End of training

README.md

@@ -1,9 +1,24 @@
 ---
 tags:
 - generated_from_trainer
+datasets:
+- MBZUAI/Bactrian-X
+metrics:
+- accuracy
 model-index:
 - name: Mistral-Indo
-  results: []
+  results:
+  - task:
+      name: Causal Language Modeling
+      type: text-generation
+    dataset:
+      name: MBZUAI/Bactrian-X id
+      type: MBZUAI/Bactrian-X
+      args: id
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.2462286333254075
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -11,7 +26,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Mistral-Indo
 
-This model was trained from scratch on an unknown dataset.
+This model was trained from scratch on the MBZUAI/Bactrian-X id dataset.
+It achieves the following results on the evaluation set:
+- Loss: 5.4269
+- Accuracy: 0.2462
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 0.2462286333254075,
+    "eval_loss": 5.426878929138184,
+    "eval_runtime": 8.9817,
+    "eval_samples": 37,
+    "eval_samples_per_second": 4.119,
+    "eval_steps_per_second": 2.115,
+    "perplexity": 227.43828551721677,
+    "total_flos": 3.670508706988032e+16,
+    "train_loss": 3.2965724309285482,
+    "train_runtime": 9164.1975,
+    "train_samples": 767,
+    "train_samples_per_second": 1.674,
+    "train_steps_per_second": 0.21
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 20.0,
+    "eval_accuracy": 0.2462286333254075,
+    "eval_loss": 5.426878929138184,
+    "eval_runtime": 8.9817,
+    "eval_samples": 37,
+    "eval_samples_per_second": 4.119,
+    "eval_steps_per_second": 2.115,
+    "perplexity": 227.43828551721677
+}

runs/Jan30_15-35-55_b59b81b92666/events.out.tfevents.1706638215.b59b81b92666.26.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0fd3a294eb3d69f405361efb7626818d06977edd5c6dc6408984a5dea3c29522
+size 411

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 20.0,
+    "total_flos": 3.670508706988032e+16,
+    "train_loss": 3.2965724309285482,
+    "train_runtime": 9164.1975,
+    "train_samples": 767,
+    "train_samples_per_second": 1.674,
+    "train_steps_per_second": 0.21
+}

trainer_state.json

@@ -0,0 +1,84 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 20.0,
+  "eval_steps": 500,
+  "global_step": 1920,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 2.08,
+      "learning_rate": 4.4791666666666673e-05,
+      "loss": 6.9421,
+      "step": 200
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 3.958333333333333e-05,
+      "loss": 4.8424,
+      "step": 400
+    },
+    {
+      "epoch": 6.25,
+      "learning_rate": 3.4375e-05,
+      "loss": 4.0383,
+      "step": 600
+    },
+    {
+      "epoch": 8.33,
+      "learning_rate": 2.916666666666667e-05,
+      "loss": 3.4406,
+      "step": 800
+    },
+    {
+      "epoch": 10.42,
+      "learning_rate": 2.3958333333333334e-05,
+      "loss": 2.9599,
+      "step": 1000
+    },
+    {
+      "epoch": 12.5,
+      "learning_rate": 1.8750000000000002e-05,
+      "loss": 2.5537,
+      "step": 1200
+    },
+    {
+      "epoch": 14.58,
+      "learning_rate": 1.3541666666666666e-05,
+      "loss": 2.2096,
+      "step": 1400
+    },
+    {
+      "epoch": 16.67,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 1.944,
+      "step": 1600
+    },
+    {
+      "epoch": 18.75,
+      "learning_rate": 3.125e-06,
+      "loss": 1.7395,
+      "step": 1800
+    },
+    {
+      "epoch": 20.0,
+      "step": 1920,
+      "total_flos": 3.670508706988032e+16,
+      "train_loss": 3.2965724309285482,
+      "train_runtime": 9164.1975,
+      "train_samples_per_second": 1.674,
+      "train_steps_per_second": 0.21
+    }
+  ],
+  "logging_steps": 200,
+  "max_steps": 1920,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 20,
+  "save_steps": 200,
+  "total_flos": 3.670508706988032e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}