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: 3.2009
+- Loss: 3.3555
 
 ## Model description
 
@@ -48,9 +48,9 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 2.9075        | 1.0   | 210  | 3.3783          |
-| 1.7841        | 2.0   | 420  | 3.2031          |
-| 0.9611        | 3.0   | 630  | 3.2009          |
+| 6.5087        | 0.99  | 45   | 5.0172          |
+| 3.5388        | 2.0   | 91   | 3.5244          |
+| 2.4645        | 2.97  | 135  | 3.3555          |
 
 
 ### Framework versions

config.json

@@ -10,12 +10,12 @@
   "hidden_size": 1024,
   "initializer_range": 0.02,
   "intermediate_size": 3072,
-  "max_position_embeddings": 60,
+  "max_position_embeddings": 256,
   "model_type": "llama",
   "num_attention_heads": 8,
   "num_hidden_layers": 24,
   "num_key_value_heads": 8,
-  "pad_token_id": 0,
+  "pad_token_id": 50256,
   "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": 32000
+  "vocab_size": 50257
 }

generation_config.json

@@ -2,6 +2,6 @@
   "_from_model_config": true,
   "bos_token_id": 1,
   "eos_token_id": 2,
-  "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 sha256:59ad553e073602086265868ee976391542d2c79bd6b9a676bda55cc75035c179
-size 1570992472
+oid sha256:8edaec246532c355a71eed5d1e21f98b73538fff9497fab79efab02e9806d1d6
+size 1720553864

runs/Apr10_09-53-11_gcn37.local.snellius.surf.nl/events.out.tfevents.1712735600.gcn37.local.snellius.surf.nl.1085902.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c918c79e787df1bdd8c95a29b516042be44267a89a726a65f5a830048330fb4b
+size 6966

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:391459dd6269ecbb83dba9c883903f96777586da6a2d64ed6471804569bbd266
+oid sha256:ece1839304b089bc2ce12ceb44e6cc28d7b73192701f0dd19aa3fa12c7b00a73
 size 4984