Commit
•
060ec4c
1
Parent(s):
d8c4cf5
Upload folder using huggingface_hub
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- checkpoint-1000/README.md +32 -0
- checkpoint-1000/adapter_config.json +22 -0
- checkpoint-1000/adapter_model.bin +3 -0
- checkpoint-1000/merges.txt +0 -0
- checkpoint-1000/optimizer.pt +3 -0
- checkpoint-1000/rng_state.pth +3 -0
- checkpoint-1000/scheduler.pt +3 -0
- checkpoint-1000/special_tokens_map.json +6 -0
- checkpoint-1000/tokenizer.json +0 -0
- checkpoint-1000/tokenizer_config.json +9 -0
- checkpoint-1000/trainer_state.json +22 -0
- checkpoint-1000/training_args.bin +3 -0
- checkpoint-1000/vocab.json +0 -0
- checkpoint-10000/README.md +32 -0
- checkpoint-10000/adapter_config.json +22 -0
- checkpoint-10000/adapter_model.bin +3 -0
- checkpoint-10000/merges.txt +0 -0
- checkpoint-10000/optimizer.pt +3 -0
- checkpoint-10000/rng_state.pth +3 -0
- checkpoint-10000/scheduler.pt +3 -0
- checkpoint-10000/special_tokens_map.json +6 -0
- checkpoint-10000/tokenizer.json +0 -0
- checkpoint-10000/tokenizer_config.json +9 -0
- checkpoint-10000/trainer_state.json +76 -0
- checkpoint-10000/training_args.bin +3 -0
- checkpoint-10000/vocab.json +0 -0
- checkpoint-2000/README.md +32 -0
- checkpoint-2000/adapter_config.json +22 -0
- checkpoint-2000/adapter_model.bin +3 -0
- checkpoint-2000/merges.txt +0 -0
- checkpoint-2000/optimizer.pt +3 -0
- checkpoint-2000/rng_state.pth +3 -0
- checkpoint-2000/scheduler.pt +3 -0
- checkpoint-2000/special_tokens_map.json +6 -0
- checkpoint-2000/tokenizer.json +0 -0
- checkpoint-2000/tokenizer_config.json +9 -0
- checkpoint-2000/trainer_state.json +28 -0
- checkpoint-2000/training_args.bin +3 -0
- checkpoint-2000/vocab.json +0 -0
- checkpoint-3000/README.md +32 -0
- checkpoint-3000/adapter_config.json +22 -0
- checkpoint-3000/adapter_model.bin +3 -0
- checkpoint-3000/merges.txt +0 -0
- checkpoint-3000/optimizer.pt +3 -0
- checkpoint-3000/rng_state.pth +3 -0
- checkpoint-3000/scheduler.pt +3 -0
- checkpoint-3000/special_tokens_map.json +6 -0
- checkpoint-3000/tokenizer.json +0 -0
- checkpoint-3000/tokenizer_config.json +9 -0
- checkpoint-3000/trainer_state.json +34 -0
checkpoint-1000/README.md
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
---
|
4 |
+
## Training procedure
|
5 |
+
|
6 |
+
|
7 |
+
The following `bitsandbytes` quantization config was used during training:
|
8 |
+
- load_in_8bit: False
|
9 |
+
- load_in_4bit: True
|
10 |
+
- llm_int8_threshold: 6.0
|
11 |
+
- llm_int8_skip_modules: None
|
12 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
13 |
+
- llm_int8_has_fp16_weight: False
|
14 |
+
- bnb_4bit_quant_type: fp4
|
15 |
+
- bnb_4bit_use_double_quant: False
|
16 |
+
- bnb_4bit_compute_dtype: float32
|
17 |
+
|
18 |
+
The following `bitsandbytes` quantization config was used during training:
|
19 |
+
- load_in_8bit: False
|
20 |
+
- load_in_4bit: True
|
21 |
+
- llm_int8_threshold: 6.0
|
22 |
+
- llm_int8_skip_modules: None
|
23 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
24 |
+
- llm_int8_has_fp16_weight: False
|
25 |
+
- bnb_4bit_quant_type: fp4
|
26 |
+
- bnb_4bit_use_double_quant: False
|
27 |
+
- bnb_4bit_compute_dtype: float32
|
28 |
+
### Framework versions
|
29 |
+
|
30 |
+
- PEFT 0.4.0
|
31 |
+
|
32 |
+
- PEFT 0.4.0
|
checkpoint-1000/adapter_config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"auto_mapping": null,
|
3 |
+
"base_model_name_or_path": "gpt2",
|
4 |
+
"bias": "none",
|
5 |
+
"fan_in_fan_out": false,
|
6 |
+
"inference_mode": true,
|
7 |
+
"init_lora_weights": true,
|
8 |
+
"layers_pattern": null,
|
9 |
+
"layers_to_transform": null,
|
10 |
+
"lora_alpha": 16,
|
11 |
+
"lora_dropout": 0.0,
|
12 |
+
"modules_to_save": [
|
13 |
+
"scores"
|
14 |
+
],
|
15 |
+
"peft_type": "LORA",
|
16 |
+
"r": 16,
|
17 |
+
"revision": null,
|
18 |
+
"target_modules": [
|
19 |
+
"c_attn"
|
20 |
+
],
|
21 |
+
"task_type": "SEQ_CLS"
|
22 |
+
}
|
checkpoint-1000/adapter_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:41cba345a800a5eef65704aafbd271f649a7a8a5a01ba08d8a431d2920d0ff8f
|
3 |
+
size 2371004
|
checkpoint-1000/merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-1000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:765e08724d144b8d1269d1eaad6e174d8289b3d5cd746db7f2299c7b2583cb6e
|
3 |
+
size 4745758
|
checkpoint-1000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f9313f13e00b9a5a441c949a1a7a5c373d4655ac4798ddc195c721e8ee18fda0
|
3 |
+
size 14575
|
checkpoint-1000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:13a54a9f9ff528ac19a8950dbdf8c0bab16f017a057f5dc6729e41729601f5a7
|
3 |
+
size 627
|
checkpoint-1000/special_tokens_map.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<|endoftext|>",
|
3 |
+
"eos_token": "<|endoftext|>",
|
4 |
+
"pad_token": "<|endoftext|>",
|
5 |
+
"unk_token": "<|endoftext|>"
|
6 |
+
}
|
checkpoint-1000/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-1000/tokenizer_config.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"bos_token": "<|endoftext|>",
|
4 |
+
"clean_up_tokenization_spaces": true,
|
5 |
+
"eos_token": "<|endoftext|>",
|
6 |
+
"model_max_length": 1024,
|
7 |
+
"tokenizer_class": "GPT2Tokenizer",
|
8 |
+
"unk_token": "<|endoftext|>"
|
9 |
+
}
|
checkpoint-1000/trainer_state.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.02716579283366385,
|
5 |
+
"global_step": 1000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.03,
|
12 |
+
"learning_rate": 1.269e-05,
|
13 |
+
"loss": 0.9046,
|
14 |
+
"step": 1000
|
15 |
+
}
|
16 |
+
],
|
17 |
+
"max_steps": 10000,
|
18 |
+
"num_train_epochs": 1,
|
19 |
+
"total_flos": 0.0,
|
20 |
+
"trial_name": null,
|
21 |
+
"trial_params": null
|
22 |
+
}
|
checkpoint-1000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1abb0c3f9d62f18134f4bc8afde3857056da9b77545dd545c1c979170edc1b94
|
3 |
+
size 3963
|
checkpoint-1000/vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-10000/README.md
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
---
|
4 |
+
## Training procedure
|
5 |
+
|
6 |
+
|
7 |
+
The following `bitsandbytes` quantization config was used during training:
|
8 |
+
- load_in_8bit: False
|
9 |
+
- load_in_4bit: True
|
10 |
+
- llm_int8_threshold: 6.0
|
11 |
+
- llm_int8_skip_modules: None
|
12 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
13 |
+
- llm_int8_has_fp16_weight: False
|
14 |
+
- bnb_4bit_quant_type: fp4
|
15 |
+
- bnb_4bit_use_double_quant: False
|
16 |
+
- bnb_4bit_compute_dtype: float32
|
17 |
+
|
18 |
+
The following `bitsandbytes` quantization config was used during training:
|
19 |
+
- load_in_8bit: False
|
20 |
+
- load_in_4bit: True
|
21 |
+
- llm_int8_threshold: 6.0
|
22 |
+
- llm_int8_skip_modules: None
|
23 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
24 |
+
- llm_int8_has_fp16_weight: False
|
25 |
+
- bnb_4bit_quant_type: fp4
|
26 |
+
- bnb_4bit_use_double_quant: False
|
27 |
+
- bnb_4bit_compute_dtype: float32
|
28 |
+
### Framework versions
|
29 |
+
|
30 |
+
- PEFT 0.4.0
|
31 |
+
|
32 |
+
- PEFT 0.4.0
|
checkpoint-10000/adapter_config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"auto_mapping": null,
|
3 |
+
"base_model_name_or_path": "gpt2",
|
4 |
+
"bias": "none",
|
5 |
+
"fan_in_fan_out": false,
|
6 |
+
"inference_mode": true,
|
7 |
+
"init_lora_weights": true,
|
8 |
+
"layers_pattern": null,
|
9 |
+
"layers_to_transform": null,
|
10 |
+
"lora_alpha": 16,
|
11 |
+
"lora_dropout": 0.0,
|
12 |
+
"modules_to_save": [
|
13 |
+
"scores"
|
14 |
+
],
|
15 |
+
"peft_type": "LORA",
|
16 |
+
"r": 16,
|
17 |
+
"revision": null,
|
18 |
+
"target_modules": [
|
19 |
+
"c_attn"
|
20 |
+
],
|
21 |
+
"task_type": "SEQ_CLS"
|
22 |
+
}
|
checkpoint-10000/adapter_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c5e9821a962b6ee6a3e121817d225fb89109633a6bb7612635171e799443d34c
|
3 |
+
size 2371004
|
checkpoint-10000/merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-10000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fca728295163382ca2d2b6b7c37f946c2f834e2638f1223cdf8b1f5a9e9b3121
|
3 |
+
size 4745758
|
checkpoint-10000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7276943e521b5115791ab7fc8906ecb17d0589bad4fdbc78325079b27386e86b
|
3 |
+
size 14575
|
checkpoint-10000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e8e759e3641c39b005eb96acf7275213881db765bee6c1376f389c9fec7b8dac
|
3 |
+
size 627
|
checkpoint-10000/special_tokens_map.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<|endoftext|>",
|
3 |
+
"eos_token": "<|endoftext|>",
|
4 |
+
"pad_token": "<|endoftext|>",
|
5 |
+
"unk_token": "<|endoftext|>"
|
6 |
+
}
|
checkpoint-10000/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-10000/tokenizer_config.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"bos_token": "<|endoftext|>",
|
4 |
+
"clean_up_tokenization_spaces": true,
|
5 |
+
"eos_token": "<|endoftext|>",
|
6 |
+
"model_max_length": 1024,
|
7 |
+
"tokenizer_class": "GPT2Tokenizer",
|
8 |
+
"unk_token": "<|endoftext|>"
|
9 |
+
}
|
checkpoint-10000/trainer_state.json
ADDED
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.2716579283366385,
|
5 |
+
"global_step": 10000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.03,
|
12 |
+
"learning_rate": 1.269e-05,
|
13 |
+
"loss": 0.9046,
|
14 |
+
"step": 1000
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.05,
|
18 |
+
"learning_rate": 1.128e-05,
|
19 |
+
"loss": 0.783,
|
20 |
+
"step": 2000
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.08,
|
24 |
+
"learning_rate": 9.87e-06,
|
25 |
+
"loss": 0.7261,
|
26 |
+
"step": 3000
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.11,
|
30 |
+
"learning_rate": 8.46e-06,
|
31 |
+
"loss": 0.7128,
|
32 |
+
"step": 4000
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.14,
|
36 |
+
"learning_rate": 7.05e-06,
|
37 |
+
"loss": 0.7113,
|
38 |
+
"step": 5000
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.16,
|
42 |
+
"learning_rate": 5.64e-06,
|
43 |
+
"loss": 0.7047,
|
44 |
+
"step": 6000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.19,
|
48 |
+
"learning_rate": 4.23e-06,
|
49 |
+
"loss": 0.7025,
|
50 |
+
"step": 7000
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.22,
|
54 |
+
"learning_rate": 2.82e-06,
|
55 |
+
"loss": 0.7035,
|
56 |
+
"step": 8000
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.24,
|
60 |
+
"learning_rate": 1.41e-06,
|
61 |
+
"loss": 0.6979,
|
62 |
+
"step": 9000
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.27,
|
66 |
+
"learning_rate": 0.0,
|
67 |
+
"loss": 0.701,
|
68 |
+
"step": 10000
|
69 |
+
}
|
70 |
+
],
|
71 |
+
"max_steps": 10000,
|
72 |
+
"num_train_epochs": 1,
|
73 |
+
"total_flos": 0.0,
|
74 |
+
"trial_name": null,
|
75 |
+
"trial_params": null
|
76 |
+
}
|
checkpoint-10000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1abb0c3f9d62f18134f4bc8afde3857056da9b77545dd545c1c979170edc1b94
|
3 |
+
size 3963
|
checkpoint-10000/vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-2000/README.md
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
---
|
4 |
+
## Training procedure
|
5 |
+
|
6 |
+
|
7 |
+
The following `bitsandbytes` quantization config was used during training:
|
8 |
+
- load_in_8bit: False
|
9 |
+
- load_in_4bit: True
|
10 |
+
- llm_int8_threshold: 6.0
|
11 |
+
- llm_int8_skip_modules: None
|
12 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
13 |
+
- llm_int8_has_fp16_weight: False
|
14 |
+
- bnb_4bit_quant_type: fp4
|
15 |
+
- bnb_4bit_use_double_quant: False
|
16 |
+
- bnb_4bit_compute_dtype: float32
|
17 |
+
|
18 |
+
The following `bitsandbytes` quantization config was used during training:
|
19 |
+
- load_in_8bit: False
|
20 |
+
- load_in_4bit: True
|
21 |
+
- llm_int8_threshold: 6.0
|
22 |
+
- llm_int8_skip_modules: None
|
23 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
24 |
+
- llm_int8_has_fp16_weight: False
|
25 |
+
- bnb_4bit_quant_type: fp4
|
26 |
+
- bnb_4bit_use_double_quant: False
|
27 |
+
- bnb_4bit_compute_dtype: float32
|
28 |
+
### Framework versions
|
29 |
+
|
30 |
+
- PEFT 0.4.0
|
31 |
+
|
32 |
+
- PEFT 0.4.0
|
checkpoint-2000/adapter_config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"auto_mapping": null,
|
3 |
+
"base_model_name_or_path": "gpt2",
|
4 |
+
"bias": "none",
|
5 |
+
"fan_in_fan_out": false,
|
6 |
+
"inference_mode": true,
|
7 |
+
"init_lora_weights": true,
|
8 |
+
"layers_pattern": null,
|
9 |
+
"layers_to_transform": null,
|
10 |
+
"lora_alpha": 16,
|
11 |
+
"lora_dropout": 0.0,
|
12 |
+
"modules_to_save": [
|
13 |
+
"scores"
|
14 |
+
],
|
15 |
+
"peft_type": "LORA",
|
16 |
+
"r": 16,
|
17 |
+
"revision": null,
|
18 |
+
"target_modules": [
|
19 |
+
"c_attn"
|
20 |
+
],
|
21 |
+
"task_type": "SEQ_CLS"
|
22 |
+
}
|
checkpoint-2000/adapter_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ed3dfefb909c7aead095a39638a31bb66e110e53434a76ce1dd5aff3a8ee644c
|
3 |
+
size 2371004
|
checkpoint-2000/merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-2000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5e825a170cd93f04317cb9b3d8d4846ee3ea47d80cbbe6996a83ce164553ed3f
|
3 |
+
size 4745758
|
checkpoint-2000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1edc8c2a3db73059d001418cd6cd010891b7615bebb41ae75ac76aa7bb061eb1
|
3 |
+
size 14575
|
checkpoint-2000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cb93e32ef77875646479aaa56fb804f27a460c2c6ea588330341e972e405a5f8
|
3 |
+
size 627
|
checkpoint-2000/special_tokens_map.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<|endoftext|>",
|
3 |
+
"eos_token": "<|endoftext|>",
|
4 |
+
"pad_token": "<|endoftext|>",
|
5 |
+
"unk_token": "<|endoftext|>"
|
6 |
+
}
|
checkpoint-2000/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-2000/tokenizer_config.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"bos_token": "<|endoftext|>",
|
4 |
+
"clean_up_tokenization_spaces": true,
|
5 |
+
"eos_token": "<|endoftext|>",
|
6 |
+
"model_max_length": 1024,
|
7 |
+
"tokenizer_class": "GPT2Tokenizer",
|
8 |
+
"unk_token": "<|endoftext|>"
|
9 |
+
}
|
checkpoint-2000/trainer_state.json
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.0543315856673277,
|
5 |
+
"global_step": 2000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.03,
|
12 |
+
"learning_rate": 1.269e-05,
|
13 |
+
"loss": 0.9046,
|
14 |
+
"step": 1000
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.05,
|
18 |
+
"learning_rate": 1.128e-05,
|
19 |
+
"loss": 0.783,
|
20 |
+
"step": 2000
|
21 |
+
}
|
22 |
+
],
|
23 |
+
"max_steps": 10000,
|
24 |
+
"num_train_epochs": 1,
|
25 |
+
"total_flos": 0.0,
|
26 |
+
"trial_name": null,
|
27 |
+
"trial_params": null
|
28 |
+
}
|
checkpoint-2000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1abb0c3f9d62f18134f4bc8afde3857056da9b77545dd545c1c979170edc1b94
|
3 |
+
size 3963
|
checkpoint-2000/vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-3000/README.md
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
---
|
4 |
+
## Training procedure
|
5 |
+
|
6 |
+
|
7 |
+
The following `bitsandbytes` quantization config was used during training:
|
8 |
+
- load_in_8bit: False
|
9 |
+
- load_in_4bit: True
|
10 |
+
- llm_int8_threshold: 6.0
|
11 |
+
- llm_int8_skip_modules: None
|
12 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
13 |
+
- llm_int8_has_fp16_weight: False
|
14 |
+
- bnb_4bit_quant_type: fp4
|
15 |
+
- bnb_4bit_use_double_quant: False
|
16 |
+
- bnb_4bit_compute_dtype: float32
|
17 |
+
|
18 |
+
The following `bitsandbytes` quantization config was used during training:
|
19 |
+
- load_in_8bit: False
|
20 |
+
- load_in_4bit: True
|
21 |
+
- llm_int8_threshold: 6.0
|
22 |
+
- llm_int8_skip_modules: None
|
23 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
24 |
+
- llm_int8_has_fp16_weight: False
|
25 |
+
- bnb_4bit_quant_type: fp4
|
26 |
+
- bnb_4bit_use_double_quant: False
|
27 |
+
- bnb_4bit_compute_dtype: float32
|
28 |
+
### Framework versions
|
29 |
+
|
30 |
+
- PEFT 0.4.0
|
31 |
+
|
32 |
+
- PEFT 0.4.0
|
checkpoint-3000/adapter_config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"auto_mapping": null,
|
3 |
+
"base_model_name_or_path": "gpt2",
|
4 |
+
"bias": "none",
|
5 |
+
"fan_in_fan_out": false,
|
6 |
+
"inference_mode": true,
|
7 |
+
"init_lora_weights": true,
|
8 |
+
"layers_pattern": null,
|
9 |
+
"layers_to_transform": null,
|
10 |
+
"lora_alpha": 16,
|
11 |
+
"lora_dropout": 0.0,
|
12 |
+
"modules_to_save": [
|
13 |
+
"scores"
|
14 |
+
],
|
15 |
+
"peft_type": "LORA",
|
16 |
+
"r": 16,
|
17 |
+
"revision": null,
|
18 |
+
"target_modules": [
|
19 |
+
"c_attn"
|
20 |
+
],
|
21 |
+
"task_type": "SEQ_CLS"
|
22 |
+
}
|
checkpoint-3000/adapter_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2cfb125968f4f7cccbe5f0f8d6d109944d8a6fb46a33a136af2b4d0780a09ca3
|
3 |
+
size 2371004
|
checkpoint-3000/merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-3000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3d82982046678f4cef59699bf771f9eddbdbe1e8b02c4acf812af0f3c419be50
|
3 |
+
size 4745758
|
checkpoint-3000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9ef179ba8c301210c5bb50d2c5392d4aa75ae1ee71de822cc5409b78fe738d27
|
3 |
+
size 14575
|
checkpoint-3000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bfb1169a217170ed5e0b1d712ed8c02b95bdb6ebc251b96933db8b149cef96f3
|
3 |
+
size 627
|
checkpoint-3000/special_tokens_map.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<|endoftext|>",
|
3 |
+
"eos_token": "<|endoftext|>",
|
4 |
+
"pad_token": "<|endoftext|>",
|
5 |
+
"unk_token": "<|endoftext|>"
|
6 |
+
}
|
checkpoint-3000/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-3000/tokenizer_config.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"bos_token": "<|endoftext|>",
|
4 |
+
"clean_up_tokenization_spaces": true,
|
5 |
+
"eos_token": "<|endoftext|>",
|
6 |
+
"model_max_length": 1024,
|
7 |
+
"tokenizer_class": "GPT2Tokenizer",
|
8 |
+
"unk_token": "<|endoftext|>"
|
9 |
+
}
|
checkpoint-3000/trainer_state.json
ADDED
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.08149737850099155,
|
5 |
+
"global_step": 3000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.03,
|
12 |
+
"learning_rate": 1.269e-05,
|
13 |
+
"loss": 0.9046,
|
14 |
+
"step": 1000
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.05,
|
18 |
+
"learning_rate": 1.128e-05,
|
19 |
+
"loss": 0.783,
|
20 |
+
"step": 2000
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.08,
|
24 |
+
"learning_rate": 9.87e-06,
|
25 |
+
"loss": 0.7261,
|
26 |
+
"step": 3000
|
27 |
+
}
|
28 |
+
],
|
29 |
+
"max_steps": 10000,
|
30 |
+
"num_train_epochs": 1,
|
31 |
+
"total_flos": 0.0,
|
32 |
+
"trial_name": null,
|
33 |
+
"trial_params": null
|
34 |
+
}
|