Hugging Face's logo

himanshue2e
/
gemma-7b-gemma

PEFT
Safetensors
trl
sft
Generated from Trainer
Model card Files Community
Himanshu singh commited on
Commit
689a3c2
•
1 Parent(s): 0db364d

End of training

Browse files
Files changed (4) hide show
  1. README.md +9 -9
  2. adapter_config.json +4 -4
  3. adapter_model.safetensors +1 -1
  4. training_args.bin +1 -1
README.md CHANGED
@@ -20,7 +20,7 @@ should probably proofread and complete it, then remove this comment. -->
20
 
21
  This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 7.4095
24
 
25
  ## Model description
26
 
@@ -40,11 +40,11 @@ More information needed
40
 
41
  The following hyperparameters were used during training:
42
  - learning_rate: 2.5e-05
43
- - train_batch_size: 8
44
  - eval_batch_size: 8
45
  - seed: 42
46
- - gradient_accumulation_steps: 4
47
- - total_train_batch_size: 32
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
50
  - training_steps: 10
@@ -53,11 +53,11 @@ The following hyperparameters were used during training:
53
 
54
  | Training Loss | Epoch | Step | Validation Loss |
55
  |:-------------:|:-----:|:----:|:---------------:|
56
- | No log | 0.01 | 2 | 10.4741 |
57
- | No log | 0.02 | 4 | 8.8970 |
58
- | 10.571 | 0.03 | 6 | 8.0453 |
59
- | 10.571 | 0.04 | 8 | 7.5881 |
60
- | 7.8353 | 0.05 | 10 | 7.4095 |
61
 
62
 
63
  ### Framework versions
 
20
 
21
  This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 7.4237
24
 
25
  ## Model description
26
 
 
40
 
41
  The following hyperparameters were used during training:
42
  - learning_rate: 2.5e-05
43
+ - train_batch_size: 16
44
  - eval_batch_size: 8
45
  - seed: 42
46
+ - gradient_accumulation_steps: 8
47
+ - total_train_batch_size: 128
48
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
  - lr_scheduler_type: linear
50
  - training_steps: 10
 
53
 
54
  | Training Loss | Epoch | Step | Validation Loss |
55
  |:-------------:|:-----:|:----:|:---------------:|
56
+ | No log | 0.04 | 2 | 10.5195 |
57
+ | No log | 0.08 | 4 | 8.9500 |
58
+ | 10.5966 | 0.12 | 6 | 8.0804 |
59
+ | 10.5966 | 0.16 | 8 | 7.6105 |
60
+ | 7.5026 | 0.2 | 10 | 7.4237 |
61
 
62
 
63
  ### Framework versions
adapter_config.json CHANGED
@@ -20,13 +20,13 @@
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
23
- "v_proj",
24
- "up_proj",
25
  "gate_proj",
26
- "o_proj",
27
- "q_proj",
28
  "down_proj",
 
29
  "lm_head",
 
 
30
  "k_proj"
31
  ],
32
  "task_type": "CAUSAL_LM",
 
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
 
 
23
  "gate_proj",
24
+ "up_proj",
 
25
  "down_proj",
26
+ "o_proj",
27
  "lm_head",
28
+ "q_proj",
29
+ "v_proj",
30
  "k_proj"
31
  ],
32
  "task_type": "CAUSAL_LM",
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:54d5bb198983c9d907c9761315c861cd731e7b8bf80f2d256ae84ad60db2b389
3
  size 3578975160
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9916a38865a8233b9fb51a363874bc616edf51ab0d084e3e17d46df820c970e8
3
  size 3578975160
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:550f79420c52b842252bab080bc2c592a9aff576a117979c8de07ce8117bff63
3
  size 4984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d811ab69a3a0ca7aa0a1f3f4c6032757320f85af2b8954fe82fcf8448755601f
3
  size 4984