Upload folder using huggingface_hub

added_tokens.json

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

config.json

@@ -0,0 +1,138 @@
+{
+  "_name_or_path": "microsoft/Phi-3.5-mini-instruct",
+  "architectures": [
+    "Phi3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "microsoft/Phi-3.5-mini-instruct--configuration_phi3.Phi3Config",
+    "AutoModelForCausalLM": "microsoft/Phi-3.5-mini-instruct--modeling_phi3.Phi3ForCausalLM"
+  },
+  "bos_token_id": 1,
+  "embd_pdrop": 0.0,
+  "eos_token_id": 32000,
+  "hidden_act": "silu",
+  "hidden_size": 3072,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 131072,
+  "model_type": "phi3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "original_max_position_embeddings": 4096,
+  "pad_token_id": 32000,
+  "resid_pdrop": 0.0,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": {
+    "long_factor": [
+      1.0800000429153442,
+      1.1100000143051147,
+      1.1399999856948853,
+      1.340000033378601,
+      1.5899999141693115,
+      1.600000023841858,
+      1.6200000047683716,
+      2.620000123977661,
+      3.2300000190734863,
+      3.2300000190734863,
+      4.789999961853027,
+      7.400000095367432,
+      7.700000286102295,
+      9.09000015258789,
+      12.199999809265137,
+      17.670000076293945,
+      24.46000099182129,
+      28.57000160217285,
+      30.420001983642578,
+      30.840002059936523,
+      32.590003967285156,
+      32.93000411987305,
+      42.320003509521484,
+      44.96000289916992,
+      50.340003967285156,
+      50.45000457763672,
+      57.55000305175781,
+      57.93000411987305,
+      58.21000289916992,
+      60.1400032043457,
+      62.61000442504883,
+      62.62000274658203,
+      62.71000289916992,
+      63.1400032043457,
+      63.1400032043457,
+      63.77000427246094,
+      63.93000411987305,
+      63.96000289916992,
+      63.970001220703125,
+      64.02999877929688,
+      64.06999969482422,
+      64.08000183105469,
+      64.12000274658203,
+      64.41000366210938,
+      64.4800033569336,
+      64.51000213623047,
+      64.52999877929688,
+      64.83999633789062
+    ],
+    "short_factor": [
+      1.0,
+      1.0199999809265137,
+      1.0299999713897705,
+      1.0299999713897705,
+      1.0499999523162842,
+      1.0499999523162842,
+      1.0499999523162842,
+      1.0499999523162842,
+      1.0499999523162842,
+      1.0699999332427979,
+      1.0999999046325684,
+      1.1099998950958252,
+      1.1599998474121094,
+      1.1599998474121094,
+      1.1699998378753662,
+      1.2899998426437378,
+      1.339999794960022,
+      1.679999828338623,
+      1.7899998426437378,
+      1.8199998140335083,
+      1.8499997854232788,
+      1.8799997568130493,
+      1.9099997282028198,
+      1.9399996995925903,
+      1.9899996519088745,
+      2.0199997425079346,
+      2.0199997425079346,
+      2.0199997425079346,
+      2.0199997425079346,
+      2.0199997425079346,
+      2.0199997425079346,
+      2.0299997329711914,
+      2.0299997329711914,
+      2.0299997329711914,
+      2.0299997329711914,
+      2.0299997329711914,
+      2.0299997329711914,
+      2.0299997329711914,
+      2.0299997329711914,
+      2.0299997329711914,
+      2.0799996852874756,
+      2.0899996757507324,
+      2.189999580383301,
+      2.2199995517730713,
+      2.5899994373321533,
+      2.729999542236328,
+      2.749999523162842,
+      2.8399994373321533
+    ],
+    "type": "longrope"
+  },
+  "rope_theta": 10000.0,
+  "sliding_window": 262144,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.43.0",
+  "use_cache": false,
+  "vocab_size": 32064
+}

generation_config.json

@@ -0,0 +1,11 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": [
+    32007,
+    32001,
+    32000
+  ],
+  "pad_token_id": 32000,
+  "transformers_version": "4.43.0"
+}

global_step5300/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:faef1a24311539d30db55d12f29dd5efb11fa93c9c152ddac4b844b6b4f6e567
+size 5731623920

global_step5300/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3cc4ab5f7a97fcd72c488897553e3a3a27d48b4d0ef76bdf448e09151ce4cf9d
+size 5731623920

global_step5300/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5ce7a6cab5e49c922d231c45807b953091a749fe35f3f5317ebbcaa4651b2df
+size 5731623920

global_step5300/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b158f45746f6eef76be6e088ab219635b6a41a323636b325b7111d85a7ceb82d
+size 5731623920

global_step5300/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8436206151a47dab4201eec803abb7951c76af9f961ec249d0e1bec70cb3d3fb
+size 5731623920

global_step5300/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71dafcefd2df7b214d78633366085c0a97a7edb317c7dc4c06dfd343f74e7a3c
+size 5731623920

global_step5300/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05b376b0911febcd3f461709e46fecca83d667a5d6690d54330e3b44df7300fb
+size 5731623920

global_step5300/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85aa1546e557b299b60def83bf073b4ab62ddd070e326fb85cb3d6743c13026d
+size 5731623920

global_step5300/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ae92015bf64800d09f6df193133b9a8725d59e340f32442aacb6828b8c3278f
+size 109445

global_step5300/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:404d9c8e9fc67c32fbf1badeaf37337bac674fd155b1adb32fb720c35f6bd4d9
+size 109445

global_step5300/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64858182fadc84e5d4726c8fb5e95bc57cbd49f6ab9b9256fe1ac7fd8b4f6fe8
+size 109445

global_step5300/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a72371ca04ece39df30e58f0024c96939000049af09559f974159ea21fc8ac10
+size 109445

global_step5300/zero_pp_rank_4_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7107b53629f602dcae5d29af0dd4400f1adea9fda238a304264c531e589928e0
+size 109445

global_step5300/zero_pp_rank_5_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe448e245b317651f2f368e5b2516d633fba95889a7c53913679699cbc4d344
+size 109445

global_step5300/zero_pp_rank_6_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89be40a5d104566a94c6107b61bf03dcd3f006f8bba37d5a006a64b3901f2f0a
+size 109445

global_step5300/zero_pp_rank_7_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b994ae34c3b896404541217aca16651f1b0243e334c04f8146a341cf0161759
+size 109445

latest

@@ -0,0 +1 @@
+global_step5300

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b103b8f056b41aa1ab5dc60c5cb614c9aefd3e2bf322fbe9935a760ab835a23
+size 4972489328

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00e5909dc3348ea95d1c35ab08fb281dee667c1ee7d8441039d932381513eb5d
+size 2669692552

model.safetensors.index.json

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

rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3624bcbf3b2b80eb48fc69b2a097502aab1a6fc1ade816c01bc4bc97e8f71dbf
+size 15984

rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e3e4827c63b744e87a5cec7b9a6843237a006fa8952bdbe029824f038c45be8
+size 15984

rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98255e3c6a0a64a839ed2630289688899472cf9462ec6c38881e3b157ea5db9a
+size 15984

rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18c0fd1771075ebb94c7d7c3a275c21e327f2c42fc8e2048e9f6d647045e92fb
+size 15984

rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2261a70c95b0f831d329a0371c347336b1cc6c49bc51dd39536786cfb83c0efe
+size 15984

rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bce16ded967463fccf3716a13e55f084cd48e215a035683d634bed5a4d7aea3d
+size 15984

rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16d4865060de2a0d3e7e14fdf5aebf32c329573574fba9930819564a2c807632
+size 15984

rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9ffd58f2d675ad36d489e78e8761cba065070485cadfe0055277898459cc72d
+size 15984

special_tokens_map.json

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

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -0,0 +1,131 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": true,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": false
+    },
+    "32000": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32002": {
+      "content": "<|placeholder1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32003": {
+      "content": "<|placeholder2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32004": {
+      "content": "<|placeholder3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32005": {
+      "content": "<|placeholder4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32006": {
+      "content": "<|system|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32007": {
+      "content": "<|end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32008": {
+      "content": "<|placeholder5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32009": {
+      "content": "<|placeholder6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    },
+    "32010": {
+      "content": "<|user|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": true,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% for message in messages %}{% if message['role'] == 'system' and message['content'] %}{{'<|system|>\n' + message['content'] + '<|end|>\n'}}{% elif message['role'] == 'user' %}{{'<|user|>\n' + message['content'] + '<|end|>\n'}}{% elif message['role'] == 'assistant' %}{{'<|assistant|>\n' + message['content'] + '<|end|>\n'}}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>\n' }}{% else %}{{ eos_token }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "legacy": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

trainer_state.json

@@ -0,0 +1,828 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "eval_steps": 100,
+  "global_step": 5300,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.18867924528301888,
+      "grad_norm": 4.437798503453146,
+      "learning_rate": 7.341400415124174e-07,
+      "loss": 5.3755,
+      "step": 100
+    },
+    {
+      "epoch": 0.18867924528301888,
+      "eval_loss": 1.0850828886032104,
+      "eval_runtime": 149.7427,
+      "eval_samples_per_second": 452.857,
+      "eval_steps_per_second": 0.888,
+      "step": 100
+    },
+    {
+      "epoch": 0.37735849056603776,
+      "grad_norm": 1.6459101244560304,
+      "learning_rate": 8.446391282690362e-07,
+      "loss": 0.9681,
+      "step": 200
+    },
+    {
+      "epoch": 0.37735849056603776,
+      "eval_loss": 0.9421371817588806,
+      "eval_runtime": 149.748,
+      "eval_samples_per_second": 452.841,
+      "eval_steps_per_second": 0.888,
+      "step": 200
+    },
+    {
+      "epoch": 0.5660377358490566,
+      "grad_norm": 4.060133963944513,
+      "learning_rate": 9.09276950385592e-07,
+      "loss": 0.9538,
+      "step": 300
+    },
+    {
+      "epoch": 0.5660377358490566,
+      "eval_loss": 0.9655478596687317,
+      "eval_runtime": 149.5973,
+      "eval_samples_per_second": 453.297,
+      "eval_steps_per_second": 0.889,
+      "step": 300
+    },
+    {
+      "epoch": 0.7547169811320755,
+      "grad_norm": 41.89172794379802,
+      "learning_rate": 9.551382150256551e-07,
+      "loss": 0.9761,
+      "step": 400
+    },
+    {
+      "epoch": 0.7547169811320755,
+      "eval_loss": 0.976280152797699,
+      "eval_runtime": 149.6311,
+      "eval_samples_per_second": 453.195,
+      "eval_steps_per_second": 0.889,
+      "step": 400
+    },
+    {
+      "epoch": 0.9433962264150944,
+      "grad_norm": 70.58972777939259,
+      "learning_rate": 9.907109755120069e-07,
+      "loss": 0.9786,
+      "step": 500
+    },
+    {
+      "epoch": 0.9433962264150944,
+      "eval_loss": 0.9799396991729736,
+      "eval_runtime": 149.6428,
+      "eval_samples_per_second": 453.159,
+      "eval_steps_per_second": 0.889,
+      "step": 500
+    },
+    {
+      "epoch": 1.1320754716981132,
+      "grad_norm": 154.52662720587995,
+      "learning_rate": 1e-06,
+      "loss": 0.9851,
+      "step": 600
+    },
+    {
+      "epoch": 1.1320754716981132,
+      "eval_loss": 0.9846892952919006,
+      "eval_runtime": 150.8433,
+      "eval_samples_per_second": 449.553,
+      "eval_steps_per_second": 0.882,
+      "step": 600
+    },
+    {
+      "epoch": 1.320754716981132,
+      "grad_norm": 147.98867672055184,
+      "learning_rate": 1e-06,
+      "loss": 0.9876,
+      "step": 700
+    },
+    {
+      "epoch": 1.320754716981132,
+      "eval_loss": 0.9889557957649231,
+      "eval_runtime": 149.6471,
+      "eval_samples_per_second": 453.146,
+      "eval_steps_per_second": 0.889,
+      "step": 700
+    },
+    {
+      "epoch": 1.509433962264151,
+      "grad_norm": 93.50922205523368,
+      "learning_rate": 1e-06,
+      "loss": 0.9892,
+      "step": 800
+    },
+    {
+      "epoch": 1.509433962264151,
+      "eval_loss": 0.9961430430412292,
+      "eval_runtime": 149.7573,
+      "eval_samples_per_second": 452.813,
+      "eval_steps_per_second": 0.888,
+      "step": 800
+    },
+    {
+      "epoch": 1.6981132075471699,
+      "grad_norm": 53.06237347231676,
+      "learning_rate": 1e-06,
+      "loss": 0.9983,
+      "step": 900
+    },
+    {
+      "epoch": 1.6981132075471699,
+      "eval_loss": 1.0119000673294067,
+      "eval_runtime": 149.7765,
+      "eval_samples_per_second": 452.755,
+      "eval_steps_per_second": 0.888,
+      "step": 900
+    },
+    {
+      "epoch": 1.8867924528301887,
+      "grad_norm": 65.28104430915417,
+      "learning_rate": 1e-06,
+      "loss": 1.0228,
+      "step": 1000
+    },
+    {
+      "epoch": 1.8867924528301887,
+      "eval_loss": 1.027756690979004,
+      "eval_runtime": 149.4845,
+      "eval_samples_per_second": 453.639,
+      "eval_steps_per_second": 0.89,
+      "step": 1000
+    },
+    {
+      "epoch": 2.0754716981132075,
+      "grad_norm": 84.50132037152585,
+      "learning_rate": 1e-06,
+      "loss": 1.0433,
+      "step": 1100
+    },
+    {
+      "epoch": 2.0754716981132075,
+      "eval_loss": 1.059239149093628,
+      "eval_runtime": 149.7269,
+      "eval_samples_per_second": 452.905,
+      "eval_steps_per_second": 0.888,
+      "step": 1100
+    },
+    {
+      "epoch": 2.2641509433962264,
+      "grad_norm": 29.00236133301261,
+      "learning_rate": 1e-06,
+      "loss": 1.0644,
+      "step": 1200
+    },
+    {
+      "epoch": 2.2641509433962264,
+      "eval_loss": 1.0800738334655762,
+      "eval_runtime": 149.9162,
+      "eval_samples_per_second": 452.333,
+      "eval_steps_per_second": 0.887,
+      "step": 1200
+    },
+    {
+      "epoch": 2.452830188679245,
+      "grad_norm": 411.7064146634104,
+      "learning_rate": 1e-06,
+      "loss": 1.0917,
+      "step": 1300
+    },
+    {
+      "epoch": 2.452830188679245,
+      "eval_loss": 1.1134124994277954,
+      "eval_runtime": 149.7734,
+      "eval_samples_per_second": 452.764,
+      "eval_steps_per_second": 0.888,
+      "step": 1300
+    },
+    {
+      "epoch": 2.641509433962264,
+      "grad_norm": 229.88146571929195,
+      "learning_rate": 1e-06,
+      "loss": 1.1147,
+      "step": 1400
+    },
+    {
+      "epoch": 2.641509433962264,
+      "eval_loss": 1.1160598993301392,
+      "eval_runtime": 149.7257,
+      "eval_samples_per_second": 452.908,
+      "eval_steps_per_second": 0.888,
+      "step": 1400
+    },
+    {
+      "epoch": 2.830188679245283,
+      "grad_norm": 7.414451982821549,
+      "learning_rate": 1e-06,
+      "loss": 1.1581,
+      "step": 1500
+    },
+    {
+      "epoch": 2.830188679245283,
+      "eval_loss": 1.2751212120056152,
+      "eval_runtime": 149.6311,
+      "eval_samples_per_second": 453.195,
+      "eval_steps_per_second": 0.889,
+      "step": 1500
+    },
+    {
+      "epoch": 3.018867924528302,
+      "grad_norm": 51.75196724068806,
+      "learning_rate": 1e-06,
+      "loss": 1.4543,
+      "step": 1600
+    },
+    {
+      "epoch": 3.018867924528302,
+      "eval_loss": 1.5799795389175415,
+      "eval_runtime": 149.5252,
+      "eval_samples_per_second": 453.516,
+      "eval_steps_per_second": 0.889,
+      "step": 1600
+    },
+    {
+      "epoch": 3.207547169811321,
+      "grad_norm": 8.197678623516119,
+      "learning_rate": 1e-06,
+      "loss": 1.6332,
+      "step": 1700
+    },
+    {
+      "epoch": 3.207547169811321,
+      "eval_loss": 1.7048678398132324,
+      "eval_runtime": 149.7414,
+      "eval_samples_per_second": 452.861,
+      "eval_steps_per_second": 0.888,
+      "step": 1700
+    },
+    {
+      "epoch": 3.3962264150943398,
+      "grad_norm": 9.609718378646408,
+      "learning_rate": 1e-06,
+      "loss": 1.7915,
+      "step": 1800
+    },
+    {
+      "epoch": 3.3962264150943398,
+      "eval_loss": 1.8891370296478271,
+      "eval_runtime": 149.8519,
+      "eval_samples_per_second": 452.527,
+      "eval_steps_per_second": 0.888,
+      "step": 1800
+    },
+    {
+      "epoch": 3.5849056603773586,
+      "grad_norm": 87.68727169428583,
+      "learning_rate": 1e-06,
+      "loss": 1.9812,
+      "step": 1900
+    },
+    {
+      "epoch": 3.5849056603773586,
+      "eval_loss": 2.057527780532837,
+      "eval_runtime": 149.7699,
+      "eval_samples_per_second": 452.774,
+      "eval_steps_per_second": 0.888,
+      "step": 1900
+    },
+    {
+      "epoch": 3.7735849056603774,
+      "grad_norm": 12.995456781993322,
+      "learning_rate": 1e-06,
+      "loss": 2.0982,
+      "step": 2000
+    },
+    {
+      "epoch": 3.7735849056603774,
+      "eval_loss": 2.1332101821899414,
+      "eval_runtime": 149.6771,
+      "eval_samples_per_second": 453.055,
+      "eval_steps_per_second": 0.889,
+      "step": 2000
+    },
+    {
+      "epoch": 3.9622641509433962,
+      "grad_norm": 13.605685201407708,
+      "learning_rate": 1e-06,
+      "loss": 2.1784,
+      "step": 2100
+    },
+    {
+      "epoch": 3.9622641509433962,
+      "eval_loss": 2.2318856716156006,
+      "eval_runtime": 149.8102,
+      "eval_samples_per_second": 452.653,
+      "eval_steps_per_second": 0.888,
+      "step": 2100
+    },
+    {
+      "epoch": 4.150943396226415,
+      "grad_norm": 12.732750088048045,
+      "learning_rate": 1e-06,
+      "loss": 2.2661,
+      "step": 2200
+    },
+    {
+      "epoch": 4.150943396226415,
+      "eval_loss": 2.3092291355133057,
+      "eval_runtime": 149.6232,
+      "eval_samples_per_second": 453.219,
+      "eval_steps_per_second": 0.889,
+      "step": 2200
+    },
+    {
+      "epoch": 4.339622641509434,
+      "grad_norm": 13.451602224902139,
+      "learning_rate": 1e-06,
+      "loss": 2.3259,
+      "step": 2300
+    },
+    {
+      "epoch": 4.339622641509434,
+      "eval_loss": 2.3665931224823,
+      "eval_runtime": 149.8265,
+      "eval_samples_per_second": 452.604,
+      "eval_steps_per_second": 0.888,
+      "step": 2300
+    },
+    {
+      "epoch": 4.528301886792453,
+      "grad_norm": 13.331046925646502,
+      "learning_rate": 1e-06,
+      "loss": 2.3908,
+      "step": 2400
+    },
+    {
+      "epoch": 4.528301886792453,
+      "eval_loss": 2.4305644035339355,
+      "eval_runtime": 149.9235,
+      "eval_samples_per_second": 452.311,
+      "eval_steps_per_second": 0.887,
+      "step": 2400
+    },
+    {
+      "epoch": 4.716981132075472,
+      "grad_norm": 17.21909294548451,
+      "learning_rate": 1e-06,
+      "loss": 2.4521,
+      "step": 2500
+    },
+    {
+      "epoch": 4.716981132075472,
+      "eval_loss": 2.545186758041382,
+      "eval_runtime": 149.7293,
+      "eval_samples_per_second": 452.897,
+      "eval_steps_per_second": 0.888,
+      "step": 2500
+    },
+    {
+      "epoch": 4.90566037735849,
+      "grad_norm": 16.704993975768705,
+      "learning_rate": 1e-06,
+      "loss": 2.5541,
+      "step": 2600
+    },
+    {
+      "epoch": 4.90566037735849,
+      "eval_loss": 2.6449944972991943,
+      "eval_runtime": 149.6085,
+      "eval_samples_per_second": 453.263,
+      "eval_steps_per_second": 0.889,
+      "step": 2600
+    },
+    {
+      "epoch": 5.09433962264151,
+      "grad_norm": 17.23044971023092,
+      "learning_rate": 1e-06,
+      "loss": 2.7022,
+      "step": 2700
+    },
+    {
+      "epoch": 5.09433962264151,
+      "eval_loss": 2.7793149948120117,
+      "eval_runtime": 149.7512,
+      "eval_samples_per_second": 452.831,
+      "eval_steps_per_second": 0.888,
+      "step": 2700
+    },
+    {
+      "epoch": 5.283018867924528,
+      "grad_norm": 17.47208816215434,
+      "learning_rate": 1e-06,
+      "loss": 2.8302,
+      "step": 2800
+    },
+    {
+      "epoch": 5.283018867924528,
+      "eval_loss": 2.888209581375122,
+      "eval_runtime": 149.8872,
+      "eval_samples_per_second": 452.42,
+      "eval_steps_per_second": 0.887,
+      "step": 2800
+    },
+    {
+      "epoch": 5.471698113207547,
+      "grad_norm": 17.705909664870283,
+      "learning_rate": 1e-06,
+      "loss": 2.9344,
+      "step": 2900
+    },
+    {
+      "epoch": 5.471698113207547,
+      "eval_loss": 2.9951322078704834,
+      "eval_runtime": 149.9777,
+      "eval_samples_per_second": 452.147,
+      "eval_steps_per_second": 0.887,
+      "step": 2900
+    },
+    {
+      "epoch": 5.660377358490566,
+      "grad_norm": 19.101859660265248,
+      "learning_rate": 1e-06,
+      "loss": 3.0392,
+      "step": 3000
+    },
+    {
+      "epoch": 5.660377358490566,
+      "eval_loss": 3.0882513523101807,
+      "eval_runtime": 149.9153,
+      "eval_samples_per_second": 452.335,
+      "eval_steps_per_second": 0.887,
+      "step": 3000
+    },
+    {
+      "epoch": 5.849056603773585,
+      "grad_norm": 19.07389675964658,
+      "learning_rate": 1e-06,
+      "loss": 3.0997,
+      "step": 3100
+    },
+    {
+      "epoch": 5.849056603773585,
+      "eval_loss": 3.1218082904815674,
+      "eval_runtime": 149.7905,
+      "eval_samples_per_second": 452.712,
+      "eval_steps_per_second": 0.888,
+      "step": 3100
+    },
+    {
+      "epoch": 6.037735849056604,
+      "grad_norm": 18.92461175759707,
+      "learning_rate": 1e-06,
+      "loss": 3.1321,
+      "step": 3200
+    },
+    {
+      "epoch": 6.037735849056604,
+      "eval_loss": 3.153890609741211,
+      "eval_runtime": 149.8617,
+      "eval_samples_per_second": 452.497,
+      "eval_steps_per_second": 0.887,
+      "step": 3200
+    },
+    {
+      "epoch": 6.226415094339623,
+      "grad_norm": 20.348062318970786,
+      "learning_rate": 1e-06,
+      "loss": 3.1567,
+      "step": 3300
+    },
+    {
+      "epoch": 6.226415094339623,
+      "eval_loss": 3.180058240890503,
+      "eval_runtime": 149.6522,
+      "eval_samples_per_second": 453.131,
+      "eval_steps_per_second": 0.889,
+      "step": 3300
+    },
+    {
+      "epoch": 6.415094339622642,
+      "grad_norm": 19.380403417398043,
+      "learning_rate": 1e-06,
+      "loss": 3.1885,
+      "step": 3400
+    },
+    {
+      "epoch": 6.415094339622642,
+      "eval_loss": 3.2085678577423096,
+      "eval_runtime": 149.9193,
+      "eval_samples_per_second": 452.323,
+      "eval_steps_per_second": 0.887,
+      "step": 3400
+    },
+    {
+      "epoch": 6.60377358490566,
+      "grad_norm": 19.172157540450666,
+      "learning_rate": 1e-06,
+      "loss": 3.2066,
+      "step": 3500
+    },
+    {
+      "epoch": 6.60377358490566,
+      "eval_loss": 3.2430124282836914,
+      "eval_runtime": 149.5804,
+      "eval_samples_per_second": 453.348,
+      "eval_steps_per_second": 0.889,
+      "step": 3500
+    },
+    {
+      "epoch": 6.7924528301886795,
+      "grad_norm": 20.834808216022473,
+      "learning_rate": 1e-06,
+      "loss": 3.2529,
+      "step": 3600
+    },
+    {
+      "epoch": 6.7924528301886795,
+      "eval_loss": 3.2701263427734375,
+      "eval_runtime": 149.8287,
+      "eval_samples_per_second": 452.597,
+      "eval_steps_per_second": 0.888,
+      "step": 3600
+    },
+    {
+      "epoch": 6.981132075471698,
+      "grad_norm": 20.71072973537811,
+      "learning_rate": 1e-06,
+      "loss": 3.2788,
+      "step": 3700
+    },
+    {
+      "epoch": 6.981132075471698,
+      "eval_loss": 3.3030686378479004,
+      "eval_runtime": 149.8206,
+      "eval_samples_per_second": 452.621,
+      "eval_steps_per_second": 0.888,
+      "step": 3700
+    },
+    {
+      "epoch": 7.169811320754717,
+      "grad_norm": 20.550695492004113,
+      "learning_rate": 1e-06,
+      "loss": 3.3094,
+      "step": 3800
+    },
+    {
+      "epoch": 7.169811320754717,
+      "eval_loss": 3.3228628635406494,
+      "eval_runtime": 149.8243,
+      "eval_samples_per_second": 452.61,
+      "eval_steps_per_second": 0.888,
+      "step": 3800
+    },
+    {
+      "epoch": 7.3584905660377355,
+      "grad_norm": 22.750882645137363,
+      "learning_rate": 1e-06,
+      "loss": 3.3288,
+      "step": 3900
+    },
+    {
+      "epoch": 7.3584905660377355,
+      "eval_loss": 3.355827808380127,
+      "eval_runtime": 149.6529,
+      "eval_samples_per_second": 453.129,
+      "eval_steps_per_second": 0.889,
+      "step": 3900
+    },
+    {
+      "epoch": 7.547169811320755,
+      "grad_norm": 21.426519401026805,
+      "learning_rate": 1e-06,
+      "loss": 3.3727,
+      "step": 4000
+    },
+    {
+      "epoch": 7.547169811320755,
+      "eval_loss": 3.3865537643432617,
+      "eval_runtime": 150.2322,
+      "eval_samples_per_second": 451.381,
+      "eval_steps_per_second": 0.885,
+      "step": 4000
+    },
+    {
+      "epoch": 7.735849056603773,
+      "grad_norm": 21.858010980941717,
+      "learning_rate": 1e-06,
+      "loss": 3.3938,
+      "step": 4100
+    },
+    {
+      "epoch": 7.735849056603773,
+      "eval_loss": 3.4126923084259033,
+      "eval_runtime": 149.7102,
+      "eval_samples_per_second": 452.955,
+      "eval_steps_per_second": 0.888,
+      "step": 4100
+    },
+    {
+      "epoch": 7.9245283018867925,
+      "grad_norm": 21.32876029273619,
+      "learning_rate": 1e-06,
+      "loss": 3.4334,
+      "step": 4200
+    },
+    {
+      "epoch": 7.9245283018867925,
+      "eval_loss": 3.4582273960113525,
+      "eval_runtime": 149.6771,
+      "eval_samples_per_second": 453.055,
+      "eval_steps_per_second": 0.889,
+      "step": 4200
+    },
+    {
+      "epoch": 8.11320754716981,
+      "grad_norm": 23.89601752008716,
+      "learning_rate": 1e-06,
+      "loss": 3.4769,
+      "step": 4300
+    },
+    {
+      "epoch": 8.11320754716981,
+      "eval_loss": 3.505073308944702,
+      "eval_runtime": 149.8114,
+      "eval_samples_per_second": 452.649,
+      "eval_steps_per_second": 0.888,
+      "step": 4300
+    },
+    {
+      "epoch": 8.30188679245283,
+      "grad_norm": 22.427822055394444,
+      "learning_rate": 1e-06,
+      "loss": 3.5019,
+      "step": 4400
+    },
+    {
+      "epoch": 8.30188679245283,
+      "eval_loss": 3.5005218982696533,
+      "eval_runtime": 149.7083,
+      "eval_samples_per_second": 452.961,
+      "eval_steps_per_second": 0.888,
+      "step": 4400
+    },
+    {
+      "epoch": 8.49056603773585,
+      "grad_norm": 20.122035975196678,
+      "learning_rate": 1e-06,
+      "loss": 3.4841,
+      "step": 4500
+    },
+    {
+      "epoch": 8.49056603773585,
+      "eval_loss": 3.4907851219177246,
+      "eval_runtime": 149.5966,
+      "eval_samples_per_second": 453.299,
+      "eval_steps_per_second": 0.889,
+      "step": 4500
+    },
+    {
+      "epoch": 8.679245283018869,
+      "grad_norm": 7.060014135330357,
+      "learning_rate": 1e-06,
+      "loss": 3.4805,
+      "step": 4600
+    },
+    {
+      "epoch": 8.679245283018869,
+      "eval_loss": 3.2945797443389893,
+      "eval_runtime": 149.8934,
+      "eval_samples_per_second": 452.402,
+      "eval_steps_per_second": 0.887,
+      "step": 4600
+    },
+    {
+      "epoch": 8.867924528301886,
+      "grad_norm": 17.493433128940666,
+      "learning_rate": 1e-06,
+      "loss": 3.332,
+      "step": 4700
+    },
+    {
+      "epoch": 8.867924528301886,
+      "eval_loss": 3.3459222316741943,
+      "eval_runtime": 149.8771,
+      "eval_samples_per_second": 452.451,
+      "eval_steps_per_second": 0.887,
+      "step": 4700
+    },
+    {
+      "epoch": 9.056603773584905,
+      "grad_norm": 24.996170360974173,
+      "learning_rate": 1e-06,
+      "loss": 3.3797,
+      "step": 4800
+    },
+    {
+      "epoch": 9.056603773584905,
+      "eval_loss": 3.4055655002593994,
+      "eval_runtime": 149.7316,
+      "eval_samples_per_second": 452.89,
+      "eval_steps_per_second": 0.888,
+      "step": 4800
+    },
+    {
+      "epoch": 9.245283018867925,
+      "grad_norm": 215.6769864867367,
+      "learning_rate": 1e-06,
+      "loss": 3.4551,
+      "step": 4900
+    },
+    {
+      "epoch": 9.245283018867925,
+      "eval_loss": 3.5277042388916016,
+      "eval_runtime": 149.7551,
+      "eval_samples_per_second": 452.819,
+      "eval_steps_per_second": 0.888,
+      "step": 4900
+    },
+    {
+      "epoch": 9.433962264150944,
+      "grad_norm": 329.1368446406449,
+      "learning_rate": 1e-06,
+      "loss": 3.4127,
+      "step": 5000
+    },
+    {
+      "epoch": 9.433962264150944,
+      "eval_loss": 3.399594783782959,
+      "eval_runtime": 150.0027,
+      "eval_samples_per_second": 452.072,
+      "eval_steps_per_second": 0.887,
+      "step": 5000
+    },
+    {
+      "epoch": 9.622641509433961,
+      "grad_norm": 189.47715891368014,
+      "learning_rate": 1e-06,
+      "loss": 3.5646,
+      "step": 5100
+    },
+    {
+      "epoch": 9.622641509433961,
+      "eval_loss": 3.6223909854888916,
+      "eval_runtime": 149.6317,
+      "eval_samples_per_second": 453.193,
+      "eval_steps_per_second": 0.889,
+      "step": 5100
+    },
+    {
+      "epoch": 9.81132075471698,
+      "grad_norm": 154.57704389721005,
+      "learning_rate": 1e-06,
+      "loss": 3.6037,
+      "step": 5200
+    },
+    {
+      "epoch": 9.81132075471698,
+      "eval_loss": 3.698052406311035,
+      "eval_runtime": 149.5975,
+      "eval_samples_per_second": 453.296,
+      "eval_steps_per_second": 0.889,
+      "step": 5200
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 201.5213039296342,
+      "learning_rate": 1e-06,
+      "loss": 3.698,
+      "step": 5300
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 3.721698522567749,
+      "eval_runtime": 149.7798,
+      "eval_samples_per_second": 452.745,
+      "eval_steps_per_second": 0.888,
+      "step": 5300
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 5300,
+  "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": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1664568262656000.0,
+  "train_batch_size": 64,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5854041825d4e3c49b0fe41047ce3abdea7df279bd9d55b5e7da360624c358ca
+size 7608

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)