Commit
•
1524439
1
Parent(s):
beb97da
nandhakumar7/unslothtrainer
Browse files- README.md +14 -10
- adapter_config.json +8 -8
- adapter_model.safetensors +1 -1
- special_tokens_map.json +8 -2
- tokenizer_config.json +5 -4
- training_args.bin +2 -2
README.md
CHANGED
@@ -1,11 +1,12 @@
|
|
1 |
---
|
2 |
-
|
3 |
library_name: peft
|
|
|
4 |
tags:
|
5 |
- trl
|
6 |
- sft
|
|
|
7 |
- generated_from_trainer
|
8 |
-
base_model: unsloth/llama-3-8b-bnb-4bit
|
9 |
model-index:
|
10 |
- name: outputs
|
11 |
results: []
|
@@ -16,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
16 |
|
17 |
# outputs
|
18 |
|
19 |
-
This model is a fine-tuned version of [unsloth/llama-3-8b-bnb-4bit](https://huggingface.co/unsloth/llama-3-8b-bnb-4bit) on an unknown dataset.
|
20 |
|
21 |
## Model description
|
22 |
|
@@ -35,14 +36,17 @@ More information needed
|
|
35 |
### Training hyperparameters
|
36 |
|
37 |
The following hyperparameters were used during training:
|
38 |
-
- learning_rate:
|
39 |
-
- train_batch_size:
|
40 |
- eval_batch_size: 8
|
41 |
- seed: 3407
|
|
|
|
|
42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
43 |
- lr_scheduler_type: linear
|
44 |
-
- lr_scheduler_warmup_steps:
|
45 |
-
-
|
|
|
46 |
|
47 |
### Training results
|
48 |
|
@@ -51,7 +55,7 @@ The following hyperparameters were used during training:
|
|
51 |
### Framework versions
|
52 |
|
53 |
- PEFT 0.11.1
|
54 |
-
- Transformers 4.
|
55 |
-
- Pytorch 2.3.
|
56 |
-
- Datasets 2.
|
57 |
- Tokenizers 0.19.1
|
|
|
1 |
---
|
2 |
+
base_model: unsloth/llama-3-8b-instruct-bnb-4bit
|
3 |
library_name: peft
|
4 |
+
license: llama3
|
5 |
tags:
|
6 |
- trl
|
7 |
- sft
|
8 |
+
- unsloth
|
9 |
- generated_from_trainer
|
|
|
10 |
model-index:
|
11 |
- name: outputs
|
12 |
results: []
|
|
|
17 |
|
18 |
# outputs
|
19 |
|
20 |
+
This model is a fine-tuned version of [unsloth/llama-3-8b-instruct-bnb-4bit](https://huggingface.co/unsloth/llama-3-8b-instruct-bnb-4bit) on an unknown dataset.
|
21 |
|
22 |
## Model description
|
23 |
|
|
|
36 |
### Training hyperparameters
|
37 |
|
38 |
The following hyperparameters were used during training:
|
39 |
+
- learning_rate: 0.0005
|
40 |
+
- train_batch_size: 2
|
41 |
- eval_batch_size: 8
|
42 |
- seed: 3407
|
43 |
+
- gradient_accumulation_steps: 4
|
44 |
+
- total_train_batch_size: 8
|
45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
46 |
- lr_scheduler_type: linear
|
47 |
+
- lr_scheduler_warmup_steps: 2
|
48 |
+
- num_epochs: 2
|
49 |
+
- mixed_precision_training: Native AMP
|
50 |
|
51 |
### Training results
|
52 |
|
|
|
55 |
### Framework versions
|
56 |
|
57 |
- PEFT 0.11.1
|
58 |
+
- Transformers 4.42.4
|
59 |
+
- Pytorch 2.3.1
|
60 |
+
- Datasets 2.20.0
|
61 |
- Tokenizers 0.19.1
|
adapter_config.json
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
{
|
2 |
"alpha_pattern": {},
|
3 |
"auto_mapping": null,
|
4 |
-
"base_model_name_or_path": "unsloth/llama-3-8b-bnb-4bit",
|
5 |
"bias": "none",
|
6 |
"fan_in_fan_out": false,
|
7 |
"inference_mode": true,
|
@@ -9,7 +9,7 @@
|
|
9 |
"layer_replication": null,
|
10 |
"layers_pattern": null,
|
11 |
"layers_to_transform": null,
|
12 |
-
"loftq_config":
|
13 |
"lora_alpha": 16,
|
14 |
"lora_dropout": 0,
|
15 |
"megatron_config": null,
|
@@ -18,15 +18,15 @@
|
|
18 |
"peft_type": "LORA",
|
19 |
"r": 16,
|
20 |
"rank_pattern": {},
|
21 |
-
"revision":
|
22 |
"target_modules": [
|
23 |
-
"up_proj",
|
24 |
"v_proj",
|
25 |
-
"q_proj",
|
26 |
-
"down_proj",
|
27 |
-
"gate_proj",
|
28 |
"o_proj",
|
29 |
-
"k_proj"
|
|
|
|
|
|
|
|
|
30 |
],
|
31 |
"task_type": "CAUSAL_LM",
|
32 |
"use_dora": false,
|
|
|
1 |
{
|
2 |
"alpha_pattern": {},
|
3 |
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "unsloth/llama-3-8b-instruct-bnb-4bit",
|
5 |
"bias": "none",
|
6 |
"fan_in_fan_out": false,
|
7 |
"inference_mode": true,
|
|
|
9 |
"layer_replication": null,
|
10 |
"layers_pattern": null,
|
11 |
"layers_to_transform": null,
|
12 |
+
"loftq_config": {},
|
13 |
"lora_alpha": 16,
|
14 |
"lora_dropout": 0,
|
15 |
"megatron_config": null,
|
|
|
18 |
"peft_type": "LORA",
|
19 |
"r": 16,
|
20 |
"rank_pattern": {},
|
21 |
+
"revision": "unsloth",
|
22 |
"target_modules": [
|
|
|
23 |
"v_proj",
|
|
|
|
|
|
|
24 |
"o_proj",
|
25 |
+
"k_proj",
|
26 |
+
"up_proj",
|
27 |
+
"down_proj",
|
28 |
+
"q_proj",
|
29 |
+
"gate_proj"
|
30 |
],
|
31 |
"task_type": "CAUSAL_LM",
|
32 |
"use_dora": false,
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 167832240
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e9ee50554a3c2f7a98cebf8f9865d82aec93dd868868d3fd2b4c7d25d4c9bf7f
|
3 |
size 167832240
|
special_tokens_map.json
CHANGED
@@ -7,11 +7,17 @@
|
|
7 |
"single_word": false
|
8 |
},
|
9 |
"eos_token": {
|
10 |
-
"content": "<|
|
11 |
"lstrip": false,
|
12 |
"normalized": false,
|
13 |
"rstrip": false,
|
14 |
"single_word": false
|
15 |
},
|
16 |
-
"pad_token":
|
|
|
|
|
|
|
|
|
|
|
|
|
17 |
}
|
|
|
7 |
"single_word": false
|
8 |
},
|
9 |
"eos_token": {
|
10 |
+
"content": "<|eot_id|>",
|
11 |
"lstrip": false,
|
12 |
"normalized": false,
|
13 |
"rstrip": false,
|
14 |
"single_word": false
|
15 |
},
|
16 |
+
"pad_token": {
|
17 |
+
"content": "<|reserved_special_token_250|>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
}
|
23 |
}
|
tokenizer_config.json
CHANGED
@@ -2050,14 +2050,15 @@
|
|
2050 |
}
|
2051 |
},
|
2052 |
"bos_token": "<|begin_of_text|>",
|
|
|
2053 |
"clean_up_tokenization_spaces": true,
|
2054 |
-
"eos_token": "<|
|
2055 |
"model_input_names": [
|
2056 |
"input_ids",
|
2057 |
"attention_mask"
|
2058 |
],
|
2059 |
-
"model_max_length":
|
2060 |
-
"pad_token": "<|
|
2061 |
-
"padding_side": "
|
2062 |
"tokenizer_class": "PreTrainedTokenizerFast"
|
2063 |
}
|
|
|
2050 |
}
|
2051 |
},
|
2052 |
"bos_token": "<|begin_of_text|>",
|
2053 |
+
"chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
|
2054 |
"clean_up_tokenization_spaces": true,
|
2055 |
+
"eos_token": "<|eot_id|>",
|
2056 |
"model_input_names": [
|
2057 |
"input_ids",
|
2058 |
"attention_mask"
|
2059 |
],
|
2060 |
+
"model_max_length": 8192,
|
2061 |
+
"pad_token": "<|reserved_special_token_250|>",
|
2062 |
+
"padding_side": "right",
|
2063 |
"tokenizer_class": "PreTrainedTokenizerFast"
|
2064 |
}
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d6601c554304cadbb67c363c3f848816993528a21e78e82e2d52f175088ebd57
|
3 |
+
size 5112
|