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: 4.6344
+- Loss: 3.4944
 
 ## Model description
 
@@ -48,13 +48,13 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 4.5524        | 1.0   | 42   | 4.7852          |
-| 2.1679        | 2.0   | 84   | 4.1345          |
-| 1.4522        | 3.0   | 126  | 4.2880          |
-| 0.9328        | 4.0   | 168  | 4.4565          |
-| 0.5606        | 5.0   | 210  | 4.6031          |
-| 0.3514        | 6.0   | 252  | 4.6564          |
-| 0.2839        | 7.0   | 294  | 4.6344          |
+| 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          |
 
 
 ### Framework versions

config.json

@@ -10,7 +10,7 @@
   "hidden_size": 1024,
   "initializer_range": 0.02,
   "intermediate_size": 3072,
-  "max_position_embeddings": 256,
+  "max_position_embeddings": 64,
   "model_type": "llama",
   "num_attention_heads": 8,
   "num_hidden_layers": 24,
@@ -24,5 +24,5 @@
   "torch_dtype": "float32",
   "transformers_version": "4.39.1",
   "use_cache": true,
-  "vocab_size": 12214
+  "vocab_size": 32000
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:581986c8393aab378ba562c2e1515606e5f06fcb1b3cd14bf03d7ba824cd0bce
-size 1408905504
+oid sha256:59b8a5e4c46e9d9a9f3a4211804b7c30c33ff390779c553649ef74ee8ab7cb36
+size 1570992472

runs/Apr10_10-13-45_gcn68.local.snellius.surf.nl/events.out.tfevents.1712736835.gcn68.local.snellius.surf.nl.3597541.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d70e0bdd4cd3a06c0b23d8ea797205d2805135b31f5cc2e59c4d81aec544723b
+size 13007

runs/Apr10_10-19-18_gcn16.local.snellius.surf.nl/events.out.tfevents.1712737168.gcn16.local.snellius.surf.nl.150583.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d91c1da480cb97fc52467549542d5313f88afaff7f18fdfee41499544f3a13d9
+size 21137

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3ed61a658032aee43721b02ee38be59a2bf2b6b2a5447552875f86900ad7932e
+oid sha256:c909776f3d4c8ae287c4e7b022d707381710ed10e5e8ad0394e3fe7325f879d8
 size 4984