ninagroot/Llama-360Mtest

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: 3.3555
+- Loss: 4.2709
 
 ## Model description
 
@@ -48,9 +48,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 6.5087        | 0.99  | 45   | 5.0172          |
-| 3.5388        | 2.0   | 91   | 3.5244          |
-| 2.4645        | 2.97  | 135  | 3.3555          |
+| 4.5887        | 1.0   | 42   | 4.8016          |
+| 2.1371        | 2.0   | 84   | 4.1574          |
+| 1.4171        | 3.0   | 126  | 4.2709          |
 
 
 ### Framework versions

config.json

@@ -15,7 +15,7 @@
   "num_attention_heads": 8,
   "num_hidden_layers": 24,
   "num_key_value_heads": 8,
-  "pad_token_id": 50256,
+  "pad_token_id": 0,
   "pretraining_tp": 1,
   "rms_norm_eps": 1e-06,
   "rope_scaling": null,
@@ -24,5 +24,5 @@
   "torch_dtype": "float32",
   "transformers_version": "4.39.1",
   "use_cache": true,
-  "vocab_size": 50257
+  "vocab_size": 12214
 }

generation_config.json

@@ -2,6 +2,6 @@
   "_from_model_config": true,
   "bos_token_id": 1,
   "eos_token_id": 2,
-  "pad_token_id": 50256,
+  "pad_token_id": 0,
   "transformers_version": "4.39.1"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8edaec246532c355a71eed5d1e21f98b73538fff9497fab79efab02e9806d1d6
-size 1720553864
+oid sha256:794d95504e6c3eeb2e6c93a841abb188804e4df3d50372c54871dd40f3cebc5b
+size 1408905504

runs/Apr10_10-03-18_gcn25.local.snellius.surf.nl/events.out.tfevents.1712736207.gcn25.local.snellius.surf.nl.3444682.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe3ed1bfa89d176d334a07debf27f079b6433c620664341889320a91b5483540
+size 6951

special_tokens_map.json

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

tokenizer_config.json

@@ -1,42 +1,20 @@
 {
-  "add_bos_token": true,
-  "add_eos_token": false,
-  "add_prefix_space": true,
+  "add_prefix_space": false,
   "added_tokens_decoder": {
-    "0": {
-      "content": "<unk>",
+    "50256": {
+      "content": "<|endoftext|>",
       "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "1": {
-      "content": "<s>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "2": {
-      "content": "</s>",
-      "lstrip": false,
-      "normalized": false,
+      "normalized": true,
       "rstrip": false,
       "single_word": false,
       "special": true
     }
   },
   "bos_token": "<s>",
-  "clean_up_tokenization_spaces": false,
+  "clean_up_tokenization_spaces": true,
   "eos_token": "</s>",
-  "legacy": true,
-  "model_max_length": 30,
+  "model_max_length": 128,
   "pad_token": "<pad>",
-  "sp_model_kwargs": {},
-  "spaces_between_special_tokens": false,
-  "tokenizer_class": "LlamaTokenizer",
-  "unk_token": "<unk>",
-  "use_default_system_prompt": false
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ece1839304b089bc2ce12ceb44e6cc28d7b73192701f0dd19aa3fa12c7b00a73
+oid sha256:13e320fb248b2a7f327332399d659f25a2c7c8f78a3631c7a187c353d17322f0
 size 4984