tickers_78_7_Channels_with_tape model training @ 2025-10-19 22:22:50

checkpoint-500/config.json

@@ -0,0 +1,76 @@
+{
+  "_name_or_path": "time_moe_50m",
+  "apply_aux_loss": true,
+  "architectures": [
+    "TimeMoeForPrediction"
+  ],
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "Maple728/TimeMoE-50M--configuration_time_moe.TimeMoeConfig",
+    "AutoModelForCausalLM": "Maple728/TimeMoE-50M--modeling_time_moe.TimeMoeForPrediction"
+  },
+  "channel_configs": [
+    [
+      63,
+      1,
+      1
+    ],
+    [
+      6,
+      1,
+      4
+    ],
+    [
+      6,
+      1,
+      5
+    ],
+    [
+      10,
+      1,
+      1
+    ],
+    [
+      5,
+      1,
+      1
+    ],
+    [
+      5,
+      1,
+      1
+    ],
+    [
+      5,
+      1,
+      2
+    ]
+  ],
+  "embedding_hidden_size": 128,
+  "hidden_act": "silu",
+  "hidden_size": 384,
+  "horizon_lengths": [
+    1,
+    8,
+    32,
+    64
+  ],
+  "initializer_range": 0.02,
+  "input_size": 42,
+  "intermediate_size": 1536,
+  "max_position_embeddings": 4096,
+  "model_type": "time_moe",
+  "num_attention_heads": 12,
+  "num_experts": 8,
+  "num_experts_per_tok": 2,
+  "num_hidden_layers": 12,
+  "num_key_value_heads": 12,
+  "rms_norm_eps": 1e-06,
+  "rope_theta": 10000,
+  "router_aux_loss_factor": 0.02,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.1",
+  "use_cache": true,
+  "use_dense": false
+}

checkpoint-500/generation_config.json

@@ -0,0 +1,4 @@
+{
+  "_from_model_config": true,
+  "transformers_version": "4.40.1"
+}

checkpoint-500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3633c26b8e782677bdf040812ca4d90cec5bb1193143f36ee5fad386d22081a1
+size 523327312

checkpoint-500/trainer_state.json

@@ -0,0 +1,114 @@
+{
+  "best_metric": 0.35392045974731445,
+  "best_model_checkpoint": "/home/yinkiat/logs/time_moe_tickers_78_7_Channels_with_tape/checkpoint-500",
+  "epoch": 0.6172839506172839,
+  "eval_steps": 250,
+  "global_step": 500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0012345679012345679,
+      "grad_norm": 0.6619585752487183,
+      "learning_rate": 9.999995299115366e-05,
+      "loss": 0.5046,
+      "step": 1
+    },
+    {
+      "epoch": 0.06172839506172839,
+      "grad_norm": 0.05976645648479462,
+      "learning_rate": 9.988256989482915e-05,
+      "loss": 0.4091,
+      "step": 50
+    },
+    {
+      "epoch": 0.12345679012345678,
+      "grad_norm": 0.2998131215572357,
+      "learning_rate": 9.953138276568462e-05,
+      "loss": 0.3854,
+      "step": 100
+    },
+    {
+      "epoch": 0.18518518518518517,
+      "grad_norm": 0.32706254720687866,
+      "learning_rate": 9.894973780788723e-05,
+      "loss": 0.3711,
+      "step": 150
+    },
+    {
+      "epoch": 0.24691358024691357,
+      "grad_norm": 0.18154236674308777,
+      "learning_rate": 9.814309923172227e-05,
+      "loss": 0.3614,
+      "step": 200
+    },
+    {
+      "epoch": 0.30864197530864196,
+      "grad_norm": 0.0911095142364502,
+      "learning_rate": 9.711904492941645e-05,
+      "loss": 0.3544,
+      "step": 250
+    },
+    {
+      "epoch": 0.30864197530864196,
+      "eval_loss": 0.36227765679359436,
+      "eval_runtime": 141.9212,
+      "eval_samples_per_second": 531.563,
+      "eval_steps_per_second": 16.615,
+      "step": 250
+    },
+    {
+      "epoch": 0.37037037037037035,
+      "grad_norm": 0.352092981338501,
+      "learning_rate": 9.588719528532342e-05,
+      "loss": 0.3503,
+      "step": 300
+    },
+    {
+      "epoch": 0.43209876543209874,
+      "grad_norm": 0.10818850994110107,
+      "learning_rate": 9.4459122798104e-05,
+      "loss": 0.3477,
+      "step": 350
+    },
+    {
+      "epoch": 0.49382716049382713,
+      "grad_norm": 0.09456054121255875,
+      "learning_rate": 9.284824336394749e-05,
+      "loss": 0.3425,
+      "step": 400
+    },
+    {
+      "epoch": 0.5555555555555556,
+      "grad_norm": 0.27488818764686584,
+      "learning_rate": 9.106969024216348e-05,
+      "loss": 0.3428,
+      "step": 450
+    },
+    {
+      "epoch": 0.6172839506172839,
+      "grad_norm": 0.10880504548549652,
+      "learning_rate": 8.914017188716347e-05,
+      "loss": 0.3406,
+      "step": 500
+    },
+    {
+      "epoch": 0.6172839506172839,
+      "eval_loss": 0.35392045974731445,
+      "eval_runtime": 145.9692,
+      "eval_samples_per_second": 516.821,
+      "eval_steps_per_second": 16.154,
+      "step": 500
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 1620,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 250,
+  "total_flos": 3164638740480000.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5edc06d66733730a64bef13b56324503876e24bbf51a066bf9182c51e90acc77
+size 5585

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:70c9a4f3d616b9b829abe11cd40619e0f19743fb43719e6eb9d1eb6e0da318a7
+oid sha256:3633c26b8e782677bdf040812ca4d90cec5bb1193143f36ee5fad386d22081a1
 size 523327312

tb_logs/events.out.tfevents.1760881013.luyao1.106761.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ea08c84f4b9d300b5dfd7bd474ffb3e5cf720b71c7a034f605d0f4678588973
+size 12624