Model save

README.md

@@ -1,4 +1,5 @@
 ---
+license: apache-2.0
 library_name: peft
 tags:
 - trl
@@ -6,7 +7,7 @@ tags:
 - generated_from_trainer
 datasets:
 - generator
-base_model: data/Tukan-1.1B-Chat-reasoning-sft-COLA_merged_0
+base_model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
 model-index:
 - name: Tukan-1.1B-Chat-reasoning-sft-COLA
   results: []
@@ -17,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # Tukan-1.1B-Chat-reasoning-sft-COLA
 
-This model was trained from scratch on the generator dataset.
+This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-Chat-v1.0](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0) on the generator dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.5347
+- Loss: 1.5464
 
 ## 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": "TinyLlama/TinyLlama-1.1B-Chat-v1.0",
   "bias": "none",
   "fan_in_fan_out": false,
   "inference_mode": true,
@@ -16,14 +16,14 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "k_proj",
+    "o_proj",
     "v_proj",
     "up_proj",
+    "gate_proj",
     "lm_head",
-    "o_proj",
-    "down_proj",
     "q_proj",
-    "gate_proj"
+    "k_proj",
+    "down_proj"
   ],
   "task_type": "CAUSAL_LM"
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:faec55aa5e087bfcfcc1a48ca7befe73c653e838144f524d8570f3d354d9d2f8
+oid sha256:68b264ea9c5c8573c5bac38b0d63c8d49f62b4bc8f9b8b97dcb0cff3e141bf3a
 size 26361536

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 0.95,
-    "eval_loss": 1.5346883535385132,
-    "eval_runtime": 1.3685,
+    "eval_loss": 1.5464149713516235,
+    "eval_runtime": 1.3787,
     "eval_samples": 36,
-    "eval_samples_per_second": 5.115,
-    "eval_steps_per_second": 2.192,
-    "train_loss": 1.555079698562622,
-    "train_runtime": 77.8333,
+    "eval_samples_per_second": 5.077,
+    "eval_steps_per_second": 2.176,
+    "train_loss": 1.5683434009552002,
+    "train_runtime": 86.2408,
     "train_samples": 682,
-    "train_samples_per_second": 1.567,
-    "train_steps_per_second": 0.013
+    "train_samples_per_second": 1.415,
+    "train_steps_per_second": 0.012
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 0.95,
-    "eval_loss": 1.5346883535385132,
-    "eval_runtime": 1.3685,
+    "eval_loss": 1.5464149713516235,
+    "eval_runtime": 1.3787,
     "eval_samples": 36,
-    "eval_samples_per_second": 5.115,
-    "eval_steps_per_second": 2.192
+    "eval_samples_per_second": 5.077,
+    "eval_steps_per_second": 2.176
 }

runs/Jan21_18-49-28_98f107f1aa39/events.out.tfevents.1705862986.98f107f1aa39.101023.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e32d6af1d9d9c738b51ec8e56a1b167019e7a8199d06254b97de0dec6e6c90fc
+size 5042

runs/Jan21_18-49-28_98f107f1aa39/events.out.tfevents.1705863074.98f107f1aa39.101023.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a15dfe8e10f5a0d0a7b43b116ec8129f944940a9cd2f3b2d652ba7a7c801b34e
+size 354

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 0.95,
-    "train_loss": 1.555079698562622,
-    "train_runtime": 77.8333,
+    "train_loss": 1.5683434009552002,
+    "train_runtime": 86.2408,
     "train_samples": 682,
-    "train_samples_per_second": 1.567,
-    "train_steps_per_second": 0.013
+    "train_samples_per_second": 1.415,
+    "train_steps_per_second": 0.012
 }

trainer_state.json

@@ -11,17 +11,17 @@
     {
       "epoch": 0.95,
       "learning_rate": 0.0,
-      "loss": 1.5551,
+      "loss": 1.5683,
       "step": 1
     },
     {
       "epoch": 0.95,
       "step": 1,
       "total_flos": 1535153516052480.0,
-      "train_loss": 1.555079698562622,
-      "train_runtime": 77.8333,
-      "train_samples_per_second": 1.567,
-      "train_steps_per_second": 0.013
+      "train_loss": 1.5683434009552002,
+      "train_runtime": 86.2408,
+      "train_samples_per_second": 1.415,
+      "train_steps_per_second": 0.012
     }
   ],
   "logging_steps": 5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aeb0ef68df7eba155819ecadf94529b54aabf6815f0cfce9a25ffa9c9040fe91
+oid sha256:54992a2c8472fa25aaf5bd77b4f99dc364639b438e0687926d0334ae0c178a23
 size 4792