Upload folder using huggingface_hub

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "/mnt/data/tungtran/output_model/irish_llama2_data_v3/checkpoint-2200",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 5120,
+  "initializer_range": 0.02,
+  "intermediate_size": 13824,
+  "max_position_embeddings": 4096,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 40,
+  "num_hidden_layers": 40,
+  "num_key_value_heads": 40,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.3",
+  "use_cache": true,
+  "vocab_size": 35483
+}

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/generation_config.json

@@ -0,0 +1,10 @@
+{
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": 2,
+  "max_length": 4096,
+  "pad_token_id": 0,
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.46.3"
+}

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/latest

@@ -0,0 +1 @@
+global_step624

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/model-00001-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b82608b73af1630f7ec9af236bbbfe1947b0be2c00fbc0954923f672569ce0a
+size 4961502800

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/model-00002-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8a71801cee916a77a0c67ee2372e687e7110fe7340ba71b084e8978d353db54
+size 4970422232

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/model-00003-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6aaa9c6184e45f53efb72aaaa8112127b9772c7af87f98a7e1cd6d9c4ea3c08
+size 4881272584

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/model-00004-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35dfda0dd3e0ae4c7d66d066bd68333dee7b04445da805c5d32ea958c7ee87cb
+size 4933722216

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/model-00005-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:025d430bc53976d95aacaa1616ee456009fd541885bd7830cf9051ae8876e4bf
+size 4933722208

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/model-00006-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d11f72a32e119049321b1d8d9ab06c8de75936fdb0d16f1983d792d416ff592
+size 1422460712

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/model.safetensors.index.json

@@ -0,0 +1,370 @@
+{
+  "metadata": {
+    "total_size": 26103060480
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00006-of-00006.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00006.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00006.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00006.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00006.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00003-of-00006.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00006.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00006.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00004-of-00006.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00006.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00004-of-00006.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.36.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.36.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.36.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.36.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.36.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.36.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.37.mlp.down_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.37.mlp.gate_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.37.mlp.up_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00005-of-00006.safetensors",
+    "model.layers.37.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.37.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.37.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00006-of-00006.safetensors",
+    "model.layers.38.mlp.down_proj.weight": "model-00006-of-00006.safetensors",
+    "model.layers.38.mlp.gate_proj.weight": "model-00006-of-00006.safetensors",
+    "model.layers.38.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
+    "model.layers.38.self_attn.k_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.38.self_attn.q_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.38.self_attn.v_proj.weight": "model-00005-of-00006.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00006-of-00006.safetensors",
+    "model.layers.39.mlp.down_proj.weight": "model-00006-of-00006.safetensors",
+    "model.layers.39.mlp.gate_proj.weight": "model-00006-of-00006.safetensors",
+    "model.layers.39.mlp.up_proj.weight": "model-00006-of-00006.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00006-of-00006.safetensors",
+    "model.layers.39.self_attn.k_proj.weight": "model-00006-of-00006.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model-00006-of-00006.safetensors",
+    "model.layers.39.self_attn.q_proj.weight": "model-00006-of-00006.safetensors",
+    "model.layers.39.self_attn.v_proj.weight": "model-00006-of-00006.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00006.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00006.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00006.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00006.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00006.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00006.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00006.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00006.safetensors",
+    "model.norm.weight": "model-00006-of-00006.safetensors"
+  }
+}

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:994d0a15e56df433a908d139d0b7caea59dbf6eb9e109191d3df763b430de5e7
+size 15984

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:480b1090b0e942f0994b95fcb4e2d0fd8effca2892f351ff441a70d9143b06a1
+size 15984

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f47f7d835b04b9510790640491f19ac37f3c0ee7f9720eea68e19f9a59f001be
+size 15984

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:179ed45c9b836e74595a5ee6959682569c15e17748ff046457798aba5998c99b
+size 15984

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc9918d8e550835f34df7f403c41b632703a623c1cce2bc21b2e28df805b1646
+size 15984

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e604fb5f644f8e388e5522212cf891a6f957b1a30b9e4a282a72c84bf68b615f
+size 15984

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5df3cd0dacde7148beb677c7d6a1bfc9b7a20b4c8e6818da432855cfe7ef80ac
+size 15984

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00a4d1cc819f2b1045d34a95b3504b61b35cfdcc4ee5cba30b1708f4e3cfc599
+size 15984

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b1e57508d3ad0384901652c36aeb55b1e17a89f8371cd3bd79dd388951df1b8
+size 1064

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d1f5d0342153f3e3bbb37b2026ba64d0b25583df351345f87cd8b9a5658c2fb
+size 558602

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/tokenizer_config.json

@@ -0,0 +1,43 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": true,
+  "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": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% set system_message = 'You are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe.' %}{% endif %}{% for message in loop_messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if loop.index0 == 0 and system_message != false %}{% set content = '<<SYS>>\\n' + system_message + '\\n<</SYS>>\\n\\n' + message['content'] %}{% else %}{% set content = message['content'] %}{% endif %}{% if message['role'] == 'user' %}{{ bos_token + '[INST] ' + content.strip() + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ ' '  + content.strip() + ' ' + eos_token }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/trainer_state.json

@@ -0,0 +1,2224 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.9920127795527156,
+  "eval_steps": 500,
+  "global_step": 624,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.006389776357827476,
+      "grad_norm": 2.055291493195234,
+      "learning_rate": 3.125e-06,
+      "loss": 1.695,
+      "step": 1
+    },
+    {
+      "epoch": 0.012779552715654952,
+      "grad_norm": 2.0685233500522586,
+      "learning_rate": 6.25e-06,
+      "loss": 1.6748,
+      "step": 2
+    },
+    {
+      "epoch": 0.025559105431309903,
+      "grad_norm": 2.325735299422439,
+      "learning_rate": 1.25e-05,
+      "loss": 1.6964,
+      "step": 4
+    },
+    {
+      "epoch": 0.038338658146964855,
+      "grad_norm": 0.4729866673863026,
+      "learning_rate": 1.8750000000000002e-05,
+      "loss": 1.4325,
+      "step": 6
+    },
+    {
+      "epoch": 0.051118210862619806,
+      "grad_norm": 0.482620239981458,
+      "learning_rate": 2.5e-05,
+      "loss": 1.3874,
+      "step": 8
+    },
+    {
+      "epoch": 0.06389776357827476,
+      "grad_norm": 1.6728433474079003,
+      "learning_rate": 3.125e-05,
+      "loss": 1.4689,
+      "step": 10
+    },
+    {
+      "epoch": 0.07667731629392971,
+      "grad_norm": 0.3405987431283081,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 1.3127,
+      "step": 12
+    },
+    {
+      "epoch": 0.08945686900958466,
+      "grad_norm": 0.2323496464888272,
+      "learning_rate": 4.375e-05,
+      "loss": 1.2639,
+      "step": 14
+    },
+    {
+      "epoch": 0.10223642172523961,
+      "grad_norm": 0.18809974511784008,
+      "learning_rate": 5e-05,
+      "loss": 1.2401,
+      "step": 16
+    },
+    {
+      "epoch": 0.11501597444089456,
+      "grad_norm": 0.18997340619225084,
+      "learning_rate": 5.6250000000000005e-05,
+      "loss": 1.2084,
+      "step": 18
+    },
+    {
+      "epoch": 0.12779552715654952,
+      "grad_norm": 0.15504216343509883,
+      "learning_rate": 6.25e-05,
+      "loss": 1.1855,
+      "step": 20
+    },
+    {
+      "epoch": 0.14057507987220447,
+      "grad_norm": 0.12848416587626313,
+      "learning_rate": 6.875e-05,
+      "loss": 1.146,
+      "step": 22
+    },
+    {
+      "epoch": 0.15335463258785942,
+      "grad_norm": 0.09889252813730416,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 1.1357,
+      "step": 24
+    },
+    {
+      "epoch": 0.16613418530351437,
+      "grad_norm": 0.09024188902019939,
+      "learning_rate": 8.125000000000001e-05,
+      "loss": 1.1096,
+      "step": 26
+    },
+    {
+      "epoch": 0.17891373801916932,
+      "grad_norm": 0.08133676595279006,
+      "learning_rate": 8.75e-05,
+      "loss": 1.0913,
+      "step": 28
+    },
+    {
+      "epoch": 0.19169329073482427,
+      "grad_norm": 0.0978463769637292,
+      "learning_rate": 9.375e-05,
+      "loss": 1.0679,
+      "step": 30
+    },
+    {
+      "epoch": 0.20447284345047922,
+      "grad_norm": 0.07943889170723487,
+      "learning_rate": 0.0001,
+      "loss": 1.075,
+      "step": 32
+    },
+    {
+      "epoch": 0.21725239616613418,
+      "grad_norm": 0.08240884428512509,
+      "learning_rate": 9.99971838728789e-05,
+      "loss": 1.075,
+      "step": 34
+    },
+    {
+      "epoch": 0.23003194888178913,
+      "grad_norm": 0.08253986997481327,
+      "learning_rate": 9.998873580873848e-05,
+      "loss": 1.0652,
+      "step": 36
+    },
+    {
+      "epoch": 0.24281150159744408,
+      "grad_norm": 0.07954648039103362,
+      "learning_rate": 9.997465675921163e-05,
+      "loss": 1.0519,
+      "step": 38
+    },
+    {
+      "epoch": 0.25559105431309903,
+      "grad_norm": 0.0776223200815433,
+      "learning_rate": 9.995494831023409e-05,
+      "loss": 1.0094,
+      "step": 40
+    },
+    {
+      "epoch": 0.268370607028754,
+      "grad_norm": 0.08000844411167178,
+      "learning_rate": 9.992961268186573e-05,
+      "loss": 1.0074,
+      "step": 42
+    },
+    {
+      "epoch": 0.28115015974440893,
+      "grad_norm": 0.0689657212250583,
+      "learning_rate": 9.989865272804063e-05,
+      "loss": 1.0087,
+      "step": 44
+    },
+    {
+      "epoch": 0.2939297124600639,
+      "grad_norm": 0.0722150479128947,
+      "learning_rate": 9.986207193624536e-05,
+      "loss": 1.0067,
+      "step": 46
+    },
+    {
+      "epoch": 0.30670926517571884,
+      "grad_norm": 0.06646168454668608,
+      "learning_rate": 9.981987442712633e-05,
+      "loss": 0.9837,
+      "step": 48
+    },
+    {
+      "epoch": 0.3194888178913738,
+      "grad_norm": 0.06815852582234988,
+      "learning_rate": 9.977206495402554e-05,
+      "loss": 1.0024,
+      "step": 50
+    },
+    {
+      "epoch": 0.33226837060702874,
+      "grad_norm": 0.07469571057420442,
+      "learning_rate": 9.971864890244513e-05,
+      "loss": 0.9606,
+      "step": 52
+    },
+    {
+      "epoch": 0.3450479233226837,
+      "grad_norm": 0.07160841663430713,
+      "learning_rate": 9.965963228944078e-05,
+      "loss": 0.9681,
+      "step": 54
+    },
+    {
+      "epoch": 0.35782747603833864,
+      "grad_norm": 0.06954866095292117,
+      "learning_rate": 9.959502176294383e-05,
+      "loss": 0.951,
+      "step": 56
+    },
+    {
+      "epoch": 0.3706070287539936,
+      "grad_norm": 0.06598684065212063,
+      "learning_rate": 9.95248246010126e-05,
+      "loss": 0.9501,
+      "step": 58
+    },
+    {
+      "epoch": 0.38338658146964855,
+      "grad_norm": 0.12103302407814338,
+      "learning_rate": 9.944904871101228e-05,
+      "loss": 0.9713,
+      "step": 60
+    },
+    {
+      "epoch": 0.3961661341853035,
+      "grad_norm": 0.07330981053456032,
+      "learning_rate": 9.936770262872443e-05,
+      "loss": 0.9283,
+      "step": 62
+    },
+    {
+      "epoch": 0.40894568690095845,
+      "grad_norm": 0.06537535724415816,
+      "learning_rate": 9.928079551738543e-05,
+      "loss": 0.9118,
+      "step": 64
+    },
+    {
+      "epoch": 0.4217252396166134,
+      "grad_norm": 0.07457609795137939,
+      "learning_rate": 9.918833716665419e-05,
+      "loss": 0.9279,
+      "step": 66
+    },
+    {
+      "epoch": 0.43450479233226835,
+      "grad_norm": 0.07491122165043795,
+      "learning_rate": 9.909033799150946e-05,
+      "loss": 0.935,
+      "step": 68
+    },
+    {
+      "epoch": 0.4472843450479233,
+      "grad_norm": 0.06781283989008571,
+      "learning_rate": 9.898680903107666e-05,
+      "loss": 0.9361,
+      "step": 70
+    },
+    {
+      "epoch": 0.46006389776357826,
+      "grad_norm": 0.07160916695151898,
+      "learning_rate": 9.887776194738432e-05,
+      "loss": 0.9159,
+      "step": 72
+    },
+    {
+      "epoch": 0.4728434504792332,
+      "grad_norm": 0.0681941013678725,
+      "learning_rate": 9.876320902405042e-05,
+      "loss": 0.8779,
+      "step": 74
+    },
+    {
+      "epoch": 0.48562300319488816,
+      "grad_norm": 0.07482319269062407,
+      "learning_rate": 9.864316316489873e-05,
+      "loss": 0.8825,
+      "step": 76
+    },
+    {
+      "epoch": 0.4984025559105431,
+      "grad_norm": 0.08697975313543096,
+      "learning_rate": 9.851763789250525e-05,
+      "loss": 0.922,
+      "step": 78
+    },
+    {
+      "epoch": 0.5111821086261981,
+      "grad_norm": 0.09978612068745818,
+      "learning_rate": 9.838664734667495e-05,
+      "loss": 0.8894,
+      "step": 80
+    },
+    {
+      "epoch": 0.5239616613418531,
+      "grad_norm": 0.09384667638421956,
+      "learning_rate": 9.825020628284896e-05,
+      "loss": 0.8593,
+      "step": 82
+    },
+    {
+      "epoch": 0.536741214057508,
+      "grad_norm": 0.06932081799385038,
+      "learning_rate": 9.810833007044247e-05,
+      "loss": 0.8662,
+      "step": 84
+    },
+    {
+      "epoch": 0.549520766773163,
+      "grad_norm": 0.10358699944795004,
+      "learning_rate": 9.796103469111351e-05,
+      "loss": 0.8723,
+      "step": 86
+    },
+    {
+      "epoch": 0.5623003194888179,
+      "grad_norm": 0.07169243369499742,
+      "learning_rate": 9.780833673696254e-05,
+      "loss": 0.8482,
+      "step": 88
+    },
+    {
+      "epoch": 0.5750798722044729,
+      "grad_norm": 0.1050406308556227,
+      "learning_rate": 9.76502534086636e-05,
+      "loss": 0.8496,
+      "step": 90
+    },
+    {
+      "epoch": 0.5878594249201278,
+      "grad_norm": 0.07201905690967678,
+      "learning_rate": 9.74868025135266e-05,
+      "loss": 0.8291,
+      "step": 92
+    },
+    {
+      "epoch": 0.6006389776357828,
+      "grad_norm": 1.2625349021090781,
+      "learning_rate": 9.731800246349148e-05,
+      "loss": 0.8503,
+      "step": 94
+    },
+    {
+      "epoch": 0.6134185303514377,
+      "grad_norm": 0.17981258022070712,
+      "learning_rate": 9.714387227305422e-05,
+      "loss": 0.8231,
+      "step": 96
+    },
+    {
+      "epoch": 0.6261980830670927,
+      "grad_norm": 0.07561478832740967,
+      "learning_rate": 9.696443155712486e-05,
+      "loss": 0.8119,
+      "step": 98
+    },
+    {
+      "epoch": 0.6389776357827476,
+      "grad_norm": 0.08195686915168865,
+      "learning_rate": 9.67797005288181e-05,
+      "loss": 0.7926,
+      "step": 100
+    },
+    {
+      "epoch": 0.6517571884984026,
+      "grad_norm": 0.0890476280007116,
+      "learning_rate": 9.65896999971763e-05,
+      "loss": 0.8039,
+      "step": 102
+    },
+    {
+      "epoch": 0.6645367412140575,
+      "grad_norm": 0.07738578891457887,
+      "learning_rate": 9.639445136482548e-05,
+      "loss": 0.7721,
+      "step": 104
+    },
+    {
+      "epoch": 0.6773162939297125,
+      "grad_norm": 0.0743037172920425,
+      "learning_rate": 9.619397662556435e-05,
+      "loss": 0.794,
+      "step": 106
+    },
+    {
+      "epoch": 0.6900958466453674,
+      "grad_norm": 0.08803835897602165,
+      "learning_rate": 9.598829836188694e-05,
+      "loss": 0.7721,
+      "step": 108
+    },
+    {
+      "epoch": 0.7028753993610224,
+      "grad_norm": 0.07702819696223887,
+      "learning_rate": 9.577743974243874e-05,
+      "loss": 0.7765,
+      "step": 110
+    },
+    {
+      "epoch": 0.7156549520766773,
+      "grad_norm": 0.07473535070111323,
+      "learning_rate": 9.55614245194068e-05,
+      "loss": 0.7598,
+      "step": 112
+    },
+    {
+      "epoch": 0.7284345047923323,
+      "grad_norm": 0.08433756541496004,
+      "learning_rate": 9.534027702584425e-05,
+      "loss": 0.7727,
+      "step": 114
+    },
+    {
+      "epoch": 0.7412140575079872,
+      "grad_norm": 0.07483257658817612,
+      "learning_rate": 9.511402217292926e-05,
+      "loss": 0.7465,
+      "step": 116
+    },
+    {
+      "epoch": 0.7539936102236422,
+      "grad_norm": 0.0880318685591304,
+      "learning_rate": 9.488268544715896e-05,
+      "loss": 0.7321,
+      "step": 118
+    },
+    {
+      "epoch": 0.7667731629392971,
+      "grad_norm": 0.07719604899450865,
+      "learning_rate": 9.464629290747842e-05,
+      "loss": 0.7624,
+      "step": 120
+    },
+    {
+      "epoch": 0.7795527156549521,
+      "grad_norm": 0.0733176421376437,
+      "learning_rate": 9.440487118234535e-05,
+      "loss": 0.6975,
+      "step": 122
+    },
+    {
+      "epoch": 0.792332268370607,
+      "grad_norm": 0.07051701385784455,
+      "learning_rate": 9.415844746673047e-05,
+      "loss": 0.7127,
+      "step": 124
+    },
+    {
+      "epoch": 0.805111821086262,
+      "grad_norm": 0.0729787562181977,
+      "learning_rate": 9.390704951905411e-05,
+      "loss": 0.7503,
+      "step": 126
+    },
+    {
+      "epoch": 0.8178913738019169,
+      "grad_norm": 0.07128874732953779,
+      "learning_rate": 9.365070565805941e-05,
+      "loss": 0.6941,
+      "step": 128
+    },
+    {
+      "epoch": 0.8306709265175719,
+      "grad_norm": 0.07804844381711577,
+      "learning_rate": 9.338944475962237e-05,
+      "loss": 0.7197,
+      "step": 130
+    },
+    {
+      "epoch": 0.8434504792332268,
+      "grad_norm": 0.08207580744924538,
+      "learning_rate": 9.312329625349902e-05,
+      "loss": 0.7134,
+      "step": 132
+    },
+    {
+      "epoch": 0.8562300319488818,
+      "grad_norm": 0.10268159904999394,
+      "learning_rate": 9.285229012001047e-05,
+      "loss": 0.705,
+      "step": 134
+    },
+    {
+      "epoch": 0.8690095846645367,
+      "grad_norm": 0.07097527094154266,
+      "learning_rate": 9.257645688666556e-05,
+      "loss": 0.7036,
+      "step": 136
+    },
+    {
+      "epoch": 0.8817891373801917,
+      "grad_norm": 0.07284443178958877,
+      "learning_rate": 9.22958276247223e-05,
+      "loss": 0.7313,
+      "step": 138
+    },
+    {
+      "epoch": 0.8945686900958466,
+      "grad_norm": 0.07294697279525543,
+      "learning_rate": 9.201043394568773e-05,
+      "loss": 0.6847,
+      "step": 140
+    },
+    {
+      "epoch": 0.9073482428115016,
+      "grad_norm": 0.0725032039002937,
+      "learning_rate": 9.172030799775699e-05,
+      "loss": 0.6877,
+      "step": 142
+    },
+    {
+      "epoch": 0.9201277955271565,
+      "grad_norm": 0.06708836437156662,
+      "learning_rate": 9.142548246219212e-05,
+      "loss": 0.6837,
+      "step": 144
+    },
+    {
+      "epoch": 0.9329073482428115,
+      "grad_norm": 0.07361178534656698,
+      "learning_rate": 9.112599054964057e-05,
+      "loss": 0.6522,
+      "step": 146
+    },
+    {
+      "epoch": 0.9456869009584664,
+      "grad_norm": 0.06961060997060975,
+      "learning_rate": 9.082186599639428e-05,
+      "loss": 0.6732,
+      "step": 148
+    },
+    {
+      "epoch": 0.9584664536741214,
+      "grad_norm": 0.06369267112915664,
+      "learning_rate": 9.051314306058933e-05,
+      "loss": 0.6615,
+      "step": 150
+    },
+    {
+      "epoch": 0.9712460063897763,
+      "grad_norm": 0.06667729772792583,
+      "learning_rate": 9.019985651834703e-05,
+      "loss": 0.6742,
+      "step": 152
+    },
+    {
+      "epoch": 0.9840255591054313,
+      "grad_norm": 0.07052786453330319,
+      "learning_rate": 8.988204165985649e-05,
+      "loss": 0.6365,
+      "step": 154
+    },
+    {
+      "epoch": 0.9968051118210862,
+      "grad_norm": 0.06352217971127558,
+      "learning_rate": 8.955973428539944e-05,
+      "loss": 0.6531,
+      "step": 156
+    },
+    {
+      "epoch": 1.011182108626198,
+      "grad_norm": 0.0907023898699884,
+      "learning_rate": 8.923297070131737e-05,
+      "loss": 0.6986,
+      "step": 158
+    },
+    {
+      "epoch": 1.023961661341853,
+      "grad_norm": 0.06588723514264389,
+      "learning_rate": 8.890178771592199e-05,
+      "loss": 0.4221,
+      "step": 160
+    },
+    {
+      "epoch": 1.036741214057508,
+      "grad_norm": 0.07457104912562523,
+      "learning_rate": 8.856622263534875e-05,
+      "loss": 0.4375,
+      "step": 162
+    },
+    {
+      "epoch": 1.049520766773163,
+      "grad_norm": 0.08716030746078077,
+      "learning_rate": 8.822631325935463e-05,
+      "loss": 0.4633,
+      "step": 164
+    },
+    {
+      "epoch": 1.0623003194888179,
+      "grad_norm": 0.07564657660605784,
+      "learning_rate": 8.788209787706015e-05,
+      "loss": 0.4149,
+      "step": 166
+    },
+    {
+      "epoch": 1.0750798722044728,
+      "grad_norm": 0.2601478494309565,
+      "learning_rate": 8.753361526263621e-05,
+      "loss": 0.4644,
+      "step": 168
+    },
+    {
+      "epoch": 1.0878594249201279,
+      "grad_norm": 0.07236244361689621,
+      "learning_rate": 8.718090467093654e-05,
+      "loss": 0.445,
+      "step": 170
+    },
+    {
+      "epoch": 1.1006389776357828,
+      "grad_norm": 0.07360308087849284,
+      "learning_rate": 8.682400583307562e-05,
+      "loss": 0.4189,
+      "step": 172
+    },
+    {
+      "epoch": 1.1134185303514377,
+      "grad_norm": 0.06934965586236702,
+      "learning_rate": 8.646295895195333e-05,
+      "loss": 0.4168,
+      "step": 174
+    },
+    {
+      "epoch": 1.1261980830670926,
+      "grad_norm": 0.06652725595095291,
+      "learning_rate": 8.609780469772623e-05,
+      "loss": 0.4332,
+      "step": 176
+    },
+    {
+      "epoch": 1.1389776357827477,
+      "grad_norm": 0.06493423808775205,
+      "learning_rate": 8.572858420322627e-05,
+      "loss": 0.4126,
+      "step": 178
+    },
+    {
+      "epoch": 1.1517571884984026,
+      "grad_norm": 0.07224306242862681,
+      "learning_rate": 8.535533905932738e-05,
+      "loss": 0.4639,
+      "step": 180
+    },
+    {
+      "epoch": 1.1645367412140575,
+      "grad_norm": 0.06325420247080109,
+      "learning_rate": 8.497811131026046e-05,
+      "loss": 0.4097,
+      "step": 182
+    },
+    {
+      "epoch": 1.1773162939297124,
+      "grad_norm": 0.05960690196531746,
+      "learning_rate": 8.459694344887732e-05,
+      "loss": 0.4258,
+      "step": 184
+    },
+    {
+      "epoch": 1.1900958466453675,
+      "grad_norm": 0.06526403248406679,
+      "learning_rate": 8.421187841186402e-05,
+      "loss": 0.4453,
+      "step": 186
+    },
+    {
+      "epoch": 1.2028753993610224,
+      "grad_norm": 0.06754636177295095,
+      "learning_rate": 8.382295957490436e-05,
+      "loss": 0.4277,
+      "step": 188
+    },
+    {
+      "epoch": 1.2156549520766773,
+      "grad_norm": 0.11883404710840821,
+      "learning_rate": 8.343023074779368e-05,
+      "loss": 0.4386,
+      "step": 190
+    },
+    {
+      "epoch": 1.2284345047923322,
+      "grad_norm": 0.07793571463351197,
+      "learning_rate": 8.303373616950408e-05,
+      "loss": 0.4072,
+      "step": 192
+    },
+    {
+      "epoch": 1.2412140575079873,
+      "grad_norm": 0.06518657342856102,
+      "learning_rate": 8.263352050320094e-05,
+      "loss": 0.4396,
+      "step": 194
+    },
+    {
+      "epoch": 1.2539936102236422,
+      "grad_norm": 0.05974282037032855,
+      "learning_rate": 8.222962883121196e-05,
+      "loss": 0.4016,
+      "step": 196
+    },
+    {
+      "epoch": 1.266773162939297,
+      "grad_norm": 0.0693639502217822,
+      "learning_rate": 8.182210664994878e-05,
+      "loss": 0.3808,
+      "step": 198
+    },
+    {
+      "epoch": 1.279552715654952,
+      "grad_norm": 0.06127831754623801,
+      "learning_rate": 8.141099986478212e-05,
+      "loss": 0.3961,
+      "step": 200
+    },
+    {
+      "epoch": 1.292332268370607,
+      "grad_norm": 0.06755312065722066,
+      "learning_rate": 8.099635478487064e-05,
+      "loss": 0.3894,
+      "step": 202
+    },
+    {
+      "epoch": 1.305111821086262,
+      "grad_norm": 0.0584212869146413,
+      "learning_rate": 8.057821811794458e-05,
+      "loss": 0.414,
+      "step": 204
+    },
+    {
+      "epoch": 1.317891373801917,
+      "grad_norm": 0.05983512956529008,
+      "learning_rate": 8.015663696504422e-05,
+      "loss": 0.3634,
+      "step": 206
+    },
+    {
+      "epoch": 1.330670926517572,
+      "grad_norm": 0.05778218969166584,
+      "learning_rate": 7.973165881521434e-05,
+      "loss": 0.4233,
+      "step": 208
+    },
+    {
+      "epoch": 1.343450479233227,
+      "grad_norm": 0.058310021079803646,
+      "learning_rate": 7.930333154015466e-05,
+      "loss": 0.4061,
+      "step": 210
+    },
+    {
+      "epoch": 1.3562300319488818,
+      "grad_norm": 0.0642143238679532,
+      "learning_rate": 7.88717033888274e-05,
+      "loss": 0.4083,
+      "step": 212
+    },
+    {
+      "epoch": 1.3690095846645367,
+      "grad_norm": 0.05656381877721736,
+      "learning_rate": 7.843682298202235e-05,
+      "loss": 0.4033,
+      "step": 214
+    },
+    {
+      "epoch": 1.3817891373801916,
+      "grad_norm": 0.05518190162844295,
+      "learning_rate": 7.799873930687978e-05,
+      "loss": 0.3953,
+      "step": 216
+    },
+    {
+      "epoch": 1.3945686900958467,
+      "grad_norm": 0.05903661851778338,
+      "learning_rate": 7.755750171137246e-05,
+      "loss": 0.4096,
+      "step": 218
+    },
+    {
+      "epoch": 1.4073482428115016,
+      "grad_norm": 0.05833074145436464,
+      "learning_rate": 7.711315989874677e-05,
+      "loss": 0.4151,
+      "step": 220
+    },
+    {
+      "epoch": 1.4201277955271565,
+      "grad_norm": 0.05919878363690307,
+      "learning_rate": 7.666576392192389e-05,
+      "loss": 0.39,
+      "step": 222
+    },
+    {
+      "epoch": 1.4329073482428116,
+      "grad_norm": 0.05913664327254173,
+      "learning_rate": 7.621536417786159e-05,
+      "loss": 0.4005,
+      "step": 224
+    },
+    {
+      "epoch": 1.4456869009584665,
+      "grad_norm": 0.0640842931075253,
+      "learning_rate": 7.576201140187727e-05,
+      "loss": 0.4165,
+      "step": 226
+    },
+    {
+      "epoch": 1.4584664536741214,
+      "grad_norm": 0.062131879810909965,
+      "learning_rate": 7.530575666193283e-05,
+      "loss": 0.3891,
+      "step": 228
+    },
+    {
+      "epoch": 1.4712460063897763,
+      "grad_norm": 0.06992276137309804,
+      "learning_rate": 7.484665135288213e-05,
+      "loss": 0.3971,
+      "step": 230
+    },
+    {
+      "epoch": 1.4840255591054312,
+      "grad_norm": 0.06078790664861669,
+      "learning_rate": 7.438474719068173e-05,
+      "loss": 0.3961,
+      "step": 232
+    },
+    {
+      "epoch": 1.4968051118210863,
+      "grad_norm": 0.06922648734675908,
+      "learning_rate": 7.392009620656513e-05,
+      "loss": 0.4331,
+      "step": 234
+    },
+    {
+      "epoch": 1.5095846645367412,
+      "grad_norm": 0.05766139832102871,
+      "learning_rate": 7.345275074118185e-05,
+      "loss": 0.4182,
+      "step": 236
+    },
+    {
+      "epoch": 1.5223642172523961,
+      "grad_norm": 0.06292873231888371,
+      "learning_rate": 7.298276343870151e-05,
+      "loss": 0.4061,
+      "step": 238
+    },
+    {
+      "epoch": 1.5351437699680512,
+      "grad_norm": 0.06000860844713537,
+      "learning_rate": 7.251018724088367e-05,
+      "loss": 0.4023,
+      "step": 240
+    },
+    {
+      "epoch": 1.547923322683706,
+      "grad_norm": 0.0585777107714916,
+      "learning_rate": 7.203507538111423e-05,
+      "loss": 0.3855,
+      "step": 242
+    },
+    {
+      "epoch": 1.560702875399361,
+      "grad_norm": 0.0571671995255021,
+      "learning_rate": 7.155748137840892e-05,
+      "loss": 0.3951,
+      "step": 244
+    },
+    {
+      "epoch": 1.573482428115016,
+      "grad_norm": 0.053447175708899994,
+      "learning_rate": 7.107745903138472e-05,
+      "loss": 0.3745,
+      "step": 246
+    },
+    {
+      "epoch": 1.5862619808306708,
+      "grad_norm": 0.055736902711725635,
+      "learning_rate": 7.059506241219965e-05,
+      "loss": 0.3911,
+      "step": 248
+    },
+    {
+      "epoch": 1.599041533546326,
+      "grad_norm": 0.05715355824554817,
+      "learning_rate": 7.011034586046176e-05,
+      "loss": 0.4043,
+      "step": 250
+    },
+    {
+      "epoch": 1.6118210862619808,
+      "grad_norm": 0.06030447320081754,
+      "learning_rate": 6.962336397710819e-05,
+      "loss": 0.3899,
+      "step": 252
+    },
+    {
+      "epoch": 1.6246006389776357,
+      "grad_norm": 0.061239135474291606,
+      "learning_rate": 6.91341716182545e-05,
+      "loss": 0.4246,
+      "step": 254
+    },
+    {
+      "epoch": 1.6373801916932909,
+      "grad_norm": 0.05695235071864785,
+      "learning_rate": 6.864282388901544e-05,
+      "loss": 0.3953,
+      "step": 256
+    },
+    {
+      "epoch": 1.6501597444089455,
+      "grad_norm": 0.05308868251491366,
+      "learning_rate": 6.814937613729766e-05,
+      "loss": 0.4103,
+      "step": 258
+    },
+    {
+      "epoch": 1.6629392971246006,
+      "grad_norm": 0.054046791633493914,
+      "learning_rate": 6.765388394756504e-05,
+      "loss": 0.4059,
+      "step": 260
+    },
+    {
+      "epoch": 1.6757188498402555,
+      "grad_norm": 0.05148697040730548,
+      "learning_rate": 6.715640313457733e-05,
+      "loss": 0.3767,
+      "step": 262
+    },
+    {
+      "epoch": 1.6884984025559104,
+      "grad_norm": 0.05318569591896447,
+      "learning_rate": 6.665698973710288e-05,
+      "loss": 0.3708,
+      "step": 264
+    },
+    {
+      "epoch": 1.7012779552715656,
+      "grad_norm": 0.05196719070381999,
+      "learning_rate": 6.615570001160626e-05,
+      "loss": 0.4042,
+      "step": 266
+    },
+    {
+      "epoch": 1.7140575079872205,
+      "grad_norm": 0.05632881769869459,
+      "learning_rate": 6.565259042591113e-05,
+      "loss": 0.3987,
+      "step": 268
+    },
+    {
+      "epoch": 1.7268370607028753,
+      "grad_norm": 0.05470059818193366,
+      "learning_rate": 6.514771765283942e-05,
+      "loss": 0.3973,
+      "step": 270
+    },
+    {
+      "epoch": 1.7396166134185305,
+      "grad_norm": 0.056351811449582394,
+      "learning_rate": 6.464113856382752e-05,
+      "loss": 0.3864,
+      "step": 272
+    },
+    {
+      "epoch": 1.7523961661341851,
+      "grad_norm": 0.05831258279981057,
+      "learning_rate": 6.413291022251989e-05,
+      "loss": 0.4041,
+      "step": 274
+    },
+    {
+      "epoch": 1.7651757188498403,
+      "grad_norm": 0.053467450310740065,
+      "learning_rate": 6.362308987834115e-05,
+      "loss": 0.3814,
+      "step": 276
+    },
+    {
+      "epoch": 1.7779552715654952,
+      "grad_norm": 0.051287152623381335,
+      "learning_rate": 6.311173496004723e-05,
+      "loss": 0.395,
+      "step": 278
+    },
+    {
+      "epoch": 1.79073482428115,
+      "grad_norm": 0.05429714498773308,
+      "learning_rate": 6.259890306925627e-05,
+      "loss": 0.3821,
+      "step": 280
+    },
+    {
+      "epoch": 1.8035143769968052,
+      "grad_norm": 0.057523653580626326,
+      "learning_rate": 6.208465197396013e-05,
+      "loss": 0.3984,
+      "step": 282
+    },
+    {
+      "epoch": 1.81629392971246,
+      "grad_norm": 0.05724842136937287,
+      "learning_rate": 6.156903960201709e-05,
+      "loss": 0.4181,
+      "step": 284
+    },
+    {
+      "epoch": 1.829073482428115,
+      "grad_norm": 0.052227309043480996,
+      "learning_rate": 6.105212403462651e-05,
+      "loss": 0.4049,
+      "step": 286
+    },
+    {
+      "epoch": 1.84185303514377,
+      "grad_norm": 0.04967908325326877,
+      "learning_rate": 6.0533963499786314e-05,
+      "loss": 0.4117,
+      "step": 288
+    },
+    {
+      "epoch": 1.854632587859425,
+      "grad_norm": 0.05539898234566285,
+      "learning_rate": 6.001461636573397e-05,
+      "loss": 0.4006,
+      "step": 290
+    },
+    {
+      "epoch": 1.8674121405750799,
+      "grad_norm": 0.05795414669880149,
+      "learning_rate": 5.949414113437142e-05,
+      "loss": 0.386,
+      "step": 292
+    },
+    {
+      "epoch": 1.880191693290735,
+      "grad_norm": 0.050446841270231885,
+      "learning_rate": 5.897259643467527e-05,
+      "loss": 0.3842,
+      "step": 294
+    },
+    {
+      "epoch": 1.8929712460063897,
+      "grad_norm": 0.052453051506198604,
+      "learning_rate": 5.8450041016092464e-05,
+      "loss": 0.3525,
+      "step": 296
+    },
+    {
+      "epoch": 1.9057507987220448,
+      "grad_norm": 0.052803823491155276,
+      "learning_rate": 5.792653374192245e-05,
+      "loss": 0.3963,
+      "step": 298
+    },
+    {
+      "epoch": 1.9185303514376997,
+      "grad_norm": 0.05180901601155745,
+      "learning_rate": 5.7402133582686576e-05,
+      "loss": 0.3798,
+      "step": 300
+    },
+    {
+      "epoch": 1.9313099041533546,
+      "grad_norm": 0.05166645429890597,
+      "learning_rate": 5.6876899609485256e-05,
+      "loss": 0.3838,
+      "step": 302
+    },
+    {
+      "epoch": 1.9440894568690097,
+      "grad_norm": 0.05306354741968808,
+      "learning_rate": 5.6350890987343944e-05,
+      "loss": 0.4165,
+      "step": 304
+    },
+    {
+      "epoch": 1.9568690095846646,
+      "grad_norm": 0.0860975722690725,
+      "learning_rate": 5.582416696854853e-05,
+      "loss": 0.3737,
+      "step": 306
+    },
+    {
+      "epoch": 1.9696485623003195,
+      "grad_norm": 0.05323286133666828,
+      "learning_rate": 5.5296786885970805e-05,
+      "loss": 0.3889,
+      "step": 308
+    },
+    {
+      "epoch": 1.9824281150159746,
+      "grad_norm": 0.05299665331057226,
+      "learning_rate": 5.476881014638491e-05,
+      "loss": 0.3896,
+      "step": 310
+    },
+    {
+      "epoch": 1.9952076677316293,
+      "grad_norm": 0.05157945275339266,
+      "learning_rate": 5.4240296223775465e-05,
+      "loss": 0.3637,
+      "step": 312
+    },
+    {
+      "epoch": 2.009584664536741,
+      "grad_norm": 0.09139947660133817,
+      "learning_rate": 5.3711304652638126e-05,
+      "loss": 0.3775,
+      "step": 314
+    },
+    {
+      "epoch": 2.022364217252396,
+      "grad_norm": 0.10130414532724454,
+      "learning_rate": 5.318189502127332e-05,
+      "loss": 0.2112,
+      "step": 316
+    },
+    {
+      "epoch": 2.0351437699680512,
+      "grad_norm": 0.0633333619180165,
+      "learning_rate": 5.265212696507387e-05,
+      "loss": 0.2004,
+      "step": 318
+    },
+    {
+      "epoch": 2.047923322683706,
+      "grad_norm": 0.0668276114954086,
+      "learning_rate": 5.212206015980742e-05,
+      "loss": 0.2019,
+      "step": 320
+    },
+    {
+      "epoch": 2.060702875399361,
+      "grad_norm": 0.05942503367303514,
+      "learning_rate": 5.159175431489424e-05,
+      "loss": 0.1978,
+      "step": 322
+    },
+    {
+      "epoch": 2.073482428115016,
+      "grad_norm": 0.07284145764738766,
+      "learning_rate": 5.1061269166681183e-05,
+      "loss": 0.1935,
+      "step": 324
+    },
+    {
+      "epoch": 2.086261980830671,
+      "grad_norm": 0.052260140697323494,
+      "learning_rate": 5.053066447171282e-05,
+      "loss": 0.1854,
+      "step": 326
+    },
+    {
+      "epoch": 2.099041533546326,
+      "grad_norm": 0.05754923159453662,
+      "learning_rate": 5e-05,
+      "loss": 0.1965,
+      "step": 328
+    },
+    {
+      "epoch": 2.1118210862619806,
+      "grad_norm": 0.05500397186780569,
+      "learning_rate": 4.94693355282872e-05,
+      "loss": 0.1827,
+      "step": 330
+    },
+    {
+      "epoch": 2.1246006389776357,
+      "grad_norm": 0.061606661346763424,
+      "learning_rate": 4.893873083331882e-05,
+      "loss": 0.2008,
+      "step": 332
+    },
+    {
+      "epoch": 2.137380191693291,
+      "grad_norm": 0.05678242709297541,
+      "learning_rate": 4.840824568510579e-05,
+      "loss": 0.1853,
+      "step": 334
+    },
+    {
+      "epoch": 2.1501597444089455,
+      "grad_norm": 0.054080318070508115,
+      "learning_rate": 4.78779398401926e-05,
+      "loss": 0.1952,
+      "step": 336
+    },
+    {
+      "epoch": 2.1629392971246006,
+      "grad_norm": 0.057204881343756786,
+      "learning_rate": 4.734787303492615e-05,
+      "loss": 0.1778,
+      "step": 338
+    },
+    {
+      "epoch": 2.1757188498402558,
+      "grad_norm": 0.6941487667655994,
+      "learning_rate": 4.6818104978726685e-05,
+      "loss": 0.219,
+      "step": 340
+    },
+    {
+      "epoch": 2.1884984025559104,
+      "grad_norm": 0.06999590590614403,
+      "learning_rate": 4.628869534736187e-05,
+      "loss": 0.181,
+      "step": 342
+    },
+    {
+      "epoch": 2.2012779552715656,
+      "grad_norm": 0.07558854262088241,
+      "learning_rate": 4.575970377622456e-05,
+      "loss": 0.2349,
+      "step": 344
+    },
+    {
+      "epoch": 2.2140575079872207,
+      "grad_norm": 0.07120027160683609,
+      "learning_rate": 4.52311898536151e-05,
+      "loss": 0.1993,
+      "step": 346
+    },
+    {
+      "epoch": 2.2268370607028753,
+      "grad_norm": 0.05697032990090494,
+      "learning_rate": 4.47032131140292e-05,
+      "loss": 0.1739,
+      "step": 348
+    },
+    {
+      "epoch": 2.2396166134185305,
+      "grad_norm": 0.06092977319118132,
+      "learning_rate": 4.4175833031451473e-05,
+      "loss": 0.188,
+      "step": 350
+    },
+    {
+      "epoch": 2.252396166134185,
+      "grad_norm": 0.05900721095602371,
+      "learning_rate": 4.364910901265606e-05,
+      "loss": 0.1778,
+      "step": 352
+    },
+    {
+      "epoch": 2.2651757188498403,
+      "grad_norm": 0.08992850669862418,
+      "learning_rate": 4.3123100390514756e-05,
+      "loss": 0.1838,
+      "step": 354
+    },
+    {
+      "epoch": 2.2779552715654954,
+      "grad_norm": 0.059213794143429914,
+      "learning_rate": 4.2597866417313436e-05,
+      "loss": 0.1902,
+      "step": 356
+    },
+    {
+      "epoch": 2.29073482428115,
+      "grad_norm": 0.051525349318871976,
+      "learning_rate": 4.207346625807756e-05,
+      "loss": 0.1784,
+      "step": 358
+    },
+    {
+      "epoch": 2.303514376996805,
+      "grad_norm": 0.055922862481655594,
+      "learning_rate": 4.1549958983907555e-05,
+      "loss": 0.1827,
+      "step": 360
+    },
+    {
+      "epoch": 2.31629392971246,
+      "grad_norm": 0.054189632126131766,
+      "learning_rate": 4.102740356532473e-05,
+      "loss": 0.186,
+      "step": 362
+    },
+    {
+      "epoch": 2.329073482428115,
+      "grad_norm": 0.06298745746452741,
+      "learning_rate": 4.050585886562858e-05,
+      "loss": 0.1854,
+      "step": 364
+    },
+    {
+      "epoch": 2.34185303514377,
+      "grad_norm": 0.06476475169367538,
+      "learning_rate": 3.998538363426605e-05,
+      "loss": 0.1794,
+      "step": 366
+    },
+    {
+      "epoch": 2.3546325878594248,
+      "grad_norm": 0.05187178001518817,
+      "learning_rate": 3.94660365002137e-05,
+      "loss": 0.1817,
+      "step": 368
+    },
+    {
+      "epoch": 2.36741214057508,
+      "grad_norm": 0.05110076217610542,
+      "learning_rate": 3.894787596537352e-05,
+      "loss": 0.1757,
+      "step": 370
+    },
+    {
+      "epoch": 2.380191693290735,
+      "grad_norm": 0.061027606854849537,
+      "learning_rate": 3.843096039798293e-05,
+      "loss": 0.1888,
+      "step": 372
+    },
+    {
+      "epoch": 2.3929712460063897,
+      "grad_norm": 0.05689282057128392,
+      "learning_rate": 3.791534802603988e-05,
+      "loss": 0.1972,
+      "step": 374
+    },
+    {
+      "epoch": 2.405750798722045,
+      "grad_norm": 0.05144327012401281,
+      "learning_rate": 3.740109693074375e-05,
+      "loss": 0.1975,
+      "step": 376
+    },
+    {
+      "epoch": 2.4185303514377,
+      "grad_norm": 0.07243681779987425,
+      "learning_rate": 3.68882650399528e-05,
+      "loss": 0.1865,
+      "step": 378
+    },
+    {
+      "epoch": 2.4313099041533546,
+      "grad_norm": 0.11601839655528177,
+      "learning_rate": 3.637691012165886e-05,
+      "loss": 0.1977,
+      "step": 380
+    },
+    {
+      "epoch": 2.4440894568690097,
+      "grad_norm": 0.05323975029748036,
+      "learning_rate": 3.586708977748012e-05,
+      "loss": 0.1873,
+      "step": 382
+    },
+    {
+      "epoch": 2.4568690095846644,
+      "grad_norm": 0.0499469664737551,
+      "learning_rate": 3.5358861436172485e-05,
+      "loss": 0.1832,
+      "step": 384
+    },
+    {
+      "epoch": 2.4696485623003195,
+      "grad_norm": 0.05043024991533826,
+      "learning_rate": 3.485228234716058e-05,
+      "loss": 0.1821,
+      "step": 386
+    },
+    {
+      "epoch": 2.4824281150159746,
+      "grad_norm": 0.054685112352780986,
+      "learning_rate": 3.434740957408889e-05,
+      "loss": 0.1816,
+      "step": 388
+    },
+    {
+      "epoch": 2.4952076677316293,
+      "grad_norm": 0.057237969167094144,
+      "learning_rate": 3.3844299988393755e-05,
+      "loss": 0.1909,
+      "step": 390
+    },
+    {
+      "epoch": 2.5079872204472844,
+      "grad_norm": 0.05134273506646416,
+      "learning_rate": 3.334301026289712e-05,
+      "loss": 0.1782,
+      "step": 392
+    },
+    {
+      "epoch": 2.520766773162939,
+      "grad_norm": 0.049993934417102925,
+      "learning_rate": 3.284359686542269e-05,
+      "loss": 0.1928,
+      "step": 394
+    },
+    {
+      "epoch": 2.533546325878594,
+      "grad_norm": 0.06457823051474779,
+      "learning_rate": 3.234611605243496e-05,
+      "loss": 0.196,
+      "step": 396
+    },
+    {
+      "epoch": 2.5463258785942493,
+      "grad_norm": 0.051805062617152425,
+      "learning_rate": 3.1850623862702344e-05,
+      "loss": 0.1881,
+      "step": 398
+    },
+    {
+      "epoch": 2.559105431309904,
+      "grad_norm": 0.049188541484928724,
+      "learning_rate": 3.135717611098458e-05,
+      "loss": 0.1806,
+      "step": 400
+    },
+    {
+      "epoch": 2.571884984025559,
+      "grad_norm": 0.05687592017078177,
+      "learning_rate": 3.086582838174551e-05,
+      "loss": 0.1784,
+      "step": 402
+    },
+    {
+      "epoch": 2.584664536741214,
+      "grad_norm": 0.05098573657706369,
+      "learning_rate": 3.0376636022891812e-05,
+      "loss": 0.1932,
+      "step": 404
+    },
+    {
+      "epoch": 2.597444089456869,
+      "grad_norm": 0.052376381772842893,
+      "learning_rate": 2.9889654139538246e-05,
+      "loss": 0.1889,
+      "step": 406
+    },
+    {
+      "epoch": 2.610223642172524,
+      "grad_norm": 0.05031660077056393,
+      "learning_rate": 2.9404937587800375e-05,
+      "loss": 0.1769,
+      "step": 408
+    },
+    {
+      "epoch": 2.623003194888179,
+      "grad_norm": 0.04930354808056054,
+      "learning_rate": 2.8922540968615286e-05,
+      "loss": 0.1685,
+      "step": 410
+    },
+    {
+      "epoch": 2.635782747603834,
+      "grad_norm": 0.06709139465230578,
+      "learning_rate": 2.8442518621591086e-05,
+      "loss": 0.1785,
+      "step": 412
+    },
+    {
+      "epoch": 2.648562300319489,
+      "grad_norm": 0.0503489735828908,
+      "learning_rate": 2.7964924618885778e-05,
+      "loss": 0.1689,
+      "step": 414
+    },
+    {
+      "epoch": 2.661341853035144,
+      "grad_norm": 0.05047783892143097,
+      "learning_rate": 2.748981275911633e-05,
+      "loss": 0.1808,
+      "step": 416
+    },
+    {
+      "epoch": 2.6741214057507987,
+      "grad_norm": 0.04955419672921838,
+      "learning_rate": 2.701723656129851e-05,
+      "loss": 0.1727,
+      "step": 418
+    },
+    {
+      "epoch": 2.686900958466454,
+      "grad_norm": 0.04769759775271665,
+      "learning_rate": 2.6547249258818164e-05,
+      "loss": 0.1708,
+      "step": 420
+    },
+    {
+      "epoch": 2.6996805111821085,
+      "grad_norm": 0.050324696099913684,
+      "learning_rate": 2.607990379343489e-05,
+      "loss": 0.1817,
+      "step": 422
+    },
+    {
+      "epoch": 2.7124600638977636,
+      "grad_norm": 0.05249496210993974,
+      "learning_rate": 2.5615252809318284e-05,
+      "loss": 0.1836,
+      "step": 424
+    },
+    {
+      "epoch": 2.7252396166134183,
+      "grad_norm": 0.0472378188955872,
+      "learning_rate": 2.5153348647117857e-05,
+      "loss": 0.1736,
+      "step": 426
+    },
+    {
+      "epoch": 2.7380191693290734,
+      "grad_norm": 0.049243154928981264,
+      "learning_rate": 2.469424333806718e-05,
+      "loss": 0.1675,
+      "step": 428
+    },
+    {
+      "epoch": 2.7507987220447285,
+      "grad_norm": 0.05096273109137321,
+      "learning_rate": 2.4237988598122752e-05,
+      "loss": 0.1658,
+      "step": 430
+    },
+    {
+      "epoch": 2.763578274760383,
+      "grad_norm": 0.0514806212844811,
+      "learning_rate": 2.3784635822138424e-05,
+      "loss": 0.1922,
+      "step": 432
+    },
+    {
+      "epoch": 2.7763578274760383,
+      "grad_norm": 0.05006269553229606,
+      "learning_rate": 2.333423607807613e-05,
+      "loss": 0.1887,
+      "step": 434
+    },
+    {
+      "epoch": 2.7891373801916934,
+      "grad_norm": 0.04935551516167026,
+      "learning_rate": 2.288684010125325e-05,
+      "loss": 0.1763,
+      "step": 436
+    },
+    {
+      "epoch": 2.801916932907348,
+      "grad_norm": 0.05353903496894845,
+      "learning_rate": 2.2442498288627556e-05,
+      "loss": 0.1944,
+      "step": 438
+    },
+    {
+      "epoch": 2.8146964856230032,
+      "grad_norm": 0.04697149845887787,
+      "learning_rate": 2.2001260693120233e-05,
+      "loss": 0.1672,
+      "step": 440
+    },
+    {
+      "epoch": 2.8274760383386583,
+      "grad_norm": 0.054384654770629585,
+      "learning_rate": 2.156317701797766e-05,
+      "loss": 0.1807,
+      "step": 442
+    },
+    {
+      "epoch": 2.840255591054313,
+      "grad_norm": 0.04684823569442938,
+      "learning_rate": 2.1128296611172593e-05,
+      "loss": 0.171,
+      "step": 444
+    },
+    {
+      "epoch": 2.853035143769968,
+      "grad_norm": 0.0498371244165766,
+      "learning_rate": 2.0696668459845355e-05,
+      "loss": 0.1827,
+      "step": 446
+    },
+    {
+      "epoch": 2.8658146964856233,
+      "grad_norm": 0.04969475724913098,
+      "learning_rate": 2.026834118478567e-05,
+      "loss": 0.1749,
+      "step": 448
+    },
+    {
+      "epoch": 2.878594249201278,
+      "grad_norm": 0.051902756416916496,
+      "learning_rate": 1.98433630349558e-05,
+      "loss": 0.1891,
+      "step": 450
+    },
+    {
+      "epoch": 2.891373801916933,
+      "grad_norm": 0.05102564026340021,
+      "learning_rate": 1.9421781882055444e-05,
+      "loss": 0.1849,
+      "step": 452
+    },
+    {
+      "epoch": 2.9041533546325877,
+      "grad_norm": 0.05200929870376942,
+      "learning_rate": 1.9003645215129355e-05,
+      "loss": 0.1891,
+      "step": 454
+    },
+    {
+      "epoch": 2.916932907348243,
+      "grad_norm": 0.05083154953396676,
+      "learning_rate": 1.858900013521788e-05,
+      "loss": 0.179,
+      "step": 456
+    },
+    {
+      "epoch": 2.9297124600638975,
+      "grad_norm": 0.049127219472404525,
+      "learning_rate": 1.817789335005121e-05,
+      "loss": 0.17,
+      "step": 458
+    },
+    {
+      "epoch": 2.9424920127795526,
+      "grad_norm": 0.049677004679461886,
+      "learning_rate": 1.777037116878804e-05,
+      "loss": 0.1831,
+      "step": 460
+    },
+    {
+      "epoch": 2.9552715654952078,
+      "grad_norm": 0.054496479788018075,
+      "learning_rate": 1.7366479496799077e-05,
+      "loss": 0.1843,
+      "step": 462
+    },
+    {
+      "epoch": 2.9680511182108624,
+      "grad_norm": 0.04820092295738451,
+      "learning_rate": 1.6966263830495936e-05,
+      "loss": 0.1685,
+      "step": 464
+    },
+    {
+      "epoch": 2.9808306709265175,
+      "grad_norm": 0.04915420841884947,
+      "learning_rate": 1.656976925220633e-05,
+      "loss": 0.1875,
+      "step": 466
+    },
+    {
+      "epoch": 2.9936102236421727,
+      "grad_norm": 0.07661474807504913,
+      "learning_rate": 1.6177040425095662e-05,
+      "loss": 0.1891,
+      "step": 468
+    },
+    {
+      "epoch": 3.0079872204472844,
+      "grad_norm": 0.07655695803476695,
+      "learning_rate": 1.5788121588135975e-05,
+      "loss": 0.1837,
+      "step": 470
+    },
+    {
+      "epoch": 3.0207667731629395,
+      "grad_norm": 0.060916330302725,
+      "learning_rate": 1.5403056551122697e-05,
+      "loss": 0.0872,
+      "step": 472
+    },
+    {
+      "epoch": 3.033546325878594,
+      "grad_norm": 0.052542395235648506,
+      "learning_rate": 1.5021888689739549e-05,
+      "loss": 0.0778,
+      "step": 474
+    },
+    {
+      "epoch": 3.0463258785942493,
+      "grad_norm": 0.20368087770560855,
+      "learning_rate": 1.4644660940672627e-05,
+      "loss": 0.102,
+      "step": 476
+    },
+    {
+      "epoch": 3.059105431309904,
+      "grad_norm": 0.10396707161226072,
+      "learning_rate": 1.427141579677374e-05,
+      "loss": 0.083,
+      "step": 478
+    },
+    {
+      "epoch": 3.071884984025559,
+      "grad_norm": 0.04599720220665865,
+      "learning_rate": 1.3902195302273779e-05,
+      "loss": 0.0757,
+      "step": 480
+    },
+    {
+      "epoch": 3.084664536741214,
+      "grad_norm": 0.056109340867354925,
+      "learning_rate": 1.3537041048046695e-05,
+      "loss": 0.081,
+      "step": 482
+    },
+    {
+      "epoch": 3.097444089456869,
+      "grad_norm": 0.048015102375770044,
+      "learning_rate": 1.3175994166924394e-05,
+      "loss": 0.0802,
+      "step": 484
+    },
+    {
+      "epoch": 3.110223642172524,
+      "grad_norm": 0.04645228076024571,
+      "learning_rate": 1.2819095329063469e-05,
+      "loss": 0.0787,
+      "step": 486
+    },
+    {
+      "epoch": 3.123003194888179,
+      "grad_norm": 0.04637085498651796,
+      "learning_rate": 1.246638473736378e-05,
+      "loss": 0.0839,
+      "step": 488
+    },
+    {
+      "epoch": 3.135782747603834,
+      "grad_norm": 0.05039074009256794,
+      "learning_rate": 1.2117902122939861e-05,
+      "loss": 0.0812,
+      "step": 490
+    },
+    {
+      "epoch": 3.148562300319489,
+      "grad_norm": 0.05079569512274489,
+      "learning_rate": 1.1773686740645384e-05,
+      "loss": 0.0797,
+      "step": 492
+    },
+    {
+      "epoch": 3.1613418530351436,
+      "grad_norm": 0.04286375870307716,
+      "learning_rate": 1.1433777364651271e-05,
+      "loss": 0.0737,
+      "step": 494
+    },
+    {
+      "epoch": 3.1741214057507987,
+      "grad_norm": 0.03982951021947898,
+      "learning_rate": 1.1098212284078036e-05,
+      "loss": 0.0722,
+      "step": 496
+    },
+    {
+      "epoch": 3.186900958466454,
+      "grad_norm": 0.0446624849328897,
+      "learning_rate": 1.076702929868264e-05,
+      "loss": 0.079,
+      "step": 498
+    },
+    {
+      "epoch": 3.1996805111821085,
+      "grad_norm": 0.04376807908723891,
+      "learning_rate": 1.0440265714600572e-05,
+      "loss": 0.0837,
+      "step": 500
+    },
+    {
+      "epoch": 3.2124600638977636,
+      "grad_norm": 0.04087367539850916,
+      "learning_rate": 1.0117958340143507e-05,
+      "loss": 0.076,
+      "step": 502
+    },
+    {
+      "epoch": 3.2252396166134187,
+      "grad_norm": 0.04066584417219993,
+      "learning_rate": 9.800143481652979e-06,
+      "loss": 0.0701,
+      "step": 504
+    },
+    {
+      "epoch": 3.2380191693290734,
+      "grad_norm": 0.08215263649470263,
+      "learning_rate": 9.48685693941067e-06,
+      "loss": 0.0776,
+      "step": 506
+    },
+    {
+      "epoch": 3.2507987220447285,
+      "grad_norm": 0.0437601284673361,
+      "learning_rate": 9.17813400360572e-06,
+      "loss": 0.0764,
+      "step": 508
+    },
+    {
+      "epoch": 3.263578274760383,
+      "grad_norm": 0.04382435518426366,
+      "learning_rate": 8.874009450359427e-06,
+      "loss": 0.0826,
+      "step": 510
+    },
+    {
+      "epoch": 3.2763578274760383,
+      "grad_norm": 0.04095610913441161,
+      "learning_rate": 8.574517537807897e-06,
+      "loss": 0.0753,
+      "step": 512
+    },
+    {
+      "epoch": 3.2891373801916934,
+      "grad_norm": 0.040525949300480126,
+      "learning_rate": 8.279692002243027e-06,
+      "loss": 0.0764,
+      "step": 514
+    },
+    {
+      "epoch": 3.301916932907348,
+      "grad_norm": 0.043675586209021296,
+      "learning_rate": 7.989566054312287e-06,
+      "loss": 0.0817,
+      "step": 516
+    },
+    {
+      "epoch": 3.3146964856230032,
+      "grad_norm": 0.04319420448553361,
+      "learning_rate": 7.704172375277691e-06,
+      "loss": 0.0759,
+      "step": 518
+    },
+    {
+      "epoch": 3.3274760383386583,
+      "grad_norm": 0.044446852802239034,
+      "learning_rate": 7.423543113334436e-06,
+      "loss": 0.0813,
+      "step": 520
+    },
+    {
+      "epoch": 3.340255591054313,
+      "grad_norm": 0.09121973616663154,
+      "learning_rate": 7.14770987998954e-06,
+      "loss": 0.0838,
+      "step": 522
+    },
+    {
+      "epoch": 3.353035143769968,
+      "grad_norm": 0.05879997473879583,
+      "learning_rate": 6.876703746500984e-06,
+      "loss": 0.0738,
+      "step": 524
+    },
+    {
+      "epoch": 3.365814696485623,
+      "grad_norm": 0.04667273388126841,
+      "learning_rate": 6.610555240377652e-06,
+      "loss": 0.0787,
+      "step": 526
+    },
+    {
+      "epoch": 3.378594249201278,
+      "grad_norm": 0.042105033545020404,
+      "learning_rate": 6.349294341940593e-06,
+      "loss": 0.0801,
+      "step": 528
+    },
+    {
+      "epoch": 3.391373801916933,
+      "grad_norm": 0.0407975465413022,
+      "learning_rate": 6.092950480945897e-06,
+      "loss": 0.0735,
+      "step": 530
+    },
+    {
+      "epoch": 3.4041533546325877,
+      "grad_norm": 0.04234912863253251,
+      "learning_rate": 5.841552533269534e-06,
+      "loss": 0.0772,
+      "step": 532
+    },
+    {
+      "epoch": 3.416932907348243,
+      "grad_norm": 0.04032120711392374,
+      "learning_rate": 5.595128817654638e-06,
+      "loss": 0.0749,
+      "step": 534
+    },
+    {
+      "epoch": 3.4297124600638975,
+      "grad_norm": 0.041050930036482094,
+      "learning_rate": 5.353707092521582e-06,
+      "loss": 0.0769,
+      "step": 536
+    },
+    {
+      "epoch": 3.4424920127795526,
+      "grad_norm": 0.043382176933190755,
+      "learning_rate": 5.117314552841052e-06,
+      "loss": 0.0767,
+      "step": 538
+    },
+    {
+      "epoch": 3.4552715654952078,
+      "grad_norm": 0.039240502138117625,
+      "learning_rate": 4.885977827070748e-06,
+      "loss": 0.0721,
+      "step": 540
+    },
+    {
+      "epoch": 3.4680511182108624,
+      "grad_norm": 0.040812347040587296,
+      "learning_rate": 4.659722974155767e-06,
+      "loss": 0.1114,
+      "step": 542
+    },
+    {
+      "epoch": 3.4808306709265175,
+      "grad_norm": 0.0423787622918925,
+      "learning_rate": 4.43857548059321e-06,
+      "loss": 0.0778,
+      "step": 544
+    },
+    {
+      "epoch": 3.4936102236421727,
+      "grad_norm": 0.042228923687598445,
+      "learning_rate": 4.2225602575612755e-06,
+      "loss": 0.0814,
+      "step": 546
+    },
+    {
+      "epoch": 3.5063897763578273,
+      "grad_norm": 0.0407267289339222,
+      "learning_rate": 4.011701638113063e-06,
+      "loss": 0.0782,
+      "step": 548
+    },
+    {
+      "epoch": 3.5191693290734825,
+      "grad_norm": 0.0389855165359938,
+      "learning_rate": 3.8060233744356633e-06,
+      "loss": 0.0789,
+      "step": 550
+    },
+    {
+      "epoch": 3.5319488817891376,
+      "grad_norm": 0.040904703617676376,
+      "learning_rate": 3.605548635174533e-06,
+      "loss": 0.078,
+      "step": 552
+    },
+    {
+      "epoch": 3.5447284345047922,
+      "grad_norm": 0.04093280012624571,
+      "learning_rate": 3.410300002823691e-06,
+      "loss": 0.0777,
+      "step": 554
+    },
+    {
+      "epoch": 3.5575079872204474,
+      "grad_norm": 0.042904856841507744,
+      "learning_rate": 3.220299471181898e-06,
+      "loss": 0.0757,
+      "step": 556
+    },
+    {
+      "epoch": 3.5702875399361025,
+      "grad_norm": 0.0436449067886704,
+      "learning_rate": 3.035568442875136e-06,
+      "loss": 0.0798,
+      "step": 558
+    },
+    {
+      "epoch": 3.583067092651757,
+      "grad_norm": 0.035664776931118955,
+      "learning_rate": 2.85612772694579e-06,
+      "loss": 0.0632,
+      "step": 560
+    },
+    {
+      "epoch": 3.5958466453674123,
+      "grad_norm": 0.03847526723825484,
+      "learning_rate": 2.6819975365085237e-06,
+      "loss": 0.0744,
+      "step": 562
+    },
+    {
+      "epoch": 3.608626198083067,
+      "grad_norm": 0.039939236612970476,
+      "learning_rate": 2.5131974864734066e-06,
+      "loss": 0.0794,
+      "step": 564
+    },
+    {
+      "epoch": 3.621405750798722,
+      "grad_norm": 0.040388305870748,
+      "learning_rate": 2.349746591336405e-06,
+      "loss": 0.0718,
+      "step": 566
+    },
+    {
+      "epoch": 3.6341853035143767,
+      "grad_norm": 0.04232813426430434,
+      "learning_rate": 2.191663263037458e-06,
+      "loss": 0.0769,
+      "step": 568
+    },
+    {
+      "epoch": 3.646964856230032,
+      "grad_norm": 0.04213845492527589,
+      "learning_rate": 2.0389653088865036e-06,
+      "loss": 0.0728,
+      "step": 570
+    },
+    {
+      "epoch": 3.659744408945687,
+      "grad_norm": 0.04098999517730541,
+      "learning_rate": 1.8916699295575324e-06,
+      "loss": 0.0724,
+      "step": 572
+    },
+    {
+      "epoch": 3.6725239616613417,
+      "grad_norm": 0.037533240934183365,
+      "learning_rate": 1.7497937171510547e-06,
+      "loss": 0.0709,
+      "step": 574
+    },
+    {
+      "epoch": 3.6853035143769968,
+      "grad_norm": 0.039040304607963414,
+      "learning_rate": 1.6133526533250565e-06,
+      "loss": 0.0756,
+      "step": 576
+    },
+    {
+      "epoch": 3.698083067092652,
+      "grad_norm": 0.04065729024121047,
+      "learning_rate": 1.4823621074947503e-06,
+      "loss": 0.0774,
+      "step": 578
+    },
+    {
+      "epoch": 3.7108626198083066,
+      "grad_norm": 0.04252602887603373,
+      "learning_rate": 1.3568368351012717e-06,
+      "loss": 0.0824,
+      "step": 580
+    },
+    {
+      "epoch": 3.7236421725239617,
+      "grad_norm": 0.04343672134882273,
+      "learning_rate": 1.236790975949592e-06,
+      "loss": 0.074,
+      "step": 582
+    },
+    {
+      "epoch": 3.736421725239617,
+      "grad_norm": 0.0403766223584342,
+      "learning_rate": 1.1222380526156928e-06,
+      "loss": 0.0755,
+      "step": 584
+    },
+    {
+      "epoch": 3.7492012779552715,
+      "grad_norm": 0.04234625541762105,
+      "learning_rate": 1.0131909689233442e-06,
+      "loss": 0.0814,
+      "step": 586
+    },
+    {
+      "epoch": 3.7619808306709266,
+      "grad_norm": 0.03861507912847567,
+      "learning_rate": 9.096620084905472e-07,
+      "loss": 0.0664,
+      "step": 588
+    },
+    {
+      "epoch": 3.7747603833865817,
+      "grad_norm": 0.041892733843973705,
+      "learning_rate": 8.11662833345822e-07,
+      "loss": 0.0832,
+      "step": 590
+    },
+    {
+      "epoch": 3.7875399361022364,
+      "grad_norm": 0.05413889236863839,
+      "learning_rate": 7.192044826145771e-07,
+      "loss": 0.0921,
+      "step": 592
+    },
+    {
+      "epoch": 3.8003194888178915,
+      "grad_norm": 0.04010280150213322,
+      "learning_rate": 6.322973712755697e-07,
+      "loss": 0.0752,
+      "step": 594
+    },
+    {
+      "epoch": 3.813099041533546,
+      "grad_norm": 0.04321423418162425,
+      "learning_rate": 5.509512889877333e-07,
+      "loss": 0.0781,
+      "step": 596
+    },
+    {
+      "epoch": 3.8258785942492013,
+      "grad_norm": 0.04049679761598481,
+      "learning_rate": 4.7517539898741524e-07,
+      "loss": 0.0694,
+      "step": 598
+    },
+    {
+      "epoch": 3.838658146964856,
+      "grad_norm": 0.04258434666712487,
+      "learning_rate": 4.049782370561583e-07,
+      "loss": 0.0756,
+      "step": 600
+    },
+    {
+      "epoch": 3.851437699680511,
+      "grad_norm": 0.03927978960342531,
+      "learning_rate": 3.4036771055923066e-07,
+      "loss": 0.075,
+      "step": 602
+    },
+    {
+      "epoch": 3.864217252396166,
+      "grad_norm": 0.04093422273122725,
+      "learning_rate": 2.813510975548772e-07,
+      "loss": 0.0793,
+      "step": 604
+    },
+    {
+      "epoch": 3.876996805111821,
+      "grad_norm": 0.0433141394014271,
+      "learning_rate": 2.2793504597447002e-07,
+      "loss": 0.0796,
+      "step": 606
+    },
+    {
+      "epoch": 3.889776357827476,
+      "grad_norm": 0.04198937065288365,
+      "learning_rate": 1.8012557287367392e-07,
+      "loss": 0.0753,
+      "step": 608
+    },
+    {
+      "epoch": 3.902555910543131,
+      "grad_norm": 0.043002763720086865,
+      "learning_rate": 1.379280637546443e-07,
+      "loss": 0.0917,
+      "step": 610
+    },
+    {
+      "epoch": 3.915335463258786,
+      "grad_norm": 0.042376326172823574,
+      "learning_rate": 1.0134727195937333e-07,
+      "loss": 0.0747,
+      "step": 612
+    },
+    {
+      "epoch": 3.928115015974441,
+      "grad_norm": 0.040600294489722695,
+      "learning_rate": 7.038731813426291e-08,
+      "loss": 0.0714,
+      "step": 614
+    },
+    {
+      "epoch": 3.940894568690096,
+      "grad_norm": 0.0395711578920217,
+      "learning_rate": 4.5051689765929214e-08,
+      "loss": 0.0801,
+      "step": 616
+    },
+    {
+      "epoch": 3.9536741214057507,
+      "grad_norm": 0.03943811103107873,
+      "learning_rate": 2.534324078837802e-08,
+      "loss": 0.074,
+      "step": 618
+    },
+    {
+      "epoch": 3.966453674121406,
+      "grad_norm": 0.041905340316324986,
+      "learning_rate": 1.1264191261528557e-08,
+      "loss": 0.0771,
+      "step": 620
+    },
+    {
+      "epoch": 3.979233226837061,
+      "grad_norm": 0.06686218682304546,
+      "learning_rate": 2.8161271211024633e-09,
+      "loss": 0.0811,
+      "step": 622
+    },
+    {
+      "epoch": 3.9920127795527156,
+      "grad_norm": 0.03876030687059135,
+      "learning_rate": 0.0,
+      "loss": 0.076,
+      "step": 624
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 624,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.5262683933881926e+19,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fbc6c15cc6de09ed10b88b2483f84e85b7b1119b7dd63c1e2d29d8ad02f02dab
+size 7352

uccix_v2_instruct_191224_lr1e-4/checkpoint-624/zero_to_fp32.py

@@ -0,0 +1,592 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example: python zero_to_fp32.py . pytorch_model.bin
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+
+    total_files = len(files)
+    state_dicts = []
+    for f in files:
+        state_dict = torch.load(f, map_location=device)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    if zero_stage <= 2:
+        fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    elif zero_stage == 3:
+        # if there is more than one param group, there will be multiple flattened tensors - one
+        # flattened tensor per group - for simplicity merge them into a single tensor
+        #
+        # XXX: could make the script more memory efficient for when there are multiple groups - it
+        # will require matching the sub-lists of param_shapes for each param group flattened tensor
+
+        fp32_flat_groups = [
+            torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
+        ]
+
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = fp32_flat_groups[0].numel() * world_size
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    for name, shape in param_shapes.items():
+
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # XXX: memory usage doubles here
+        state_dict[name] = torch.cat(
+            tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
+            0).narrow(0, 0, unpartitioned_numel).view(shape)
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    Note: this approach may not work if your application doesn't have sufficient free CPU memory and
+    you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+    """
+
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+    print(f"Saving fp32 state dict to {output_file}")
+    torch.save(state_dict, output_file)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument(
+        "output_file",
+        type=str,
+        help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, args.output_file, tag=args.tag)