Model save

README.md

@@ -2,15 +2,11 @@
 license: mit
 base_model: gpt2
 tags:
-- alignment-handbook
-- trl
-- sft
-- generated_from_trainer
 - trl
 - sft
 - generated_from_trainer
 datasets:
-- appliedai-qx/sample-dataset-ah
+- generator
 model-index:
 - name: gpt2
   results: []
@@ -21,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # gpt2
 
-This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the appliedai-qx/sample-dataset-ah dataset.
+This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the generator dataset.
 
 ## Model description
 
@@ -45,9 +41,9 @@ The following hyperparameters were used during training:
 - eval_batch_size: 8
 - seed: 42
 - distributed_type: multi-GPU
-- num_devices: 4
-- total_train_batch_size: 64
-- total_eval_batch_size: 32
+- num_devices: 32
+- total_train_batch_size: 512
+- total_eval_batch_size: 256
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.1

all_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
-    "total_flos": 7357983621120000.0,
-    "train_loss": 1.3830752210183577,
-    "train_runtime": 55.0309,
-    "train_samples": 10000,
-    "train_samples_per_second": 255.239,
-    "train_steps_per_second": 3.998
+    "total_flos": 1.011923420184576e+18,
+    "train_loss": 0.9601045426144794,
+    "train_runtime": 898.0661,
+    "train_samples": 1371223,
+    "train_samples_per_second": 2156.094,
+    "train_steps_per_second": 4.211
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb8e6337b30209ac71373e81fd96da57437e0dd62d8e0bf87cccc75d7c16df40
+oid sha256:2aa31b4c25100273b321a8bd3f669cc58c946babae8edc15a5e09670ddcdf324
 size 248894656

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
-    "total_flos": 7357983621120000.0,
-    "train_loss": 1.3830752210183577,
-    "train_runtime": 55.0309,
-    "train_samples": 10000,
-    "train_samples_per_second": 255.239,
-    "train_steps_per_second": 3.998
+    "total_flos": 1.011923420184576e+18,
+    "train_loss": 0.9601045426144794,
+    "train_runtime": 898.0661,
+    "train_samples": 1371223,
+    "train_samples_per_second": 2156.094,
+    "train_steps_per_second": 4.211
 }