ale-bay commited on
Commit
7145fc6
1 Parent(s): c56e3a0

End of training

Browse files
README.md CHANGED
@@ -1,10 +1,15 @@
1
  ---
 
2
  tags:
 
 
 
 
3
  - trl
4
  - sft
5
  - generated_from_trainer
6
  datasets:
7
- - generator
8
  model-index:
9
  - name: zephyr-2b-gemma-dft-debug
10
  results: []
@@ -15,7 +20,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # zephyr-2b-gemma-dft-debug
17
 
18
- This model was trained from scratch on the generator dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 5.7648
21
 
 
1
  ---
2
+ base_model: data/gemma-2b
3
  tags:
4
+ - alignment-handbook
5
+ - trl
6
+ - sft
7
+ - generated_from_trainer
8
  - trl
9
  - sft
10
  - generated_from_trainer
11
  datasets:
12
+ - argilla/dpo-mix-7k
13
  model-index:
14
  - name: zephyr-2b-gemma-dft-debug
15
  results: []
 
20
 
21
  # zephyr-2b-gemma-dft-debug
22
 
23
+ This model is a fine-tuned version of [data/gemma-2b](https://huggingface.co/data/gemma-2b) on the argilla/dpo-mix-7k dataset.
24
  It achieves the following results on the evaluation set:
25
  - Loss: 5.7648
26
 
all_results.json CHANGED
@@ -1,5 +1,10 @@
1
  {
2
  "epoch": 0.9981515711645101,
 
 
 
 
 
3
  "total_flos": 5.260333472022528e+16,
4
  "train_loss": 7.369897298459653,
5
  "train_runtime": 2195.5957,
 
1
  {
2
  "epoch": 0.9981515711645101,
3
+ "eval_loss": 5.764816761016846,
4
+ "eval_runtime": 363.9913,
5
+ "eval_samples": 750,
6
+ "eval_samples_per_second": 2.668,
7
+ "eval_steps_per_second": 0.335,
8
  "total_flos": 5.260333472022528e+16,
9
  "train_loss": 7.369897298459653,
10
  "train_runtime": 2195.5957,
config.json CHANGED
@@ -24,6 +24,6 @@
24
  "rope_theta": 10000.0,
25
  "torch_dtype": "bfloat16",
26
  "transformers_version": "4.43.3",
27
- "use_cache": false,
28
  "vocab_size": 256000
29
  }
 
24
  "rope_theta": 10000.0,
25
  "torch_dtype": "bfloat16",
26
  "transformers_version": "4.43.3",
27
+ "use_cache": true,
28
  "vocab_size": 256000
29
  }
eval_results.json CHANGED
@@ -1,8 +1,8 @@
1
  {
2
  "epoch": 0.9981515711645101,
3
- "eval_loss": 5.75302791595459,
4
- "eval_runtime": 422.4972,
5
  "eval_samples": 750,
6
- "eval_samples_per_second": 2.298,
7
- "eval_steps_per_second": 0.289
8
  }
 
1
  {
2
  "epoch": 0.9981515711645101,
3
+ "eval_loss": 5.764816761016846,
4
+ "eval_runtime": 363.9913,
5
  "eval_samples": 750,
6
+ "eval_samples_per_second": 2.668,
7
+ "eval_steps_per_second": 0.335
8
  }
runs/Aug02_11-59-21_ale-distillm-8-0-0/events.out.tfevents.1722599098.ale-distillm-8-0-0.10851.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5953baedf1481273660c9a3aab7da30f5e9bd8a5467f95e180aa6745b446881
3
+ size 359