add evaluations

evals/mmlu/mmlu_nl_Llama-2-13b-chat-dutch.json

@@ -0,0 +1,23 @@
+{
+  "results": {
+    "mmlu_nl": {
+      "acc": 0.4120057676254079,
+      "acc_stderr": 0.00428791717360262,
+      "acc_norm": 0.35387417469833804,
+      "acc_norm_stderr": 0.0041657307070805455
+    }
+  },
+  "versions": {
+    "mmlu_nl": 0
+  },
+  "config": {
+    "model": "hf-auto",
+    "model_args": "pretrained=BramVanroy/Llama-2-13b-chat-dutch,use_accelerate=True,device_map_option=auto,dtype=bfloat16,load_in_8bit=True",
+    "batch_size": 1,
+    "device": "cuda",
+    "no_cache": false,
+    "limit": null,
+    "bootstrap_iters": 100000,
+    "description_dict": {}
+  }
+}

evals/mmlu/mmlu_nl_Llama-2-13b-chat-hf.json

@@ -0,0 +1,23 @@
+{
+  "results": {
+    "mmlu_nl": {
+      "acc": 0.4193670790012901,
+      "acc_stderr": 0.004298888710440378,
+      "acc_norm": 0.37398497381801626,
+      "acc_norm_stderr": 0.004215291981440015
+    }
+  },
+  "versions": {
+    "mmlu_nl": 0
+  },
+  "config": {
+    "model": "hf-auto",
+    "model_args": "pretrained=meta-llama/Llama-2-13b-chat-hf,use_accelerate=True,device_map_option=auto,dtype=bfloat16,load_in_8bit=True",
+    "batch_size": 1,
+    "device": "cuda",
+    "no_cache": false,
+    "limit": null,
+    "bootstrap_iters": 100000,
+    "description_dict": {}
+  }
+}

evals/mmlu/mmlu_nl_Llama-2-13b-hf.json

@@ -0,0 +1,23 @@
+{
+  "results": {
+    "mmlu_nl": {
+      "acc": 0.416938605145329,
+      "acc_stderr": 0.004295378164911688,
+      "acc_norm": 0.35637853836229794,
+      "acc_norm_stderr": 0.0041723356313098465
+    }
+  },
+  "versions": {
+    "mmlu_nl": 0
+  },
+  "config": {
+    "model": "hf-auto",
+    "model_args": "pretrained=meta-llama/Llama-2-13b-hf,use_accelerate=True,device_map_option=auto,dtype=bfloat16,load_in_8bit=True",
+    "batch_size": 1,
+    "device": "cuda",
+    "no_cache": false,
+    "limit": null,
+    "bootstrap_iters": 100000,
+    "description_dict": {}
+  }
+}

evals/mmlu/mmlu_nl_llama2-13b-ft-mc4_nl_cleaned_tiny.json

@@ -0,0 +1,23 @@
+{
+  "results": {
+    "mmlu_nl": {
+      "acc": 0.4082871670334674,
+      "acc_stderr": 0.004281999043287171,
+      "acc_norm": 0.35190103969036957,
+      "acc_norm_stderr": 0.004160438860806826
+    }
+  },
+  "versions": {
+    "mmlu_nl": 0
+  },
+  "config": {
+    "model": "hf-auto",
+    "model_args": "pretrained=BramVanroy/llama2-13b-ft-mc4_nl_cleaned_tiny,use_accelerate=True,device_map_option=auto,dtype=bfloat16,load_in_8bit=True",
+    "batch_size": 1,
+    "device": "cuda",
+    "no_cache": false,
+    "limit": null,
+    "bootstrap_iters": 100000,
+    "description_dict": {}
+  }
+}