Upload folder using huggingface_hub

uccix_instruct_191224_lr1e-5/checkpoint-780/config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "ReliableAI/UCCIX-Llama2-13B",
+  "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_instruct_191224_lr1e-5/checkpoint-780/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_instruct_191224_lr1e-5/checkpoint-780/latest

@@ -0,0 +1 @@
+global_step780

uccix_instruct_191224_lr1e-5/checkpoint-780/model-00001-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94965cf498cb0f139ec3e2dba2a39a5783dcb09b7e8c3f92a268db18c7a5e70f
+size 4961502800

uccix_instruct_191224_lr1e-5/checkpoint-780/model-00002-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6dae0127fe777d3dda1e7b996bf7a8f989ef5e1341813449c1bc17a1731b5218
+size 4970422232

uccix_instruct_191224_lr1e-5/checkpoint-780/model-00003-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ff6d509711cc88ba6760fcde5f2356ec328e340d069ccf53a8f2c3912de3091
+size 4881272584

uccix_instruct_191224_lr1e-5/checkpoint-780/model-00004-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88177016a275975476dccde391c6e4c6a40fc135b40b7c3be507f3765b0b7005
+size 4933722216

uccix_instruct_191224_lr1e-5/checkpoint-780/model-00005-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5c1c2740a0eb1dbdf876fd161dbe9a49f9d7eee88898792421b3b5abbb9cedc
+size 4933722208

uccix_instruct_191224_lr1e-5/checkpoint-780/model-00006-of-00006.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:58734ae52c39ae93621ff4ad05a156ab41c86a6ce2d8bed2b9d6a1200110e27e
+size 1422460712

uccix_instruct_191224_lr1e-5/checkpoint-780/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_instruct_191224_lr1e-5/checkpoint-780/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00dfeb21ba336403a7e72c8f77c9778a257aad0edee2feae2d340cf03caca787
+size 15984

uccix_instruct_191224_lr1e-5/checkpoint-780/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50465e9d75c92e96fd7519f629c79aead5d2cc9b6088eee8fa9736b132e587d6
+size 15984

uccix_instruct_191224_lr1e-5/checkpoint-780/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3cf4c7fbb993fef05bd7f7fb8a7f87bad0c0e0da1bbeb38e720877b83d2e86b4
+size 15984

uccix_instruct_191224_lr1e-5/checkpoint-780/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fdab190afc0f03dae6afda58a9cbb58b2316be270b8410333bd454c4545a72e
+size 15984

uccix_instruct_191224_lr1e-5/checkpoint-780/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:011bd910deb7f0f8d5392fefc20e75941dfb3224ef0ddddf1618e98d9f28760d
+size 15984

uccix_instruct_191224_lr1e-5/checkpoint-780/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab705de1b950b842a5f0ddeed8c2e279a4e5f557eaa09d848925c1c5af78a2cf
+size 15984

uccix_instruct_191224_lr1e-5/checkpoint-780/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03e7ca626152124bd7340d340c0be0769a6a733a8a52b8861de45b114fdb043a
+size 15984

uccix_instruct_191224_lr1e-5/checkpoint-780/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77b33bae6d2188390ebc6dfdd276a1f09b13f49680aa59e33f8ab221e3b919bd
+size 15984

uccix_instruct_191224_lr1e-5/checkpoint-780/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46f217040728795cee3845514c2663ea7abe3dcb3034ff5b61525c8b41f46289
+size 1064

uccix_instruct_191224_lr1e-5/checkpoint-780/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_instruct_191224_lr1e-5/checkpoint-780/tokenizer.model

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

uccix_instruct_191224_lr1e-5/checkpoint-780/tokenizer_config.json

@@ -0,0 +1,43 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": 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 = false %}{% 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_instruct_191224_lr1e-5/checkpoint-780/trainer_state.json

@@ -0,0 +1,2770 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "eval_steps": 500,
+  "global_step": 780,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00641025641025641,
+      "grad_norm": 3.8148568052575884,
+      "learning_rate": 1.282051282051282e-07,
+      "loss": 4.889,
+      "step": 1
+    },
+    {
+      "epoch": 0.01282051282051282,
+      "grad_norm": 4.453444589892027,
+      "learning_rate": 2.564102564102564e-07,
+      "loss": 4.9097,
+      "step": 2
+    },
+    {
+      "epoch": 0.02564102564102564,
+      "grad_norm": 4.896614258621833,
+      "learning_rate": 5.128205128205128e-07,
+      "loss": 4.9099,
+      "step": 4
+    },
+    {
+      "epoch": 0.038461538461538464,
+      "grad_norm": 4.456576485464451,
+      "learning_rate": 7.692307692307694e-07,
+      "loss": 4.9102,
+      "step": 6
+    },
+    {
+      "epoch": 0.05128205128205128,
+      "grad_norm": 4.193427815120892,
+      "learning_rate": 1.0256410256410257e-06,
+      "loss": 4.8924,
+      "step": 8
+    },
+    {
+      "epoch": 0.0641025641025641,
+      "grad_norm": 3.6726747534666555,
+      "learning_rate": 1.282051282051282e-06,
+      "loss": 4.8372,
+      "step": 10
+    },
+    {
+      "epoch": 0.07692307692307693,
+      "grad_norm": 3.337981680961211,
+      "learning_rate": 1.5384615384615387e-06,
+      "loss": 4.7794,
+      "step": 12
+    },
+    {
+      "epoch": 0.08974358974358974,
+      "grad_norm": 2.675890453922504,
+      "learning_rate": 1.794871794871795e-06,
+      "loss": 4.6191,
+      "step": 14
+    },
+    {
+      "epoch": 0.10256410256410256,
+      "grad_norm": 2.398848700299253,
+      "learning_rate": 2.0512820512820513e-06,
+      "loss": 4.5723,
+      "step": 16
+    },
+    {
+      "epoch": 0.11538461538461539,
+      "grad_norm": 1.8159784961859098,
+      "learning_rate": 2.307692307692308e-06,
+      "loss": 4.3568,
+      "step": 18
+    },
+    {
+      "epoch": 0.1282051282051282,
+      "grad_norm": 1.6094220673057946,
+      "learning_rate": 2.564102564102564e-06,
+      "loss": 4.2686,
+      "step": 20
+    },
+    {
+      "epoch": 0.14102564102564102,
+      "grad_norm": 1.4349818434671497,
+      "learning_rate": 2.8205128205128207e-06,
+      "loss": 4.169,
+      "step": 22
+    },
+    {
+      "epoch": 0.15384615384615385,
+      "grad_norm": 1.4412559958198408,
+      "learning_rate": 3.0769230769230774e-06,
+      "loss": 4.0415,
+      "step": 24
+    },
+    {
+      "epoch": 0.16666666666666666,
+      "grad_norm": 1.3626982007755366,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 3.8569,
+      "step": 26
+    },
+    {
+      "epoch": 0.1794871794871795,
+      "grad_norm": 1.3679096739652512,
+      "learning_rate": 3.58974358974359e-06,
+      "loss": 3.7409,
+      "step": 28
+    },
+    {
+      "epoch": 0.19230769230769232,
+      "grad_norm": 1.3396391976584703,
+      "learning_rate": 3.846153846153847e-06,
+      "loss": 3.6585,
+      "step": 30
+    },
+    {
+      "epoch": 0.20512820512820512,
+      "grad_norm": 1.294876480457606,
+      "learning_rate": 4.102564102564103e-06,
+      "loss": 3.4961,
+      "step": 32
+    },
+    {
+      "epoch": 0.21794871794871795,
+      "grad_norm": 1.103820056614455,
+      "learning_rate": 4.358974358974359e-06,
+      "loss": 3.3518,
+      "step": 34
+    },
+    {
+      "epoch": 0.23076923076923078,
+      "grad_norm": 1.0522131115906572,
+      "learning_rate": 4.615384615384616e-06,
+      "loss": 3.1984,
+      "step": 36
+    },
+    {
+      "epoch": 0.24358974358974358,
+      "grad_norm": 1.0081732884085817,
+      "learning_rate": 4.871794871794872e-06,
+      "loss": 3.054,
+      "step": 38
+    },
+    {
+      "epoch": 0.2564102564102564,
+      "grad_norm": 0.9214039999549644,
+      "learning_rate": 5.128205128205128e-06,
+      "loss": 2.8628,
+      "step": 40
+    },
+    {
+      "epoch": 0.2692307692307692,
+      "grad_norm": 0.8143994876297143,
+      "learning_rate": 5.384615384615385e-06,
+      "loss": 2.7475,
+      "step": 42
+    },
+    {
+      "epoch": 0.28205128205128205,
+      "grad_norm": 0.700891765547207,
+      "learning_rate": 5.641025641025641e-06,
+      "loss": 2.5869,
+      "step": 44
+    },
+    {
+      "epoch": 0.2948717948717949,
+      "grad_norm": 0.7510674065754775,
+      "learning_rate": 5.897435897435898e-06,
+      "loss": 2.4461,
+      "step": 46
+    },
+    {
+      "epoch": 0.3076923076923077,
+      "grad_norm": 0.6794074940373539,
+      "learning_rate": 6.153846153846155e-06,
+      "loss": 2.3477,
+      "step": 48
+    },
+    {
+      "epoch": 0.32051282051282054,
+      "grad_norm": 0.5162215042692575,
+      "learning_rate": 6.410256410256412e-06,
+      "loss": 2.2152,
+      "step": 50
+    },
+    {
+      "epoch": 0.3333333333333333,
+      "grad_norm": 0.5146975027904754,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 2.1975,
+      "step": 52
+    },
+    {
+      "epoch": 0.34615384615384615,
+      "grad_norm": 0.4474574545979082,
+      "learning_rate": 6.923076923076923e-06,
+      "loss": 2.0824,
+      "step": 54
+    },
+    {
+      "epoch": 0.358974358974359,
+      "grad_norm": 0.40379510918119965,
+      "learning_rate": 7.17948717948718e-06,
+      "loss": 2.0388,
+      "step": 56
+    },
+    {
+      "epoch": 0.3717948717948718,
+      "grad_norm": 0.4109144194248555,
+      "learning_rate": 7.435897435897437e-06,
+      "loss": 1.9699,
+      "step": 58
+    },
+    {
+      "epoch": 0.38461538461538464,
+      "grad_norm": 0.36878556755849573,
+      "learning_rate": 7.692307692307694e-06,
+      "loss": 1.9252,
+      "step": 60
+    },
+    {
+      "epoch": 0.3974358974358974,
+      "grad_norm": 0.33951214974325605,
+      "learning_rate": 7.948717948717949e-06,
+      "loss": 1.8773,
+      "step": 62
+    },
+    {
+      "epoch": 0.41025641025641024,
+      "grad_norm": 0.31625266306424027,
+      "learning_rate": 8.205128205128205e-06,
+      "loss": 1.7966,
+      "step": 64
+    },
+    {
+      "epoch": 0.4230769230769231,
+      "grad_norm": 0.7180890498799148,
+      "learning_rate": 8.461538461538462e-06,
+      "loss": 1.8108,
+      "step": 66
+    },
+    {
+      "epoch": 0.4358974358974359,
+      "grad_norm": 0.33704662479371716,
+      "learning_rate": 8.717948717948719e-06,
+      "loss": 1.7498,
+      "step": 68
+    },
+    {
+      "epoch": 0.44871794871794873,
+      "grad_norm": 0.2761824271642518,
+      "learning_rate": 8.974358974358976e-06,
+      "loss": 1.7124,
+      "step": 70
+    },
+    {
+      "epoch": 0.46153846153846156,
+      "grad_norm": 0.24386286193528572,
+      "learning_rate": 9.230769230769232e-06,
+      "loss": 1.6382,
+      "step": 72
+    },
+    {
+      "epoch": 0.47435897435897434,
+      "grad_norm": 0.25885451676676363,
+      "learning_rate": 9.487179487179487e-06,
+      "loss": 1.6588,
+      "step": 74
+    },
+    {
+      "epoch": 0.48717948717948717,
+      "grad_norm": 0.3040030663690383,
+      "learning_rate": 9.743589743589744e-06,
+      "loss": 1.6209,
+      "step": 76
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.26598080566137733,
+      "learning_rate": 1e-05,
+      "loss": 1.6294,
+      "step": 78
+    },
+    {
+      "epoch": 0.5128205128205128,
+      "grad_norm": 0.22696288673824674,
+      "learning_rate": 9.99995506314361e-06,
+      "loss": 1.58,
+      "step": 80
+    },
+    {
+      "epoch": 0.5256410256410257,
+      "grad_norm": 0.21242259411358655,
+      "learning_rate": 9.99982025338217e-06,
+      "loss": 1.5439,
+      "step": 82
+    },
+    {
+      "epoch": 0.5384615384615384,
+      "grad_norm": 0.20291826899403465,
+      "learning_rate": 9.999595573138845e-06,
+      "loss": 1.5274,
+      "step": 84
+    },
+    {
+      "epoch": 0.5512820512820513,
+      "grad_norm": 0.1855444412322797,
+      "learning_rate": 9.99928102645221e-06,
+      "loss": 1.5161,
+      "step": 86
+    },
+    {
+      "epoch": 0.5641025641025641,
+      "grad_norm": 0.17883874148398324,
+      "learning_rate": 9.99887661897616e-06,
+      "loss": 1.4916,
+      "step": 88
+    },
+    {
+      "epoch": 0.5769230769230769,
+      "grad_norm": 0.17041478792908024,
+      "learning_rate": 9.99838235797981e-06,
+      "loss": 1.4679,
+      "step": 90
+    },
+    {
+      "epoch": 0.5897435897435898,
+      "grad_norm": 0.1904762198987749,
+      "learning_rate": 9.997798252347382e-06,
+      "loss": 1.471,
+      "step": 92
+    },
+    {
+      "epoch": 0.6025641025641025,
+      "grad_norm": 0.19077041355708335,
+      "learning_rate": 9.99712431257802e-06,
+      "loss": 1.4672,
+      "step": 94
+    },
+    {
+      "epoch": 0.6153846153846154,
+      "grad_norm": 0.1702104328191874,
+      "learning_rate": 9.996360550785619e-06,
+      "loss": 1.4455,
+      "step": 96
+    },
+    {
+      "epoch": 0.6282051282051282,
+      "grad_norm": 0.19039133859515542,
+      "learning_rate": 9.9955069806986e-06,
+      "loss": 1.4727,
+      "step": 98
+    },
+    {
+      "epoch": 0.6410256410256411,
+      "grad_norm": 0.15448238517128507,
+      "learning_rate": 9.994563617659665e-06,
+      "loss": 1.4257,
+      "step": 100
+    },
+    {
+      "epoch": 0.6538461538461539,
+      "grad_norm": 0.15202351051018634,
+      "learning_rate": 9.993530478625524e-06,
+      "loss": 1.4214,
+      "step": 102
+    },
+    {
+      "epoch": 0.6666666666666666,
+      "grad_norm": 0.16296598133044526,
+      "learning_rate": 9.992407582166582e-06,
+      "loss": 1.4213,
+      "step": 104
+    },
+    {
+      "epoch": 0.6794871794871795,
+      "grad_norm": 0.1462038294164801,
+      "learning_rate": 9.991194948466615e-06,
+      "loss": 1.3993,
+      "step": 106
+    },
+    {
+      "epoch": 0.6923076923076923,
+      "grad_norm": 0.14470989191451086,
+      "learning_rate": 9.989892599322404e-06,
+      "loss": 1.4014,
+      "step": 108
+    },
+    {
+      "epoch": 0.7051282051282052,
+      "grad_norm": 0.15440545758233384,
+      "learning_rate": 9.988500558143337e-06,
+      "loss": 1.3878,
+      "step": 110
+    },
+    {
+      "epoch": 0.717948717948718,
+      "grad_norm": 0.1412948019214843,
+      "learning_rate": 9.987018849950996e-06,
+      "loss": 1.355,
+      "step": 112
+    },
+    {
+      "epoch": 0.7307692307692307,
+      "grad_norm": 0.15156074653795895,
+      "learning_rate": 9.985447501378706e-06,
+      "loss": 1.3642,
+      "step": 114
+    },
+    {
+      "epoch": 0.7435897435897436,
+      "grad_norm": 0.3875845143038168,
+      "learning_rate": 9.983786540671052e-06,
+      "loss": 1.3797,
+      "step": 116
+    },
+    {
+      "epoch": 0.7564102564102564,
+      "grad_norm": 0.15788537547887518,
+      "learning_rate": 9.982035997683372e-06,
+      "loss": 1.3388,
+      "step": 118
+    },
+    {
+      "epoch": 0.7692307692307693,
+      "grad_norm": 0.15056320914445512,
+      "learning_rate": 9.980195903881231e-06,
+      "loss": 1.343,
+      "step": 120
+    },
+    {
+      "epoch": 0.782051282051282,
+      "grad_norm": 0.1555129283317706,
+      "learning_rate": 9.978266292339838e-06,
+      "loss": 1.328,
+      "step": 122
+    },
+    {
+      "epoch": 0.7948717948717948,
+      "grad_norm": 0.14999182496915453,
+      "learning_rate": 9.976247197743465e-06,
+      "loss": 1.352,
+      "step": 124
+    },
+    {
+      "epoch": 0.8076923076923077,
+      "grad_norm": 0.14124313426191026,
+      "learning_rate": 9.974138656384815e-06,
+      "loss": 1.3243,
+      "step": 126
+    },
+    {
+      "epoch": 0.8205128205128205,
+      "grad_norm": 0.1378326204862212,
+      "learning_rate": 9.97194070616438e-06,
+      "loss": 1.3241,
+      "step": 128
+    },
+    {
+      "epoch": 0.8333333333333334,
+      "grad_norm": 0.14227960534974604,
+      "learning_rate": 9.969653386589749e-06,
+      "loss": 1.3219,
+      "step": 130
+    },
+    {
+      "epoch": 0.8461538461538461,
+      "grad_norm": 0.12713543749272155,
+      "learning_rate": 9.967276738774897e-06,
+      "loss": 1.3096,
+      "step": 132
+    },
+    {
+      "epoch": 0.8589743589743589,
+      "grad_norm": 0.15061232362563903,
+      "learning_rate": 9.964810805439464e-06,
+      "loss": 1.3011,
+      "step": 134
+    },
+    {
+      "epoch": 0.8717948717948718,
+      "grad_norm": 0.14361563348990292,
+      "learning_rate": 9.962255630907964e-06,
+      "loss": 1.2827,
+      "step": 136
+    },
+    {
+      "epoch": 0.8846153846153846,
+      "grad_norm": 0.17754387209035652,
+      "learning_rate": 9.959611261108999e-06,
+      "loss": 1.3185,
+      "step": 138
+    },
+    {
+      "epoch": 0.8974358974358975,
+      "grad_norm": 0.1458623897430443,
+      "learning_rate": 9.956877743574437e-06,
+      "loss": 1.3286,
+      "step": 140
+    },
+    {
+      "epoch": 0.9102564102564102,
+      "grad_norm": 0.14084398418567437,
+      "learning_rate": 9.954055127438554e-06,
+      "loss": 1.3005,
+      "step": 142
+    },
+    {
+      "epoch": 0.9230769230769231,
+      "grad_norm": 0.13580861113069753,
+      "learning_rate": 9.951143463437145e-06,
+      "loss": 1.3165,
+      "step": 144
+    },
+    {
+      "epoch": 0.9358974358974359,
+      "grad_norm": 0.13622051889734035,
+      "learning_rate": 9.948142803906623e-06,
+      "loss": 1.2929,
+      "step": 146
+    },
+    {
+      "epoch": 0.9487179487179487,
+      "grad_norm": 0.12679082371935066,
+      "learning_rate": 9.94505320278307e-06,
+      "loss": 1.2833,
+      "step": 148
+    },
+    {
+      "epoch": 0.9615384615384616,
+      "grad_norm": 0.11939382079952243,
+      "learning_rate": 9.94187471560127e-06,
+      "loss": 1.2851,
+      "step": 150
+    },
+    {
+      "epoch": 0.9743589743589743,
+      "grad_norm": 0.11752490134274678,
+      "learning_rate": 9.938607399493714e-06,
+      "loss": 1.2559,
+      "step": 152
+    },
+    {
+      "epoch": 0.9871794871794872,
+      "grad_norm": 0.11807212671773365,
+      "learning_rate": 9.935251313189564e-06,
+      "loss": 1.285,
+      "step": 154
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.1120761333795772,
+      "learning_rate": 9.931806517013612e-06,
+      "loss": 1.2491,
+      "step": 156
+    },
+    {
+      "epoch": 1.0128205128205128,
+      "grad_norm": 0.10750345822189263,
+      "learning_rate": 9.92827307288518e-06,
+      "loss": 1.2442,
+      "step": 158
+    },
+    {
+      "epoch": 1.0256410256410255,
+      "grad_norm": 0.10918642022881683,
+      "learning_rate": 9.924651044317017e-06,
+      "loss": 1.2286,
+      "step": 160
+    },
+    {
+      "epoch": 1.0384615384615385,
+      "grad_norm": 0.11225330042691335,
+      "learning_rate": 9.920940496414153e-06,
+      "loss": 1.2158,
+      "step": 162
+    },
+    {
+      "epoch": 1.0512820512820513,
+      "grad_norm": 0.11366482652198566,
+      "learning_rate": 9.917141495872733e-06,
+      "loss": 1.2074,
+      "step": 164
+    },
+    {
+      "epoch": 1.064102564102564,
+      "grad_norm": 0.12295651003296312,
+      "learning_rate": 9.913254110978812e-06,
+      "loss": 1.2003,
+      "step": 166
+    },
+    {
+      "epoch": 1.0769230769230769,
+      "grad_norm": 0.1144456030840293,
+      "learning_rate": 9.909278411607134e-06,
+      "loss": 1.206,
+      "step": 168
+    },
+    {
+      "epoch": 1.0897435897435896,
+      "grad_norm": 0.2468334129961725,
+      "learning_rate": 9.90521446921987e-06,
+      "loss": 1.2235,
+      "step": 170
+    },
+    {
+      "epoch": 1.1025641025641026,
+      "grad_norm": 0.127278158070263,
+      "learning_rate": 9.90106235686534e-06,
+      "loss": 1.1928,
+      "step": 172
+    },
+    {
+      "epoch": 1.1153846153846154,
+      "grad_norm": 0.1280282060730887,
+      "learning_rate": 9.896822149176695e-06,
+      "loss": 1.2068,
+      "step": 174
+    },
+    {
+      "epoch": 1.1282051282051282,
+      "grad_norm": 0.1142922422404122,
+      "learning_rate": 9.892493922370575e-06,
+      "loss": 1.217,
+      "step": 176
+    },
+    {
+      "epoch": 1.141025641025641,
+      "grad_norm": 0.17470470224878323,
+      "learning_rate": 9.888077754245741e-06,
+      "loss": 1.2099,
+      "step": 178
+    },
+    {
+      "epoch": 1.1538461538461537,
+      "grad_norm": 0.10477882692325258,
+      "learning_rate": 9.883573724181683e-06,
+      "loss": 1.1944,
+      "step": 180
+    },
+    {
+      "epoch": 1.1666666666666667,
+      "grad_norm": 0.114790034377695,
+      "learning_rate": 9.878981913137178e-06,
+      "loss": 1.172,
+      "step": 182
+    },
+    {
+      "epoch": 1.1794871794871795,
+      "grad_norm": 0.1044922535107306,
+      "learning_rate": 9.87430240364885e-06,
+      "loss": 1.2147,
+      "step": 184
+    },
+    {
+      "epoch": 1.1923076923076923,
+      "grad_norm": 0.09771283060341285,
+      "learning_rate": 9.869535279829674e-06,
+      "loss": 1.173,
+      "step": 186
+    },
+    {
+      "epoch": 1.205128205128205,
+      "grad_norm": 0.1013995999635824,
+      "learning_rate": 9.864680627367476e-06,
+      "loss": 1.2023,
+      "step": 188
+    },
+    {
+      "epoch": 1.217948717948718,
+      "grad_norm": 0.10273326452887067,
+      "learning_rate": 9.859738533523384e-06,
+      "loss": 1.1732,
+      "step": 190
+    },
+    {
+      "epoch": 1.2307692307692308,
+      "grad_norm": 0.09684048616936082,
+      "learning_rate": 9.854709087130261e-06,
+      "loss": 1.1952,
+      "step": 192
+    },
+    {
+      "epoch": 1.2435897435897436,
+      "grad_norm": 0.10827760658070901,
+      "learning_rate": 9.849592378591113e-06,
+      "loss": 1.1864,
+      "step": 194
+    },
+    {
+      "epoch": 1.2564102564102564,
+      "grad_norm": 0.09989527940011267,
+      "learning_rate": 9.844388499877457e-06,
+      "loss": 1.2016,
+      "step": 196
+    },
+    {
+      "epoch": 1.2692307692307692,
+      "grad_norm": 0.09930771667309381,
+      "learning_rate": 9.839097544527674e-06,
+      "loss": 1.1738,
+      "step": 198
+    },
+    {
+      "epoch": 1.282051282051282,
+      "grad_norm": 0.1032001919164007,
+      "learning_rate": 9.833719607645325e-06,
+      "loss": 1.176,
+      "step": 200
+    },
+    {
+      "epoch": 1.294871794871795,
+      "grad_norm": 0.09859412157061716,
+      "learning_rate": 9.82825478589744e-06,
+      "loss": 1.1682,
+      "step": 202
+    },
+    {
+      "epoch": 1.3076923076923077,
+      "grad_norm": 0.09558235334437347,
+      "learning_rate": 9.822703177512783e-06,
+      "loss": 1.181,
+      "step": 204
+    },
+    {
+      "epoch": 1.3205128205128205,
+      "grad_norm": 0.08733478657745303,
+      "learning_rate": 9.817064882280085e-06,
+      "loss": 1.1686,
+      "step": 206
+    },
+    {
+      "epoch": 1.3333333333333333,
+      "grad_norm": 0.09397505343456257,
+      "learning_rate": 9.811340001546252e-06,
+      "loss": 1.1778,
+      "step": 208
+    },
+    {
+      "epoch": 1.3461538461538463,
+      "grad_norm": 0.09590407825516856,
+      "learning_rate": 9.805528638214543e-06,
+      "loss": 1.1542,
+      "step": 210
+    },
+    {
+      "epoch": 1.358974358974359,
+      "grad_norm": 0.0912508440064145,
+      "learning_rate": 9.799630896742716e-06,
+      "loss": 1.1643,
+      "step": 212
+    },
+    {
+      "epoch": 1.3717948717948718,
+      "grad_norm": 0.09258955107744923,
+      "learning_rate": 9.793646883141155e-06,
+      "loss": 1.1686,
+      "step": 214
+    },
+    {
+      "epoch": 1.3846153846153846,
+      "grad_norm": 0.09889457149777804,
+      "learning_rate": 9.787576704970965e-06,
+      "loss": 1.1677,
+      "step": 216
+    },
+    {
+      "epoch": 1.3974358974358974,
+      "grad_norm": 0.09374670756166416,
+      "learning_rate": 9.781420471342035e-06,
+      "loss": 1.146,
+      "step": 218
+    },
+    {
+      "epoch": 1.4102564102564101,
+      "grad_norm": 0.09136677460744856,
+      "learning_rate": 9.77517829291108e-06,
+      "loss": 1.1594,
+      "step": 220
+    },
+    {
+      "epoch": 1.4230769230769231,
+      "grad_norm": 0.10584946030378292,
+      "learning_rate": 9.768850281879651e-06,
+      "loss": 1.1865,
+      "step": 222
+    },
+    {
+      "epoch": 1.435897435897436,
+      "grad_norm": 0.09187981607301214,
+      "learning_rate": 9.762436551992117e-06,
+      "loss": 1.1606,
+      "step": 224
+    },
+    {
+      "epoch": 1.4487179487179487,
+      "grad_norm": 0.09880449655805854,
+      "learning_rate": 9.755937218533622e-06,
+      "loss": 1.1586,
+      "step": 226
+    },
+    {
+      "epoch": 1.4615384615384617,
+      "grad_norm": 0.08704607108972029,
+      "learning_rate": 9.74935239832801e-06,
+      "loss": 1.1746,
+      "step": 228
+    },
+    {
+      "epoch": 1.4743589743589745,
+      "grad_norm": 0.08909112778091671,
+      "learning_rate": 9.742682209735727e-06,
+      "loss": 1.1575,
+      "step": 230
+    },
+    {
+      "epoch": 1.4871794871794872,
+      "grad_norm": 0.09035998053799675,
+      "learning_rate": 9.735926772651703e-06,
+      "loss": 1.1678,
+      "step": 232
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 0.09500864788295198,
+      "learning_rate": 9.729086208503174e-06,
+      "loss": 1.1466,
+      "step": 234
+    },
+    {
+      "epoch": 1.5128205128205128,
+      "grad_norm": 0.09247434213683463,
+      "learning_rate": 9.722160640247523e-06,
+      "loss": 1.1687,
+      "step": 236
+    },
+    {
+      "epoch": 1.5256410256410255,
+      "grad_norm": 0.09322212100100113,
+      "learning_rate": 9.715150192370054e-06,
+      "loss": 1.1376,
+      "step": 238
+    },
+    {
+      "epoch": 1.5384615384615383,
+      "grad_norm": 0.08824919508271642,
+      "learning_rate": 9.708054990881763e-06,
+      "loss": 1.1523,
+      "step": 240
+    },
+    {
+      "epoch": 1.5512820512820513,
+      "grad_norm": 0.25559730635424294,
+      "learning_rate": 9.700875163317072e-06,
+      "loss": 1.1488,
+      "step": 242
+    },
+    {
+      "epoch": 1.564102564102564,
+      "grad_norm": 0.2487505162861363,
+      "learning_rate": 9.693610838731532e-06,
+      "loss": 1.1481,
+      "step": 244
+    },
+    {
+      "epoch": 1.5769230769230769,
+      "grad_norm": 0.12151469789600829,
+      "learning_rate": 9.686262147699507e-06,
+      "loss": 1.1483,
+      "step": 246
+    },
+    {
+      "epoch": 1.5897435897435899,
+      "grad_norm": 0.10407519891252137,
+      "learning_rate": 9.678829222311827e-06,
+      "loss": 1.13,
+      "step": 248
+    },
+    {
+      "epoch": 1.6025641025641026,
+      "grad_norm": 0.11236395690738615,
+      "learning_rate": 9.671312196173413e-06,
+      "loss": 1.1493,
+      "step": 250
+    },
+    {
+      "epoch": 1.6153846153846154,
+      "grad_norm": 0.1012523372817843,
+      "learning_rate": 9.663711204400872e-06,
+      "loss": 1.148,
+      "step": 252
+    },
+    {
+      "epoch": 1.6282051282051282,
+      "grad_norm": 0.09652583778417714,
+      "learning_rate": 9.656026383620076e-06,
+      "loss": 1.1074,
+      "step": 254
+    },
+    {
+      "epoch": 1.641025641025641,
+      "grad_norm": 0.09448533541138639,
+      "learning_rate": 9.6482578719637e-06,
+      "loss": 1.1486,
+      "step": 256
+    },
+    {
+      "epoch": 1.6538461538461537,
+      "grad_norm": 0.09453430664055591,
+      "learning_rate": 9.640405809068743e-06,
+      "loss": 1.1197,
+      "step": 258
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.0952812616531032,
+      "learning_rate": 9.632470336074009e-06,
+      "loss": 1.1337,
+      "step": 260
+    },
+    {
+      "epoch": 1.6794871794871795,
+      "grad_norm": 0.09048018082770859,
+      "learning_rate": 9.624451595617588e-06,
+      "loss": 1.0885,
+      "step": 262
+    },
+    {
+      "epoch": 1.6923076923076923,
+      "grad_norm": 0.0922717302732401,
+      "learning_rate": 9.616349731834271e-06,
+      "loss": 1.1294,
+      "step": 264
+    },
+    {
+      "epoch": 1.7051282051282053,
+      "grad_norm": 0.09113342238000427,
+      "learning_rate": 9.608164890352977e-06,
+      "loss": 1.0871,
+      "step": 266
+    },
+    {
+      "epoch": 1.717948717948718,
+      "grad_norm": 0.10188653395954697,
+      "learning_rate": 9.599897218294122e-06,
+      "loss": 1.1237,
+      "step": 268
+    },
+    {
+      "epoch": 1.7307692307692308,
+      "grad_norm": 0.08946291041522332,
+      "learning_rate": 9.591546864266983e-06,
+      "loss": 1.1129,
+      "step": 270
+    },
+    {
+      "epoch": 1.7435897435897436,
+      "grad_norm": 0.092702242157672,
+      "learning_rate": 9.583113978367026e-06,
+      "loss": 1.1089,
+      "step": 272
+    },
+    {
+      "epoch": 1.7564102564102564,
+      "grad_norm": 0.1140491779513373,
+      "learning_rate": 9.574598712173202e-06,
+      "loss": 1.1286,
+      "step": 274
+    },
+    {
+      "epoch": 1.7692307692307692,
+      "grad_norm": 0.09516237353719291,
+      "learning_rate": 9.56600121874523e-06,
+      "loss": 1.1122,
+      "step": 276
+    },
+    {
+      "epoch": 1.782051282051282,
+      "grad_norm": 0.08916708413619781,
+      "learning_rate": 9.557321652620839e-06,
+      "loss": 1.1048,
+      "step": 278
+    },
+    {
+      "epoch": 1.7948717948717947,
+      "grad_norm": 0.09140805156925046,
+      "learning_rate": 9.548560169812997e-06,
+      "loss": 1.1058,
+      "step": 280
+    },
+    {
+      "epoch": 1.8076923076923077,
+      "grad_norm": 0.08683635001330178,
+      "learning_rate": 9.539716927807102e-06,
+      "loss": 1.0925,
+      "step": 282
+    },
+    {
+      "epoch": 1.8205128205128205,
+      "grad_norm": 0.09284148179598711,
+      "learning_rate": 9.530792085558151e-06,
+      "loss": 1.0948,
+      "step": 284
+    },
+    {
+      "epoch": 1.8333333333333335,
+      "grad_norm": 0.08800610945553744,
+      "learning_rate": 9.521785803487888e-06,
+      "loss": 1.1116,
+      "step": 286
+    },
+    {
+      "epoch": 1.8461538461538463,
+      "grad_norm": 0.08758546749473674,
+      "learning_rate": 9.512698243481914e-06,
+      "loss": 1.1059,
+      "step": 288
+    },
+    {
+      "epoch": 1.858974358974359,
+      "grad_norm": 0.08336608124209365,
+      "learning_rate": 9.50352956888678e-06,
+      "loss": 1.1015,
+      "step": 290
+    },
+    {
+      "epoch": 1.8717948717948718,
+      "grad_norm": 0.09199580396288136,
+      "learning_rate": 9.49427994450705e-06,
+      "loss": 1.0828,
+      "step": 292
+    },
+    {
+      "epoch": 1.8846153846153846,
+      "grad_norm": 0.5410940704298627,
+      "learning_rate": 9.484949536602343e-06,
+      "loss": 1.1412,
+      "step": 294
+    },
+    {
+      "epoch": 1.8974358974358974,
+      "grad_norm": 0.08913430120295451,
+      "learning_rate": 9.47553851288434e-06,
+      "loss": 1.1073,
+      "step": 296
+    },
+    {
+      "epoch": 1.9102564102564101,
+      "grad_norm": 0.09420167495815907,
+      "learning_rate": 9.466047042513767e-06,
+      "loss": 1.0957,
+      "step": 298
+    },
+    {
+      "epoch": 1.9230769230769231,
+      "grad_norm": 0.08189970955203785,
+      "learning_rate": 9.45647529609736e-06,
+      "loss": 1.0909,
+      "step": 300
+    },
+    {
+      "epoch": 1.935897435897436,
+      "grad_norm": 0.09065809775757692,
+      "learning_rate": 9.4468234456848e-06,
+      "loss": 1.0896,
+      "step": 302
+    },
+    {
+      "epoch": 1.9487179487179487,
+      "grad_norm": 0.08763498764491487,
+      "learning_rate": 9.437091664765611e-06,
+      "loss": 1.1099,
+      "step": 304
+    },
+    {
+      "epoch": 1.9615384615384617,
+      "grad_norm": 0.09257403574026254,
+      "learning_rate": 9.427280128266049e-06,
+      "loss": 1.1236,
+      "step": 306
+    },
+    {
+      "epoch": 1.9743589743589745,
+      "grad_norm": 0.08983923370086075,
+      "learning_rate": 9.41738901254596e-06,
+      "loss": 1.0909,
+      "step": 308
+    },
+    {
+      "epoch": 1.9871794871794872,
+      "grad_norm": 0.086289850522152,
+      "learning_rate": 9.4074184953956e-06,
+      "loss": 1.0942,
+      "step": 310
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.0874296283040965,
+      "learning_rate": 9.397368756032445e-06,
+      "loss": 1.0651,
+      "step": 312
+    },
+    {
+      "epoch": 2.0128205128205128,
+      "grad_norm": 0.0848953888966574,
+      "learning_rate": 9.38723997509798e-06,
+      "loss": 1.0569,
+      "step": 314
+    },
+    {
+      "epoch": 2.0256410256410255,
+      "grad_norm": 0.08790616172980993,
+      "learning_rate": 9.37703233465443e-06,
+      "loss": 1.035,
+      "step": 316
+    },
+    {
+      "epoch": 2.0384615384615383,
+      "grad_norm": 0.08376355574572536,
+      "learning_rate": 9.366746018181503e-06,
+      "loss": 1.0379,
+      "step": 318
+    },
+    {
+      "epoch": 2.051282051282051,
+      "grad_norm": 0.7353839032057593,
+      "learning_rate": 9.356381210573092e-06,
+      "loss": 1.0623,
+      "step": 320
+    },
+    {
+      "epoch": 2.064102564102564,
+      "grad_norm": 0.09158722362975955,
+      "learning_rate": 9.345938098133946e-06,
+      "loss": 1.0264,
+      "step": 322
+    },
+    {
+      "epoch": 2.076923076923077,
+      "grad_norm": 0.08819422670959466,
+      "learning_rate": 9.33541686857632e-06,
+      "loss": 1.0456,
+      "step": 324
+    },
+    {
+      "epoch": 2.08974358974359,
+      "grad_norm": 0.0905819981621342,
+      "learning_rate": 9.324817711016609e-06,
+      "loss": 1.0239,
+      "step": 326
+    },
+    {
+      "epoch": 2.1025641025641026,
+      "grad_norm": 0.08799589635983858,
+      "learning_rate": 9.31414081597194e-06,
+      "loss": 1.0498,
+      "step": 328
+    },
+    {
+      "epoch": 2.1153846153846154,
+      "grad_norm": 0.0847927160084877,
+      "learning_rate": 9.303386375356752e-06,
+      "loss": 1.0163,
+      "step": 330
+    },
+    {
+      "epoch": 2.128205128205128,
+      "grad_norm": 0.09169187613815971,
+      "learning_rate": 9.292554582479349e-06,
+      "loss": 1.0054,
+      "step": 332
+    },
+    {
+      "epoch": 2.141025641025641,
+      "grad_norm": 0.08905293788047657,
+      "learning_rate": 9.281645632038417e-06,
+      "loss": 1.062,
+      "step": 334
+    },
+    {
+      "epoch": 2.1538461538461537,
+      "grad_norm": 0.09229173633666073,
+      "learning_rate": 9.270659720119533e-06,
+      "loss": 1.039,
+      "step": 336
+    },
+    {
+      "epoch": 2.1666666666666665,
+      "grad_norm": 0.08430144514732368,
+      "learning_rate": 9.259597044191635e-06,
+      "loss": 1.0268,
+      "step": 338
+    },
+    {
+      "epoch": 2.1794871794871793,
+      "grad_norm": 0.08706427078942988,
+      "learning_rate": 9.248457803103476e-06,
+      "loss": 1.0038,
+      "step": 340
+    },
+    {
+      "epoch": 2.1923076923076925,
+      "grad_norm": 0.0851666955740436,
+      "learning_rate": 9.237242197080045e-06,
+      "loss": 1.0218,
+      "step": 342
+    },
+    {
+      "epoch": 2.2051282051282053,
+      "grad_norm": 0.08446573269728049,
+      "learning_rate": 9.225950427718974e-06,
+      "loss": 1.0254,
+      "step": 344
+    },
+    {
+      "epoch": 2.217948717948718,
+      "grad_norm": 0.08907279788471897,
+      "learning_rate": 9.21458269798691e-06,
+      "loss": 0.9916,
+      "step": 346
+    },
+    {
+      "epoch": 2.230769230769231,
+      "grad_norm": 0.09072043470187022,
+      "learning_rate": 9.203139212215868e-06,
+      "loss": 1.0103,
+      "step": 348
+    },
+    {
+      "epoch": 2.2435897435897436,
+      "grad_norm": 0.08618586552830075,
+      "learning_rate": 9.191620176099559e-06,
+      "loss": 0.9995,
+      "step": 350
+    },
+    {
+      "epoch": 2.2564102564102564,
+      "grad_norm": 0.09111342426909275,
+      "learning_rate": 9.180025796689692e-06,
+      "loss": 1.0292,
+      "step": 352
+    },
+    {
+      "epoch": 2.269230769230769,
+      "grad_norm": 0.2022564482536435,
+      "learning_rate": 9.168356282392253e-06,
+      "loss": 1.0226,
+      "step": 354
+    },
+    {
+      "epoch": 2.282051282051282,
+      "grad_norm": 0.1039362123101456,
+      "learning_rate": 9.156611842963753e-06,
+      "loss": 1.0152,
+      "step": 356
+    },
+    {
+      "epoch": 2.2948717948717947,
+      "grad_norm": 0.10035717927769394,
+      "learning_rate": 9.144792689507471e-06,
+      "loss": 1.0049,
+      "step": 358
+    },
+    {
+      "epoch": 2.3076923076923075,
+      "grad_norm": 0.08924064734394851,
+      "learning_rate": 9.132899034469648e-06,
+      "loss": 0.9962,
+      "step": 360
+    },
+    {
+      "epoch": 2.3205128205128207,
+      "grad_norm": 0.09443040073005612,
+      "learning_rate": 9.120931091635669e-06,
+      "loss": 0.9976,
+      "step": 362
+    },
+    {
+      "epoch": 2.3333333333333335,
+      "grad_norm": 0.09377508422363312,
+      "learning_rate": 9.108889076126226e-06,
+      "loss": 1.0306,
+      "step": 364
+    },
+    {
+      "epoch": 2.3461538461538463,
+      "grad_norm": 0.0895229930946655,
+      "learning_rate": 9.09677320439345e-06,
+      "loss": 1.0126,
+      "step": 366
+    },
+    {
+      "epoch": 2.358974358974359,
+      "grad_norm": 0.08795872722111464,
+      "learning_rate": 9.084583694217012e-06,
+      "loss": 0.9926,
+      "step": 368
+    },
+    {
+      "epoch": 2.371794871794872,
+      "grad_norm": 0.08704560136887454,
+      "learning_rate": 9.072320764700223e-06,
+      "loss": 0.9978,
+      "step": 370
+    },
+    {
+      "epoch": 2.3846153846153846,
+      "grad_norm": 0.0898387630341298,
+      "learning_rate": 9.059984636266082e-06,
+      "loss": 1.0042,
+      "step": 372
+    },
+    {
+      "epoch": 2.3974358974358974,
+      "grad_norm": 0.08357247562762515,
+      "learning_rate": 9.047575530653324e-06,
+      "loss": 1.0094,
+      "step": 374
+    },
+    {
+      "epoch": 2.41025641025641,
+      "grad_norm": 0.0843437057196144,
+      "learning_rate": 9.035093670912424e-06,
+      "loss": 0.9966,
+      "step": 376
+    },
+    {
+      "epoch": 2.423076923076923,
+      "grad_norm": 0.08357196997203281,
+      "learning_rate": 9.022539281401601e-06,
+      "loss": 1.0038,
+      "step": 378
+    },
+    {
+      "epoch": 2.435897435897436,
+      "grad_norm": 0.08859683961596204,
+      "learning_rate": 9.009912587782772e-06,
+      "loss": 1.0133,
+      "step": 380
+    },
+    {
+      "epoch": 2.448717948717949,
+      "grad_norm": 0.09024266497375917,
+      "learning_rate": 8.997213817017508e-06,
+      "loss": 0.9782,
+      "step": 382
+    },
+    {
+      "epoch": 2.4615384615384617,
+      "grad_norm": 0.0960929339414081,
+      "learning_rate": 8.984443197362938e-06,
+      "loss": 1.0013,
+      "step": 384
+    },
+    {
+      "epoch": 2.4743589743589745,
+      "grad_norm": 0.08862629313408348,
+      "learning_rate": 8.971600958367668e-06,
+      "loss": 1.0059,
+      "step": 386
+    },
+    {
+      "epoch": 2.4871794871794872,
+      "grad_norm": 0.09201716039902362,
+      "learning_rate": 8.958687330867634e-06,
+      "loss": 1.0263,
+      "step": 388
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.08694363384662504,
+      "learning_rate": 8.94570254698197e-06,
+      "loss": 1.0163,
+      "step": 390
+    },
+    {
+      "epoch": 2.5128205128205128,
+      "grad_norm": 0.09205164914341211,
+      "learning_rate": 8.932646840108818e-06,
+      "loss": 0.9865,
+      "step": 392
+    },
+    {
+      "epoch": 2.5256410256410255,
+      "grad_norm": 0.09081872370987605,
+      "learning_rate": 8.919520444921153e-06,
+      "loss": 0.9819,
+      "step": 394
+    },
+    {
+      "epoch": 2.5384615384615383,
+      "grad_norm": 0.08905442630582544,
+      "learning_rate": 8.906323597362547e-06,
+      "loss": 1.0171,
+      "step": 396
+    },
+    {
+      "epoch": 2.551282051282051,
+      "grad_norm": 0.08717951944686292,
+      "learning_rate": 8.893056534642938e-06,
+      "loss": 1.0244,
+      "step": 398
+    },
+    {
+      "epoch": 2.564102564102564,
+      "grad_norm": 0.09573458066741532,
+      "learning_rate": 8.879719495234363e-06,
+      "loss": 0.9848,
+      "step": 400
+    },
+    {
+      "epoch": 2.5769230769230766,
+      "grad_norm": 0.0898624666623644,
+      "learning_rate": 8.866312718866669e-06,
+      "loss": 0.982,
+      "step": 402
+    },
+    {
+      "epoch": 2.58974358974359,
+      "grad_norm": 0.09305658353350323,
+      "learning_rate": 8.852836446523213e-06,
+      "loss": 0.9742,
+      "step": 404
+    },
+    {
+      "epoch": 2.6025641025641026,
+      "grad_norm": 0.08663704229153721,
+      "learning_rate": 8.83929092043652e-06,
+      "loss": 0.9783,
+      "step": 406
+    },
+    {
+      "epoch": 2.6153846153846154,
+      "grad_norm": 0.08983846726156959,
+      "learning_rate": 8.825676384083936e-06,
+      "loss": 0.998,
+      "step": 408
+    },
+    {
+      "epoch": 2.628205128205128,
+      "grad_norm": 0.09388895481313425,
+      "learning_rate": 8.811993082183243e-06,
+      "loss": 1.0005,
+      "step": 410
+    },
+    {
+      "epoch": 2.641025641025641,
+      "grad_norm": 0.09226783931828283,
+      "learning_rate": 8.798241260688273e-06,
+      "loss": 1.0055,
+      "step": 412
+    },
+    {
+      "epoch": 2.6538461538461537,
+      "grad_norm": 0.09021054214140613,
+      "learning_rate": 8.784421166784476e-06,
+      "loss": 0.9981,
+      "step": 414
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.0860573848233807,
+      "learning_rate": 8.770533048884483e-06,
+      "loss": 1.0017,
+      "step": 416
+    },
+    {
+      "epoch": 2.6794871794871797,
+      "grad_norm": 0.0880124822318372,
+      "learning_rate": 8.756577156623636e-06,
+      "loss": 0.9834,
+      "step": 418
+    },
+    {
+      "epoch": 2.6923076923076925,
+      "grad_norm": 0.0867421199146975,
+      "learning_rate": 8.742553740855507e-06,
+      "loss": 0.9983,
+      "step": 420
+    },
+    {
+      "epoch": 2.7051282051282053,
+      "grad_norm": 0.09006077507273828,
+      "learning_rate": 8.728463053647382e-06,
+      "loss": 0.9702,
+      "step": 422
+    },
+    {
+      "epoch": 2.717948717948718,
+      "grad_norm": 0.08669250030062742,
+      "learning_rate": 8.71430534827574e-06,
+      "loss": 0.9952,
+      "step": 424
+    },
+    {
+      "epoch": 2.730769230769231,
+      "grad_norm": 0.09026424854741899,
+      "learning_rate": 8.700080879221689e-06,
+      "loss": 1.0054,
+      "step": 426
+    },
+    {
+      "epoch": 2.7435897435897436,
+      "grad_norm": 0.087975640704094,
+      "learning_rate": 8.685789902166395e-06,
+      "loss": 0.9845,
+      "step": 428
+    },
+    {
+      "epoch": 2.7564102564102564,
+      "grad_norm": 0.08642431755631451,
+      "learning_rate": 8.671432673986493e-06,
+      "loss": 0.9791,
+      "step": 430
+    },
+    {
+      "epoch": 2.769230769230769,
+      "grad_norm": 0.08649701419340423,
+      "learning_rate": 8.657009452749466e-06,
+      "loss": 0.9752,
+      "step": 432
+    },
+    {
+      "epoch": 2.782051282051282,
+      "grad_norm": 0.0879183947838203,
+      "learning_rate": 8.642520497709001e-06,
+      "loss": 0.9788,
+      "step": 434
+    },
+    {
+      "epoch": 2.7948717948717947,
+      "grad_norm": 0.08596416297337815,
+      "learning_rate": 8.627966069300332e-06,
+      "loss": 0.9807,
+      "step": 436
+    },
+    {
+      "epoch": 2.8076923076923075,
+      "grad_norm": 0.08918860363970792,
+      "learning_rate": 8.613346429135567e-06,
+      "loss": 0.9958,
+      "step": 438
+    },
+    {
+      "epoch": 2.8205128205128203,
+      "grad_norm": 0.08972585580799317,
+      "learning_rate": 8.598661839998972e-06,
+      "loss": 0.9895,
+      "step": 440
+    },
+    {
+      "epoch": 2.8333333333333335,
+      "grad_norm": 0.08703685151364528,
+      "learning_rate": 8.583912565842258e-06,
+      "loss": 0.9652,
+      "step": 442
+    },
+    {
+      "epoch": 2.8461538461538463,
+      "grad_norm": 0.08688465565057563,
+      "learning_rate": 8.569098871779828e-06,
+      "loss": 0.9984,
+      "step": 444
+    },
+    {
+      "epoch": 2.858974358974359,
+      "grad_norm": 0.08809758545326962,
+      "learning_rate": 8.554221024084019e-06,
+      "loss": 0.9905,
+      "step": 446
+    },
+    {
+      "epoch": 2.871794871794872,
+      "grad_norm": 0.08572911529655777,
+      "learning_rate": 8.539279290180315e-06,
+      "loss": 0.9692,
+      "step": 448
+    },
+    {
+      "epoch": 2.8846153846153846,
+      "grad_norm": 0.08836722634323343,
+      "learning_rate": 8.524273938642539e-06,
+      "loss": 0.9547,
+      "step": 450
+    },
+    {
+      "epoch": 2.8974358974358974,
+      "grad_norm": 0.09242854914045788,
+      "learning_rate": 8.509205239188017e-06,
+      "loss": 0.9838,
+      "step": 452
+    },
+    {
+      "epoch": 2.91025641025641,
+      "grad_norm": 0.08849881930024005,
+      "learning_rate": 8.494073462672743e-06,
+      "loss": 0.9615,
+      "step": 454
+    },
+    {
+      "epoch": 2.9230769230769234,
+      "grad_norm": 0.08854620618403236,
+      "learning_rate": 8.478878881086505e-06,
+      "loss": 0.9977,
+      "step": 456
+    },
+    {
+      "epoch": 2.935897435897436,
+      "grad_norm": 0.094665430731143,
+      "learning_rate": 8.463621767547998e-06,
+      "loss": 0.9927,
+      "step": 458
+    },
+    {
+      "epoch": 2.948717948717949,
+      "grad_norm": 0.09196410792880014,
+      "learning_rate": 8.448302396299906e-06,
+      "loss": 1.0113,
+      "step": 460
+    },
+    {
+      "epoch": 2.9615384615384617,
+      "grad_norm": 0.09036486236859728,
+      "learning_rate": 8.432921042703985e-06,
+      "loss": 0.9457,
+      "step": 462
+    },
+    {
+      "epoch": 2.9743589743589745,
+      "grad_norm": 0.08576032950610284,
+      "learning_rate": 8.417477983236107e-06,
+      "loss": 0.9645,
+      "step": 464
+    },
+    {
+      "epoch": 2.9871794871794872,
+      "grad_norm": 0.08403590001526823,
+      "learning_rate": 8.401973495481289e-06,
+      "loss": 0.9544,
+      "step": 466
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.09355532269950335,
+      "learning_rate": 8.386407858128707e-06,
+      "loss": 0.9719,
+      "step": 468
+    },
+    {
+      "epoch": 3.0128205128205128,
+      "grad_norm": 0.08685232548889178,
+      "learning_rate": 8.370781350966683e-06,
+      "loss": 0.8933,
+      "step": 470
+    },
+    {
+      "epoch": 3.0256410256410255,
+      "grad_norm": 0.10917681684685593,
+      "learning_rate": 8.355094254877665e-06,
+      "loss": 0.9222,
+      "step": 472
+    },
+    {
+      "epoch": 3.0384615384615383,
+      "grad_norm": 0.09821414680349456,
+      "learning_rate": 8.339346851833163e-06,
+      "loss": 0.9187,
+      "step": 474
+    },
+    {
+      "epoch": 3.051282051282051,
+      "grad_norm": 0.0953257584501641,
+      "learning_rate": 8.323539424888695e-06,
+      "loss": 0.9068,
+      "step": 476
+    },
+    {
+      "epoch": 3.064102564102564,
+      "grad_norm": 0.10096821936698265,
+      "learning_rate": 8.30767225817869e-06,
+      "loss": 0.9005,
+      "step": 478
+    },
+    {
+      "epoch": 3.076923076923077,
+      "grad_norm": 0.09745049198474258,
+      "learning_rate": 8.291745636911382e-06,
+      "loss": 0.8955,
+      "step": 480
+    },
+    {
+      "epoch": 3.08974358974359,
+      "grad_norm": 0.09581071499737452,
+      "learning_rate": 8.27575984736369e-06,
+      "loss": 0.9034,
+      "step": 482
+    },
+    {
+      "epoch": 3.1025641025641026,
+      "grad_norm": 0.09048589565605356,
+      "learning_rate": 8.259715176876069e-06,
+      "loss": 0.8964,
+      "step": 484
+    },
+    {
+      "epoch": 3.1153846153846154,
+      "grad_norm": 0.09408149538192938,
+      "learning_rate": 8.243611913847337e-06,
+      "loss": 0.9157,
+      "step": 486
+    },
+    {
+      "epoch": 3.128205128205128,
+      "grad_norm": 0.0947487050346647,
+      "learning_rate": 8.2274503477295e-06,
+      "loss": 0.9053,
+      "step": 488
+    },
+    {
+      "epoch": 3.141025641025641,
+      "grad_norm": 0.09366500902355888,
+      "learning_rate": 8.211230769022552e-06,
+      "loss": 0.8925,
+      "step": 490
+    },
+    {
+      "epoch": 3.1538461538461537,
+      "grad_norm": 0.09167161100151112,
+      "learning_rate": 8.19495346926924e-06,
+      "loss": 0.9165,
+      "step": 492
+    },
+    {
+      "epoch": 3.1666666666666665,
+      "grad_norm": 0.09307041831758973,
+      "learning_rate": 8.178618741049841e-06,
+      "loss": 0.8989,
+      "step": 494
+    },
+    {
+      "epoch": 3.1794871794871793,
+      "grad_norm": 0.09585560939367876,
+      "learning_rate": 8.162226877976886e-06,
+      "loss": 0.9147,
+      "step": 496
+    },
+    {
+      "epoch": 3.1923076923076925,
+      "grad_norm": 0.09180060088840723,
+      "learning_rate": 8.145778174689897e-06,
+      "loss": 0.8882,
+      "step": 498
+    },
+    {
+      "epoch": 3.2051282051282053,
+      "grad_norm": 0.09609878354099273,
+      "learning_rate": 8.129272926850079e-06,
+      "loss": 0.8744,
+      "step": 500
+    },
+    {
+      "epoch": 3.217948717948718,
+      "grad_norm": 0.09691473472460625,
+      "learning_rate": 8.112711431135014e-06,
+      "loss": 0.8736,
+      "step": 502
+    },
+    {
+      "epoch": 3.230769230769231,
+      "grad_norm": 0.09236636322834278,
+      "learning_rate": 8.096093985233323e-06,
+      "loss": 0.848,
+      "step": 504
+    },
+    {
+      "epoch": 3.2435897435897436,
+      "grad_norm": 0.09704717599279773,
+      "learning_rate": 8.079420887839316e-06,
+      "loss": 0.8844,
+      "step": 506
+    },
+    {
+      "epoch": 3.2564102564102564,
+      "grad_norm": 0.09939291409466518,
+      "learning_rate": 8.062692438647628e-06,
+      "loss": 0.8866,
+      "step": 508
+    },
+    {
+      "epoch": 3.269230769230769,
+      "grad_norm": 0.09353962075083472,
+      "learning_rate": 8.045908938347828e-06,
+      "loss": 0.8742,
+      "step": 510
+    },
+    {
+      "epoch": 3.282051282051282,
+      "grad_norm": 0.09465310178443197,
+      "learning_rate": 8.029070688619013e-06,
+      "loss": 0.8833,
+      "step": 512
+    },
+    {
+      "epoch": 3.2948717948717947,
+      "grad_norm": 0.09443637715651476,
+      "learning_rate": 8.012177992124385e-06,
+      "loss": 0.8794,
+      "step": 514
+    },
+    {
+      "epoch": 3.3076923076923075,
+      "grad_norm": 0.09728431520292821,
+      "learning_rate": 7.995231152505815e-06,
+      "loss": 0.8732,
+      "step": 516
+    },
+    {
+      "epoch": 3.3205128205128207,
+      "grad_norm": 0.09428493650909285,
+      "learning_rate": 7.978230474378383e-06,
+      "loss": 0.8597,
+      "step": 518
+    },
+    {
+      "epoch": 3.3333333333333335,
+      "grad_norm": 0.09850772889396305,
+      "learning_rate": 7.961176263324902e-06,
+      "loss": 0.8624,
+      "step": 520
+    },
+    {
+      "epoch": 3.3461538461538463,
+      "grad_norm": 0.09087037549609535,
+      "learning_rate": 7.944068825890424e-06,
+      "loss": 0.8821,
+      "step": 522
+    },
+    {
+      "epoch": 3.358974358974359,
+      "grad_norm": 0.09180369503983593,
+      "learning_rate": 7.92690846957673e-06,
+      "loss": 0.8688,
+      "step": 524
+    },
+    {
+      "epoch": 3.371794871794872,
+      "grad_norm": 0.09491604280681391,
+      "learning_rate": 7.909695502836814e-06,
+      "loss": 0.8647,
+      "step": 526
+    },
+    {
+      "epoch": 3.3846153846153846,
+      "grad_norm": 0.09921876854138406,
+      "learning_rate": 7.892430235069317e-06,
+      "loss": 0.8869,
+      "step": 528
+    },
+    {
+      "epoch": 3.3974358974358974,
+      "grad_norm": 0.09457741703712105,
+      "learning_rate": 7.875112976612984e-06,
+      "loss": 0.8639,
+      "step": 530
+    },
+    {
+      "epoch": 3.41025641025641,
+      "grad_norm": 0.09583219613481893,
+      "learning_rate": 7.857744038741076e-06,
+      "loss": 0.8805,
+      "step": 532
+    },
+    {
+      "epoch": 3.423076923076923,
+      "grad_norm": 0.09260516206658106,
+      "learning_rate": 7.84032373365578e-06,
+      "loss": 0.8603,
+      "step": 534
+    },
+    {
+      "epoch": 3.435897435897436,
+      "grad_norm": 0.09932108403192164,
+      "learning_rate": 7.822852374482597e-06,
+      "loss": 0.8658,
+      "step": 536
+    },
+    {
+      "epoch": 3.448717948717949,
+      "grad_norm": 0.09728531208245553,
+      "learning_rate": 7.805330275264707e-06,
+      "loss": 0.8536,
+      "step": 538
+    },
+    {
+      "epoch": 3.4615384615384617,
+      "grad_norm": 0.09952432033061036,
+      "learning_rate": 7.787757750957335e-06,
+      "loss": 0.8763,
+      "step": 540
+    },
+    {
+      "epoch": 3.4743589743589745,
+      "grad_norm": 0.09845329832112057,
+      "learning_rate": 7.77013511742208e-06,
+      "loss": 0.8658,
+      "step": 542
+    },
+    {
+      "epoch": 3.4871794871794872,
+      "grad_norm": 0.10349699075619775,
+      "learning_rate": 7.752462691421245e-06,
+      "loss": 0.8538,
+      "step": 544
+    },
+    {
+      "epoch": 3.5,
+      "grad_norm": 0.15469316317671902,
+      "learning_rate": 7.734740790612137e-06,
+      "loss": 0.8644,
+      "step": 546
+    },
+    {
+      "epoch": 3.5128205128205128,
+      "grad_norm": 0.09649309700047885,
+      "learning_rate": 7.716969733541357e-06,
+      "loss": 0.8755,
+      "step": 548
+    },
+    {
+      "epoch": 3.5256410256410255,
+      "grad_norm": 0.09860823779259517,
+      "learning_rate": 7.699149839639086e-06,
+      "loss": 0.8471,
+      "step": 550
+    },
+    {
+      "epoch": 3.5384615384615383,
+      "grad_norm": 0.09867635522074884,
+      "learning_rate": 7.681281429213328e-06,
+      "loss": 0.8512,
+      "step": 552
+    },
+    {
+      "epoch": 3.551282051282051,
+      "grad_norm": 0.09856703594780034,
+      "learning_rate": 7.663364823444157e-06,
+      "loss": 0.8581,
+      "step": 554
+    },
+    {
+      "epoch": 3.564102564102564,
+      "grad_norm": 0.10120010505390695,
+      "learning_rate": 7.645400344377953e-06,
+      "loss": 0.8647,
+      "step": 556
+    },
+    {
+      "epoch": 3.5769230769230766,
+      "grad_norm": 0.09353647856294549,
+      "learning_rate": 7.627388314921602e-06,
+      "loss": 0.8563,
+      "step": 558
+    },
+    {
+      "epoch": 3.58974358974359,
+      "grad_norm": 0.097727849555005,
+      "learning_rate": 7.609329058836694e-06,
+      "loss": 0.8629,
+      "step": 560
+    },
+    {
+      "epoch": 3.6025641025641026,
+      "grad_norm": 0.09185843649741915,
+      "learning_rate": 7.59122290073371e-06,
+      "loss": 0.8517,
+      "step": 562
+    },
+    {
+      "epoch": 3.6153846153846154,
+      "grad_norm": 0.16467906411387448,
+      "learning_rate": 7.5730701660661795e-06,
+      "loss": 0.8588,
+      "step": 564
+    },
+    {
+      "epoch": 3.628205128205128,
+      "grad_norm": 0.10490078157659109,
+      "learning_rate": 7.554871181124836e-06,
+      "loss": 0.8916,
+      "step": 566
+    },
+    {
+      "epoch": 3.641025641025641,
+      "grad_norm": 0.09862237486460196,
+      "learning_rate": 7.536626273031747e-06,
+      "loss": 0.8486,
+      "step": 568
+    },
+    {
+      "epoch": 3.6538461538461537,
+      "grad_norm": 0.09855168103779419,
+      "learning_rate": 7.5183357697344395e-06,
+      "loss": 0.8532,
+      "step": 570
+    },
+    {
+      "epoch": 3.6666666666666665,
+      "grad_norm": 0.09943631897387811,
+      "learning_rate": 7.500000000000001e-06,
+      "loss": 0.8643,
+      "step": 572
+    },
+    {
+      "epoch": 3.6794871794871797,
+      "grad_norm": 0.09470558794565637,
+      "learning_rate": 7.481619293409173e-06,
+      "loss": 0.8705,
+      "step": 574
+    },
+    {
+      "epoch": 3.6923076923076925,
+      "grad_norm": 0.09434833275033037,
+      "learning_rate": 7.4631939803504215e-06,
+      "loss": 0.8597,
+      "step": 576
+    },
+    {
+      "epoch": 3.7051282051282053,
+      "grad_norm": 0.09852625213361811,
+      "learning_rate": 7.44472439201401e-06,
+      "loss": 0.8665,
+      "step": 578
+    },
+    {
+      "epoch": 3.717948717948718,
+      "grad_norm": 0.09522012579767557,
+      "learning_rate": 7.426210860386032e-06,
+      "loss": 0.8373,
+      "step": 580
+    },
+    {
+      "epoch": 3.730769230769231,
+      "grad_norm": 0.09872214935386595,
+      "learning_rate": 7.407653718242449e-06,
+      "loss": 0.8266,
+      "step": 582
+    },
+    {
+      "epoch": 3.7435897435897436,
+      "grad_norm": 0.09611754066886699,
+      "learning_rate": 7.3890532991431174e-06,
+      "loss": 0.8422,
+      "step": 584
+    },
+    {
+      "epoch": 3.7564102564102564,
+      "grad_norm": 0.09430702389773353,
+      "learning_rate": 7.370409937425781e-06,
+      "loss": 0.8349,
+      "step": 586
+    },
+    {
+      "epoch": 3.769230769230769,
+      "grad_norm": 0.10000120202753963,
+      "learning_rate": 7.3517239682000675e-06,
+      "loss": 0.8589,
+      "step": 588
+    },
+    {
+      "epoch": 3.782051282051282,
+      "grad_norm": 0.09477208728170344,
+      "learning_rate": 7.332995727341462e-06,
+      "loss": 0.8591,
+      "step": 590
+    },
+    {
+      "epoch": 3.7948717948717947,
+      "grad_norm": 0.09696166000717225,
+      "learning_rate": 7.314225551485273e-06,
+      "loss": 0.8397,
+      "step": 592
+    },
+    {
+      "epoch": 3.8076923076923075,
+      "grad_norm": 0.09621353397155066,
+      "learning_rate": 7.295413778020579e-06,
+      "loss": 0.8166,
+      "step": 594
+    },
+    {
+      "epoch": 3.8205128205128203,
+      "grad_norm": 0.09692687114207367,
+      "learning_rate": 7.276560745084167e-06,
+      "loss": 0.8521,
+      "step": 596
+    },
+    {
+      "epoch": 3.8333333333333335,
+      "grad_norm": 0.09885126357081214,
+      "learning_rate": 7.257666791554448e-06,
+      "loss": 0.8416,
+      "step": 598
+    },
+    {
+      "epoch": 3.8461538461538463,
+      "grad_norm": 0.10239714078021848,
+      "learning_rate": 7.2387322570453724e-06,
+      "loss": 0.8324,
+      "step": 600
+    },
+    {
+      "epoch": 3.858974358974359,
+      "grad_norm": 0.11251898784242197,
+      "learning_rate": 7.219757481900325e-06,
+      "loss": 0.835,
+      "step": 602
+    },
+    {
+      "epoch": 3.871794871794872,
+      "grad_norm": 0.1005799166719958,
+      "learning_rate": 7.2007428071860045e-06,
+      "loss": 0.8035,
+      "step": 604
+    },
+    {
+      "epoch": 3.8846153846153846,
+      "grad_norm": 0.10103534145014936,
+      "learning_rate": 7.181688574686292e-06,
+      "loss": 0.8709,
+      "step": 606
+    },
+    {
+      "epoch": 3.8974358974358974,
+      "grad_norm": 0.10027552225015914,
+      "learning_rate": 7.162595126896111e-06,
+      "loss": 0.8319,
+      "step": 608
+    },
+    {
+      "epoch": 3.91025641025641,
+      "grad_norm": 0.10075780749863547,
+      "learning_rate": 7.143462807015271e-06,
+      "loss": 0.8323,
+      "step": 610
+    },
+    {
+      "epoch": 3.9230769230769234,
+      "grad_norm": 0.09472929060217589,
+      "learning_rate": 7.1242919589422974e-06,
+      "loss": 0.8185,
+      "step": 612
+    },
+    {
+      "epoch": 3.935897435897436,
+      "grad_norm": 0.09472378350788888,
+      "learning_rate": 7.105082927268247e-06,
+      "loss": 0.8304,
+      "step": 614
+    },
+    {
+      "epoch": 3.948717948717949,
+      "grad_norm": 0.10337359146731352,
+      "learning_rate": 7.085836057270521e-06,
+      "loss": 0.8174,
+      "step": 616
+    },
+    {
+      "epoch": 3.9615384615384617,
+      "grad_norm": 0.0983672088113577,
+      "learning_rate": 7.066551694906651e-06,
+      "loss": 0.8322,
+      "step": 618
+    },
+    {
+      "epoch": 3.9743589743589745,
+      "grad_norm": 0.1019500525911841,
+      "learning_rate": 7.047230186808085e-06,
+      "loss": 0.8021,
+      "step": 620
+    },
+    {
+      "epoch": 3.9871794871794872,
+      "grad_norm": 0.09750574300751329,
+      "learning_rate": 7.027871880273959e-06,
+      "loss": 0.7983,
+      "step": 622
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.10208128186441004,
+      "learning_rate": 7.008477123264849e-06,
+      "loss": 0.8239,
+      "step": 624
+    },
+    {
+      "epoch": 4.012820512820513,
+      "grad_norm": 0.10734300522197977,
+      "learning_rate": 6.989046264396516e-06,
+      "loss": 0.7678,
+      "step": 626
+    },
+    {
+      "epoch": 4.0256410256410255,
+      "grad_norm": 0.102980617519378,
+      "learning_rate": 6.96957965293365e-06,
+      "loss": 0.7377,
+      "step": 628
+    },
+    {
+      "epoch": 4.038461538461538,
+      "grad_norm": 0.12545611352143843,
+      "learning_rate": 6.9500776387835785e-06,
+      "loss": 0.7581,
+      "step": 630
+    },
+    {
+      "epoch": 4.051282051282051,
+      "grad_norm": 0.122707057481331,
+      "learning_rate": 6.9305405724899876e-06,
+      "loss": 0.7399,
+      "step": 632
+    },
+    {
+      "epoch": 4.064102564102564,
+      "grad_norm": 0.11397293701236821,
+      "learning_rate": 6.91096880522661e-06,
+      "loss": 0.7447,
+      "step": 634
+    },
+    {
+      "epoch": 4.076923076923077,
+      "grad_norm": 0.13487306338562802,
+      "learning_rate": 6.891362688790925e-06,
+      "loss": 0.7546,
+      "step": 636
+    },
+    {
+      "epoch": 4.089743589743589,
+      "grad_norm": 0.10896326697255375,
+      "learning_rate": 6.871722575597829e-06,
+      "loss": 0.7579,
+      "step": 638
+    },
+    {
+      "epoch": 4.102564102564102,
+      "grad_norm": 0.11165624709106642,
+      "learning_rate": 6.8520488186733e-06,
+      "loss": 0.7517,
+      "step": 640
+    },
+    {
+      "epoch": 4.115384615384615,
+      "grad_norm": 0.11518303790398043,
+      "learning_rate": 6.832341771648057e-06,
+      "loss": 0.7459,
+      "step": 642
+    },
+    {
+      "epoch": 4.128205128205128,
+      "grad_norm": 0.11119475069076129,
+      "learning_rate": 6.812601788751192e-06,
+      "loss": 0.7825,
+      "step": 644
+    },
+    {
+      "epoch": 4.141025641025641,
+      "grad_norm": 0.10743202492055963,
+      "learning_rate": 6.792829224803816e-06,
+      "loss": 0.7395,
+      "step": 646
+    },
+    {
+      "epoch": 4.153846153846154,
+      "grad_norm": 0.10582985557685172,
+      "learning_rate": 6.773024435212678e-06,
+      "loss": 0.7637,
+      "step": 648
+    },
+    {
+      "epoch": 4.166666666666667,
+      "grad_norm": 0.10835455281881788,
+      "learning_rate": 6.753187775963773e-06,
+      "loss": 0.7576,
+      "step": 650
+    },
+    {
+      "epoch": 4.17948717948718,
+      "grad_norm": 0.1107213708183791,
+      "learning_rate": 6.733319603615941e-06,
+      "loss": 0.7519,
+      "step": 652
+    },
+    {
+      "epoch": 4.1923076923076925,
+      "grad_norm": 0.11143239841237282,
+      "learning_rate": 6.713420275294467e-06,
+      "loss": 0.7421,
+      "step": 654
+    },
+    {
+      "epoch": 4.205128205128205,
+      "grad_norm": 0.10135913047939792,
+      "learning_rate": 6.693490148684654e-06,
+      "loss": 0.7503,
+      "step": 656
+    },
+    {
+      "epoch": 4.217948717948718,
+      "grad_norm": 0.10935890173613132,
+      "learning_rate": 6.673529582025398e-06,
+      "loss": 0.7469,
+      "step": 658
+    },
+    {
+      "epoch": 4.230769230769231,
+      "grad_norm": 0.10682800250997206,
+      "learning_rate": 6.653538934102743e-06,
+      "loss": 0.7519,
+      "step": 660
+    },
+    {
+      "epoch": 4.243589743589744,
+      "grad_norm": 0.11174312070750286,
+      "learning_rate": 6.633518564243442e-06,
+      "loss": 0.7388,
+      "step": 662
+    },
+    {
+      "epoch": 4.256410256410256,
+      "grad_norm": 0.10996882792698588,
+      "learning_rate": 6.6134688323084884e-06,
+      "loss": 0.735,
+      "step": 664
+    },
+    {
+      "epoch": 4.269230769230769,
+      "grad_norm": 0.11513381552989353,
+      "learning_rate": 6.593390098686653e-06,
+      "loss": 0.7266,
+      "step": 666
+    },
+    {
+      "epoch": 4.282051282051282,
+      "grad_norm": 0.10383307615951057,
+      "learning_rate": 6.573282724288001e-06,
+      "loss": 0.7354,
+      "step": 668
+    },
+    {
+      "epoch": 4.294871794871795,
+      "grad_norm": 0.10064526192695795,
+      "learning_rate": 6.553147070537413e-06,
+      "loss": 0.7316,
+      "step": 670
+    },
+    {
+      "epoch": 4.3076923076923075,
+      "grad_norm": 0.10546529880700707,
+      "learning_rate": 6.532983499368078e-06,
+      "loss": 0.7345,
+      "step": 672
+    },
+    {
+      "epoch": 4.32051282051282,
+      "grad_norm": 0.10452514955349174,
+      "learning_rate": 6.512792373215e-06,
+      "loss": 0.7552,
+      "step": 674
+    },
+    {
+      "epoch": 4.333333333333333,
+      "grad_norm": 0.10501851155628895,
+      "learning_rate": 6.492574055008474e-06,
+      "loss": 0.715,
+      "step": 676
+    },
+    {
+      "epoch": 4.346153846153846,
+      "grad_norm": 0.10411036896818421,
+      "learning_rate": 6.472328908167562e-06,
+      "loss": 0.729,
+      "step": 678
+    },
+    {
+      "epoch": 4.358974358974359,
+      "grad_norm": 0.11127049713049718,
+      "learning_rate": 6.452057296593568e-06,
+      "loss": 0.744,
+      "step": 680
+    },
+    {
+      "epoch": 4.371794871794872,
+      "grad_norm": 0.12676881136201423,
+      "learning_rate": 6.431759584663492e-06,
+      "loss": 0.7588,
+      "step": 682
+    },
+    {
+      "epoch": 4.384615384615385,
+      "grad_norm": 0.105870619579206,
+      "learning_rate": 6.411436137223479e-06,
+      "loss": 0.7247,
+      "step": 684
+    },
+    {
+      "epoch": 4.397435897435898,
+      "grad_norm": 0.10374120826824249,
+      "learning_rate": 6.391087319582264e-06,
+      "loss": 0.7309,
+      "step": 686
+    },
+    {
+      "epoch": 4.410256410256411,
+      "grad_norm": 0.10865846153479375,
+      "learning_rate": 6.370713497504607e-06,
+      "loss": 0.7482,
+      "step": 688
+    },
+    {
+      "epoch": 4.423076923076923,
+      "grad_norm": 0.11160085810481411,
+      "learning_rate": 6.350315037204714e-06,
+      "loss": 0.7254,
+      "step": 690
+    },
+    {
+      "epoch": 4.435897435897436,
+      "grad_norm": 0.10544486611527323,
+      "learning_rate": 6.329892305339659e-06,
+      "loss": 0.7053,
+      "step": 692
+    },
+    {
+      "epoch": 4.448717948717949,
+      "grad_norm": 0.10611707780750092,
+      "learning_rate": 6.309445669002787e-06,
+      "loss": 0.7078,
+      "step": 694
+    },
+    {
+      "epoch": 4.461538461538462,
+      "grad_norm": 0.10588157071847835,
+      "learning_rate": 6.288975495717124e-06,
+      "loss": 0.7412,
+      "step": 696
+    },
+    {
+      "epoch": 4.4743589743589745,
+      "grad_norm": 0.10785564192135899,
+      "learning_rate": 6.268482153428763e-06,
+      "loss": 0.7289,
+      "step": 698
+    },
+    {
+      "epoch": 4.487179487179487,
+      "grad_norm": 0.10456174831291559,
+      "learning_rate": 6.247966010500258e-06,
+      "loss": 0.7233,
+      "step": 700
+    },
+    {
+      "epoch": 4.5,
+      "grad_norm": 0.10739198046560715,
+      "learning_rate": 6.227427435703997e-06,
+      "loss": 0.7308,
+      "step": 702
+    },
+    {
+      "epoch": 4.512820512820513,
+      "grad_norm": 0.11062331534549659,
+      "learning_rate": 6.206866798215571e-06,
+      "loss": 0.7188,
+      "step": 704
+    },
+    {
+      "epoch": 4.5256410256410255,
+      "grad_norm": 0.1120412879852177,
+      "learning_rate": 6.186284467607149e-06,
+      "loss": 0.7149,
+      "step": 706
+    },
+    {
+      "epoch": 4.538461538461538,
+      "grad_norm": 0.10581044212948068,
+      "learning_rate": 6.165680813840822e-06,
+      "loss": 0.7286,
+      "step": 708
+    },
+    {
+      "epoch": 4.551282051282051,
+      "grad_norm": 0.10581925858925155,
+      "learning_rate": 6.1450562072619635e-06,
+      "loss": 0.6854,
+      "step": 710
+    },
+    {
+      "epoch": 4.564102564102564,
+      "grad_norm": 0.11850925857398317,
+      "learning_rate": 6.124411018592568e-06,
+      "loss": 0.7215,
+      "step": 712
+    },
+    {
+      "epoch": 4.576923076923077,
+      "grad_norm": 0.12029983367038724,
+      "learning_rate": 6.103745618924587e-06,
+      "loss": 0.7142,
+      "step": 714
+    },
+    {
+      "epoch": 4.589743589743589,
+      "grad_norm": 0.10567103079137533,
+      "learning_rate": 6.0830603797132574e-06,
+      "loss": 0.7162,
+      "step": 716
+    },
+    {
+      "epoch": 4.602564102564102,
+      "grad_norm": 0.10836686741052724,
+      "learning_rate": 6.0623556727704306e-06,
+      "loss": 0.7165,
+      "step": 718
+    },
+    {
+      "epoch": 4.615384615384615,
+      "grad_norm": 0.11249604087548312,
+      "learning_rate": 6.041631870257882e-06,
+      "loss": 0.7383,
+      "step": 720
+    },
+    {
+      "epoch": 4.628205128205128,
+      "grad_norm": 0.1082063599668396,
+      "learning_rate": 6.020889344680627e-06,
+      "loss": 0.6952,
+      "step": 722
+    },
+    {
+      "epoch": 4.641025641025641,
+      "grad_norm": 0.10282892185990167,
+      "learning_rate": 6.000128468880223e-06,
+      "loss": 0.7167,
+      "step": 724
+    },
+    {
+      "epoch": 4.653846153846154,
+      "grad_norm": 0.14775806059988206,
+      "learning_rate": 5.979349616028067e-06,
+      "loss": 0.7015,
+      "step": 726
+    },
+    {
+      "epoch": 4.666666666666667,
+      "grad_norm": 0.1146560861251404,
+      "learning_rate": 5.958553159618693e-06,
+      "loss": 0.7213,
+      "step": 728
+    },
+    {
+      "epoch": 4.67948717948718,
+      "grad_norm": 0.10561771243314702,
+      "learning_rate": 5.937739473463047e-06,
+      "loss": 0.7296,
+      "step": 730
+    },
+    {
+      "epoch": 4.6923076923076925,
+      "grad_norm": 0.1030552904773058,
+      "learning_rate": 5.916908931681781e-06,
+      "loss": 0.7123,
+      "step": 732
+    },
+    {
+      "epoch": 4.705128205128205,
+      "grad_norm": 0.11007539115142843,
+      "learning_rate": 5.896061908698521e-06,
+      "loss": 0.7048,
+      "step": 734
+    },
+    {
+      "epoch": 4.717948717948718,
+      "grad_norm": 0.11416376306689043,
+      "learning_rate": 5.8751987792331365e-06,
+      "loss": 0.7137,
+      "step": 736
+    },
+    {
+      "epoch": 4.730769230769231,
+      "grad_norm": 0.10152180107259362,
+      "learning_rate": 5.854319918295012e-06,
+      "loss": 0.7051,
+      "step": 738
+    },
+    {
+      "epoch": 4.743589743589744,
+      "grad_norm": 0.11206883891832514,
+      "learning_rate": 5.833425701176294e-06,
+      "loss": 0.6923,
+      "step": 740
+    },
+    {
+      "epoch": 4.756410256410256,
+      "grad_norm": 0.10804199427828234,
+      "learning_rate": 5.812516503445158e-06,
+      "loss": 0.6955,
+      "step": 742
+    },
+    {
+      "epoch": 4.769230769230769,
+      "grad_norm": 0.10618536471151145,
+      "learning_rate": 5.79159270093905e-06,
+      "loss": 0.7051,
+      "step": 744
+    },
+    {
+      "epoch": 4.782051282051282,
+      "grad_norm": 0.112445946670164,
+      "learning_rate": 5.770654669757935e-06,
+      "loss": 0.6862,
+      "step": 746
+    },
+    {
+      "epoch": 4.794871794871795,
+      "grad_norm": 0.10623939719616725,
+      "learning_rate": 5.749702786257529e-06,
+      "loss": 0.7021,
+      "step": 748
+    },
+    {
+      "epoch": 4.8076923076923075,
+      "grad_norm": 0.11066503537728437,
+      "learning_rate": 5.7287374270425475e-06,
+      "loss": 0.7083,
+      "step": 750
+    },
+    {
+      "epoch": 4.82051282051282,
+      "grad_norm": 0.11956485729401434,
+      "learning_rate": 5.707758968959923e-06,
+      "loss": 0.7052,
+      "step": 752
+    },
+    {
+      "epoch": 4.833333333333333,
+      "grad_norm": 0.11607859173183654,
+      "learning_rate": 5.686767789092041e-06,
+      "loss": 0.7114,
+      "step": 754
+    },
+    {
+      "epoch": 4.846153846153846,
+      "grad_norm": 0.10875829497210732,
+      "learning_rate": 5.6657642647499545e-06,
+      "loss": 0.7159,
+      "step": 756
+    },
+    {
+      "epoch": 4.858974358974359,
+      "grad_norm": 0.10952816111674243,
+      "learning_rate": 5.644748773466606e-06,
+      "loss": 0.7036,
+      "step": 758
+    },
+    {
+      "epoch": 4.871794871794872,
+      "grad_norm": 0.10684780948629531,
+      "learning_rate": 5.62372169299004e-06,
+      "loss": 0.7225,
+      "step": 760
+    },
+    {
+      "epoch": 4.884615384615385,
+      "grad_norm": 0.1047662448976948,
+      "learning_rate": 5.6026834012766155e-06,
+      "loss": 0.6805,
+      "step": 762
+    },
+    {
+      "epoch": 4.897435897435898,
+      "grad_norm": 0.10955003114927836,
+      "learning_rate": 5.581634276484211e-06,
+      "loss": 0.6792,
+      "step": 764
+    },
+    {
+      "epoch": 4.910256410256411,
+      "grad_norm": 0.10878018550941551,
+      "learning_rate": 5.560574696965425e-06,
+      "loss": 0.6921,
+      "step": 766
+    },
+    {
+      "epoch": 4.923076923076923,
+      "grad_norm": 0.11093790171018045,
+      "learning_rate": 5.539505041260779e-06,
+      "loss": 0.6956,
+      "step": 768
+    },
+    {
+      "epoch": 4.935897435897436,
+      "grad_norm": 0.1115655815203421,
+      "learning_rate": 5.518425688091906e-06,
+      "loss": 0.7024,
+      "step": 770
+    },
+    {
+      "epoch": 4.948717948717949,
+      "grad_norm": 0.1131005595068268,
+      "learning_rate": 5.497337016354757e-06,
+      "loss": 0.7148,
+      "step": 772
+    },
+    {
+      "epoch": 4.961538461538462,
+      "grad_norm": 0.11347516336979874,
+      "learning_rate": 5.476239405112775e-06,
+      "loss": 0.6816,
+      "step": 774
+    },
+    {
+      "epoch": 4.9743589743589745,
+      "grad_norm": 0.10898186232548415,
+      "learning_rate": 5.45513323359009e-06,
+      "loss": 0.7273,
+      "step": 776
+    },
+    {
+      "epoch": 4.987179487179487,
+      "grad_norm": 0.11549198646562549,
+      "learning_rate": 5.434018881164702e-06,
+      "loss": 0.6917,
+      "step": 778
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.10772346133987304,
+      "learning_rate": 5.412896727361663e-06,
+      "loss": 0.6863,
+      "step": 780
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 1560,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.1578354913325875e+19,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

uccix_instruct_191224_lr1e-5/checkpoint-780/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d03657189395bc20a651edd32d0f180bbab64bb031be8ead254fb2e38ffc37e
+size 7288

uccix_instruct_191224_lr1e-5/checkpoint-780/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)