Upload checkpoint-600

.gitattributes

@@ -35,3 +35,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 ARM-7B-Test/checkpoint-5/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 ARM-7B-Test/checkpoint-6/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+ARM-7B-Pruned/checkpoint-600/tokenizer.json filter=lfs diff=lfs merge=lfs -text

ARM-7B-Pruned/checkpoint-600/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

ARM-7B-Pruned/checkpoint-600/config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "arm-team/ARM-7B",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "max_position_embeddings": 131072,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "pad_token_id": 151643,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.1",
+  "use_cache": false,
+  "use_mrope": false,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}

ARM-7B-Pruned/checkpoint-600/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "eos_token_id": 151643,
+  "pad_token_id": 151643,
+  "transformers_version": "4.46.1"
+}

ARM-7B-Pruned/checkpoint-600/global_step600/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e5787d7fbd89f2b8aaa2b2ec9062eecbef407117b9aed364900639f6666f370
+size 45693708006

ARM-7B-Pruned/checkpoint-600/global_step600/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4967d68c29b707d82754ad12c05a2d0512607b92ca28c5362288387ab9908ca4
+size 45693708006

ARM-7B-Pruned/checkpoint-600/global_step600/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd4746e892d0c29df6548515f35a25fb82c768981ee39da0de6c02d30d001764
+size 168277

ARM-7B-Pruned/checkpoint-600/global_step600/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33b24fa9d7f397981ba3b63d93dadf9f6f5ea3549e8c4a5f1362081657e3a96c
+size 168277

ARM-7B-Pruned/checkpoint-600/latest

@@ -0,0 +1 @@
+global_step600

ARM-7B-Pruned/checkpoint-600/model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a7c65ac4614fb2823e2c74fe56b5a85f915ee7f9be2b1fe4cc8c2e2c7bd680d
+size 4877660776

ARM-7B-Pruned/checkpoint-600/model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1e7b49ee50721dfae01e9705127ddf94e0178d56666efdf8d6875a510fc2baa
+size 4932751008

ARM-7B-Pruned/checkpoint-600/model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18a42b7f5a1bd1db6756e8d6bb9d98bbc0ac1e967a4844937c5e50993611f7e5
+size 4330865200

ARM-7B-Pruned/checkpoint-600/model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de4062ce4cc52d6628a1838480ee1ba2e673dc1a4ac78f19a13c0c1ee1151689
+size 1089994880

ARM-7B-Pruned/checkpoint-600/model.safetensors.index.json

@@ -0,0 +1,346 @@
+{
+  "metadata": {
+    "total_size": 15231233024
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00004-of-00004.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.norm.weight": "model-00003-of-00004.safetensors"
+  }
+}

ARM-7B-Pruned/checkpoint-600/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8d6a959372d5e0c2ea025dd26c9d0ad2046fce19352056cae8074dcbd0a6fd4
+size 14512

ARM-7B-Pruned/checkpoint-600/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f68a37892a1b445d21bb35cc10bf7a058a6f9ec8c363f5ed156ff4f49d90fb6
+size 14512

ARM-7B-Pruned/checkpoint-600/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00011a11caf446f8f330335e6ee23f7ebb2deff91b76544883e322fad1e22f1d
+size 1064

ARM-7B-Pruned/checkpoint-600/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "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
+  }
+}

ARM-7B-Pruned/checkpoint-600/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

ARM-7B-Pruned/checkpoint-600/tokenizer_config.json

@@ -0,0 +1,209 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message + '\n' }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ 'Human: ' + content + '\nAssistant:' }}{% elif message['role'] == 'assistant' %}{{ content + '<|endoftext|>' + '\n' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

ARM-7B-Pruned/checkpoint-600/trainer_state.json

@@ -0,0 +1,2133 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.06400512040963277,
+  "eval_steps": 50000,
+  "global_step": 600,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00021335040136544257,
+      "grad_norm": 42.684086970967954,
+      "learning_rate": 4.264392324093816e-09,
+      "loss": 0.5986,
+      "step": 2
+    },
+    {
+      "epoch": 0.00042670080273088514,
+      "grad_norm": 42.537918257537875,
+      "learning_rate": 8.528784648187632e-09,
+      "loss": 0.6957,
+      "step": 4
+    },
+    {
+      "epoch": 0.0006400512040963277,
+      "grad_norm": 58.765999906046574,
+      "learning_rate": 1.2793176972281448e-08,
+      "loss": 0.047,
+      "step": 6
+    },
+    {
+      "epoch": 0.0008534016054617703,
+      "grad_norm": 154.70890416675192,
+      "learning_rate": 1.7057569296375264e-08,
+      "loss": -0.6682,
+      "step": 8
+    },
+    {
+      "epoch": 0.0010667520068272129,
+      "grad_norm": 315.97286091458943,
+      "learning_rate": 2.1321961620469082e-08,
+      "loss": 0.0971,
+      "step": 10
+    },
+    {
+      "epoch": 0.0012801024081926554,
+      "grad_norm": 102.37154058019046,
+      "learning_rate": 2.5586353944562897e-08,
+      "loss": 1.0541,
+      "step": 12
+    },
+    {
+      "epoch": 0.001493452809558098,
+      "grad_norm": 105.127529829733,
+      "learning_rate": 2.9850746268656714e-08,
+      "loss": 0.4648,
+      "step": 14
+    },
+    {
+      "epoch": 0.0017068032109235406,
+      "grad_norm": 48.013196159706226,
+      "learning_rate": 3.411513859275053e-08,
+      "loss": -0.4529,
+      "step": 16
+    },
+    {
+      "epoch": 0.0019201536122889832,
+      "grad_norm": 55.69084727420936,
+      "learning_rate": 3.837953091684435e-08,
+      "loss": -0.1927,
+      "step": 18
+    },
+    {
+      "epoch": 0.0021335040136544257,
+      "grad_norm": 33.25875332287273,
+      "learning_rate": 4.2643923240938164e-08,
+      "loss": 0.2517,
+      "step": 20
+    },
+    {
+      "epoch": 0.002346854415019868,
+      "grad_norm": 245.94604658609376,
+      "learning_rate": 4.6908315565031985e-08,
+      "loss": -0.8646,
+      "step": 22
+    },
+    {
+      "epoch": 0.002560204816385311,
+      "grad_norm": 33.35694659794871,
+      "learning_rate": 5.117270788912579e-08,
+      "loss": 0.7383,
+      "step": 24
+    },
+    {
+      "epoch": 0.0027735552177507532,
+      "grad_norm": 193.28196705366508,
+      "learning_rate": 5.5437100213219614e-08,
+      "loss": -0.9655,
+      "step": 26
+    },
+    {
+      "epoch": 0.002986905619116196,
+      "grad_norm": 50.147398314622684,
+      "learning_rate": 5.970149253731343e-08,
+      "loss": -0.5218,
+      "step": 28
+    },
+    {
+      "epoch": 0.0032002560204816384,
+      "grad_norm": 75.21413198410397,
+      "learning_rate": 6.396588486140725e-08,
+      "loss": 0.1629,
+      "step": 30
+    },
+    {
+      "epoch": 0.003413606421847081,
+      "grad_norm": 103.01315028563378,
+      "learning_rate": 6.823027718550106e-08,
+      "loss": -0.7119,
+      "step": 32
+    },
+    {
+      "epoch": 0.0036269568232125235,
+      "grad_norm": 36.63545743331723,
+      "learning_rate": 7.249466950959488e-08,
+      "loss": 0.3538,
+      "step": 34
+    },
+    {
+      "epoch": 0.0038403072245779663,
+      "grad_norm": 39.002709650329486,
+      "learning_rate": 7.67590618336887e-08,
+      "loss": -0.6754,
+      "step": 36
+    },
+    {
+      "epoch": 0.004053657625943409,
+      "grad_norm": 23.502647009611,
+      "learning_rate": 8.102345415778252e-08,
+      "loss": 0.1865,
+      "step": 38
+    },
+    {
+      "epoch": 0.0042670080273088514,
+      "grad_norm": 26.139670413986497,
+      "learning_rate": 8.528784648187633e-08,
+      "loss": -0.3739,
+      "step": 40
+    },
+    {
+      "epoch": 0.004480358428674294,
+      "grad_norm": 30.02660527440867,
+      "learning_rate": 8.955223880597014e-08,
+      "loss": -0.0359,
+      "step": 42
+    },
+    {
+      "epoch": 0.004693708830039736,
+      "grad_norm": 13.980962042551367,
+      "learning_rate": 9.381663113006397e-08,
+      "loss": 0.4169,
+      "step": 44
+    },
+    {
+      "epoch": 0.004907059231405179,
+      "grad_norm": 94.42191623111091,
+      "learning_rate": 9.808102345415778e-08,
+      "loss": -1.4962,
+      "step": 46
+    },
+    {
+      "epoch": 0.005120409632770622,
+      "grad_norm": 36.146349198898875,
+      "learning_rate": 1.0234541577825159e-07,
+      "loss": 0.8652,
+      "step": 48
+    },
+    {
+      "epoch": 0.0053337600341360645,
+      "grad_norm": 34.112375662585286,
+      "learning_rate": 1.0660980810234541e-07,
+      "loss": 0.4556,
+      "step": 50
+    },
+    {
+      "epoch": 0.0055471104355015064,
+      "grad_norm": 54.41510979168854,
+      "learning_rate": 1.1087420042643923e-07,
+      "loss": 0.2452,
+      "step": 52
+    },
+    {
+      "epoch": 0.005760460836866949,
+      "grad_norm": 27.31961709504184,
+      "learning_rate": 1.1513859275053305e-07,
+      "loss": 0.3131,
+      "step": 54
+    },
+    {
+      "epoch": 0.005973811238232392,
+      "grad_norm": 77.00896363994629,
+      "learning_rate": 1.1940298507462686e-07,
+      "loss": -0.0289,
+      "step": 56
+    },
+    {
+      "epoch": 0.006187161639597835,
+      "grad_norm": 21.8781170195382,
+      "learning_rate": 1.2366737739872068e-07,
+      "loss": -0.6188,
+      "step": 58
+    },
+    {
+      "epoch": 0.006400512040963277,
+      "grad_norm": 16.368843761461356,
+      "learning_rate": 1.279317697228145e-07,
+      "loss": 0.2618,
+      "step": 60
+    },
+    {
+      "epoch": 0.0066138624423287195,
+      "grad_norm": 17.051752922162727,
+      "learning_rate": 1.3219616204690832e-07,
+      "loss": -0.1717,
+      "step": 62
+    },
+    {
+      "epoch": 0.006827212843694162,
+      "grad_norm": 99.13064855085295,
+      "learning_rate": 1.3646055437100212e-07,
+      "loss": -0.449,
+      "step": 64
+    },
+    {
+      "epoch": 0.007040563245059605,
+      "grad_norm": 136.3486847826033,
+      "learning_rate": 1.4072494669509594e-07,
+      "loss": 0.2984,
+      "step": 66
+    },
+    {
+      "epoch": 0.007253913646425047,
+      "grad_norm": 62.554463250965604,
+      "learning_rate": 1.4498933901918976e-07,
+      "loss": -0.2482,
+      "step": 68
+    },
+    {
+      "epoch": 0.00746726404779049,
+      "grad_norm": 51.515904581548206,
+      "learning_rate": 1.4925373134328355e-07,
+      "loss": -0.8223,
+      "step": 70
+    },
+    {
+      "epoch": 0.007680614449155933,
+      "grad_norm": 113.86732871533368,
+      "learning_rate": 1.535181236673774e-07,
+      "loss": -0.0383,
+      "step": 72
+    },
+    {
+      "epoch": 0.007893964850521375,
+      "grad_norm": 26.087446116741575,
+      "learning_rate": 1.5778251599147122e-07,
+      "loss": 0.7253,
+      "step": 74
+    },
+    {
+      "epoch": 0.008107315251886817,
+      "grad_norm": 24.131096654425452,
+      "learning_rate": 1.6204690831556504e-07,
+      "loss": -0.0834,
+      "step": 76
+    },
+    {
+      "epoch": 0.008320665653252261,
+      "grad_norm": 74.42536774731673,
+      "learning_rate": 1.6631130063965884e-07,
+      "loss": -0.2997,
+      "step": 78
+    },
+    {
+      "epoch": 0.008534016054617703,
+      "grad_norm": 71.52938681964571,
+      "learning_rate": 1.7057569296375266e-07,
+      "loss": 0.0944,
+      "step": 80
+    },
+    {
+      "epoch": 0.008747366455983145,
+      "grad_norm": 29.241243114282817,
+      "learning_rate": 1.7484008528784648e-07,
+      "loss": -0.7054,
+      "step": 82
+    },
+    {
+      "epoch": 0.008960716857348588,
+      "grad_norm": 134.58638764428474,
+      "learning_rate": 1.7910447761194027e-07,
+      "loss": -0.338,
+      "step": 84
+    },
+    {
+      "epoch": 0.00917406725871403,
+      "grad_norm": 109.66885361053902,
+      "learning_rate": 1.8336886993603412e-07,
+      "loss": -0.1817,
+      "step": 86
+    },
+    {
+      "epoch": 0.009387417660079472,
+      "grad_norm": 27.969535103995202,
+      "learning_rate": 1.8763326226012794e-07,
+      "loss": -0.9258,
+      "step": 88
+    },
+    {
+      "epoch": 0.009600768061444916,
+      "grad_norm": 27.49676445944368,
+      "learning_rate": 1.9189765458422174e-07,
+      "loss": 0.3309,
+      "step": 90
+    },
+    {
+      "epoch": 0.009814118462810358,
+      "grad_norm": 25.544238337228855,
+      "learning_rate": 1.9616204690831556e-07,
+      "loss": -0.0894,
+      "step": 92
+    },
+    {
+      "epoch": 0.010027468864175802,
+      "grad_norm": 26.09909186001115,
+      "learning_rate": 2.0042643923240938e-07,
+      "loss": -0.1209,
+      "step": 94
+    },
+    {
+      "epoch": 0.010240819265541243,
+      "grad_norm": 32.632466828944544,
+      "learning_rate": 2.0469083155650317e-07,
+      "loss": 0.4736,
+      "step": 96
+    },
+    {
+      "epoch": 0.010454169666906685,
+      "grad_norm": 50.68481961380247,
+      "learning_rate": 2.08955223880597e-07,
+      "loss": -0.6321,
+      "step": 98
+    },
+    {
+      "epoch": 0.010667520068272129,
+      "grad_norm": 25.00088670078768,
+      "learning_rate": 2.1321961620469082e-07,
+      "loss": -0.3787,
+      "step": 100
+    },
+    {
+      "epoch": 0.010880870469637571,
+      "grad_norm": 21.25939551515176,
+      "learning_rate": 2.1748400852878466e-07,
+      "loss": 0.1969,
+      "step": 102
+    },
+    {
+      "epoch": 0.011094220871003013,
+      "grad_norm": 23.873098687717384,
+      "learning_rate": 2.2174840085287846e-07,
+      "loss": -0.0885,
+      "step": 104
+    },
+    {
+      "epoch": 0.011307571272368457,
+      "grad_norm": 17.01811754764777,
+      "learning_rate": 2.2601279317697228e-07,
+      "loss": 0.0203,
+      "step": 106
+    },
+    {
+      "epoch": 0.011520921673733898,
+      "grad_norm": 31.009663505522784,
+      "learning_rate": 2.302771855010661e-07,
+      "loss": -0.817,
+      "step": 108
+    },
+    {
+      "epoch": 0.011734272075099342,
+      "grad_norm": 33.200887615423774,
+      "learning_rate": 2.345415778251599e-07,
+      "loss": 0.0471,
+      "step": 110
+    },
+    {
+      "epoch": 0.011947622476464784,
+      "grad_norm": 32.63042973424626,
+      "learning_rate": 2.388059701492537e-07,
+      "loss": -1.5996,
+      "step": 112
+    },
+    {
+      "epoch": 0.012160972877830226,
+      "grad_norm": 30.22918533639733,
+      "learning_rate": 2.4307036247334754e-07,
+      "loss": -0.2466,
+      "step": 114
+    },
+    {
+      "epoch": 0.01237432327919567,
+      "grad_norm": 27.145400816572458,
+      "learning_rate": 2.4733475479744136e-07,
+      "loss": -0.3294,
+      "step": 116
+    },
+    {
+      "epoch": 0.012587673680561112,
+      "grad_norm": 19.75203303790041,
+      "learning_rate": 2.515991471215352e-07,
+      "loss": -0.7427,
+      "step": 118
+    },
+    {
+      "epoch": 0.012801024081926553,
+      "grad_norm": 16.441388594403527,
+      "learning_rate": 2.55863539445629e-07,
+      "loss": -0.2637,
+      "step": 120
+    },
+    {
+      "epoch": 0.013014374483291997,
+      "grad_norm": 23.81603976262189,
+      "learning_rate": 2.601279317697228e-07,
+      "loss": -0.07,
+      "step": 122
+    },
+    {
+      "epoch": 0.013227724884657439,
+      "grad_norm": 80.42145460011395,
+      "learning_rate": 2.6439232409381664e-07,
+      "loss": -0.1647,
+      "step": 124
+    },
+    {
+      "epoch": 0.013441075286022881,
+      "grad_norm": 28.328954293896295,
+      "learning_rate": 2.686567164179104e-07,
+      "loss": -0.1786,
+      "step": 126
+    },
+    {
+      "epoch": 0.013654425687388325,
+      "grad_norm": 44.29342497352258,
+      "learning_rate": 2.7292110874200423e-07,
+      "loss": 0.6337,
+      "step": 128
+    },
+    {
+      "epoch": 0.013867776088753767,
+      "grad_norm": 14.247744370026695,
+      "learning_rate": 2.7718550106609805e-07,
+      "loss": -0.7468,
+      "step": 130
+    },
+    {
+      "epoch": 0.01408112649011921,
+      "grad_norm": 30.909935252567514,
+      "learning_rate": 2.8144989339019187e-07,
+      "loss": 0.2444,
+      "step": 132
+    },
+    {
+      "epoch": 0.014294476891484652,
+      "grad_norm": 19.951382058960338,
+      "learning_rate": 2.857142857142857e-07,
+      "loss": -0.0888,
+      "step": 134
+    },
+    {
+      "epoch": 0.014507827292850094,
+      "grad_norm": 19.451243521855947,
+      "learning_rate": 2.899786780383795e-07,
+      "loss": 0.0622,
+      "step": 136
+    },
+    {
+      "epoch": 0.014721177694215538,
+      "grad_norm": 44.543818203511506,
+      "learning_rate": 2.9424307036247334e-07,
+      "loss": -0.5024,
+      "step": 138
+    },
+    {
+      "epoch": 0.01493452809558098,
+      "grad_norm": 33.99708212815377,
+      "learning_rate": 2.985074626865671e-07,
+      "loss": 0.3096,
+      "step": 140
+    },
+    {
+      "epoch": 0.015147878496946422,
+      "grad_norm": 23.05447566893894,
+      "learning_rate": 3.02771855010661e-07,
+      "loss": 0.3329,
+      "step": 142
+    },
+    {
+      "epoch": 0.015361228898311865,
+      "grad_norm": 16.88015370164898,
+      "learning_rate": 3.070362473347548e-07,
+      "loss": -0.1371,
+      "step": 144
+    },
+    {
+      "epoch": 0.015574579299677307,
+      "grad_norm": 26.13573562942401,
+      "learning_rate": 3.113006396588486e-07,
+      "loss": -0.4183,
+      "step": 146
+    },
+    {
+      "epoch": 0.01578792970104275,
+      "grad_norm": 22.309803740611596,
+      "learning_rate": 3.1556503198294244e-07,
+      "loss": -0.3246,
+      "step": 148
+    },
+    {
+      "epoch": 0.016001280102408193,
+      "grad_norm": 22.52244155087573,
+      "learning_rate": 3.1982942430703626e-07,
+      "loss": -0.5349,
+      "step": 150
+    },
+    {
+      "epoch": 0.016214630503773635,
+      "grad_norm": 18.473624086858514,
+      "learning_rate": 3.240938166311301e-07,
+      "loss": 0.2352,
+      "step": 152
+    },
+    {
+      "epoch": 0.016427980905139077,
+      "grad_norm": 7.7748886163241355,
+      "learning_rate": 3.2835820895522385e-07,
+      "loss": 0.2151,
+      "step": 154
+    },
+    {
+      "epoch": 0.016641331306504522,
+      "grad_norm": 10.802467734897878,
+      "learning_rate": 3.3262260127931767e-07,
+      "loss": 0.2848,
+      "step": 156
+    },
+    {
+      "epoch": 0.016854681707869964,
+      "grad_norm": 18.162628483420374,
+      "learning_rate": 3.368869936034115e-07,
+      "loss": -0.168,
+      "step": 158
+    },
+    {
+      "epoch": 0.017068032109235406,
+      "grad_norm": 40.528567461071695,
+      "learning_rate": 3.411513859275053e-07,
+      "loss": -0.4126,
+      "step": 160
+    },
+    {
+      "epoch": 0.017281382510600848,
+      "grad_norm": 35.36550572570942,
+      "learning_rate": 3.4541577825159914e-07,
+      "loss": 0.0287,
+      "step": 162
+    },
+    {
+      "epoch": 0.01749473291196629,
+      "grad_norm": 10.892875403462355,
+      "learning_rate": 3.4968017057569296e-07,
+      "loss": -0.613,
+      "step": 164
+    },
+    {
+      "epoch": 0.01770808331333173,
+      "grad_norm": 27.141686915753116,
+      "learning_rate": 3.539445628997867e-07,
+      "loss": -0.9305,
+      "step": 166
+    },
+    {
+      "epoch": 0.017921433714697177,
+      "grad_norm": 37.37206055456951,
+      "learning_rate": 3.5820895522388055e-07,
+      "loss": -0.3178,
+      "step": 168
+    },
+    {
+      "epoch": 0.01813478411606262,
+      "grad_norm": 52.89469341474386,
+      "learning_rate": 3.6247334754797437e-07,
+      "loss": -0.6419,
+      "step": 170
+    },
+    {
+      "epoch": 0.01834813451742806,
+      "grad_norm": 23.215765487324806,
+      "learning_rate": 3.6673773987206824e-07,
+      "loss": -0.0533,
+      "step": 172
+    },
+    {
+      "epoch": 0.018561484918793503,
+      "grad_norm": 21.171847082083467,
+      "learning_rate": 3.7100213219616206e-07,
+      "loss": -1.0652,
+      "step": 174
+    },
+    {
+      "epoch": 0.018774835320158945,
+      "grad_norm": 17.335722663927694,
+      "learning_rate": 3.752665245202559e-07,
+      "loss": -0.169,
+      "step": 176
+    },
+    {
+      "epoch": 0.01898818572152439,
+      "grad_norm": 13.15039858318736,
+      "learning_rate": 3.795309168443497e-07,
+      "loss": 0.4414,
+      "step": 178
+    },
+    {
+      "epoch": 0.019201536122889832,
+      "grad_norm": 9.293181092571949,
+      "learning_rate": 3.8379530916844347e-07,
+      "loss": 0.1714,
+      "step": 180
+    },
+    {
+      "epoch": 0.019414886524255274,
+      "grad_norm": 27.253514499053658,
+      "learning_rate": 3.880597014925373e-07,
+      "loss": -0.3719,
+      "step": 182
+    },
+    {
+      "epoch": 0.019628236925620716,
+      "grad_norm": 11.179212679495947,
+      "learning_rate": 3.923240938166311e-07,
+      "loss": -1.5775,
+      "step": 184
+    },
+    {
+      "epoch": 0.019841587326986158,
+      "grad_norm": 28.002238164167814,
+      "learning_rate": 3.9658848614072494e-07,
+      "loss": -0.9116,
+      "step": 186
+    },
+    {
+      "epoch": 0.020054937728351603,
+      "grad_norm": 20.304811169854457,
+      "learning_rate": 4.0085287846481876e-07,
+      "loss": -0.6315,
+      "step": 188
+    },
+    {
+      "epoch": 0.020268288129717045,
+      "grad_norm": 43.26972545321108,
+      "learning_rate": 4.051172707889126e-07,
+      "loss": 0.0818,
+      "step": 190
+    },
+    {
+      "epoch": 0.020481638531082487,
+      "grad_norm": 23.280256775268967,
+      "learning_rate": 4.0938166311300635e-07,
+      "loss": -0.2754,
+      "step": 192
+    },
+    {
+      "epoch": 0.02069498893244793,
+      "grad_norm": 25.086595962784372,
+      "learning_rate": 4.1364605543710017e-07,
+      "loss": 0.1783,
+      "step": 194
+    },
+    {
+      "epoch": 0.02090833933381337,
+      "grad_norm": 17.929854339567687,
+      "learning_rate": 4.17910447761194e-07,
+      "loss": -0.3967,
+      "step": 196
+    },
+    {
+      "epoch": 0.021121689735178813,
+      "grad_norm": 26.684280905377257,
+      "learning_rate": 4.221748400852878e-07,
+      "loss": -0.2432,
+      "step": 198
+    },
+    {
+      "epoch": 0.021335040136544258,
+      "grad_norm": 20.811875320249584,
+      "learning_rate": 4.2643923240938163e-07,
+      "loss": -0.6011,
+      "step": 200
+    },
+    {
+      "epoch": 0.0215483905379097,
+      "grad_norm": 28.776367397287462,
+      "learning_rate": 4.3070362473347545e-07,
+      "loss": -1.6853,
+      "step": 202
+    },
+    {
+      "epoch": 0.021761740939275142,
+      "grad_norm": 22.23424675819631,
+      "learning_rate": 4.349680170575693e-07,
+      "loss": 0.2222,
+      "step": 204
+    },
+    {
+      "epoch": 0.021975091340640584,
+      "grad_norm": 15.408003573608147,
+      "learning_rate": 4.392324093816631e-07,
+      "loss": -0.0164,
+      "step": 206
+    },
+    {
+      "epoch": 0.022188441742006026,
+      "grad_norm": 8.60164597442568,
+      "learning_rate": 4.434968017057569e-07,
+      "loss": -0.5671,
+      "step": 208
+    },
+    {
+      "epoch": 0.02240179214337147,
+      "grad_norm": 10.343966416597475,
+      "learning_rate": 4.4776119402985074e-07,
+      "loss": -0.815,
+      "step": 210
+    },
+    {
+      "epoch": 0.022615142544736913,
+      "grad_norm": 15.741330211675757,
+      "learning_rate": 4.5202558635394456e-07,
+      "loss": -1.0932,
+      "step": 212
+    },
+    {
+      "epoch": 0.022828492946102355,
+      "grad_norm": 16.39054521840376,
+      "learning_rate": 4.562899786780384e-07,
+      "loss": -1.7177,
+      "step": 214
+    },
+    {
+      "epoch": 0.023041843347467797,
+      "grad_norm": 11.781875373813582,
+      "learning_rate": 4.605543710021322e-07,
+      "loss": 0.3606,
+      "step": 216
+    },
+    {
+      "epoch": 0.02325519374883324,
+      "grad_norm": 10.097449683599512,
+      "learning_rate": 4.64818763326226e-07,
+      "loss": -0.7045,
+      "step": 218
+    },
+    {
+      "epoch": 0.023468544150198684,
+      "grad_norm": 15.715275415653137,
+      "learning_rate": 4.690831556503198e-07,
+      "loss": -0.0658,
+      "step": 220
+    },
+    {
+      "epoch": 0.023681894551564126,
+      "grad_norm": 26.933594441832767,
+      "learning_rate": 4.733475479744136e-07,
+      "loss": -0.1298,
+      "step": 222
+    },
+    {
+      "epoch": 0.023895244952929568,
+      "grad_norm": 8.712912536861461,
+      "learning_rate": 4.776119402985074e-07,
+      "loss": 0.095,
+      "step": 224
+    },
+    {
+      "epoch": 0.02410859535429501,
+      "grad_norm": 21.120479987680937,
+      "learning_rate": 4.818763326226012e-07,
+      "loss": -0.0818,
+      "step": 226
+    },
+    {
+      "epoch": 0.024321945755660452,
+      "grad_norm": 10.071445207074044,
+      "learning_rate": 4.861407249466951e-07,
+      "loss": -0.1419,
+      "step": 228
+    },
+    {
+      "epoch": 0.024535296157025894,
+      "grad_norm": 10.801668877306875,
+      "learning_rate": 4.904051172707888e-07,
+      "loss": -0.3432,
+      "step": 230
+    },
+    {
+      "epoch": 0.02474864655839134,
+      "grad_norm": 13.692305863205236,
+      "learning_rate": 4.946695095948827e-07,
+      "loss": -0.1519,
+      "step": 232
+    },
+    {
+      "epoch": 0.02496199695975678,
+      "grad_norm": 22.11353305341118,
+      "learning_rate": 4.989339019189765e-07,
+      "loss": -0.0445,
+      "step": 234
+    },
+    {
+      "epoch": 0.025175347361122223,
+      "grad_norm": 25.883156385523233,
+      "learning_rate": 5.031982942430704e-07,
+      "loss": -0.4949,
+      "step": 236
+    },
+    {
+      "epoch": 0.025388697762487665,
+      "grad_norm": 61.080118987133744,
+      "learning_rate": 5.074626865671642e-07,
+      "loss": -1.1921,
+      "step": 238
+    },
+    {
+      "epoch": 0.025602048163853107,
+      "grad_norm": 9.282902236974593,
+      "learning_rate": 5.11727078891258e-07,
+      "loss": 0.1989,
+      "step": 240
+    },
+    {
+      "epoch": 0.025815398565218552,
+      "grad_norm": 9.781179043931736,
+      "learning_rate": 5.159914712153518e-07,
+      "loss": 0.9967,
+      "step": 242
+    },
+    {
+      "epoch": 0.026028748966583994,
+      "grad_norm": 12.216217874660604,
+      "learning_rate": 5.202558635394456e-07,
+      "loss": -0.3518,
+      "step": 244
+    },
+    {
+      "epoch": 0.026242099367949436,
+      "grad_norm": 24.42587736751593,
+      "learning_rate": 5.245202558635394e-07,
+      "loss": -1.0814,
+      "step": 246
+    },
+    {
+      "epoch": 0.026455449769314878,
+      "grad_norm": 38.12352896449889,
+      "learning_rate": 5.287846481876333e-07,
+      "loss": 0.0811,
+      "step": 248
+    },
+    {
+      "epoch": 0.02666880017068032,
+      "grad_norm": 24.52163737825959,
+      "learning_rate": 5.33049040511727e-07,
+      "loss": -0.8824,
+      "step": 250
+    },
+    {
+      "epoch": 0.026882150572045762,
+      "grad_norm": 19.231993611076483,
+      "learning_rate": 5.373134328358208e-07,
+      "loss": 0.5571,
+      "step": 252
+    },
+    {
+      "epoch": 0.027095500973411207,
+      "grad_norm": 17.129505372417224,
+      "learning_rate": 5.415778251599147e-07,
+      "loss": -0.6865,
+      "step": 254
+    },
+    {
+      "epoch": 0.02730885137477665,
+      "grad_norm": 19.406038647540843,
+      "learning_rate": 5.458422174840085e-07,
+      "loss": -0.4991,
+      "step": 256
+    },
+    {
+      "epoch": 0.02752220177614209,
+      "grad_norm": 6.847722005977987,
+      "learning_rate": 5.501066098081023e-07,
+      "loss": 0.1979,
+      "step": 258
+    },
+    {
+      "epoch": 0.027735552177507533,
+      "grad_norm": 13.405288361475698,
+      "learning_rate": 5.543710021321961e-07,
+      "loss": -0.3812,
+      "step": 260
+    },
+    {
+      "epoch": 0.027948902578872975,
+      "grad_norm": 21.529785897303153,
+      "learning_rate": 5.5863539445629e-07,
+      "loss": -0.6946,
+      "step": 262
+    },
+    {
+      "epoch": 0.02816225298023842,
+      "grad_norm": 14.20190258955381,
+      "learning_rate": 5.628997867803837e-07,
+      "loss": -0.7182,
+      "step": 264
+    },
+    {
+      "epoch": 0.028375603381603862,
+      "grad_norm": 20.09792579726273,
+      "learning_rate": 5.671641791044775e-07,
+      "loss": -0.2583,
+      "step": 266
+    },
+    {
+      "epoch": 0.028588953782969304,
+      "grad_norm": 17.213177542585473,
+      "learning_rate": 5.714285714285714e-07,
+      "loss": 0.3218,
+      "step": 268
+    },
+    {
+      "epoch": 0.028802304184334746,
+      "grad_norm": 11.185587231889192,
+      "learning_rate": 5.756929637526652e-07,
+      "loss": -0.0663,
+      "step": 270
+    },
+    {
+      "epoch": 0.029015654585700188,
+      "grad_norm": 27.722334316990437,
+      "learning_rate": 5.79957356076759e-07,
+      "loss": 0.1254,
+      "step": 272
+    },
+    {
+      "epoch": 0.029229004987065633,
+      "grad_norm": 20.49167387185228,
+      "learning_rate": 5.842217484008528e-07,
+      "loss": -0.3077,
+      "step": 274
+    },
+    {
+      "epoch": 0.029442355388431075,
+      "grad_norm": 17.084103781873182,
+      "learning_rate": 5.884861407249467e-07,
+      "loss": -1.0397,
+      "step": 276
+    },
+    {
+      "epoch": 0.029655705789796517,
+      "grad_norm": 23.870809301137577,
+      "learning_rate": 5.927505330490404e-07,
+      "loss": -0.401,
+      "step": 278
+    },
+    {
+      "epoch": 0.02986905619116196,
+      "grad_norm": 11.313699213427757,
+      "learning_rate": 5.970149253731342e-07,
+      "loss": -1.1555,
+      "step": 280
+    },
+    {
+      "epoch": 0.0300824065925274,
+      "grad_norm": 9.331131563295791,
+      "learning_rate": 6.012793176972282e-07,
+      "loss": -0.5128,
+      "step": 282
+    },
+    {
+      "epoch": 0.030295756993892843,
+      "grad_norm": 10.71153796479133,
+      "learning_rate": 6.05543710021322e-07,
+      "loss": 0.239,
+      "step": 284
+    },
+    {
+      "epoch": 0.03050910739525829,
+      "grad_norm": 28.31484795145744,
+      "learning_rate": 6.098081023454158e-07,
+      "loss": -0.2084,
+      "step": 286
+    },
+    {
+      "epoch": 0.03072245779662373,
+      "grad_norm": 11.917796878106765,
+      "learning_rate": 6.140724946695096e-07,
+      "loss": -0.2701,
+      "step": 288
+    },
+    {
+      "epoch": 0.030935808197989172,
+      "grad_norm": 15.31345988143714,
+      "learning_rate": 6.183368869936035e-07,
+      "loss": -0.9096,
+      "step": 290
+    },
+    {
+      "epoch": 0.031149158599354614,
+      "grad_norm": 49.77491988047747,
+      "learning_rate": 6.226012793176972e-07,
+      "loss": 0.1465,
+      "step": 292
+    },
+    {
+      "epoch": 0.03136250900072006,
+      "grad_norm": 12.884205486410131,
+      "learning_rate": 6.26865671641791e-07,
+      "loss": 0.5899,
+      "step": 294
+    },
+    {
+      "epoch": 0.0315758594020855,
+      "grad_norm": 22.77696725291766,
+      "learning_rate": 6.311300639658849e-07,
+      "loss": -0.3255,
+      "step": 296
+    },
+    {
+      "epoch": 0.031789209803450943,
+      "grad_norm": 14.760034845954518,
+      "learning_rate": 6.353944562899787e-07,
+      "loss": 0.2191,
+      "step": 298
+    },
+    {
+      "epoch": 0.032002560204816385,
+      "grad_norm": 16.04873208614678,
+      "learning_rate": 6.396588486140725e-07,
+      "loss": 0.9425,
+      "step": 300
+    },
+    {
+      "epoch": 0.03221591060618183,
+      "grad_norm": 21.948692800323336,
+      "learning_rate": 6.439232409381663e-07,
+      "loss": 0.4721,
+      "step": 302
+    },
+    {
+      "epoch": 0.03242926100754727,
+      "grad_norm": 17.475770533963207,
+      "learning_rate": 6.481876332622602e-07,
+      "loss": -0.8516,
+      "step": 304
+    },
+    {
+      "epoch": 0.03264261140891271,
+      "grad_norm": 21.938554689368157,
+      "learning_rate": 6.524520255863539e-07,
+      "loss": 0.1675,
+      "step": 306
+    },
+    {
+      "epoch": 0.03285596181027815,
+      "grad_norm": 46.95903237109459,
+      "learning_rate": 6.567164179104477e-07,
+      "loss": 0.7013,
+      "step": 308
+    },
+    {
+      "epoch": 0.033069312211643595,
+      "grad_norm": 21.7738424183671,
+      "learning_rate": 6.609808102345416e-07,
+      "loss": -0.8641,
+      "step": 310
+    },
+    {
+      "epoch": 0.033282662613009044,
+      "grad_norm": 20.818736283324778,
+      "learning_rate": 6.652452025586353e-07,
+      "loss": -0.2203,
+      "step": 312
+    },
+    {
+      "epoch": 0.033496013014374486,
+      "grad_norm": 12.191027612540129,
+      "learning_rate": 6.695095948827292e-07,
+      "loss": -0.2385,
+      "step": 314
+    },
+    {
+      "epoch": 0.03370936341573993,
+      "grad_norm": 17.485900290878302,
+      "learning_rate": 6.73773987206823e-07,
+      "loss": -0.6482,
+      "step": 316
+    },
+    {
+      "epoch": 0.03392271381710537,
+      "grad_norm": 12.365985112237825,
+      "learning_rate": 6.780383795309168e-07,
+      "loss": -0.0618,
+      "step": 318
+    },
+    {
+      "epoch": 0.03413606421847081,
+      "grad_norm": 19.889441682727426,
+      "learning_rate": 6.823027718550106e-07,
+      "loss": 0.255,
+      "step": 320
+    },
+    {
+      "epoch": 0.034349414619836253,
+      "grad_norm": 19.5015710993416,
+      "learning_rate": 6.865671641791044e-07,
+      "loss": -0.6494,
+      "step": 322
+    },
+    {
+      "epoch": 0.034562765021201695,
+      "grad_norm": 21.45614945706602,
+      "learning_rate": 6.908315565031983e-07,
+      "loss": -0.4695,
+      "step": 324
+    },
+    {
+      "epoch": 0.03477611542256714,
+      "grad_norm": 11.714892632525226,
+      "learning_rate": 6.95095948827292e-07,
+      "loss": -0.6134,
+      "step": 326
+    },
+    {
+      "epoch": 0.03498946582393258,
+      "grad_norm": 17.08341230963109,
+      "learning_rate": 6.993603411513859e-07,
+      "loss": -0.9212,
+      "step": 328
+    },
+    {
+      "epoch": 0.03520281622529802,
+      "grad_norm": 18.71372293407707,
+      "learning_rate": 7.036247334754797e-07,
+      "loss": -0.1783,
+      "step": 330
+    },
+    {
+      "epoch": 0.03541616662666346,
+      "grad_norm": 40.394000763247476,
+      "learning_rate": 7.078891257995734e-07,
+      "loss": -0.1696,
+      "step": 332
+    },
+    {
+      "epoch": 0.03562951702802891,
+      "grad_norm": 16.851717713795054,
+      "learning_rate": 7.121535181236673e-07,
+      "loss": -0.168,
+      "step": 334
+    },
+    {
+      "epoch": 0.035842867429394354,
+      "grad_norm": 20.980121165459167,
+      "learning_rate": 7.164179104477611e-07,
+      "loss": -0.1898,
+      "step": 336
+    },
+    {
+      "epoch": 0.036056217830759796,
+      "grad_norm": 5.128039309618372,
+      "learning_rate": 7.20682302771855e-07,
+      "loss": -0.3406,
+      "step": 338
+    },
+    {
+      "epoch": 0.03626956823212524,
+      "grad_norm": 17.130506401349365,
+      "learning_rate": 7.249466950959487e-07,
+      "loss": -0.2751,
+      "step": 340
+    },
+    {
+      "epoch": 0.03648291863349068,
+      "grad_norm": 17.383816800029777,
+      "learning_rate": 7.292110874200426e-07,
+      "loss": -0.0005,
+      "step": 342
+    },
+    {
+      "epoch": 0.03669626903485612,
+      "grad_norm": 43.198248806456,
+      "learning_rate": 7.334754797441365e-07,
+      "loss": -0.2181,
+      "step": 344
+    },
+    {
+      "epoch": 0.036909619436221563,
+      "grad_norm": 16.46060851996884,
+      "learning_rate": 7.377398720682303e-07,
+      "loss": 0.2075,
+      "step": 346
+    },
+    {
+      "epoch": 0.037122969837587005,
+      "grad_norm": 12.986350336767204,
+      "learning_rate": 7.420042643923241e-07,
+      "loss": -0.5395,
+      "step": 348
+    },
+    {
+      "epoch": 0.03733632023895245,
+      "grad_norm": 19.883036986347154,
+      "learning_rate": 7.462686567164179e-07,
+      "loss": -0.047,
+      "step": 350
+    },
+    {
+      "epoch": 0.03754967064031789,
+      "grad_norm": 8.289068924690984,
+      "learning_rate": 7.505330490405118e-07,
+      "loss": 0.9832,
+      "step": 352
+    },
+    {
+      "epoch": 0.03776302104168334,
+      "grad_norm": 10.173680365988295,
+      "learning_rate": 7.547974413646055e-07,
+      "loss": -0.4718,
+      "step": 354
+    },
+    {
+      "epoch": 0.03797637144304878,
+      "grad_norm": 29.406476571062164,
+      "learning_rate": 7.590618336886994e-07,
+      "loss": -0.3167,
+      "step": 356
+    },
+    {
+      "epoch": 0.03818972184441422,
+      "grad_norm": 27.934726009197963,
+      "learning_rate": 7.633262260127932e-07,
+      "loss": -0.1972,
+      "step": 358
+    },
+    {
+      "epoch": 0.038403072245779664,
+      "grad_norm": 25.076744001467002,
+      "learning_rate": 7.675906183368869e-07,
+      "loss": 0.2433,
+      "step": 360
+    },
+    {
+      "epoch": 0.038616422647145106,
+      "grad_norm": 17.034753146666695,
+      "learning_rate": 7.718550106609808e-07,
+      "loss": -0.6862,
+      "step": 362
+    },
+    {
+      "epoch": 0.03882977304851055,
+      "grad_norm": 30.28266067136412,
+      "learning_rate": 7.761194029850746e-07,
+      "loss": 0.0555,
+      "step": 364
+    },
+    {
+      "epoch": 0.03904312344987599,
+      "grad_norm": 15.433190582955902,
+      "learning_rate": 7.803837953091685e-07,
+      "loss": -0.6999,
+      "step": 366
+    },
+    {
+      "epoch": 0.03925647385124143,
+      "grad_norm": 16.705763291209895,
+      "learning_rate": 7.846481876332622e-07,
+      "loss": 0.145,
+      "step": 368
+    },
+    {
+      "epoch": 0.039469824252606873,
+      "grad_norm": 7.207700811021795,
+      "learning_rate": 7.889125799573561e-07,
+      "loss": 0.4911,
+      "step": 370
+    },
+    {
+      "epoch": 0.039683174653972315,
+      "grad_norm": 12.58245126026889,
+      "learning_rate": 7.931769722814499e-07,
+      "loss": -0.4105,
+      "step": 372
+    },
+    {
+      "epoch": 0.03989652505533776,
+      "grad_norm": 9.727684047283205,
+      "learning_rate": 7.974413646055436e-07,
+      "loss": -0.4456,
+      "step": 374
+    },
+    {
+      "epoch": 0.040109875456703206,
+      "grad_norm": 37.82883818815438,
+      "learning_rate": 8.017057569296375e-07,
+      "loss": -1.003,
+      "step": 376
+    },
+    {
+      "epoch": 0.04032322585806865,
+      "grad_norm": 9.205831543930358,
+      "learning_rate": 8.059701492537313e-07,
+      "loss": -0.9034,
+      "step": 378
+    },
+    {
+      "epoch": 0.04053657625943409,
+      "grad_norm": 17.5439627015,
+      "learning_rate": 8.102345415778252e-07,
+      "loss": -0.3568,
+      "step": 380
+    },
+    {
+      "epoch": 0.04074992666079953,
+      "grad_norm": 15.109693571978122,
+      "learning_rate": 8.144989339019189e-07,
+      "loss": -0.3776,
+      "step": 382
+    },
+    {
+      "epoch": 0.040963277062164974,
+      "grad_norm": 10.552742457862886,
+      "learning_rate": 8.187633262260127e-07,
+      "loss": 0.2015,
+      "step": 384
+    },
+    {
+      "epoch": 0.041176627463530416,
+      "grad_norm": 19.08793342231466,
+      "learning_rate": 8.230277185501066e-07,
+      "loss": -1.4421,
+      "step": 386
+    },
+    {
+      "epoch": 0.04138997786489586,
+      "grad_norm": 34.056443898662884,
+      "learning_rate": 8.272921108742003e-07,
+      "loss": -0.197,
+      "step": 388
+    },
+    {
+      "epoch": 0.0416033282662613,
+      "grad_norm": 7.384540107176898,
+      "learning_rate": 8.315565031982942e-07,
+      "loss": 0.7544,
+      "step": 390
+    },
+    {
+      "epoch": 0.04181667866762674,
+      "grad_norm": 23.683836639585067,
+      "learning_rate": 8.35820895522388e-07,
+      "loss": -0.3735,
+      "step": 392
+    },
+    {
+      "epoch": 0.042030029068992183,
+      "grad_norm": 19.629725497894157,
+      "learning_rate": 8.400852878464819e-07,
+      "loss": -0.6826,
+      "step": 394
+    },
+    {
+      "epoch": 0.042243379470357625,
+      "grad_norm": 13.85103939830437,
+      "learning_rate": 8.443496801705756e-07,
+      "loss": -0.2708,
+      "step": 396
+    },
+    {
+      "epoch": 0.042456729871723074,
+      "grad_norm": 19.710815160612455,
+      "learning_rate": 8.486140724946694e-07,
+      "loss": 0.0877,
+      "step": 398
+    },
+    {
+      "epoch": 0.042670080273088516,
+      "grad_norm": 4.706110486614041,
+      "learning_rate": 8.528784648187633e-07,
+      "loss": 0.1916,
+      "step": 400
+    },
+    {
+      "epoch": 0.04288343067445396,
+      "grad_norm": 12.864617887211866,
+      "learning_rate": 8.57142857142857e-07,
+      "loss": 0.0179,
+      "step": 402
+    },
+    {
+      "epoch": 0.0430967810758194,
+      "grad_norm": 9.560466147390757,
+      "learning_rate": 8.614072494669509e-07,
+      "loss": -0.0813,
+      "step": 404
+    },
+    {
+      "epoch": 0.04331013147718484,
+      "grad_norm": 21.130991357317583,
+      "learning_rate": 8.656716417910447e-07,
+      "loss": -1.1384,
+      "step": 406
+    },
+    {
+      "epoch": 0.043523481878550284,
+      "grad_norm": 19.392285709372985,
+      "learning_rate": 8.699360341151387e-07,
+      "loss": 0.1794,
+      "step": 408
+    },
+    {
+      "epoch": 0.043736832279915726,
+      "grad_norm": 9.266238784116867,
+      "learning_rate": 8.742004264392324e-07,
+      "loss": -1.768,
+      "step": 410
+    },
+    {
+      "epoch": 0.04395018268128117,
+      "grad_norm": 40.16756332202219,
+      "learning_rate": 8.784648187633262e-07,
+      "loss": -0.8232,
+      "step": 412
+    },
+    {
+      "epoch": 0.04416353308264661,
+      "grad_norm": 39.24330802424692,
+      "learning_rate": 8.827292110874201e-07,
+      "loss": -1.6178,
+      "step": 414
+    },
+    {
+      "epoch": 0.04437688348401205,
+      "grad_norm": 11.88082478788392,
+      "learning_rate": 8.869936034115138e-07,
+      "loss": -0.3053,
+      "step": 416
+    },
+    {
+      "epoch": 0.044590233885377493,
+      "grad_norm": 6.357526673930185,
+      "learning_rate": 8.912579957356077e-07,
+      "loss": -1.0471,
+      "step": 418
+    },
+    {
+      "epoch": 0.04480358428674294,
+      "grad_norm": 16.303453044925423,
+      "learning_rate": 8.955223880597015e-07,
+      "loss": -0.1262,
+      "step": 420
+    },
+    {
+      "epoch": 0.045016934688108384,
+      "grad_norm": 14.6518934802552,
+      "learning_rate": 8.997867803837953e-07,
+      "loss": -0.2323,
+      "step": 422
+    },
+    {
+      "epoch": 0.045230285089473826,
+      "grad_norm": 8.02837999022306,
+      "learning_rate": 9.040511727078891e-07,
+      "loss": 0.1657,
+      "step": 424
+    },
+    {
+      "epoch": 0.04544363549083927,
+      "grad_norm": 15.64582877736265,
+      "learning_rate": 9.083155650319829e-07,
+      "loss": -0.1882,
+      "step": 426
+    },
+    {
+      "epoch": 0.04565698589220471,
+      "grad_norm": 16.363233536467853,
+      "learning_rate": 9.125799573560768e-07,
+      "loss": -0.4399,
+      "step": 428
+    },
+    {
+      "epoch": 0.04587033629357015,
+      "grad_norm": 15.548489913061752,
+      "learning_rate": 9.168443496801705e-07,
+      "loss": 0.3602,
+      "step": 430
+    },
+    {
+      "epoch": 0.046083686694935594,
+      "grad_norm": 7.007708536573836,
+      "learning_rate": 9.211087420042644e-07,
+      "loss": -0.2646,
+      "step": 432
+    },
+    {
+      "epoch": 0.046297037096301036,
+      "grad_norm": 13.687256516704583,
+      "learning_rate": 9.253731343283582e-07,
+      "loss": -1.0053,
+      "step": 434
+    },
+    {
+      "epoch": 0.04651038749766648,
+      "grad_norm": 30.306488264414572,
+      "learning_rate": 9.29637526652452e-07,
+      "loss": -0.3203,
+      "step": 436
+    },
+    {
+      "epoch": 0.04672373789903192,
+      "grad_norm": 26.094659474249983,
+      "learning_rate": 9.339019189765458e-07,
+      "loss": -0.6359,
+      "step": 438
+    },
+    {
+      "epoch": 0.04693708830039737,
+      "grad_norm": 13.796812579064204,
+      "learning_rate": 9.381663113006396e-07,
+      "loss": -0.3471,
+      "step": 440
+    },
+    {
+      "epoch": 0.04715043870176281,
+      "grad_norm": 14.69389520446514,
+      "learning_rate": 9.424307036247334e-07,
+      "loss": -0.8145,
+      "step": 442
+    },
+    {
+      "epoch": 0.04736378910312825,
+      "grad_norm": 16.918575972701632,
+      "learning_rate": 9.466950959488272e-07,
+      "loss": 0.2185,
+      "step": 444
+    },
+    {
+      "epoch": 0.047577139504493694,
+      "grad_norm": 10.422480971977198,
+      "learning_rate": 9.509594882729211e-07,
+      "loss": -0.4649,
+      "step": 446
+    },
+    {
+      "epoch": 0.047790489905859136,
+      "grad_norm": 7.03443725132983,
+      "learning_rate": 9.552238805970149e-07,
+      "loss": -0.3627,
+      "step": 448
+    },
+    {
+      "epoch": 0.04800384030722458,
+      "grad_norm": 18.359610314742397,
+      "learning_rate": 9.594882729211086e-07,
+      "loss": -0.9404,
+      "step": 450
+    },
+    {
+      "epoch": 0.04821719070859002,
+      "grad_norm": 8.277072501759134,
+      "learning_rate": 9.637526652452024e-07,
+      "loss": 0.2409,
+      "step": 452
+    },
+    {
+      "epoch": 0.04843054110995546,
+      "grad_norm": 13.071634752922243,
+      "learning_rate": 9.680170575692964e-07,
+      "loss": 0.3506,
+      "step": 454
+    },
+    {
+      "epoch": 0.048643891511320904,
+      "grad_norm": 36.801112890968085,
+      "learning_rate": 9.722814498933901e-07,
+      "loss": -0.3961,
+      "step": 456
+    },
+    {
+      "epoch": 0.048857241912686346,
+      "grad_norm": 40.38991240984631,
+      "learning_rate": 9.76545842217484e-07,
+      "loss": -0.3497,
+      "step": 458
+    },
+    {
+      "epoch": 0.04907059231405179,
+      "grad_norm": 8.414591188760289,
+      "learning_rate": 9.808102345415777e-07,
+      "loss": -0.5775,
+      "step": 460
+    },
+    {
+      "epoch": 0.04928394271541724,
+      "grad_norm": 8.76108225719733,
+      "learning_rate": 9.850746268656714e-07,
+      "loss": 0.0839,
+      "step": 462
+    },
+    {
+      "epoch": 0.04949729311678268,
+      "grad_norm": 9.275399140799998,
+      "learning_rate": 9.893390191897654e-07,
+      "loss": 0.0137,
+      "step": 464
+    },
+    {
+      "epoch": 0.04971064351814812,
+      "grad_norm": 9.832501914571974,
+      "learning_rate": 9.936034115138592e-07,
+      "loss": 0.1014,
+      "step": 466
+    },
+    {
+      "epoch": 0.04992399391951356,
+      "grad_norm": 19.204369086700787,
+      "learning_rate": 9.97867803837953e-07,
+      "loss": -0.31,
+      "step": 468
+    },
+    {
+      "epoch": 0.050137344320879004,
+      "grad_norm": 6.109980991190154,
+      "learning_rate": 1.0021321961620467e-06,
+      "loss": -0.26,
+      "step": 470
+    },
+    {
+      "epoch": 0.050350694722244446,
+      "grad_norm": 31.865612558858906,
+      "learning_rate": 1.0063965884861407e-06,
+      "loss": -0.1208,
+      "step": 472
+    },
+    {
+      "epoch": 0.05056404512360989,
+      "grad_norm": 10.044184014650897,
+      "learning_rate": 1.0106609808102345e-06,
+      "loss": -0.2205,
+      "step": 474
+    },
+    {
+      "epoch": 0.05077739552497533,
+      "grad_norm": 21.875745094199374,
+      "learning_rate": 1.0149253731343285e-06,
+      "loss": -0.0812,
+      "step": 476
+    },
+    {
+      "epoch": 0.05099074592634077,
+      "grad_norm": 11.995975513229988,
+      "learning_rate": 1.019189765458422e-06,
+      "loss": -1.7783,
+      "step": 478
+    },
+    {
+      "epoch": 0.051204096327706214,
+      "grad_norm": 7.696707350691998,
+      "learning_rate": 1.023454157782516e-06,
+      "loss": 0.2012,
+      "step": 480
+    },
+    {
+      "epoch": 0.051417446729071656,
+      "grad_norm": 72.33916217166278,
+      "learning_rate": 1.0277185501066098e-06,
+      "loss": 0.0427,
+      "step": 482
+    },
+    {
+      "epoch": 0.051630797130437105,
+      "grad_norm": 13.569652394718384,
+      "learning_rate": 1.0319829424307035e-06,
+      "loss": 0.0953,
+      "step": 484
+    },
+    {
+      "epoch": 0.05184414753180255,
+      "grad_norm": 6.640484086255635,
+      "learning_rate": 1.0362473347547973e-06,
+      "loss": -1.0771,
+      "step": 486
+    },
+    {
+      "epoch": 0.05205749793316799,
+      "grad_norm": 16.284392321889953,
+      "learning_rate": 1.0405117270788913e-06,
+      "loss": -0.661,
+      "step": 488
+    },
+    {
+      "epoch": 0.05227084833453343,
+      "grad_norm": 10.396345506952937,
+      "learning_rate": 1.0447761194029848e-06,
+      "loss": -0.3095,
+      "step": 490
+    },
+    {
+      "epoch": 0.05248419873589887,
+      "grad_norm": 12.867700984841619,
+      "learning_rate": 1.0490405117270788e-06,
+      "loss": 0.4289,
+      "step": 492
+    },
+    {
+      "epoch": 0.052697549137264314,
+      "grad_norm": 18.31825168601856,
+      "learning_rate": 1.0533049040511726e-06,
+      "loss": 0.1729,
+      "step": 494
+    },
+    {
+      "epoch": 0.052910899538629756,
+      "grad_norm": 17.248690968213506,
+      "learning_rate": 1.0575692963752666e-06,
+      "loss": -1.1607,
+      "step": 496
+    },
+    {
+      "epoch": 0.0531242499399952,
+      "grad_norm": 8.51928074224459,
+      "learning_rate": 1.0618336886993601e-06,
+      "loss": -0.3654,
+      "step": 498
+    },
+    {
+      "epoch": 0.05333760034136064,
+      "grad_norm": 20.639043194792887,
+      "learning_rate": 1.066098081023454e-06,
+      "loss": -0.4893,
+      "step": 500
+    },
+    {
+      "epoch": 0.05355095074272608,
+      "grad_norm": 14.990753207547238,
+      "learning_rate": 1.070362473347548e-06,
+      "loss": -0.468,
+      "step": 502
+    },
+    {
+      "epoch": 0.053764301144091524,
+      "grad_norm": 24.695774003905925,
+      "learning_rate": 1.0746268656716416e-06,
+      "loss": -1.2465,
+      "step": 504
+    },
+    {
+      "epoch": 0.05397765154545697,
+      "grad_norm": 17.81220746000941,
+      "learning_rate": 1.0788912579957356e-06,
+      "loss": -0.232,
+      "step": 506
+    },
+    {
+      "epoch": 0.054191001946822415,
+      "grad_norm": 7.6840011548352924,
+      "learning_rate": 1.0831556503198294e-06,
+      "loss": 0.1394,
+      "step": 508
+    },
+    {
+      "epoch": 0.05440435234818786,
+      "grad_norm": 17.135280175028917,
+      "learning_rate": 1.0874200426439234e-06,
+      "loss": -0.1113,
+      "step": 510
+    },
+    {
+      "epoch": 0.0546177027495533,
+      "grad_norm": 21.577778290948245,
+      "learning_rate": 1.091684434968017e-06,
+      "loss": 0.1389,
+      "step": 512
+    },
+    {
+      "epoch": 0.05483105315091874,
+      "grad_norm": 26.22719409585662,
+      "learning_rate": 1.095948827292111e-06,
+      "loss": -0.1216,
+      "step": 514
+    },
+    {
+      "epoch": 0.05504440355228418,
+      "grad_norm": 8.763214398721058,
+      "learning_rate": 1.1002132196162047e-06,
+      "loss": -0.6157,
+      "step": 516
+    },
+    {
+      "epoch": 0.055257753953649624,
+      "grad_norm": 14.925448860683707,
+      "learning_rate": 1.1044776119402984e-06,
+      "loss": 0.0356,
+      "step": 518
+    },
+    {
+      "epoch": 0.055471104355015066,
+      "grad_norm": 14.365407282055516,
+      "learning_rate": 1.1087420042643922e-06,
+      "loss": -0.0624,
+      "step": 520
+    },
+    {
+      "epoch": 0.05568445475638051,
+      "grad_norm": 8.802366004297667,
+      "learning_rate": 1.1130063965884862e-06,
+      "loss": 0.5658,
+      "step": 522
+    },
+    {
+      "epoch": 0.05589780515774595,
+      "grad_norm": 24.251203850442998,
+      "learning_rate": 1.11727078891258e-06,
+      "loss": -0.0644,
+      "step": 524
+    },
+    {
+      "epoch": 0.0561111555591114,
+      "grad_norm": 16.632568797956132,
+      "learning_rate": 1.1215351812366737e-06,
+      "loss": -1.0315,
+      "step": 526
+    },
+    {
+      "epoch": 0.05632450596047684,
+      "grad_norm": 15.777059103835,
+      "learning_rate": 1.1257995735607675e-06,
+      "loss": 0.6824,
+      "step": 528
+    },
+    {
+      "epoch": 0.05653785636184228,
+      "grad_norm": 6.521931225844872,
+      "learning_rate": 1.1300639658848615e-06,
+      "loss": -0.2547,
+      "step": 530
+    },
+    {
+      "epoch": 0.056751206763207725,
+      "grad_norm": 16.18907262330339,
+      "learning_rate": 1.134328358208955e-06,
+      "loss": -0.714,
+      "step": 532
+    },
+    {
+      "epoch": 0.05696455716457317,
+      "grad_norm": 17.05716262320914,
+      "learning_rate": 1.138592750533049e-06,
+      "loss": -1.1937,
+      "step": 534
+    },
+    {
+      "epoch": 0.05717790756593861,
+      "grad_norm": 8.560189061728243,
+      "learning_rate": 1.1428571428571428e-06,
+      "loss": -0.3703,
+      "step": 536
+    },
+    {
+      "epoch": 0.05739125796730405,
+      "grad_norm": 11.56228500281614,
+      "learning_rate": 1.1471215351812368e-06,
+      "loss": -0.8073,
+      "step": 538
+    },
+    {
+      "epoch": 0.05760460836866949,
+      "grad_norm": 13.24672543046646,
+      "learning_rate": 1.1513859275053303e-06,
+      "loss": -0.3222,
+      "step": 540
+    },
+    {
+      "epoch": 0.057817958770034934,
+      "grad_norm": 14.035481073008786,
+      "learning_rate": 1.1556503198294243e-06,
+      "loss": 0.286,
+      "step": 542
+    },
+    {
+      "epoch": 0.058031309171400376,
+      "grad_norm": 28.95516126438781,
+      "learning_rate": 1.159914712153518e-06,
+      "loss": -1.189,
+      "step": 544
+    },
+    {
+      "epoch": 0.05824465957276582,
+      "grad_norm": 20.850060598425877,
+      "learning_rate": 1.1641791044776118e-06,
+      "loss": -0.6204,
+      "step": 546
+    },
+    {
+      "epoch": 0.05845800997413127,
+      "grad_norm": 21.211499139014034,
+      "learning_rate": 1.1684434968017056e-06,
+      "loss": -0.5428,
+      "step": 548
+    },
+    {
+      "epoch": 0.05867136037549671,
+      "grad_norm": 16.101447224508938,
+      "learning_rate": 1.1727078891257996e-06,
+      "loss": -0.7822,
+      "step": 550
+    },
+    {
+      "epoch": 0.05888471077686215,
+      "grad_norm": 60.24925771229807,
+      "learning_rate": 1.1769722814498933e-06,
+      "loss": -0.5338,
+      "step": 552
+    },
+    {
+      "epoch": 0.05909806117822759,
+      "grad_norm": 10.911944544305852,
+      "learning_rate": 1.1812366737739871e-06,
+      "loss": -0.1546,
+      "step": 554
+    },
+    {
+      "epoch": 0.059311411579593035,
+      "grad_norm": 20.123496073532923,
+      "learning_rate": 1.1855010660980809e-06,
+      "loss": 0.1078,
+      "step": 556
+    },
+    {
+      "epoch": 0.05952476198095848,
+      "grad_norm": 8.077431875506283,
+      "learning_rate": 1.1897654584221749e-06,
+      "loss": 0.6775,
+      "step": 558
+    },
+    {
+      "epoch": 0.05973811238232392,
+      "grad_norm": 20.69430693950377,
+      "learning_rate": 1.1940298507462684e-06,
+      "loss": -0.1531,
+      "step": 560
+    },
+    {
+      "epoch": 0.05995146278368936,
+      "grad_norm": 9.149229771649699,
+      "learning_rate": 1.1982942430703624e-06,
+      "loss": 0.2026,
+      "step": 562
+    },
+    {
+      "epoch": 0.0601648131850548,
+      "grad_norm": 35.409517307149535,
+      "learning_rate": 1.2025586353944564e-06,
+      "loss": 0.2081,
+      "step": 564
+    },
+    {
+      "epoch": 0.060378163586420244,
+      "grad_norm": 18.732961050530974,
+      "learning_rate": 1.2068230277185501e-06,
+      "loss": -0.4564,
+      "step": 566
+    },
+    {
+      "epoch": 0.060591513987785686,
+      "grad_norm": 18.17521906003365,
+      "learning_rate": 1.211087420042644e-06,
+      "loss": -0.4694,
+      "step": 568
+    },
+    {
+      "epoch": 0.060804864389151135,
+      "grad_norm": 15.950240307469345,
+      "learning_rate": 1.2153518123667377e-06,
+      "loss": -0.7607,
+      "step": 570
+    },
+    {
+      "epoch": 0.06101821479051658,
+      "grad_norm": 27.667171128676177,
+      "learning_rate": 1.2196162046908317e-06,
+      "loss": -0.4962,
+      "step": 572
+    },
+    {
+      "epoch": 0.06123156519188202,
+      "grad_norm": 6.875981558346762,
+      "learning_rate": 1.2238805970149252e-06,
+      "loss": 0.1071,
+      "step": 574
+    },
+    {
+      "epoch": 0.06144491559324746,
+      "grad_norm": 23.330801092034456,
+      "learning_rate": 1.2281449893390192e-06,
+      "loss": -1.1081,
+      "step": 576
+    },
+    {
+      "epoch": 0.0616582659946129,
+      "grad_norm": 18.097087985455573,
+      "learning_rate": 1.232409381663113e-06,
+      "loss": 0.292,
+      "step": 578
+    },
+    {
+      "epoch": 0.061871616395978345,
+      "grad_norm": 44.96075507661794,
+      "learning_rate": 1.236673773987207e-06,
+      "loss": -1.8344,
+      "step": 580
+    },
+    {
+      "epoch": 0.06208496679734379,
+      "grad_norm": 11.33849063045387,
+      "learning_rate": 1.2409381663113005e-06,
+      "loss": -0.5339,
+      "step": 582
+    },
+    {
+      "epoch": 0.06229831719870923,
+      "grad_norm": 25.655929914012017,
+      "learning_rate": 1.2452025586353945e-06,
+      "loss": 0.2773,
+      "step": 584
+    },
+    {
+      "epoch": 0.06251166760007468,
+      "grad_norm": 19.518509371933693,
+      "learning_rate": 1.2494669509594882e-06,
+      "loss": -0.3635,
+      "step": 586
+    },
+    {
+      "epoch": 0.06272501800144012,
+      "grad_norm": 19.337178836658445,
+      "learning_rate": 1.253731343283582e-06,
+      "loss": -0.4397,
+      "step": 588
+    },
+    {
+      "epoch": 0.06293836840280556,
+      "grad_norm": 15.629253302757428,
+      "learning_rate": 1.2579957356076758e-06,
+      "loss": -0.3098,
+      "step": 590
+    },
+    {
+      "epoch": 0.063151718804171,
+      "grad_norm": 35.95259160255183,
+      "learning_rate": 1.2622601279317698e-06,
+      "loss": -0.9708,
+      "step": 592
+    },
+    {
+      "epoch": 0.06336506920553645,
+      "grad_norm": 27.574715730226895,
+      "learning_rate": 1.2665245202558633e-06,
+      "loss": -0.5131,
+      "step": 594
+    },
+    {
+      "epoch": 0.06357841960690189,
+      "grad_norm": 47.74900017442824,
+      "learning_rate": 1.2707889125799573e-06,
+      "loss": 0.3332,
+      "step": 596
+    },
+    {
+      "epoch": 0.06379177000826733,
+      "grad_norm": 20.309365479498744,
+      "learning_rate": 1.275053304904051e-06,
+      "loss": -1.2309,
+      "step": 598
+    },
+    {
+      "epoch": 0.06400512040963277,
+      "grad_norm": 11.40208623029672,
+      "learning_rate": 1.279317697228145e-06,
+      "loss": -0.5857,
+      "step": 600
+    }
+  ],
+  "logging_steps": 2,
+  "max_steps": 9374,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 600,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 51407077736448.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

ARM-7B-Pruned/checkpoint-600/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a81db8d15143df8bb94c1c2a8705c726095471d7f688ad728817d29193bb0eba
+size 7096

ARM-7B-Pruned/checkpoint-600/zero_to_fp32.py

@@ -0,0 +1,760 @@
+#!/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 . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import gc
+import json
+import numpy as np
+from tqdm import tqdm
+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, weights_only=False)
+
+        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 tqdm(files, desc='Loading checkpoint shards'):
+        state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
+        # 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 ZERO_STAGE not 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}")
+
+    fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] 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, exclude_frozen_parameters):
+    """
+    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,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+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,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _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")
+
+
+class GatheredTensor:
+    """
+    A pseudo tensor that collects partitioned weights.
+    It is more memory efficient when there are multiple groups.
+    """
+
+    def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
+        self.flat_groups = flat_groups
+        self.flat_groups_offset = flat_groups_offset
+        self.offset = offset
+        self.partitioned_numel = partitioned_numel
+        self.shape = shape
+        self.dtype = self.flat_groups[0][0].dtype
+
+    def contiguous(self):
+        """
+        Merge partitioned weights from flat_groups into a single tensor.
+        """
+        end_idx = self.offset + self.partitioned_numel
+        world_size = len(self.flat_groups)
+        pad_flat_param_chunks = []
+
+        for rank_i in range(world_size):
+            # for each rank, we need to collect weights from related group/groups
+            flat_groups_at_rank_i = self.flat_groups[rank_i]
+            start_group_id = None
+            end_group_id = None
+            for group_id in range(len(self.flat_groups_offset)):
+                if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
+                    start_group_id = group_id
+                if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
+                    end_group_id = group_id
+                    break
+            # collect weights from related group/groups
+            for group_id in range(start_group_id, end_group_id + 1):
+                flat_tensor = flat_groups_at_rank_i[group_id]
+                start_offset = self.offset - self.flat_groups_offset[group_id]
+                end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
+                pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
+
+        # collect weights from all ranks
+        pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
+        param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
+        return param
+
+
+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 = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * 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
+    flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
+        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}"
+            )
+
+        # memory efficient tensor
+        tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
+        state_dict[name] = tensor
+        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,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _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 to_torch_tensor(state_dict, return_empty_tensor=False):
+    """
+    Convert state_dict of GatheredTensor to torch tensor
+    """
+    torch_state_dict = {}
+    converted_tensors = {}
+    for name, tensor in state_dict.items():
+        tensor_id = id(tensor)
+        if tensor_id in converted_tensors:  # shared tensors
+            shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
+            torch_state_dict[name] = shared_tensor
+        else:
+            converted_tensors[tensor_id] = name
+            if return_empty_tensor:
+                torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
+            else:
+                torch_state_dict[name] = tensor.contiguous()
+    return torch_state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                             tag=None,
+                                             exclude_frozen_parameters=False,
+                                             lazy_mode=False):
+    """
+    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``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+        - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
+          Convert the pesduo tensor to torch tensor by ``.contiguous()``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    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.
+
+    Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
+    You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint. Or you can load state_dict in lazy mode ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
+        for name, lazy_tensor in state_dict.item():
+            tensor = lazy_tensor.contiguous()  # to cpu
+            print(name, tensor)
+            # del tensor to release memory if it no longer in use
+    """
+    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")
+
+    state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+    if lazy_mode:
+        return state_dict
+    else:
+        return to_torch_tensor(state_dict)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               tag=None,
+                                               exclude_frozen_parameters=False):
+    """
+    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_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``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``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                                          tag,
+                                                          exclude_frozen_parameters,
+                                                          lazy_mode=True)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        # an memory-efficient approach for sharding
+        empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
+        state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model by shard
+    os.makedirs(output_dir, exist_ok=True)
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
+        shard_state_dict = to_torch_tensor(shard_state_dict)
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard_state_dict, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard_state_dict, output_path)
+        # release the memory of current shard
+        for tensor_name in list(shard_state_dict.keys()):
+            del state_dict[tensor_name]
+            del shard_state_dict[tensor_name]
+        del shard_state_dict
+        gc.collect()
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+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("Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info("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_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    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("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    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_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)