ale-bay commited on
Commit
fdadbc0
1 Parent(s): d304fa9

End of training

Browse files
README.md CHANGED
@@ -2,10 +2,16 @@
2
  license: apache-2.0
3
  library_name: peft
4
  tags:
 
 
 
 
5
  - trl
6
  - dpo
7
  - generated_from_trainer
8
  base_model: mistralai/Mistral-7B-v0.1
 
 
9
  model-index:
10
  - name: zephyr-7b-dpo-qlora
11
  results: []
@@ -16,17 +22,17 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  # zephyr-7b-dpo-qlora
18
 
19
- This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.4921
22
  - Rewards/chosen: -2.5098
23
- - Rewards/rejected: -3.5900
24
- - Rewards/accuracies: 0.7550
25
- - Rewards/margins: 1.0803
26
- - Logps/rejected: -600.2638
27
- - Logps/chosen: -516.2831
28
  - Logits/rejected: 2.5098
29
- - Logits/chosen: 2.2971
30
 
31
  ## Model description
32
 
 
2
  license: apache-2.0
3
  library_name: peft
4
  tags:
5
+ - alignment-handbook
6
+ - trl
7
+ - dpo
8
+ - generated_from_trainer
9
  - trl
10
  - dpo
11
  - generated_from_trainer
12
  base_model: mistralai/Mistral-7B-v0.1
13
+ datasets:
14
+ - HuggingFaceH4/ultrafeedback_binarized
15
  model-index:
16
  - name: zephyr-7b-dpo-qlora
17
  results: []
 
22
 
23
  # zephyr-7b-dpo-qlora
24
 
25
+ This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-qlora](https://huggingface.co/alignment-handbook/zephyr-7b-sft-qlora) on the HuggingFaceH4/ultrafeedback_binarized dataset.
26
  It achieves the following results on the evaluation set:
27
+ - Loss: 0.4920
28
  - Rewards/chosen: -2.5098
29
+ - Rewards/rejected: -3.5905
30
+ - Rewards/accuracies: 0.7560
31
+ - Rewards/margins: 1.0807
32
+ - Logps/rejected: -600.3103
33
+ - Logps/chosen: -516.2818
34
  - Logits/rejected: 2.5098
35
+ - Logits/chosen: 2.2972
36
 
37
  ## Model description
38
 
all_results.json CHANGED
@@ -1,5 +1,18 @@
1
  {
2
  "epoch": 1.0,
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  "train_loss": 0.528951391499704,
4
  "train_runtime": 37787.7404,
5
  "train_samples": 61135,
 
1
  {
2
  "epoch": 1.0,
3
+ "eval_logits/chosen": 2.2971765995025635,
4
+ "eval_logits/rejected": 2.509765625,
5
+ "eval_logps/chosen": -516.2817993164062,
6
+ "eval_logps/rejected": -600.310302734375,
7
+ "eval_loss": 0.4919675290584564,
8
+ "eval_rewards/accuracies": 0.7559999823570251,
9
+ "eval_rewards/chosen": -2.5097556114196777,
10
+ "eval_rewards/margins": 1.0807288885116577,
11
+ "eval_rewards/rejected": -3.590484380722046,
12
+ "eval_runtime": 449.3054,
13
+ "eval_samples": 2000,
14
+ "eval_samples_per_second": 4.451,
15
+ "eval_steps_per_second": 0.278,
16
  "train_loss": 0.528951391499704,
17
  "train_runtime": 37787.7404,
18
  "train_samples": 61135,
config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "mistralai/Mistral-7B-v0.1",
3
+ "architectures": [
4
+ "MistralForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 1,
8
+ "eos_token_id": 2,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 4096,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 14336,
13
+ "max_position_embeddings": 32768,
14
+ "model_type": "mistral",
15
+ "num_attention_heads": 32,
16
+ "num_hidden_layers": 32,
17
+ "num_key_value_heads": 8,
18
+ "quantization_config": {
19
+ "_load_in_4bit": true,
20
+ "_load_in_8bit": false,
21
+ "bnb_4bit_compute_dtype": "bfloat16",
22
+ "bnb_4bit_quant_type": "nf4",
23
+ "bnb_4bit_use_double_quant": false,
24
+ "llm_int8_enable_fp32_cpu_offload": false,
25
+ "llm_int8_has_fp16_weight": false,
26
+ "llm_int8_skip_modules": null,
27
+ "llm_int8_threshold": 6.0,
28
+ "load_in_4bit": true,
29
+ "load_in_8bit": false,
30
+ "quant_method": "bitsandbytes"
31
+ },
32
+ "rms_norm_eps": 1e-05,
33
+ "rope_theta": 10000.0,
34
+ "sliding_window": 4096,
35
+ "tie_word_embeddings": false,
36
+ "torch_dtype": "bfloat16",
37
+ "transformers_version": "4.38.2",
38
+ "use_cache": true,
39
+ "vocab_size": 32000
40
+ }
eval_results.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "eval_logits/chosen": 2.2971765995025635,
4
+ "eval_logits/rejected": 2.509765625,
5
+ "eval_logps/chosen": -516.2817993164062,
6
+ "eval_logps/rejected": -600.310302734375,
7
+ "eval_loss": 0.4919675290584564,
8
+ "eval_rewards/accuracies": 0.7559999823570251,
9
+ "eval_rewards/chosen": -2.5097556114196777,
10
+ "eval_rewards/margins": 1.0807288885116577,
11
+ "eval_rewards/rejected": -3.590484380722046,
12
+ "eval_runtime": 449.3054,
13
+ "eval_samples": 2000,
14
+ "eval_samples_per_second": 4.451,
15
+ "eval_steps_per_second": 0.278
16
+ }
runs/Mar07_10-37-47_ale-dpo-1-0-0/events.out.tfevents.1709846645.ale-dpo-1-0-0.1134.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b1c391730d9bd5ec37b05c71a93210760587fdd6ca385a28eef9ba593ec7bbc7
3
+ size 828