chauhoang commited on
Commit
8b0ae51
·
verified ·
1 Parent(s): ba959aa

End of training

Browse files
Files changed (2) hide show
  1. README.md +11 -4
  2. adapter_model.bin +1 -1
README.md CHANGED
@@ -65,7 +65,7 @@ lora_model_dir: null
65
  lora_r: 8
66
  lora_target_linear: true
67
  lr_scheduler: cosine
68
- max_steps: 1
69
  micro_batch_size: 2
70
  mlflow_experiment_name: /tmp/50a3279bbe0d6c57_train_data.json
71
  model_type: AutoModelForCausalLM
@@ -90,7 +90,7 @@ wandb_name: b1558c1d-92f1-4acc-948a-bb5d77f97503
90
  wandb_project: Gradients-On-Demand
91
  wandb_run: your_name
92
  wandb_runid: b1558c1d-92f1-4acc-948a-bb5d77f97503
93
- warmup_steps: 1
94
  weight_decay: 0.0
95
  xformers_attention: null
96
 
@@ -101,6 +101,8 @@ xformers_attention: null
101
  # b674cfc8-d0a7-94a7-52df-8ce4be32eb9c
102
 
103
  This model is a fine-tuned version of [unsloth/Llama-3.2-3B](https://huggingface.co/unsloth/Llama-3.2-3B) on the None dataset.
 
 
104
 
105
  ## Model description
106
 
@@ -127,14 +129,19 @@ The following hyperparameters were used during training:
127
  - total_train_batch_size: 8
128
  - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
129
  - lr_scheduler_type: cosine
130
- - lr_scheduler_warmup_steps: 2
131
- - training_steps: 1
132
 
133
  ### Training results
134
 
135
  | Training Loss | Epoch | Step | Validation Loss |
136
  |:-------------:|:------:|:----:|:---------------:|
137
  | No log | 0.0001 | 1 | 2.5663 |
 
 
 
 
 
138
 
139
 
140
  ### Framework versions
 
65
  lora_r: 8
66
  lora_target_linear: true
67
  lr_scheduler: cosine
68
+ max_steps: 50
69
  micro_batch_size: 2
70
  mlflow_experiment_name: /tmp/50a3279bbe0d6c57_train_data.json
71
  model_type: AutoModelForCausalLM
 
90
  wandb_project: Gradients-On-Demand
91
  wandb_run: your_name
92
  wandb_runid: b1558c1d-92f1-4acc-948a-bb5d77f97503
93
+ warmup_steps: 10
94
  weight_decay: 0.0
95
  xformers_attention: null
96
 
 
101
  # b674cfc8-d0a7-94a7-52df-8ce4be32eb9c
102
 
103
  This model is a fine-tuned version of [unsloth/Llama-3.2-3B](https://huggingface.co/unsloth/Llama-3.2-3B) on the None dataset.
104
+ It achieves the following results on the evaluation set:
105
+ - Loss: 2.2525
106
 
107
  ## Model description
108
 
 
129
  - total_train_batch_size: 8
130
  - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
131
  - lr_scheduler_type: cosine
132
+ - lr_scheduler_warmup_steps: 10
133
+ - training_steps: 50
134
 
135
  ### Training results
136
 
137
  | Training Loss | Epoch | Step | Validation Loss |
138
  |:-------------:|:------:|:----:|:---------------:|
139
  | No log | 0.0001 | 1 | 2.5663 |
140
+ | 1.518 | 0.0013 | 10 | 2.4846 |
141
+ | 1.6358 | 0.0025 | 20 | 2.3336 |
142
+ | 1.8006 | 0.0038 | 30 | 2.2754 |
143
+ | 1.5463 | 0.0051 | 40 | 2.2546 |
144
+ | 1.9366 | 0.0064 | 50 | 2.2525 |
145
 
146
 
147
  ### Framework versions
adapter_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6dd48165e4026e5ad75b692bb03a9812cfd8c038ea9238c991953284a4247500
3
  size 48768810
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f055c1f133cd6879532029fc6fa9cff38faa35bf0d28ac7023dc5970a877cd8f
3
  size 48768810