Hugging Face's logo

rbos
/
outputs

PEFT
Safetensors
trl
sft
unsloth
Generated from Trainer
Model card Files
xet
 Community
rbos commited on
Commit
02f5032
·
verified ·
1 Parent(s): b52125f

rbos/ACLED_DeepSeek_Fine-Tuned_Classifier-unsloth-DeepSeek-R1-Distill-Llama-8B-unsloth-bnb-4bit

Browse files
Files changed (4) hide show
  1. README.md +18 -18
  2. adapter_config.json +6 -6
  3. adapter_model.safetensors +1 -1
  4. training_args.bin +1 -1
README.md CHANGED
@@ -19,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
19
 
20
  This model is a fine-tuned version of [unsloth/DeepSeek-R1-Distill-Llama-8B-unsloth-bnb-4bit](https://huggingface.co/unsloth/DeepSeek-R1-Distill-Llama-8B-unsloth-bnb-4bit) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
- - Loss: 0.3521
23
 
24
  ## Model description
25
 
@@ -46,29 +46,29 @@ The following hyperparameters were used during training:
46
  - total_train_batch_size: 32
47
  - optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
48
  - lr_scheduler_type: linear
49
- - lr_scheduler_warmup_steps: 10
50
  - num_epochs: 3
51
 
52
  ### Training results
53
 
54
  | Training Loss | Epoch | Step | Validation Loss |
55
  |:-------------:|:------:|:----:|:---------------:|
56
- | 0.8251 | 0.1773 | 50 | 0.6503 |
57
- | 0.5281 | 0.3546 | 100 | 0.4812 |
58
- | 0.452 | 0.5319 | 150 | 0.4312 |
59
- | 0.4297 | 0.7092 | 200 | 0.4092 |
60
- | 0.4191 | 0.8865 | 250 | 0.3957 |
61
- | 0.3891 | 1.0638 | 300 | 0.3862 |
62
- | 0.3824 | 1.2411 | 350 | 0.3793 |
63
- | 0.3762 | 1.4184 | 400 | 0.3729 |
64
- | 0.3661 | 1.5957 | 450 | 0.3690 |
65
- | 0.3778 | 1.7730 | 500 | 0.3652 |
66
- | 0.3724 | 1.9504 | 550 | 0.3612 |
67
- | 0.351 | 2.1277 | 600 | 0.3591 |
68
- | 0.3407 | 2.3050 | 650 | 0.3581 |
69
- | 0.3485 | 2.4823 | 700 | 0.3547 |
70
- | 0.3411 | 2.6596 | 750 | 0.3532 |
71
- | 0.343 | 2.8369 | 800 | 0.3521 |
72
 
73
 
74
  ### Framework versions
 
19
 
20
  This model is a fine-tuned version of [unsloth/DeepSeek-R1-Distill-Llama-8B-unsloth-bnb-4bit](https://huggingface.co/unsloth/DeepSeek-R1-Distill-Llama-8B-unsloth-bnb-4bit) on the None dataset.
21
  It achieves the following results on the evaluation set:
22
+ - Loss: 0.3463
23
 
24
  ## Model description
25
 
 
46
  - total_train_batch_size: 32
47
  - optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
48
  - lr_scheduler_type: linear
49
+ - lr_scheduler_warmup_steps: 20
50
  - num_epochs: 3
51
 
52
  ### Training results
53
 
54
  | Training Loss | Epoch | Step | Validation Loss |
55
  |:-------------:|:------:|:----:|:---------------:|
56
+ | 0.9219 | 0.1773 | 50 | 0.6719 |
57
+ | 0.529 | 0.3546 | 100 | 0.4809 |
58
+ | 0.448 | 0.5319 | 150 | 0.4260 |
59
+ | 0.4242 | 0.7092 | 200 | 0.4043 |
60
+ | 0.4133 | 0.8865 | 250 | 0.3898 |
61
+ | 0.3839 | 1.0638 | 300 | 0.3803 |
62
+ | 0.3767 | 1.2411 | 350 | 0.3735 |
63
+ | 0.3708 | 1.4184 | 400 | 0.3673 |
64
+ | 0.3605 | 1.5957 | 450 | 0.3632 |
65
+ | 0.3721 | 1.7730 | 500 | 0.3593 |
66
+ | 0.3668 | 1.9504 | 550 | 0.3554 |
67
+ | 0.3457 | 2.1277 | 600 | 0.3536 |
68
+ | 0.3352 | 2.3050 | 650 | 0.3518 |
69
+ | 0.3426 | 2.4823 | 700 | 0.3490 |
70
+ | 0.3356 | 2.6596 | 750 | 0.3478 |
71
+ | 0.3376 | 2.8369 | 800 | 0.3463 |
72
 
73
 
74
  ### Framework versions
adapter_config.json CHANGED
@@ -23,13 +23,13 @@
23
  "rank_pattern": {},
24
  "revision": null,
25
  "target_modules": [
26
- "v_proj",
27
- "q_proj",
28
- "k_proj",
29
- "gate_proj",
30
- "o_proj",
31
  "up_proj",
32
- "down_proj"
 
 
 
 
 
33
  ],
34
  "task_type": "CAUSAL_LM",
35
  "use_dora": false,
 
23
  "rank_pattern": {},
24
  "revision": null,
25
  "target_modules": [
 
 
 
 
 
26
  "up_proj",
27
+ "o_proj",
28
+ "gate_proj",
29
+ "k_proj",
30
+ "q_proj",
31
+ "down_proj",
32
+ "v_proj"
33
  ],
34
  "task_type": "CAUSAL_LM",
35
  "use_dora": false,
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2ac29f9c910ce443e22b95c5d22065a4494d01b206e3b95a640628efdfd4b3a3
3
  size 83945296
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c8e43546f153b860ce564218512d310c5e1ade0db8a908613bf5f14aa7bb8068
3
  size 83945296
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:33d19787cd888e8059b7579e941ea460799765e929a9ff1b09b8607fcb61f2f4
3
  size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b0a9ceb383c83e1a491d1dd6926482ad0e135551942eba2577aae492dbcdac4
3
  size 5304