Model save

README.md

@@ -6,7 +6,7 @@ tags:
 - generated_from_trainer
 datasets:
 - generator
-base_model: data/Tukan-1.1B-Chat-reasoning-sft-COLA_merged_0
+base_model: data/Tukan-1.1B-Chat-reasoning-sft-COLA_merged_1
 model-index:
 - name: Tukan-1.1B-Chat-reasoning-sft-COLA
   results: []
@@ -19,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model was trained from scratch on the generator dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.5414
+- Loss: 1.5347
 
 ## Model description
 

adapter_config.json

@@ -1,7 +1,7 @@
 {
   "alpha_pattern": {},
   "auto_mapping": null,
-  "base_model_name_or_path": "data/Tukan-1.1B-Chat-reasoning-sft-COLA_merged_0",
+  "base_model_name_or_path": "data/Tukan-1.1B-Chat-reasoning-sft-COLA_merged_1",
   "bias": "none",
   "fan_in_fan_out": false,
   "inference_mode": true,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:57448e5cb7a06f63d75e80343ec9dff64ed69de618f3355b139bf7ce1d2a649b
+oid sha256:faec55aa5e087bfcfcc1a48ca7befe73c653e838144f524d8570f3d354d9d2f8
 size 26361536

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 0.95,
-    "eval_loss": 1.541361689567566,
-    "eval_runtime": 1.365,
+    "eval_loss": 1.5346883535385132,
+    "eval_runtime": 1.3676,
     "eval_samples": 36,
-    "eval_samples_per_second": 5.128,
-    "eval_steps_per_second": 2.198,
-    "train_loss": 1.5626062154769897,
-    "train_runtime": 77.8859,
+    "eval_samples_per_second": 5.119,
+    "eval_steps_per_second": 2.194,
+    "train_loss": 1.555079698562622,
+    "train_runtime": 77.9815,
     "train_samples": 682,
-    "train_samples_per_second": 1.566,
+    "train_samples_per_second": 1.564,
     "train_steps_per_second": 0.013
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 0.95,
-    "eval_loss": 1.541361689567566,
-    "eval_runtime": 1.365,
+    "eval_loss": 1.5346883535385132,
+    "eval_runtime": 1.3676,
     "eval_samples": 36,
-    "eval_samples_per_second": 5.128,
-    "eval_steps_per_second": 2.198
+    "eval_samples_per_second": 5.119,
+    "eval_steps_per_second": 2.194
 }

runs/Jan21_18-49-28_98f107f1aa39/events.out.tfevents.1705863201.98f107f1aa39.101023.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c153299128922add5cbebb8480945ab662afdd4fa707f360bc48ac1401c1dd53
+size 5056

runs/Jan21_18-49-28_98f107f1aa39/events.out.tfevents.1705863280.98f107f1aa39.101023.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d056605e40207e1443cef7eac0ee125738a21f86296cd4fc3bcf73a0ea4362d
+size 354

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 0.95,
-    "train_loss": 1.5626062154769897,
-    "train_runtime": 77.8859,
+    "train_loss": 1.555079698562622,
+    "train_runtime": 77.9815,
     "train_samples": 682,
-    "train_samples_per_second": 1.566,
+    "train_samples_per_second": 1.564,
     "train_steps_per_second": 0.013
 }

trainer_state.json

@@ -11,16 +11,16 @@
     {
       "epoch": 0.95,
       "learning_rate": 0.0,
-      "loss": 1.5626,
+      "loss": 1.5551,
       "step": 1
     },
     {
       "epoch": 0.95,
       "step": 1,
       "total_flos": 1535153516052480.0,
-      "train_loss": 1.5626062154769897,
-      "train_runtime": 77.8859,
-      "train_samples_per_second": 1.566,
+      "train_loss": 1.555079698562622,
+      "train_runtime": 77.9815,
+      "train_samples_per_second": 1.564,
       "train_steps_per_second": 0.013
     }
   ],