musiclang/musiclang-v2-bpe

README.md

@@ -1,5 +1,4 @@
 ---
-base_model: musiclang/musiclang-v2-xl
 tags:
 - generated_from_trainer
 model-index:
@@ -12,9 +11,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # model
 
-This model is a fine-tuned version of [musiclang/musiclang-v2-xl](https://huggingface.co/musiclang/musiclang-v2-xl) on the None dataset.
+This model is a fine-tuned version of [](https://huggingface.co/) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2930
+- Loss: 0.9350
 
 ## Model description
 
@@ -34,8 +33,8 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0001
-- train_batch_size: 24
-- eval_batch_size: 24
+- train_batch_size: 16
+- eval_batch_size: 16
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine_with_restarts
@@ -47,36 +46,20 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step  | Validation Loss |
 |:-------------:|:-----:|:-----:|:---------------:|
-| 0.4098        | 0.03  | 2000  | 0.3828          |
-| 0.3901        | 0.07  | 4000  | 0.3684          |
-| 0.3737        | 0.1   | 6000  | 0.3569          |
-| 0.3652        | 0.13  | 8000  | 0.3489          |
-| 0.356         | 0.17  | 10000 | 0.3393          |
-| 0.35          | 0.2   | 12000 | 0.3342          |
-| 0.3443        | 0.23  | 14000 | 0.3282          |
-| 0.3371        | 0.26  | 16000 | 0.3235          |
-| 0.3361        | 0.3   | 18000 | 0.3201          |
-| 0.3301        | 0.33  | 20000 | 0.3160          |
-| 0.3253        | 0.36  | 22000 | 0.3131          |
-| 0.327         | 0.4   | 24000 | 0.3109          |
-| 0.3225        | 0.43  | 26000 | 0.3089          |
-| 0.3156        | 0.46  | 28000 | 0.3066          |
-| 0.3147        | 0.5   | 30000 | 0.3045          |
-| 0.3182        | 0.53  | 32000 | 0.3026          |
-| 0.3129        | 0.56  | 34000 | 0.3017          |
-| 0.3132        | 0.59  | 36000 | 0.3008          |
-| 0.3109        | 0.63  | 38000 | 0.2987          |
-| 0.3092        | 0.66  | 40000 | 0.2972          |
-| 0.3091        | 0.69  | 42000 | 0.2963          |
-| 0.3034        | 0.73  | 44000 | 0.2960          |
-| 0.3061        | 0.76  | 46000 | 0.2956          |
-| 0.3044        | 0.79  | 48000 | 0.2946          |
-| 0.3036        | 0.83  | 50000 | 0.2940          |
-| 0.3003        | 0.86  | 52000 | 0.2939          |
-| 0.303         | 0.89  | 54000 | 0.2934          |
-| 0.3007        | 0.93  | 56000 | 0.2932          |
-| 0.3009        | 0.96  | 58000 | 0.2930          |
-| 0.3           | 0.99  | 60000 | 0.2930          |
+| 2.799         | 0.07  | 2000  | 2.6594          |
+| 1.5819        | 0.13  | 4000  | 1.4878          |
+| 1.3252        | 0.2   | 6000  | 1.2718          |
+| 1.2293        | 0.27  | 8000  | 1.1748          |
+| 1.141         | 0.34  | 10000 | 1.1004          |
+| 1.093         | 0.4   | 12000 | 1.0582          |
+| 1.0601        | 0.47  | 14000 | 1.0282          |
+| 1.0285        | 0.54  | 16000 | 0.9957          |
+| 1.002         | 0.61  | 18000 | 0.9794          |
+| 0.9876        | 0.67  | 20000 | 0.9605          |
+| 0.9903        | 0.74  | 22000 | 0.9489          |
+| 0.9698        | 0.81  | 24000 | 0.9418          |
+| 0.962         | 0.88  | 26000 | 0.9370          |
+| 0.9598        | 0.94  | 28000 | 0.9350          |
 
 
 ### Framework versions

config.json

@@ -1,22 +1,21 @@
 {
-  "_name_or_path": "musiclang/musiclang-v2-xl",
   "activation_function": "gelu_new",
   "architectures": [
     "GPT2LMHeadModel"
   ],
   "attn_pdrop": 0.1,
-  "bos_token_id": 5,
+  "bos_token_id": 30000,
   "embd_pdrop": 0.1,
-  "eos_token_id": 374,
+  "eos_token_id": 30000,
   "initializer_range": 0.02,
   "layer_norm_epsilon": 1e-05,
   "model_type": "gpt2",
-  "n_embd": 300,
+  "n_embd": 600,
   "n_head": 10,
   "n_inner": null,
   "n_layer": 10,
   "n_positions": 4096,
-  "padding_token_id": 3,
+  "padding_token_id": 30000,
   "reorder_and_upcast_attn": false,
   "resid_pdrop": 0.1,
   "scale_attn_by_inverse_layer_idx": false,
@@ -29,5 +28,5 @@
   "torch_dtype": "float32",
   "transformers_version": "4.37.2",
   "use_cache": true,
-  "vocab_size": 374
+  "vocab_size": 30001
 }

generation_config.json

@@ -1,6 +1,6 @@
 {
   "_from_model_config": true,
-  "bos_token_id": 5,
-  "eos_token_id": 374,
+  "bos_token_id": 30000,
+  "eos_token_id": 30000,
   "transformers_version": "4.37.2"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0edaec61be7b684821241f0871d5e8a7b48d865f5ad1aee2495c9dc9f54290a8
-size 48734696
+oid sha256:22726d5d220ff7c2d8b4bda44044f456543da712c6f13abceb7ebfbdf291e29c
+size 254962056

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ded85577c88c0610c6cc2a534d279f3cac2c97cc898df360f85bd6ae1a843d78
+oid sha256:db2daf1ba377eb851dad0905d389222e965f048ff071e14d1039b7e459c05c67
 size 4664