Hugging Face's logo

Casper0508
/
MSc_llama3_finetuned_model_secondData

PEFT
TensorBoard
Safetensors
Generated from Trainer
Model card Files Metrics Community
Casper0508 commited on
Commit
f0ae726
•
1 Parent(s): ea8328e

End of training

Browse files
Files changed (5) hide show
  1. README.md +29 -29
  2. adapter_model.safetensors +1 -1
  3. emissions.csv +1 -1
  4. runs/Jul17_19-43-13_msc-modeltrain-pod/events.out.tfevents.1721245397.msc-modeltrain-pod.2588.0 +3 -0
  5. training_args.bin +1 -1
README.md CHANGED
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 1.5993
20
 
21
  ## Model description
22
 
@@ -49,11 +49,11 @@ The following `bitsandbytes` quantization config was used during training:
49
  ### Training hyperparameters
50
 
51
  The following hyperparameters were used during training:
52
- - learning_rate: 0.0002
53
- - train_batch_size: 16
54
  - eval_batch_size: 8
55
  - seed: 42
56
- - gradient_accumulation_steps: 4
57
  - total_train_batch_size: 64
58
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
59
  - lr_scheduler_type: cosine
@@ -64,31 +64,31 @@ The following hyperparameters were used during training:
64
 
65
  | Training Loss | Epoch | Step | Validation Loss |
66
  |:-------------:|:-----:|:----:|:---------------:|
67
- | 3.329 | 1.33 | 10 | 1.8003 |
68
- | 1.296 | 2.67 | 20 | 1.0774 |
69
- | 0.9489 | 4.0 | 30 | 0.9022 |
70
- | 0.7167 | 5.33 | 40 | 0.7270 |
71
- | 0.552 | 6.67 | 50 | 0.7372 |
72
- | 0.4766 | 8.0 | 60 | 0.7281 |
73
- | 0.4153 | 9.33 | 70 | 0.7673 |
74
- | 0.3614 | 10.67 | 80 | 0.8597 |
75
- | 0.3238 | 12.0 | 90 | 0.8915 |
76
- | 0.2923 | 13.33 | 100 | 0.9281 |
77
- | 0.2648 | 14.67 | 110 | 1.0239 |
78
- | 0.2483 | 16.0 | 120 | 1.0198 |
79
- | 0.2311 | 17.33 | 130 | 1.1314 |
80
- | 0.2196 | 18.67 | 140 | 1.2578 |
81
- | 0.2109 | 20.0 | 150 | 1.3155 |
82
- | 0.1997 | 21.33 | 160 | 1.2602 |
83
- | 0.1927 | 22.67 | 170 | 1.4758 |
84
- | 0.191 | 24.0 | 180 | 1.4080 |
85
- | 0.1834 | 25.33 | 190 | 1.4783 |
86
- | 0.1799 | 26.67 | 200 | 1.5217 |
87
- | 0.1796 | 28.0 | 210 | 1.5525 |
88
- | 0.1738 | 29.33 | 220 | 1.5714 |
89
- | 0.1725 | 30.67 | 230 | 1.5953 |
90
- | 0.1727 | 32.0 | 240 | 1.5980 |
91
- | 0.172 | 33.33 | 250 | 1.5993 |
92
 
93
 
94
  ### Framework versions
 
16
 
17
  This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.7622
20
 
21
  ## Model description
22
 
 
49
  ### Training hyperparameters
50
 
51
  The following hyperparameters were used during training:
52
+ - learning_rate: 5e-05
53
+ - train_batch_size: 8
54
  - eval_batch_size: 8
55
  - seed: 42
56
+ - gradient_accumulation_steps: 8
57
  - total_train_batch_size: 64
58
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
59
  - lr_scheduler_type: cosine
 
64
 
65
  | Training Loss | Epoch | Step | Validation Loss |
66
  |:-------------:|:-----:|:----:|:---------------:|
67
+ | 3.7647 | 1.36 | 10 | 3.3602 |
68
+ | 2.8 | 2.71 | 20 | 2.0480 |
69
+ | 1.5819 | 4.07 | 30 | 1.2852 |
70
+ | 1.1832 | 5.42 | 40 | 1.1025 |
71
+ | 1.0318 | 6.78 | 50 | 1.0150 |
72
+ | 0.9674 | 8.14 | 60 | 0.9718 |
73
+ | 0.8975 | 9.49 | 70 | 0.9348 |
74
+ | 0.8375 | 10.85 | 80 | 0.8912 |
75
+ | 0.7851 | 12.2 | 90 | 0.8685 |
76
+ | 0.728 | 13.56 | 100 | 0.8443 |
77
+ | 0.6804 | 14.92 | 110 | 0.8038 |
78
+ | 0.6123 | 16.27 | 120 | 0.7684 |
79
+ | 0.5536 | 17.63 | 130 | 0.7314 |
80
+ | 0.4922 | 18.98 | 140 | 0.6943 |
81
+ | 0.4738 | 20.34 | 150 | 0.7095 |
82
+ | 0.4467 | 21.69 | 160 | 0.7344 |
83
+ | 0.4452 | 23.05 | 170 | 0.7397 |
84
+ | 0.4258 | 24.41 | 180 | 0.7332 |
85
+ | 0.4179 | 25.76 | 190 | 0.7436 |
86
+ | 0.4105 | 27.12 | 200 | 0.7373 |
87
+ | 0.4081 | 28.47 | 210 | 0.7596 |
88
+ | 0.4005 | 29.83 | 220 | 0.7552 |
89
+ | 0.4001 | 31.19 | 230 | 0.7652 |
90
+ | 0.393 | 32.54 | 240 | 0.7612 |
91
+ | 0.4016 | 33.9 | 250 | 0.7622 |
92
 
93
 
94
  ### Framework versions
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cdd34584c6acc5f0d9ac55c6e1190f8f78557a7c4ce7d5d4bea77352db827001
3
  size 75523312
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff4a36e1a6dd121f9e1a7d350268e382f1c8d02a207ccedd1f6d86a1205890f1
3
  size 75523312
emissions.csv CHANGED
@@ -1,2 +1,2 @@
1
  timestamp,experiment_id,project_name,duration,emissions,energy_consumed,country_name,country_iso_code,region,on_cloud,cloud_provider,cloud_region
2
- 2024-07-17T17:04:02,acf495a5-fe7d-4741-820f-f1df91f69def,codecarbon,714.2170021533966,0.0483531364122497,0.07194239682851639,United Kingdom,GBR,scotland,N,,
 
1
  timestamp,experiment_id,project_name,duration,emissions,energy_consumed,country_name,country_iso_code,region,on_cloud,cloud_provider,cloud_region
2
+ 2024-07-17T20:05:05,a2d5e5ed-35d4-4380-9b5c-d7337e3b0d8b,codecarbon,1308.196786403656,0.08001020888288951,0.11904349179564445,United Kingdom,GBR,scotland,N,,
runs/Jul17_19-43-13_msc-modeltrain-pod/events.out.tfevents.1721245397.msc-modeltrain-pod.2588.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45c6f2c37d9b504e1fd9bb669a6cc59b1b1987e5ec96612b182f1fef8356687d
3
+ size 17467
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ef8e5dc5533306720bffeb0cf13b6ace0ad8000b5f6b5240b681a005a13a301e
3
  size 4984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c02584efc3e8ec3d6a7e29381306b66072c8cda4d47e71431afa8b88df054da1
3
  size 4984