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: 3.4944
+- Loss: 3.1541
 
 ## Model description
 
@@ -41,20 +41,16 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 100
-- num_epochs: 7
+- num_epochs: 3
 - mixed_precision_training: Native AMP
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 3.0504        | 1.0   | 197  | 3.3870          |
-| 1.9707        | 2.0   | 394  | 3.2436          |
-| 0.9107        | 3.0   | 591  | 3.3470          |
-| 0.5113        | 4.0   | 788  | 3.4348          |
-| 0.3692        | 5.0   | 985  | 3.4695          |
-| 0.2936        | 6.0   | 1182 | 3.4905          |
-| 0.2342        | 7.0   | 1379 | 3.4944          |
+| 6.439         | 0.99  | 49   | 4.6694          |
+| 3.4594        | 2.0   | 99   | 3.3145          |
+| 2.1092        | 2.97  | 147  | 3.1541          |
 
 
 ### Framework versions

config.json

@@ -10,7 +10,7 @@
   "hidden_size": 1024,
   "initializer_range": 0.02,
   "intermediate_size": 3072,
-  "max_position_embeddings": 64,
+  "max_position_embeddings": 256,
   "model_type": "llama",
   "num_attention_heads": 8,
   "num_hidden_layers": 24,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:59b8a5e4c46e9d9a9f3a4211804b7c30c33ff390779c553649ef74ee8ab7cb36
+oid sha256:d33b38d37faa26df141e2ede5d285133bdaf9d21506bc146e75a7a4da98e98fc
 size 1570992472

runs/Apr10_10-27-16_gcn13.local.snellius.surf.nl/events.out.tfevents.1712737646.gcn13.local.snellius.surf.nl.1571505.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a53271a420258bfa414c3fdbd4eb41bde06c8145743d3dcb4a7b372c27a80cdb
+size 7173

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_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
-    },
-    "12214": {
-      "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>",
-  "model_max_length": 128,
+  "legacy": true,
+  "model_max_length": 32,
   "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:c909776f3d4c8ae287c4e7b022d707381710ed10e5e8ad0394e3fe7325f879d8
+oid sha256:4cf117e95f9040050d4df8601ab7207ca1afbad249c15f452b953ceb694d4231
 size 4984