Upload 11 files

config.json

@@ -0,0 +1,36 @@
+{
+  "H_cycles": 3,
+  "L_cycles": 2,
+  "architectures": [
+    "GemmaHRMForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 2,
+  "dtype": "float32",
+  "eos_token_id": 1,
+  "head_dim": 256,
+  "hidden_act": "gelu_pytorch_tanh",
+  "hidden_size": 2048,
+  "hrm_layers": 2,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 8192,
+  "max_thoughts": 4,
+  "merged_lm_and_talk_heads": false,
+  "model_type": "gemma_hrm",
+  "num_attention_heads": 8,
+  "num_hidden_layers": 8,
+  "num_key_value_heads": 8,
+  "pad_token_id": 0,
+  "rms_norm_eps": 1e-06,
+  "rope_theta": 10000.0,
+  "sliding_window": null,
+  "transformers_version": "4.56.1",
+  "use_cache": true,
+  "use_complex_talk_head": false,
+  "use_concat_talk_head": false,
+  "use_shallow_talk": false,
+  "use_weighted_talk_head": false,
+  "vocab_size": 262152
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "eos_token_id": 1,
+  "pad_token_id": 0,
+  "transformers_version": "4.56.1"
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:226f27864a74cb7407a93c688e4da22c7b7eba4a39700fcf6a2d870945e160b9
+size 4295230024

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ebd239c38ea8f8b8b29a5a01d481497712f9142ab0161e84fd5f190a9d174b7
+size 2684802240

model.safetensors.index.json

@@ -0,0 +1,137 @@
+{
+  "metadata": {
+    "total_parameters": 1745004546,
+    "total_size": 6980018184
+  },
+  "weight_map": {
+    "H_init": "model-00001-of-00002.safetensors",
+    "H_level.layers.0.mlp.0.bias": "model-00002-of-00002.safetensors",
+    "H_level.layers.0.mlp.0.weight": "model-00002-of-00002.safetensors",
+    "H_level.layers.0.mlp.2.bias": "model-00002-of-00002.safetensors",
+    "H_level.layers.0.mlp.2.weight": "model-00002-of-00002.safetensors",
+    "H_level.layers.0.norm1.bias": "model-00002-of-00002.safetensors",
+    "H_level.layers.0.norm1.weight": "model-00002-of-00002.safetensors",
+    "H_level.layers.0.norm2.bias": "model-00002-of-00002.safetensors",
+    "H_level.layers.0.norm2.weight": "model-00002-of-00002.safetensors",
+    "H_level.layers.0.self_attn.in_proj_bias": "model-00002-of-00002.safetensors",
+    "H_level.layers.0.self_attn.in_proj_weight": "model-00002-of-00002.safetensors",
+    "H_level.layers.0.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "H_level.layers.0.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "H_level.layers.1.mlp.0.bias": "model-00002-of-00002.safetensors",
+    "H_level.layers.1.mlp.0.weight": "model-00002-of-00002.safetensors",
+    "H_level.layers.1.mlp.2.bias": "model-00002-of-00002.safetensors",
+    "H_level.layers.1.mlp.2.weight": "model-00002-of-00002.safetensors",
+    "H_level.layers.1.norm1.bias": "model-00002-of-00002.safetensors",
+    "H_level.layers.1.norm1.weight": "model-00002-of-00002.safetensors",
+    "H_level.layers.1.norm2.bias": "model-00002-of-00002.safetensors",
+    "H_level.layers.1.norm2.weight": "model-00002-of-00002.safetensors",
+    "H_level.layers.1.self_attn.in_proj_bias": "model-00002-of-00002.safetensors",
+    "H_level.layers.1.self_attn.in_proj_weight": "model-00002-of-00002.safetensors",
+    "H_level.layers.1.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "H_level.layers.1.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "L_init": "model-00001-of-00002.safetensors",
+    "L_level.layers.0.mlp.0.bias": "model-00002-of-00002.safetensors",
+    "L_level.layers.0.mlp.0.weight": "model-00002-of-00002.safetensors",
+    "L_level.layers.0.mlp.2.bias": "model-00002-of-00002.safetensors",
+    "L_level.layers.0.mlp.2.weight": "model-00002-of-00002.safetensors",
+    "L_level.layers.0.norm1.bias": "model-00002-of-00002.safetensors",
+    "L_level.layers.0.norm1.weight": "model-00002-of-00002.safetensors",
+    "L_level.layers.0.norm2.bias": "model-00002-of-00002.safetensors",
+    "L_level.layers.0.norm2.weight": "model-00002-of-00002.safetensors",
+    "L_level.layers.0.self_attn.in_proj_bias": "model-00002-of-00002.safetensors",
+    "L_level.layers.0.self_attn.in_proj_weight": "model-00002-of-00002.safetensors",
+    "L_level.layers.0.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "L_level.layers.0.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "L_level.layers.1.mlp.0.bias": "model-00002-of-00002.safetensors",
+    "L_level.layers.1.mlp.0.weight": "model-00002-of-00002.safetensors",
+    "L_level.layers.1.mlp.2.bias": "model-00002-of-00002.safetensors",
+    "L_level.layers.1.mlp.2.weight": "model-00002-of-00002.safetensors",
+    "L_level.layers.1.norm1.bias": "model-00002-of-00002.safetensors",
+    "L_level.layers.1.norm1.weight": "model-00002-of-00002.safetensors",
+    "L_level.layers.1.norm2.bias": "model-00002-of-00002.safetensors",
+    "L_level.layers.1.norm2.weight": "model-00002-of-00002.safetensors",
+    "L_level.layers.1.self_attn.in_proj_bias": "model-00002-of-00002.safetensors",
+    "L_level.layers.1.self_attn.in_proj_weight": "model-00002-of-00002.safetensors",
+    "L_level.layers.1.self_attn.out_proj.bias": "model-00002-of-00002.safetensors",
+    "L_level.layers.1.self_attn.out_proj.weight": "model-00002-of-00002.safetensors",
+    "end_embedding": "model-00001-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00001-of-00002.safetensors",
+    "q_head.bias": "model-00002-of-00002.safetensors",
+    "q_head.weight": "model-00002-of-00002.safetensors",
+    "start_embedding": "model-00001-of-00002.safetensors",
+    "talk_head.0.0.weight": "model-00002-of-00002.safetensors"
+  }
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1903d7b83b81ed5e2132614cec5490278d7de109024c700d61c73e3be0d7d6e1
+size 2762379

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:900815fdf40b0af934d38b458614657951a42df8b631ba531bb88bdee58e4e09
+size 14645

scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14ae2a2128444abab378aa06c09a61a84665f758fcc19fc46f5789b0bc1b5665
+size 1383

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4198a63d71f5202ac3b0fc1f47a7fd5027500a872f3ffac6ed3602224ced7b02
+size 1465

trainer_state.json

@@ -0,0 +1,734 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.08,
+      "grad_norm": 10.272224426269531,
+      "learning_rate": 4.5e-06,
+      "loss": 12.8379,
+      "step": 10
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 11.09077262878418,
+      "learning_rate": 9.5e-06,
+      "loss": 12.3314,
+      "step": 20
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 7.8299150466918945,
+      "learning_rate": 9.908163265306123e-06,
+      "loss": 10.8693,
+      "step": 30
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 5.720259189605713,
+      "learning_rate": 9.806122448979593e-06,
+      "loss": 9.5772,
+      "step": 40
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 4.4982500076293945,
+      "learning_rate": 9.704081632653061e-06,
+      "loss": 8.8462,
+      "step": 50
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 5.525058269500732,
+      "learning_rate": 9.60204081632653e-06,
+      "loss": 8.3404,
+      "step": 60
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 4.1277570724487305,
+      "learning_rate": 9.5e-06,
+      "loss": 8.1087,
+      "step": 70
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 4.99296760559082,
+      "learning_rate": 9.39795918367347e-06,
+      "loss": 7.9132,
+      "step": 80
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 4.5625319480896,
+      "learning_rate": 9.29591836734694e-06,
+      "loss": 7.5496,
+      "step": 90
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 4.021756649017334,
+      "learning_rate": 9.19387755102041e-06,
+      "loss": 7.6371,
+      "step": 100
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 4.0445098876953125,
+      "learning_rate": 9.091836734693878e-06,
+      "loss": 7.0593,
+      "step": 110
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 4.158232688903809,
+      "learning_rate": 8.989795918367348e-06,
+      "loss": 6.8773,
+      "step": 120
+    },
+    {
+      "epoch": 1.04,
+      "grad_norm": 4.3605546951293945,
+      "learning_rate": 8.887755102040817e-06,
+      "loss": 6.365,
+      "step": 130
+    },
+    {
+      "epoch": 1.12,
+      "grad_norm": 3.7614684104919434,
+      "learning_rate": 8.785714285714286e-06,
+      "loss": 6.3579,
+      "step": 140
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 3.5187251567840576,
+      "learning_rate": 8.683673469387755e-06,
+      "loss": 6.4043,
+      "step": 150
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 3.6821653842926025,
+      "learning_rate": 8.581632653061225e-06,
+      "loss": 6.219,
+      "step": 160
+    },
+    {
+      "epoch": 1.3599999999999999,
+      "grad_norm": 4.2614970207214355,
+      "learning_rate": 8.479591836734695e-06,
+      "loss": 6.2059,
+      "step": 170
+    },
+    {
+      "epoch": 1.44,
+      "grad_norm": 3.459669351577759,
+      "learning_rate": 8.377551020408165e-06,
+      "loss": 5.9866,
+      "step": 180
+    },
+    {
+      "epoch": 1.52,
+      "grad_norm": 3.8360798358917236,
+      "learning_rate": 8.275510204081634e-06,
+      "loss": 5.8988,
+      "step": 190
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 3.6687963008880615,
+      "learning_rate": 8.173469387755103e-06,
+      "loss": 5.603,
+      "step": 200
+    },
+    {
+      "epoch": 1.6800000000000002,
+      "grad_norm": 4.398599147796631,
+      "learning_rate": 8.071428571428572e-06,
+      "loss": 5.5413,
+      "step": 210
+    },
+    {
+      "epoch": 1.76,
+      "grad_norm": 4.392071723937988,
+      "learning_rate": 7.969387755102042e-06,
+      "loss": 5.6413,
+      "step": 220
+    },
+    {
+      "epoch": 1.8399999999999999,
+      "grad_norm": 3.310082197189331,
+      "learning_rate": 7.86734693877551e-06,
+      "loss": 5.4796,
+      "step": 230
+    },
+    {
+      "epoch": 1.92,
+      "grad_norm": 3.3684332370758057,
+      "learning_rate": 7.76530612244898e-06,
+      "loss": 5.6276,
+      "step": 240
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 4.198359966278076,
+      "learning_rate": 7.66326530612245e-06,
+      "loss": 5.3979,
+      "step": 250
+    },
+    {
+      "epoch": 2.08,
+      "grad_norm": 3.726461410522461,
+      "learning_rate": 7.561224489795919e-06,
+      "loss": 5.2551,
+      "step": 260
+    },
+    {
+      "epoch": 2.16,
+      "grad_norm": 4.147603511810303,
+      "learning_rate": 7.459183673469388e-06,
+      "loss": 5.1481,
+      "step": 270
+    },
+    {
+      "epoch": 2.24,
+      "grad_norm": 4.2914204597473145,
+      "learning_rate": 7.357142857142858e-06,
+      "loss": 5.079,
+      "step": 280
+    },
+    {
+      "epoch": 2.32,
+      "grad_norm": 3.718191146850586,
+      "learning_rate": 7.255102040816327e-06,
+      "loss": 4.9855,
+      "step": 290
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 3.8326470851898193,
+      "learning_rate": 7.153061224489797e-06,
+      "loss": 5.1467,
+      "step": 300
+    },
+    {
+      "epoch": 2.48,
+      "grad_norm": 3.780733585357666,
+      "learning_rate": 7.051020408163266e-06,
+      "loss": 4.9813,
+      "step": 310
+    },
+    {
+      "epoch": 2.56,
+      "grad_norm": 4.059634685516357,
+      "learning_rate": 6.948979591836736e-06,
+      "loss": 5.0103,
+      "step": 320
+    },
+    {
+      "epoch": 2.64,
+      "grad_norm": 4.013712406158447,
+      "learning_rate": 6.8469387755102046e-06,
+      "loss": 4.9963,
+      "step": 330
+    },
+    {
+      "epoch": 2.7199999999999998,
+      "grad_norm": 4.24052619934082,
+      "learning_rate": 6.7448979591836735e-06,
+      "loss": 4.8682,
+      "step": 340
+    },
+    {
+      "epoch": 2.8,
+      "grad_norm": 4.262918949127197,
+      "learning_rate": 6.642857142857143e-06,
+      "loss": 4.8391,
+      "step": 350
+    },
+    {
+      "epoch": 2.88,
+      "grad_norm": 4.253732681274414,
+      "learning_rate": 6.540816326530612e-06,
+      "loss": 4.8529,
+      "step": 360
+    },
+    {
+      "epoch": 2.96,
+      "grad_norm": 4.445166110992432,
+      "learning_rate": 6.438775510204082e-06,
+      "loss": 4.6523,
+      "step": 370
+    },
+    {
+      "epoch": 3.04,
+      "grad_norm": 4.086543560028076,
+      "learning_rate": 6.336734693877552e-06,
+      "loss": 4.6866,
+      "step": 380
+    },
+    {
+      "epoch": 3.12,
+      "grad_norm": 4.3391265869140625,
+      "learning_rate": 6.2346938775510215e-06,
+      "loss": 4.8265,
+      "step": 390
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 4.4690327644348145,
+      "learning_rate": 6.1326530612244905e-06,
+      "loss": 4.5729,
+      "step": 400
+    },
+    {
+      "epoch": 3.2800000000000002,
+      "grad_norm": 4.506042957305908,
+      "learning_rate": 6.03061224489796e-06,
+      "loss": 4.645,
+      "step": 410
+    },
+    {
+      "epoch": 3.36,
+      "grad_norm": 3.8643500804901123,
+      "learning_rate": 5.928571428571429e-06,
+      "loss": 4.4849,
+      "step": 420
+    },
+    {
+      "epoch": 3.44,
+      "grad_norm": 4.425049781799316,
+      "learning_rate": 5.826530612244898e-06,
+      "loss": 4.4227,
+      "step": 430
+    },
+    {
+      "epoch": 3.52,
+      "grad_norm": 4.880602836608887,
+      "learning_rate": 5.724489795918368e-06,
+      "loss": 4.5872,
+      "step": 440
+    },
+    {
+      "epoch": 3.6,
+      "grad_norm": 4.425590991973877,
+      "learning_rate": 5.622448979591837e-06,
+      "loss": 4.4351,
+      "step": 450
+    },
+    {
+      "epoch": 3.68,
+      "grad_norm": 4.5981125831604,
+      "learning_rate": 5.520408163265306e-06,
+      "loss": 4.2647,
+      "step": 460
+    },
+    {
+      "epoch": 3.76,
+      "grad_norm": 5.0482611656188965,
+      "learning_rate": 5.4183673469387755e-06,
+      "loss": 4.4483,
+      "step": 470
+    },
+    {
+      "epoch": 3.84,
+      "grad_norm": 4.352452754974365,
+      "learning_rate": 5.316326530612246e-06,
+      "loss": 4.2195,
+      "step": 480
+    },
+    {
+      "epoch": 3.92,
+      "grad_norm": 4.14626932144165,
+      "learning_rate": 5.214285714285715e-06,
+      "loss": 4.2741,
+      "step": 490
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 4.289257526397705,
+      "learning_rate": 5.112244897959184e-06,
+      "loss": 4.3662,
+      "step": 500
+    },
+    {
+      "epoch": 4.08,
+      "grad_norm": 4.666240215301514,
+      "learning_rate": 5.010204081632654e-06,
+      "loss": 4.1023,
+      "step": 510
+    },
+    {
+      "epoch": 4.16,
+      "grad_norm": 4.825807094573975,
+      "learning_rate": 4.908163265306123e-06,
+      "loss": 4.0558,
+      "step": 520
+    },
+    {
+      "epoch": 4.24,
+      "grad_norm": 5.112582683563232,
+      "learning_rate": 4.8061224489795925e-06,
+      "loss": 4.1276,
+      "step": 530
+    },
+    {
+      "epoch": 4.32,
+      "grad_norm": 5.386510848999023,
+      "learning_rate": 4.704081632653061e-06,
+      "loss": 4.133,
+      "step": 540
+    },
+    {
+      "epoch": 4.4,
+      "grad_norm": 5.501555442810059,
+      "learning_rate": 4.602040816326531e-06,
+      "loss": 4.0733,
+      "step": 550
+    },
+    {
+      "epoch": 4.48,
+      "grad_norm": 4.4781060218811035,
+      "learning_rate": 4.5e-06,
+      "loss": 4.1908,
+      "step": 560
+    },
+    {
+      "epoch": 4.5600000000000005,
+      "grad_norm": 4.725682258605957,
+      "learning_rate": 4.39795918367347e-06,
+      "loss": 4.2183,
+      "step": 570
+    },
+    {
+      "epoch": 4.64,
+      "grad_norm": 4.902810573577881,
+      "learning_rate": 4.295918367346939e-06,
+      "loss": 3.9457,
+      "step": 580
+    },
+    {
+      "epoch": 4.72,
+      "grad_norm": 4.52883243560791,
+      "learning_rate": 4.193877551020409e-06,
+      "loss": 4.1341,
+      "step": 590
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 4.9895124435424805,
+      "learning_rate": 4.091836734693878e-06,
+      "loss": 3.9592,
+      "step": 600
+    },
+    {
+      "epoch": 4.88,
+      "grad_norm": 4.645427227020264,
+      "learning_rate": 3.989795918367347e-06,
+      "loss": 4.042,
+      "step": 610
+    },
+    {
+      "epoch": 4.96,
+      "grad_norm": 4.708733558654785,
+      "learning_rate": 3.887755102040816e-06,
+      "loss": 4.2402,
+      "step": 620
+    },
+    {
+      "epoch": 5.04,
+      "grad_norm": 4.388338088989258,
+      "learning_rate": 3.785714285714286e-06,
+      "loss": 3.8437,
+      "step": 630
+    },
+    {
+      "epoch": 5.12,
+      "grad_norm": 4.25796365737915,
+      "learning_rate": 3.6836734693877554e-06,
+      "loss": 3.8926,
+      "step": 640
+    },
+    {
+      "epoch": 5.2,
+      "grad_norm": 4.4803996086120605,
+      "learning_rate": 3.581632653061225e-06,
+      "loss": 3.7283,
+      "step": 650
+    },
+    {
+      "epoch": 5.28,
+      "grad_norm": 4.613894939422607,
+      "learning_rate": 3.479591836734694e-06,
+      "loss": 4.0191,
+      "step": 660
+    },
+    {
+      "epoch": 5.36,
+      "grad_norm": 4.880568027496338,
+      "learning_rate": 3.3775510204081634e-06,
+      "loss": 3.7278,
+      "step": 670
+    },
+    {
+      "epoch": 5.44,
+      "grad_norm": 4.900217533111572,
+      "learning_rate": 3.2755102040816328e-06,
+      "loss": 3.7275,
+      "step": 680
+    },
+    {
+      "epoch": 5.52,
+      "grad_norm": 5.112405300140381,
+      "learning_rate": 3.173469387755102e-06,
+      "loss": 3.6629,
+      "step": 690
+    },
+    {
+      "epoch": 5.6,
+      "grad_norm": 5.060956954956055,
+      "learning_rate": 3.071428571428572e-06,
+      "loss": 4.1205,
+      "step": 700
+    },
+    {
+      "epoch": 5.68,
+      "grad_norm": 4.825809001922607,
+      "learning_rate": 2.9693877551020413e-06,
+      "loss": 3.6707,
+      "step": 710
+    },
+    {
+      "epoch": 5.76,
+      "grad_norm": 4.704409122467041,
+      "learning_rate": 2.86734693877551e-06,
+      "loss": 3.9554,
+      "step": 720
+    },
+    {
+      "epoch": 5.84,
+      "grad_norm": 4.779526710510254,
+      "learning_rate": 2.7653061224489795e-06,
+      "loss": 3.707,
+      "step": 730
+    },
+    {
+      "epoch": 5.92,
+      "grad_norm": 4.754640102386475,
+      "learning_rate": 2.663265306122449e-06,
+      "loss": 3.9676,
+      "step": 740
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 4.465866565704346,
+      "learning_rate": 2.5612244897959187e-06,
+      "loss": 3.5938,
+      "step": 750
+    },
+    {
+      "epoch": 6.08,
+      "grad_norm": 5.0297722816467285,
+      "learning_rate": 2.459183673469388e-06,
+      "loss": 3.4993,
+      "step": 760
+    },
+    {
+      "epoch": 6.16,
+      "grad_norm": 5.041810989379883,
+      "learning_rate": 2.3571428571428574e-06,
+      "loss": 3.2929,
+      "step": 770
+    },
+    {
+      "epoch": 6.24,
+      "grad_norm": 5.438796043395996,
+      "learning_rate": 2.2551020408163267e-06,
+      "loss": 3.724,
+      "step": 780
+    },
+    {
+      "epoch": 6.32,
+      "grad_norm": 5.112025260925293,
+      "learning_rate": 2.153061224489796e-06,
+      "loss": 3.5256,
+      "step": 790
+    },
+    {
+      "epoch": 6.4,
+      "grad_norm": 4.986135959625244,
+      "learning_rate": 2.0510204081632654e-06,
+      "loss": 3.607,
+      "step": 800
+    },
+    {
+      "epoch": 6.48,
+      "grad_norm": 4.837615489959717,
+      "learning_rate": 1.948979591836735e-06,
+      "loss": 3.7234,
+      "step": 810
+    },
+    {
+      "epoch": 6.5600000000000005,
+      "grad_norm": 4.9461750984191895,
+      "learning_rate": 1.8469387755102042e-06,
+      "loss": 3.551,
+      "step": 820
+    },
+    {
+      "epoch": 6.64,
+      "grad_norm": 5.679599285125732,
+      "learning_rate": 1.7448979591836737e-06,
+      "loss": 3.694,
+      "step": 830
+    },
+    {
+      "epoch": 6.72,
+      "grad_norm": 4.861215114593506,
+      "learning_rate": 1.642857142857143e-06,
+      "loss": 3.5914,
+      "step": 840
+    },
+    {
+      "epoch": 6.8,
+      "grad_norm": 4.849487781524658,
+      "learning_rate": 1.5408163265306122e-06,
+      "loss": 3.5168,
+      "step": 850
+    },
+    {
+      "epoch": 6.88,
+      "grad_norm": 5.109979629516602,
+      "learning_rate": 1.4387755102040818e-06,
+      "loss": 3.8957,
+      "step": 860
+    },
+    {
+      "epoch": 6.96,
+      "grad_norm": 4.744449138641357,
+      "learning_rate": 1.3367346938775511e-06,
+      "loss": 3.6189,
+      "step": 870
+    },
+    {
+      "epoch": 7.04,
+      "grad_norm": 5.0839080810546875,
+      "learning_rate": 1.2346938775510205e-06,
+      "loss": 3.5112,
+      "step": 880
+    },
+    {
+      "epoch": 7.12,
+      "grad_norm": 5.095026016235352,
+      "learning_rate": 1.1326530612244898e-06,
+      "loss": 3.4154,
+      "step": 890
+    },
+    {
+      "epoch": 7.2,
+      "grad_norm": 5.221547603607178,
+      "learning_rate": 1.0306122448979592e-06,
+      "loss": 3.5512,
+      "step": 900
+    },
+    {
+      "epoch": 7.28,
+      "grad_norm": 4.755325794219971,
+      "learning_rate": 9.285714285714287e-07,
+      "loss": 3.5292,
+      "step": 910
+    },
+    {
+      "epoch": 7.36,
+      "grad_norm": 5.137501239776611,
+      "learning_rate": 8.265306122448981e-07,
+      "loss": 3.4896,
+      "step": 920
+    },
+    {
+      "epoch": 7.44,
+      "grad_norm": 4.977344989776611,
+      "learning_rate": 7.244897959183674e-07,
+      "loss": 3.4731,
+      "step": 930
+    },
+    {
+      "epoch": 7.52,
+      "grad_norm": 4.887359142303467,
+      "learning_rate": 6.224489795918367e-07,
+      "loss": 3.7047,
+      "step": 940
+    },
+    {
+      "epoch": 7.6,
+      "grad_norm": 5.050560474395752,
+      "learning_rate": 5.204081632653062e-07,
+      "loss": 3.316,
+      "step": 950
+    },
+    {
+      "epoch": 7.68,
+      "grad_norm": 4.847559452056885,
+      "learning_rate": 4.183673469387756e-07,
+      "loss": 3.373,
+      "step": 960
+    },
+    {
+      "epoch": 7.76,
+      "grad_norm": 4.991832256317139,
+      "learning_rate": 3.163265306122449e-07,
+      "loss": 3.4879,
+      "step": 970
+    },
+    {
+      "epoch": 7.84,
+      "grad_norm": 5.611753463745117,
+      "learning_rate": 2.142857142857143e-07,
+      "loss": 3.4826,
+      "step": 980
+    },
+    {
+      "epoch": 7.92,
+      "grad_norm": 4.520301818847656,
+      "learning_rate": 1.1224489795918368e-07,
+      "loss": 3.5984,
+      "step": 990
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 5.4338178634643555,
+      "learning_rate": 1.0204081632653063e-08,
+      "loss": 3.3724,
+      "step": 1000
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 1000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.4845344768e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8948911f189f2562fd8b0ca8b0902c3023e843938c0d650710ab1f9998d54762
+size 5777