ninagroot/GPT2-705Mtest

README.md

@@ -13,7 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 3.6575
+- Loss: 3.6555
 
 ## Model description
 
@@ -33,27 +33,24 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.00025
-- train_batch_size: 1
+- train_batch_size: 16
 - eval_batch_size: 8
 - seed: 42
 - gradient_accumulation_steps: 2
-- total_train_batch_size: 2
+- total_train_batch_size: 32
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 100
-- num_epochs: 6
+- num_epochs: 3
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 2.9814        | 1.0   | 786  | 3.7914          |
-| 2.5963        | 2.0   | 1572 | 3.6927          |
-| 2.3478        | 3.0   | 2358 | 3.7172          |
-| 2.0341        | 4.0   | 3144 | 3.6845          |
-| 1.7373        | 5.0   | 3930 | 3.6596          |
-| 1.7065        | 6.0   | 4716 | 3.6575          |
+| 3.073         | 1.0   | 194  | 3.6424          |
+| 2.0148        | 2.0   | 389  | 3.6091          |
+| 1.3653        | 2.99  | 582  | 3.6555          |
 
 
 ### Framework versions

config.json

@@ -11,11 +11,11 @@
   "layer_norm_epsilon": 1e-05,
   "model_type": "gpt2",
   "n_embd": 1536,
-  "n_head": 16,
+  "n_head": 8,
   "n_inner": null,
   "n_layer": 24,
-  "n_positions": 256,
-  "pad_token_id": 0,
+  "n_positions": 60,
+  "pad_token_id": 50256,
   "reorder_and_upcast_attn": false,
   "resid_pdrop": 0.0,
   "scale_attn_by_inverse_layer_idx": false,
@@ -28,5 +28,5 @@
   "torch_dtype": "float32",
   "transformers_version": "4.39.1",
   "use_cache": true,
-  "vocab_size": 32000
+  "vocab_size": 50257
 }

generation_config.json

@@ -2,6 +2,6 @@
   "_from_model_config": true,
   "bos_token_id": 50256,
   "eos_token_id": 50256,
-  "pad_token_id": 0,
+  "pad_token_id": 50256,
   "transformers_version": "4.39.1"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6c91e31860b19345b05e0f1e9e1ba2de97ffc56b229cb792f4542af9b151e2af
-size 2918049568
+oid sha256:2abf6ffbb842f8f7acca2c6c3bd5c2c8b0897130820c1844f28256896bf9f235
+size 3029016352

runs/Apr05_15-32-40_gcn33.local.snellius.surf.nl/events.out.tfevents.1712323967.gcn33.local.snellius.surf.nl.3834045.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fcfd9e49ca484e635452b2c37963b8444882b053cdd6140ddd25243805cb1a99
+size 11944

special_tokens_map.json

@@ -2,5 +2,11 @@
   "bos_token": "<s>",
   "eos_token": "</s>",
   "pad_token": "<pad>",
-  "unk_token": "<|endoftext|>"
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
 }

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -1,8 +1,10 @@
 {
-  "add_prefix_space": false,
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": true,
   "added_tokens_decoder": {
     "0": {
-      "content": "<pad>",
+      "content": "<unk>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
@@ -24,21 +26,17 @@
       "rstrip": false,
       "single_word": false,
       "special": true
-    },
-    "4312": {
-      "content": "<|endoftext|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
     }
   },
   "bos_token": "<s>",
-  "clean_up_tokenization_spaces": true,
+  "clean_up_tokenization_spaces": false,
   "eos_token": "</s>",
+  "legacy": true,
   "model_max_length": 128,
   "pad_token": "<pad>",
-  "tokenizer_class": "GPT2Tokenizer",
-  "unk_token": "<|endoftext|>"
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aa0847ab599c58a1133dac29dd7ae134718e1285515ee60a80d4d30e0cb92086
+oid sha256:c40e2376c3eb0011bb960b01f5109ca28ebed0d38154671666787999fe15f16c
 size 4984