Upload folder using huggingface_hub
Browse files- README.md +65 -0
- adapter_config.json +31 -0
- adapter_model.safetensors +3 -0
- all_results.json +14 -0
- config.json +49 -0
- eval_results.json +8 -0
- runs/Jun21_16-30-50_lrdn0368.leonardo.local/events.out.tfevents.1718980842.lrdn0368.leonardo.local.335481.0 +3 -0
- runs/Jun21_16-30-50_lrdn0368.leonardo.local/events.out.tfevents.1719050530.lrdn0368.leonardo.local.335481.1 +3 -0
- special_tokens_map.json +30 -0
- tokenizer.json +0 -0
- tokenizer.model +3 -0
- tokenizer_config.json +51 -0
- train_results.json +9 -0
- trainer_state.json +0 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
tags:
|
4 |
+
- alignment-handbook
|
5 |
+
- generated_from_trainer
|
6 |
+
base_model: RiTA-nlp/llama3-tweety-8b-italian
|
7 |
+
datasets:
|
8 |
+
- giux78/100k-sft-ready-ultrafeedback-ita
|
9 |
+
model-index:
|
10 |
+
- name: llama3-tweety-8b-italian-sft-uf_ita
|
11 |
+
results: []
|
12 |
+
---
|
13 |
+
|
14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
15 |
+
should probably proofread and complete it, then remove this comment. -->
|
16 |
+
|
17 |
+
# llama3-tweety-8b-italian-sft-uf_ita
|
18 |
+
|
19 |
+
This model is a fine-tuned version of [RiTA-nlp/llama3-tweety-8b-italian](https://huggingface.co/RiTA-nlp/llama3-tweety-8b-italian) on the giux78/100k-sft-ready-ultrafeedback-ita dataset.
|
20 |
+
It achieves the following results on the evaluation set:
|
21 |
+
- Loss: 1.4126
|
22 |
+
|
23 |
+
## Model description
|
24 |
+
|
25 |
+
More information needed
|
26 |
+
|
27 |
+
## Intended uses & limitations
|
28 |
+
|
29 |
+
More information needed
|
30 |
+
|
31 |
+
## Training and evaluation data
|
32 |
+
|
33 |
+
More information needed
|
34 |
+
|
35 |
+
## Training procedure
|
36 |
+
|
37 |
+
### Training hyperparameters
|
38 |
+
|
39 |
+
The following hyperparameters were used during training:
|
40 |
+
- learning_rate: 0.0002
|
41 |
+
- train_batch_size: 4
|
42 |
+
- eval_batch_size: 8
|
43 |
+
- seed: 42
|
44 |
+
- distributed_type: multi-GPU
|
45 |
+
- gradient_accumulation_steps: 2
|
46 |
+
- total_train_batch_size: 8
|
47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
48 |
+
- lr_scheduler_type: cosine
|
49 |
+
- lr_scheduler_warmup_ratio: 0.1
|
50 |
+
- num_epochs: 1
|
51 |
+
|
52 |
+
### Training results
|
53 |
+
|
54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
55 |
+
|:-------------:|:------:|:----:|:---------------:|
|
56 |
+
| 1.4435 | 0.9999 | 9050 | 1.4126 |
|
57 |
+
|
58 |
+
|
59 |
+
### Framework versions
|
60 |
+
|
61 |
+
- PEFT 0.7.1
|
62 |
+
- Transformers 4.40.2
|
63 |
+
- Pytorch 2.1.2+cu121
|
64 |
+
- Datasets 2.19.1
|
65 |
+
- Tokenizers 0.19.1
|
adapter_config.json
ADDED
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "RiTA-nlp/llama3-tweety-8b-italian",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layers_pattern": null,
|
10 |
+
"layers_to_transform": null,
|
11 |
+
"loftq_config": {},
|
12 |
+
"lora_alpha": 16,
|
13 |
+
"lora_dropout": 0.05,
|
14 |
+
"megatron_config": null,
|
15 |
+
"megatron_core": "megatron.core",
|
16 |
+
"modules_to_save": null,
|
17 |
+
"peft_type": "LORA",
|
18 |
+
"r": 16,
|
19 |
+
"rank_pattern": {},
|
20 |
+
"revision": null,
|
21 |
+
"target_modules": [
|
22 |
+
"gate_proj",
|
23 |
+
"k_proj",
|
24 |
+
"v_proj",
|
25 |
+
"o_proj",
|
26 |
+
"up_proj",
|
27 |
+
"q_proj",
|
28 |
+
"down_proj"
|
29 |
+
],
|
30 |
+
"task_type": "CAUSAL_LM"
|
31 |
+
}
|
adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b7d7589a3e3c7ebd73c810d37228d952b77511575c1df866595c1bb068877d19
|
3 |
+
size 83946192
|
all_results.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 0.9999447544334568,
|
3 |
+
"eval_loss": 1.4126445055007935,
|
4 |
+
"eval_runtime": 1775.8332,
|
5 |
+
"eval_samples": 10000,
|
6 |
+
"eval_samples_per_second": 4.058,
|
7 |
+
"eval_steps_per_second": 0.507,
|
8 |
+
"total_flos": 6.363322205866033e+18,
|
9 |
+
"train_loss": 1.4676292563538524,
|
10 |
+
"train_runtime": 67905.3687,
|
11 |
+
"train_samples": 100000,
|
12 |
+
"train_samples_per_second": 1.066,
|
13 |
+
"train_steps_per_second": 0.133
|
14 |
+
}
|
config.json
ADDED
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "RiTA-nlp/llama3-tweety-8b-italian",
|
3 |
+
"additional_special_tokens_ids": [],
|
4 |
+
"architectures": [
|
5 |
+
"LlamaForCausalLM"
|
6 |
+
],
|
7 |
+
"attention_bias": false,
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"bos_token_id": 1,
|
10 |
+
"cls_token_id": null,
|
11 |
+
"eos_token_id": 2,
|
12 |
+
"hidden_act": "silu",
|
13 |
+
"hidden_size": 4096,
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"intermediate_size": 14336,
|
16 |
+
"mask_token_id": null,
|
17 |
+
"max_position_embeddings": 8192,
|
18 |
+
"model_type": "llama",
|
19 |
+
"num_attention_heads": 32,
|
20 |
+
"num_hidden_layers": 32,
|
21 |
+
"num_key_value_heads": 8,
|
22 |
+
"pad_token_id": 2,
|
23 |
+
"pretraining_tp": 1,
|
24 |
+
"quantization_config": {
|
25 |
+
"_load_in_4bit": true,
|
26 |
+
"_load_in_8bit": false,
|
27 |
+
"bnb_4bit_compute_dtype": "bfloat16",
|
28 |
+
"bnb_4bit_quant_storage": "uint8",
|
29 |
+
"bnb_4bit_quant_type": "nf4",
|
30 |
+
"bnb_4bit_use_double_quant": false,
|
31 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
32 |
+
"llm_int8_has_fp16_weight": false,
|
33 |
+
"llm_int8_skip_modules": null,
|
34 |
+
"llm_int8_threshold": 6.0,
|
35 |
+
"load_in_4bit": true,
|
36 |
+
"load_in_8bit": false,
|
37 |
+
"quant_method": "bitsandbytes"
|
38 |
+
},
|
39 |
+
"rms_norm_eps": 1e-05,
|
40 |
+
"rope_scaling": null,
|
41 |
+
"rope_theta": 500000.0,
|
42 |
+
"tie_word_embeddings": false,
|
43 |
+
"tokenizer_class": "LlamaTokenizerFast",
|
44 |
+
"torch_dtype": "bfloat16",
|
45 |
+
"transformers_version": "4.40.2",
|
46 |
+
"unk_token_id": 0,
|
47 |
+
"use_cache": true,
|
48 |
+
"vocab_size": 32000
|
49 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 0.9999447544334568,
|
3 |
+
"eval_loss": 1.4126445055007935,
|
4 |
+
"eval_runtime": 1775.8332,
|
5 |
+
"eval_samples": 10000,
|
6 |
+
"eval_samples_per_second": 4.058,
|
7 |
+
"eval_steps_per_second": 0.507
|
8 |
+
}
|
runs/Jun21_16-30-50_lrdn0368.leonardo.local/events.out.tfevents.1718980842.lrdn0368.leonardo.local.335481.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:afbad4d92e499206b4a8beb871f4f78ee0c50fae3da9959df6f8068d2ed3dae8
|
3 |
+
size 388307
|
runs/Jun21_16-30-50_lrdn0368.leonardo.local/events.out.tfevents.1719050530.lrdn0368.leonardo.local.335481.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7cae560afa6ce47a5f84fe59f90a5bcf26900d09c88532694d74c3ed6913fea7
|
3 |
+
size 359
|
special_tokens_map.json
ADDED
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<s>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"eos_token": {
|
10 |
+
"content": "</s>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"pad_token": {
|
17 |
+
"content": "</s>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
},
|
23 |
+
"unk_token": {
|
24 |
+
"content": "<unk>",
|
25 |
+
"lstrip": false,
|
26 |
+
"normalized": false,
|
27 |
+
"rstrip": false,
|
28 |
+
"single_word": false
|
29 |
+
}
|
30 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f2ddc1899cc92e7a8d539baa0107d226a195a0548881a2806c5de01eea2822e0
|
3 |
+
size 723618
|
tokenizer_config.json
ADDED
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_bos_token": true,
|
3 |
+
"add_eos_token": false,
|
4 |
+
"add_prefix_space": true,
|
5 |
+
"added_tokens_decoder": {
|
6 |
+
"0": {
|
7 |
+
"content": "<unk>",
|
8 |
+
"lstrip": false,
|
9 |
+
"normalized": false,
|
10 |
+
"rstrip": false,
|
11 |
+
"single_word": false,
|
12 |
+
"special": true
|
13 |
+
},
|
14 |
+
"1": {
|
15 |
+
"content": "<s>",
|
16 |
+
"lstrip": false,
|
17 |
+
"normalized": false,
|
18 |
+
"rstrip": false,
|
19 |
+
"single_word": false,
|
20 |
+
"special": true
|
21 |
+
},
|
22 |
+
"2": {
|
23 |
+
"content": "</s>",
|
24 |
+
"lstrip": false,
|
25 |
+
"normalized": false,
|
26 |
+
"rstrip": false,
|
27 |
+
"single_word": false,
|
28 |
+
"special": true
|
29 |
+
},
|
30 |
+
"3": {
|
31 |
+
"content": "<pad>",
|
32 |
+
"lstrip": false,
|
33 |
+
"normalized": false,
|
34 |
+
"rstrip": false,
|
35 |
+
"single_word": false,
|
36 |
+
"special": true
|
37 |
+
}
|
38 |
+
},
|
39 |
+
"bos_token": "<s>",
|
40 |
+
"chat_template": "{% for message in messages %}\n{% if message['role'] == 'user' %}\n{{ '<|user|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'system' %}\n{{ '<|system|>\n' + message['content'] + eos_token }}\n{% elif message['role'] == 'assistant' %}\n{{ '<|assistant|>\n' + message['content'] + eos_token }}\n{% endif %}\n{% if loop.last and add_generation_prompt %}\n{{ '<|assistant|>' }}\n{% endif %}\n{% endfor %}",
|
41 |
+
"clean_up_tokenization_spaces": false,
|
42 |
+
"eos_token": "</s>",
|
43 |
+
"legacy": true,
|
44 |
+
"model_max_length": 2048,
|
45 |
+
"pad_token": "</s>",
|
46 |
+
"sp_model_kwargs": {},
|
47 |
+
"spaces_between_special_tokens": false,
|
48 |
+
"tokenizer_class": "LlamaTokenizer",
|
49 |
+
"unk_token": "<unk>",
|
50 |
+
"use_default_system_prompt": false
|
51 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 0.9999447544334568,
|
3 |
+
"total_flos": 6.363322205866033e+18,
|
4 |
+
"train_loss": 1.4676292563538524,
|
5 |
+
"train_runtime": 67905.3687,
|
6 |
+
"train_samples": 100000,
|
7 |
+
"train_samples_per_second": 1.066,
|
8 |
+
"train_steps_per_second": 0.133
|
9 |
+
}
|
trainer_state.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:422a0c14aab7f1d146f2514bc85cf25350344b77222d51eba172ced958446888
|
3 |
+
size 5176
|