Upload folder using huggingface_hub

.ipynb_checkpoints/config-checkpoint.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "./output3",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 256,
+  "initializer_range": 0.02,
+  "intermediate_size": 512,
+  "max_position_embeddings": 8192,
+  "model_type": "llama",
+  "num_attention_heads": 4,
+  "num_hidden_layers": 2,
+  "num_key_value_heads": 4,
+  "pretraining_tp": 2,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.34.0.dev0",
+  "use_cache": true,
+  "vocab_size": 32000
+}

README.md

@@ -1,18 +1,18 @@
 ---
-base_model: ./output
+base_model: ./output4
 tags:
 - generated_from_trainer
 model-index:
-- name: output
+- name: output4
   results: []
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# output
+# output4
 
-This model is a fine-tuned version of [./output](https://huggingface.co/./output) on an unknown dataset.
+This model is a fine-tuned version of [./output4](https://huggingface.co/./output4) on an unknown dataset.
 
 ## Model description
 
@@ -35,8 +35,8 @@ The following hyperparameters were used during training:
 - train_batch_size: 1
 - eval_batch_size: 1
 - seed: 42
-- gradient_accumulation_steps: 4
-- total_train_batch_size: 4
+- gradient_accumulation_steps: 128
+- total_train_batch_size: 128
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: constant
 - num_epochs: 5.0

all_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 4.91,
-    "train_loss": 2.390471041644061,
-    "train_runtime": 81.0032,
-    "train_samples": 110,
-    "train_samples_per_second": 6.79,
-    "train_steps_per_second": 1.667
+    "epoch": 4.72,
+    "train_loss": 4.668148568960336,
+    "train_runtime": 1730.4041,
+    "train_samples": 1763,
+    "train_samples_per_second": 5.094,
+    "train_steps_per_second": 0.038
 }

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "./output",
+  "_name_or_path": "./output4",
   "architectures": [
     "LlamaForCausalLM"
   ],
@@ -14,7 +14,7 @@
   "num_attention_heads": 4,
   "num_hidden_layers": 2,
   "num_key_value_heads": 4,
-  "pretraining_tp": 1,
+  "pretraining_tp": 2,
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,
   "rope_theta": 10000.0,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3550fd807f67088b3db8d9b8c48f71d83695a9ce4da9795c8f34c9ce50adbab1
+oid sha256:2849613c32e353c618b1f18ae87d4b47413ee690ecec07623224d97180e03129
 size 70790896

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 4.91,
-    "train_loss": 2.390471041644061,
-    "train_runtime": 81.0032,
-    "train_samples": 110,
-    "train_samples_per_second": 6.79,
-    "train_steps_per_second": 1.667
+    "epoch": 4.72,
+    "train_loss": 4.668148568960336,
+    "train_runtime": 1730.4041,
+    "train_samples": 1763,
+    "train_samples_per_second": 5.094,
+    "train_steps_per_second": 0.038
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5dec8047718524269e1241952d3698cdbfee23464c6ef644897b4290313cfb34
+oid sha256:62ab075136e8e12e6d8a904f1b44613cd64ca0e659e605796d6ac5e0a391b8c5
 size 4027