267764419317245,
+      "step": 26
+    },
+    {
+      "epoch": 0.3557019349526554,
+      "grad_norm": 0.40187934041023254,
+      "learning_rate": 4.0717821664772124e-07,
+      "logits/chosen": 10.416847229003906,
+      "logits/rejected": 10.693917274475098,
+      "logps/chosen": -180.66729736328125,
+      "logps/ref_chosen": -174.87091064453125,
+      "logps/ref_rejected": -183.6770477294922,
+      "logps/rejected": -189.53399658203125,
+      "loss": 0.6929,
+      "rewards/accuracies": 0.4921875,
+      "rewards/chosen": -0.057963769882917404,
+      "rewards/margins": 0.0006057576974853873,
+      "rewards/rejected": -0.0585695244371891,
+      "step": 27
+    },
+    {
+      "epoch": 0.3688760806916426,
+      "grad_norm": 0.537835419178009,
+      "learning_rate": 3.978930531033806e-07,
+      "logits/chosen": 10.315240859985352,
+      "logits/rejected": 10.52540111541748,
+      "logps/chosen": -178.289794921875,
+      "logps/ref_chosen": -172.75225830078125,
+      "logps/ref_rejected": -180.3443145751953,
+      "logps/rejected": -186.29811096191406,
+      "loss": 0.6911,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.05537542328238487,
+      "rewards/margins": 0.004162484314292669,
+      "rewards/rejected": -0.05953790992498398,
+      "step": 28
+    },
+    {
+      "epoch": 0.3820502264306299,
+      "grad_norm": 0.9168035984039307,
+      "learning_rate": 3.882827885312998e-07,
+      "logits/chosen": 10.197809219360352,
+      "logits/rejected": 10.199935913085938,
+      "logps/chosen": -181.69606018066406,
+      "logps/ref_chosen": -176.17613220214844,
+      "logps/ref_rejected": -180.6341552734375,
+      "logps/rejected": -186.46124267578125,
+      "loss": 0.6917,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.05519918352365494,
+      "rewards/margins": 0.0030717107001692057,
+      "rewards/rejected": -0.0582708939909935,
+      "step": 29
+    },
+    {
+      "epoch": 0.39522437216961714,
+      "grad_norm": 0.3577282428741455,
+      "learning_rate": 3.7836854837871044e-07,
+      "logits/chosen": 10.497368812561035,
+      "logits/rejected": 10.725122451782227,
+      "logps/chosen": -184.75241088867188,
+      "logps/ref_chosen": -178.8612060546875,
+      "logps/ref_rejected": -187.1546173095703,
+      "logps/rejected": -193.5741424560547,
+      "loss": 0.6906,
+      "rewards/accuracies": 0.5859375,
+      "rewards/chosen": -0.058911994099617004,
+      "rewards/margins": 0.0052833897061645985,
+      "rewards/rejected": -0.06419539451599121,
+      "step": 30
+    },
+    {
+      "epoch": 0.4083985179086044,
+      "grad_norm": 0.46136385202407837,
+      "learning_rate": 3.681721262971413e-07,
+      "logits/chosen": 10.34453010559082,
+      "logits/rejected": 10.35668659210205,
+      "logps/chosen": -178.43930053710938,
+      "logps/ref_chosen": -172.60848999023438,
+      "logps/ref_rejected": -179.45013427734375,
+      "logps/rejected": -185.967529296875,
+      "loss": 0.6898,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.0583079531788826,
+      "rewards/margins": 0.006865999195724726,
+      "rewards/rejected": -0.06517395377159119,
+      "step": 31
+    },
+    {
+      "epoch": 0.4215726636475916,
+      "grad_norm": 0.3592880368232727,
+      "learning_rate": 3.577159362352426e-07,
+      "logits/chosen": 10.357152938842773,
+      "logits/rejected": 10.40371322631836,
+      "logps/chosen": -181.8780975341797,
+      "logps/ref_chosen": -175.29119873046875,
+      "logps/ref_rejected": -182.2250213623047,
+      "logps/rejected": -189.01063537597656,
+      "loss": 0.6923,
+      "rewards/accuracies": 0.5390625,
+      "rewards/chosen": -0.06586895883083344,
+      "rewards/margins": 0.001987436320632696,
+      "rewards/rejected": -0.06785639375448227,
+      "step": 32
+    },
+    {
+      "epoch": 0.43474680938657884,
+      "grad_norm": 0.4338357746601105,
+      "learning_rate": 3.470229631680624e-07,
+      "logits/chosen": 10.255874633789062,
+      "logits/rejected": 10.296948432922363,
+      "logps/chosen": -183.95484924316406,
+      "logps/ref_chosen": -177.42674255371094,
+      "logps/ref_rejected": -181.18801879882812,
+      "logps/rejected": -188.07952880859375,
+      "loss": 0.6914,
+      "rewards/accuracies": 0.5859375,
+      "rewards/chosen": -0.065280981361866,
+      "rewards/margins": 0.003634275868535042,
+      "rewards/rejected": -0.06891525536775589,
+      "step": 33
+    },
+    {
+      "epoch": 0.44792095512556607,
+      "grad_norm": 0.4542549252510071,
+      "learning_rate": 3.361167125710832e-07,
+      "logits/chosen": 10.056829452514648,
+      "logits/rejected": 10.121377944946289,
+      "logps/chosen": -181.4767303466797,
+      "logps/ref_chosen": -173.72467041015625,
+      "logps/ref_rejected": -180.5912322998047,
+      "logps/rejected": -188.53164672851562,
+      "loss": 0.6923,
+      "rewards/accuracies": 0.4921875,
+      "rewards/chosen": -0.07752064615488052,
+      "rewards/margins": 0.0018832057248800993,
+      "rewards/rejected": -0.0794038400053978,
+      "step": 34
+    },
+    {
+      "epoch": 0.4610951008645533,
+      "grad_norm": 0.4729287028312683,
+      "learning_rate": 3.2502115875008516e-07,
+      "logits/chosen": 10.46947193145752,
+      "logits/rejected": 10.667445182800293,
+      "logps/chosen": -182.939208984375,
+      "logps/ref_chosen": -175.0980987548828,
+      "logps/ref_rejected": -181.65821838378906,
+      "logps/rejected": -189.89309692382812,
+      "loss": 0.6913,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.07841099798679352,
+      "rewards/margins": 0.00393773103132844,
+      "rewards/rejected": -0.08234872668981552,
+      "step": 35
+    },
+    {
+      "epoch": 0.47426924660354053,
+      "grad_norm": 0.8570416569709778,
+      "learning_rate": 3.137606921404191e-07,
+      "logits/chosen": 10.280503273010254,
+      "logits/rejected": 10.623340606689453,
+      "logps/chosen": -182.2921142578125,
+      "logps/ref_chosen": -174.0312042236328,
+      "logps/ref_rejected": -181.84324645996094,
+      "logps/rejected": -190.1889190673828,
+      "loss": 0.6928,
+      "rewards/accuracies": 0.484375,
+      "rewards/chosen": -0.08260920643806458,
+      "rewards/margins": 0.0008472882909700274,
+      "rewards/rejected": -0.08345648646354675,
+      "step": 36
+    },
+    {
+      "epoch": 0.4874433923425278,
+      "grad_norm": 0.9711470007896423,
+      "learning_rate": 3.0236006569153616e-07,
+      "logits/chosen": 10.310943603515625,
+      "logits/rejected": 10.279195785522461,
+      "logps/chosen": -183.245361328125,
+      "logps/ref_chosen": -175.3839569091797,
+      "logps/ref_rejected": -182.51773071289062,
+      "logps/rejected": -190.82374572753906,
+      "loss": 0.691,
+      "rewards/accuracies": 0.5625,
+      "rewards/chosen": -0.07861402630805969,
+      "rewards/margins": 0.004445935599505901,
+      "rewards/rejected": -0.08305995911359787,
+      "step": 37
+    },
+    {
+      "epoch": 0.500617538081515,
+      "grad_norm": 0.680104672908783,
+      "learning_rate": 2.9084434045463254e-07,
+      "logits/chosen": 10.41287612915039,
+      "logits/rejected": 10.449625968933105,
+      "logps/chosen": -182.55191040039062,
+      "logps/ref_chosen": -174.5338134765625,
+      "logps/ref_rejected": -179.96829223632812,
+      "logps/rejected": -188.60166931152344,
+      "loss": 0.6902,
+      "rewards/accuracies": 0.6171875,
+      "rewards/chosen": -0.08018088340759277,
+      "rewards/margins": 0.00615295534953475,
+      "rewards/rejected": -0.08633382618427277,
+      "step": 38
+    },
+    {
+      "epoch": 0.5137916838205022,
+      "grad_norm": 0.41836705803871155,
+      "learning_rate": 2.7923883049302066e-07,
+      "logits/chosen": 10.237342834472656,
+      "logits/rejected": 10.221612930297852,
+      "logps/chosen": -190.19024658203125,
+      "logps/ref_chosen": -181.61000061035156,
+      "logps/ref_rejected": -187.48886108398438,
+      "logps/rejected": -196.35353088378906,
+      "loss": 0.6919,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": -0.08580265194177628,
+      "rewards/margins": 0.0028439394664019346,
+      "rewards/rejected": -0.08864659070968628,
+      "step": 39
+    },
+    {
+      "epoch": 0.5269658295594895,
+      "grad_norm": 0.4576071798801422,
+      "learning_rate": 2.6756904723632324e-07,
+      "logits/chosen": 10.366464614868164,
+      "logits/rejected": 10.504589080810547,
+      "logps/chosen": -181.5367431640625,
+      "logps/ref_chosen": -172.43283081054688,
+      "logps/ref_rejected": -186.99594116210938,
+      "logps/rejected": -196.2064971923828,
+      "loss": 0.6927,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": -0.0910390242934227,
+      "rewards/margins": 0.001066619879566133,
+      "rewards/rejected": -0.09210564941167831,
+      "step": 40
+    },
+    {
+      "epoch": 0.5401399752984768,
+      "grad_norm": 0.5803886651992798,
+      "learning_rate": 2.5586064340081516e-07,
+      "logits/chosen": 10.624752044677734,
+      "logits/rejected": 10.738086700439453,
+      "logps/chosen": -183.4002685546875,
+      "logps/ref_chosen": -175.37374877929688,
+      "logps/ref_rejected": -181.78671264648438,
+      "logps/rejected": -190.14678955078125,
+      "loss": 0.6916,
+      "rewards/accuracies": 0.5703125,
+      "rewards/chosen": -0.08026508241891861,
+      "rewards/margins": 0.003335628192871809,
+      "rewards/rejected": -0.083600714802742,
+      "step": 41
+    },
+    {
+      "epoch": 0.553314121037464,
+      "grad_norm": 0.48118287324905396,
+      "learning_rate": 2.4413935659918487e-07,
+      "logits/chosen": 10.010458946228027,
+      "logits/rejected": 10.1697359085083,
+      "logps/chosen": -182.1250457763672,
+      "logps/ref_chosen": -173.79501342773438,
+      "logps/ref_rejected": -179.49351501464844,
+      "logps/rejected": -188.2036895751953,
+      "loss": 0.6914,
+      "rewards/accuracies": 0.4921875,
+      "rewards/chosen": -0.08330036699771881,
+      "rewards/margins": 0.003801207058131695,
+      "rewards/rejected": -0.08710157871246338,
+      "step": 42
+    },
+    {
+      "epoch": 0.5664882667764513,
+      "grad_norm": 0.7869899868965149,
+      "learning_rate": 2.3243095276367684e-07,
+      "logits/chosen": 9.809760093688965,
+      "logits/rejected": 10.002002716064453,
+      "logps/chosen": -171.12933349609375,
+      "logps/ref_chosen": -162.35044860839844,
+      "logps/ref_rejected": -177.5733642578125,
+      "logps/rejected": -187.271240234375,
+      "loss": 0.6887,
+      "rewards/accuracies": 0.640625,
+      "rewards/chosen": -0.08778898417949677,
+      "rewards/margins": 0.009189910255372524,
+      "rewards/rejected": -0.09697890281677246,
+      "step": 43
+    },
+    {
+      "epoch": 0.5796624125154385,
+      "grad_norm": 0.46572622656822205,
+      "learning_rate": 2.2076116950697937e-07,
+      "logits/chosen": 10.310693740844727,
+      "logits/rejected": 10.41200065612793,
+      "logps/chosen": -177.54794311523438,
+      "logps/ref_chosen": -168.2965545654297,
+      "logps/ref_rejected": -176.21559143066406,
+      "logps/rejected": -185.90658569335938,
+      "loss": 0.6911,
+      "rewards/accuracies": 0.609375,
+      "rewards/chosen": -0.09251392632722855,
+      "rewards/margins": 0.004395842086523771,
+      "rewards/rejected": -0.0969097688794136,
+      "step": 44
+    },
+    {
+      "epoch": 0.5928365582544257,
+      "grad_norm": 0.4241108298301697,
+      "learning_rate": 2.091556595453674e-07,
+      "logits/chosen": 10.16834831237793,
+      "logits/rejected": 10.282001495361328,
+      "logps/chosen": -186.37091064453125,
+      "logps/ref_chosen": -177.40481567382812,
+      "logps/ref_rejected": -184.82774353027344,
+      "logps/rejected": -194.56387329101562,
+      "loss": 0.6894,
+      "rewards/accuracies": 0.5703125,
+      "rewards/chosen": -0.08966079354286194,
+      "rewards/margins": 0.007700393907725811,
+      "rewards/rejected": -0.09736119955778122,
+      "step": 45
+    },
+    {
+      "epoch": 0.606010703993413,
+      "grad_norm": 0.5031992197036743,
+      "learning_rate": 1.9763993430846392e-07,
+      "logits/chosen": 10.157776832580566,
+      "logits/rejected": 10.314593315124512,
+      "logps/chosen": -184.5338134765625,
+      "logps/ref_chosen": -176.05081176757812,
+      "logps/ref_rejected": -183.15682983398438,
+      "logps/rejected": -192.16787719726562,
+      "loss": 0.6906,
+      "rewards/accuracies": 0.53125,
+      "rewards/chosen": -0.08483003079891205,
+      "rewards/margins": 0.0052804588340222836,
+      "rewards/rejected": -0.09011048823595047,
+      "step": 46
+    },
+    {
+      "epoch": 0.6191848497324002,
+      "grad_norm": 0.5240408778190613,
+      "learning_rate": 1.862393078595809e-07,
+      "logits/chosen": 10.676742553710938,
+      "logits/rejected": 10.760393142700195,
+      "logps/chosen": -182.15650939941406,
+      "logps/ref_chosen": -172.76321411132812,
+      "logps/ref_rejected": -181.09564208984375,
+      "logps/rejected": -190.9324493408203,
+      "loss": 0.691,
+      "rewards/accuracies": 0.578125,
+      "rewards/chosen": -0.0939328670501709,
+      "rewards/margins": 0.004435211420059204,
+      "rewards/rejected": -0.0983680710196495,
+      "step": 47
+    },
+    {
+      "epoch": 0.6323589954713874,
+      "grad_norm": 0.7354983687400818,
+      "learning_rate": 1.7497884124991485e-07,
+      "logits/chosen": 10.81379508972168,
+      "logits/rejected": 10.839184761047363,
+      "logps/chosen": -183.8255157470703,
+      "logps/ref_chosen": -174.04627990722656,
+      "logps/ref_rejected": -178.9498748779297,
+      "logps/rejected": -188.9517059326172,
+      "loss": 0.6921,
+      "rewards/accuracies": 0.5078125,
+      "rewards/chosen": -0.09779238700866699,
+      "rewards/margins": 0.0022260185796767473,
+      "rewards/rejected": -0.10001839697360992,
+      "step": 48
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 75,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 12,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-60/added_tokens.json

@@ -0,0 +1,13 @@
+{
+  "<|assistant|>": 32001,
+  "<|endoftext|>": 32000,
+  "<|end|>": 32007,
+  "<|placeholder1|>": 32002,
+  "<|placeholder2|>": 32003,
+  "<|placeholder3|>": 32004,
+  "<|placeholder4|>": 32005,
+  "<|placeholder5|>": 32008,
+  "<|placeholder6|>": 32009,
+  "<|system|>": 32006,
+  "<|user|>": 32010
+}

checkpoint-60/config.json

@@ -0,0 +1,36 @@
+{
+  "_name_or_path": "/home/v-liuzichuan/huggingface/Phi-3-mini-4k-instruct",
+  "architectures": [
+    "Phi3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_phi3.Phi3Config",
+    "AutoModelForCausalLM": "modeling_phi3.Phi3ForCausalLM"
+  },
+  "bos_token_id": 1,
+  "embd_pdrop": 0.0,
+  "eos_token_id": 32000,
+  "hidden_act": "silu",
+  "hidden_size": 3072,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 4096,
+  "model_type": "phi3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "original_max_position_embeddings": 4096,
+  "pad_token_id": 32000,
+  "resid_pdrop": 0.0,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "sliding_window": 2047,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.45.2",
+  "use_cache": false,
+  "vocab_size": 32064
+}

checkpoint-60/generation_config.json

@@ -0,0 +1,11 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": [
+    32000,
+    32001,
+    32007
+  ],
+  "pad_token_id": 32000,
+  "transformers_version": "4.45.2"
+}

checkpoint-60/model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd872b829ea1f059b5b54fd96b5288c830a95d44d1ff8d84a2808a519049b2e8
+size 4972489328

checkpoint-60/model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b00adcf13869b6f62ae83406676bbc99c0ad87de6fbeffda5d8505c25ad5b538
+size 2669692552

checkpoint-60/model.safetensors.index.json

@@ -0,0 +1,202 @@
+{
+  "metadata": {
+    "total_size": 7642159104
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.22.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.23.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.24.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.25.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.26.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.27.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.qkv_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.qkv_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}

checkpoint-60/special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}