Model save

README.md

@@ -1,19 +1,11 @@
 ---
 library_name: transformers
 license: llama3.2
-base_model: meta-llama/Llama-3.2-1B
+base_model: meta-llama/Llama-3.2-1B-Instruct
 tags:
-- alignment-handbook
 - trl
 - sft
 - generated_from_trainer
-- trl
-- sft
-- generated_from_trainer
-datasets:
-- barc0/transduction_angmented_100k-gpt4-description-gpt4omini-code_generated_problems
-- barc0/transduction_angmented_100k_gpt4o-mini_generated_problems
-- barc0/transduction_rearc_dataset_400k
 model-index:
 - name: llama3.2-1b-instruct-fft-transduction-engineer_lr1e-5_epoch4
   results: []
@@ -24,9 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama3.2-1b-instruct-fft-transduction-engineer_lr1e-5_epoch4
 
-This model is a fine-tuned version of [meta-llama/Llama-3.2-1B](https://huggingface.co/meta-llama/Llama-3.2-1B) on the barc0/transduction_angmented_100k-gpt4-description-gpt4omini-code_generated_problems, the barc0/transduction_angmented_100k_gpt4o-mini_generated_problems and the barc0/transduction_rearc_dataset_400k datasets.
+This model is a fine-tuned version of [meta-llama/Llama-3.2-1B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0363
+- Loss: 0.0409
 
 ## Model description
 
@@ -50,10 +42,10 @@ The following hyperparameters were used during training:
 - eval_batch_size: 8
 - seed: 42
 - distributed_type: multi-GPU
-- num_devices: 4
+- num_devices: 8
 - gradient_accumulation_steps: 2
-- total_train_batch_size: 128
-- total_eval_batch_size: 32
+- total_train_batch_size: 256
+- total_eval_batch_size: 64
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.1
@@ -61,12 +53,12 @@ The following hyperparameters were used during training:
 
 ### Training results
 
-| Training Loss | Epoch  | Step | Validation Loss |
-|:-------------:|:------:|:----:|:---------------:|
-| 0.0584        | 0.9998 | 2251 | 0.0631          |
-| 0.0502        | 2.0    | 4503 | 0.0447          |
-| 0.0421        | 2.9998 | 6754 | 0.0367          |
-| 0.0269        | 3.9991 | 9004 | 0.0363          |
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 0.0618        | 1.0   | 1126 | 0.0657          |
+| 0.0504        | 2.0   | 2252 | 0.0494          |
+| 0.0363        | 3.0   | 3378 | 0.0418          |
+| 0.0238        | 4.0   | 4504 | 0.0409          |
 
 
 ### Framework versions

all_results.json

@@ -1,14 +1,9 @@
 {
-    "epoch": 3.9991117033089054,
-    "eval_loss": 0.03634560480713844,
-    "eval_runtime": 402.1706,
-    "eval_samples": 15166,
-    "eval_samples_per_second": 37.71,
-    "eval_steps_per_second": 1.179,
-    "total_flos": 1010387024977920.0,
-    "train_loss": 0.05092642861352228,
-    "train_runtime": 98184.5209,
+    "epoch": 4.0,
+    "total_flos": 1010769106173952.0,
+    "train_loss": 0.055861650320550765,
+    "train_runtime": 18499.8374,
     "train_samples": 288129,
-    "train_samples_per_second": 11.738,
-    "train_steps_per_second": 0.092
+    "train_samples_per_second": 62.299,
+    "train_steps_per_second": 0.243
 }

generation_config.json

@@ -1,8 +1,11 @@
 {
-  "_from_model_config": true,
   "bos_token_id": 128000,
   "do_sample": true,
-  "eos_token_id": 128001,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
   "temperature": 0.6,
   "top_p": 0.9,
   "transformers_version": "4.45.0.dev0"

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 3.9991117033089054,
-    "total_flos": 1010387024977920.0,
-    "train_loss": 0.05092642861352228,
-    "train_runtime": 98184.5209,
+    "epoch": 4.0,
+    "total_flos": 1010769106173952.0,
+    "train_loss": 0.055861650320550765,
+    "train_runtime": 18499.8374,
     "train_samples": 288129,
-    "train_samples_per_second": 11.738,
-    "train_steps_per_second": 0.092
+    "train_samples_per_second": 62.299,
+    "train_steps_per_second": 0.243
 }