ruggsea commited on
Commit
d130aeb
·
verified ·
1 Parent(s): 0c67b2f

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. README.md +59 -0
  2. checkpoint-1000/config.json +30 -0
  3. checkpoint-1000/generation_config.json +7 -0
  4. checkpoint-1000/model.safetensors +3 -0
  5. checkpoint-1000/optimizer.pt +3 -0
  6. checkpoint-1000/rng_state.pth +3 -0
  7. checkpoint-1000/scheduler.pt +3 -0
  8. checkpoint-1000/special_tokens_map.json +24 -0
  9. checkpoint-1000/tokenizer.json +0 -0
  10. checkpoint-1000/tokenizer_config.json +48 -0
  11. checkpoint-1000/trainer_state.json +0 -0
  12. checkpoint-1000/training_args.bin +3 -0
  13. checkpoint-1250/config.json +30 -0
  14. checkpoint-1250/generation_config.json +7 -0
  15. checkpoint-1250/model.safetensors +3 -0
  16. checkpoint-1250/optimizer.pt +3 -0
  17. checkpoint-1250/rng_state.pth +3 -0
  18. checkpoint-1250/scheduler.pt +3 -0
  19. checkpoint-1250/special_tokens_map.json +24 -0
  20. checkpoint-1250/tokenizer.json +0 -0
  21. checkpoint-1250/tokenizer_config.json +48 -0
  22. checkpoint-1250/trainer_state.json +0 -0
  23. checkpoint-1250/training_args.bin +3 -0
  24. checkpoint-1500/config.json +30 -0
  25. checkpoint-1500/generation_config.json +7 -0
  26. checkpoint-1500/model.safetensors +3 -0
  27. checkpoint-1500/optimizer.pt +3 -0
  28. checkpoint-1500/rng_state.pth +3 -0
  29. checkpoint-1500/scheduler.pt +3 -0
  30. checkpoint-1500/special_tokens_map.json +24 -0
  31. checkpoint-1500/tokenizer.json +0 -0
  32. checkpoint-1500/tokenizer_config.json +48 -0
  33. checkpoint-1500/trainer_state.json +0 -0
  34. checkpoint-1500/training_args.bin +3 -0
  35. checkpoint-2000/config.json +30 -0
  36. checkpoint-2000/generation_config.json +7 -0
  37. checkpoint-2000/model.safetensors +3 -0
  38. checkpoint-2000/optimizer.pt +3 -0
  39. checkpoint-2000/rng_state.pth +3 -0
  40. checkpoint-2000/scheduler.pt +3 -0
  41. checkpoint-2000/special_tokens_map.json +24 -0
  42. checkpoint-2000/tokenizer.json +0 -0
  43. checkpoint-2000/tokenizer_config.json +48 -0
  44. checkpoint-2000/trainer_state.json +0 -0
  45. checkpoint-2000/training_args.bin +3 -0
  46. checkpoint-500/config.json +30 -0
  47. checkpoint-500/generation_config.json +7 -0
  48. checkpoint-500/model.safetensors +3 -0
  49. checkpoint-500/optimizer.pt +3 -0
  50. checkpoint-500/rng_state.pth +3 -0
README.md ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Dante-Zero Fine-tuned Model
2
+
3
+ This model was fine-tuned using Reinforcement Learning with Generative Pre-trained Transformer Optimization (GRPO) to generate Dante-style poetry in endecasillabi (11-syllable lines).
4
+
5
+ ## Model Details
6
+
7
+ - **Base Model:** PleIAs/Pleias-350m-Preview
8
+ - **Training Method:** GRPO (Generative Pre-trained Transformer Optimization)
9
+ - **Training Data:** 1,000 chunks from Dante's Divine Comedy
10
+ - **Epochs:** 10
11
+ - **Trained By:** ruggsea
12
+ - **Date:** 2025-03-08
13
+ - **Run Name:** dante-zero-20250308-Pleias-350m-Preview
14
+
15
+ ## Model Description
16
+
17
+ This model is specialized in generating Italian poetry in the style of Dante Alighieri's Divine Comedy. It has been trained to:
18
+
19
+ 1. Generate proper endecasillabi (11-syllable lines)
20
+ 2. Follow the structure of Dante's poetry
21
+ 3. Avoid repetition
22
+ 4. Create original content (not plagiarize the Divine Comedy)
23
+
24
+ ## Usage
25
+
26
+ ```python
27
+ from transformers import AutoModelForCausalLM, AutoTokenizer
28
+
29
+ # Load model and tokenizer
30
+ model = AutoModelForCausalLM.from_pretrained("ruggsea/dante-zero-20250308-Pleias-350m-Preview")
31
+ tokenizer = AutoTokenizer.from_pretrained("ruggsea/dante-zero-20250308-Pleias-350m-Preview", padding_side="left")
32
+
33
+ # Ensure proper tokenizer settings
34
+ tokenizer.pad_token = tokenizer.eos_token
35
+ tokenizer.pad_token_id = tokenizer.eos_token_id
36
+
37
+ # Generate poetry
38
+ prompt = "Nel mezzo del cammin di nostra vita"
39
+ inputs = tokenizer(prompt, return_tensors="pt", padding_side="left")
40
+ outputs = model.generate(
41
+ inputs.input_ids,
42
+ max_new_tokens=200,
43
+ do_sample=True,
44
+ temperature=0.7,
45
+ top_p=0.9,
46
+ repetition_penalty=1.2
47
+ )
48
+ generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
49
+ print(generated_text)
50
+ ```
51
+
52
+ ## Reward Functions
53
+
54
+ The model was trained using several reward functions:
55
+
56
+ 1. **Endecasillabo Checker:** Rewards proper 11-syllable lines
57
+ 2. **Plagiarism Checker:** Penalizes copying from the Divine Comedy
58
+ 3. **Verse Structure Checker:** Encourages verse-like structure
59
+ 4. **Repetition Penalty:** Discourages repetitive text
checkpoint-1000/config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "PleIAs/Pleias-350m-Preview",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "head_dim": 64,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 1024,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 2560,
15
+ "max_position_embeddings": 2048,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "num_attention_heads": 16,
19
+ "num_hidden_layers": 26,
20
+ "num_key_value_heads": 8,
21
+ "pretraining_tp": 1,
22
+ "rms_norm_eps": 1e-05,
23
+ "rope_scaling": null,
24
+ "rope_theta": 10000,
25
+ "tie_word_embeddings": true,
26
+ "torch_dtype": "bfloat16",
27
+ "transformers_version": "4.49.0",
28
+ "use_cache": true,
29
+ "vocab_size": 65536
30
+ }
checkpoint-1000/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "_padding_side": "left",
4
+ "bos_token_id": 1,
5
+ "eos_token_id": 2,
6
+ "transformers_version": "4.49.0"
7
+ }
checkpoint-1000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5b1057b8b10255e303a36a3103ee518349c9d103a6aea089c14b17cf9bc5eeac
3
+ size 706875632
checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f232f3ebd7e5d8458402b3fc31a67ba1a770a6522ac39182ebe28766924e4115
3
+ size 1413896442
checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c90ea460f09ef8b13f54f5ceece06a61a4d2357d6b18ca52f4bdea82b999736
3
+ size 14180
checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:066959694903b2428a992642efadd126fbc0e6a7300aedf33231c1641a3b801a
3
+ size 1064
checkpoint-1000/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|end_of_text|>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|end_of_text|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<|end_of_text|>",
17
+ "unk_token": {
18
+ "content": "[UNK]",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
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 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[UNK]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<|begin_of_text|>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "<|end_of_text|>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[PAD]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ }
35
+ },
36
+ "bos_token": "<|end_of_text|>",
37
+ "clean_up_tokenization_spaces": true,
38
+ "eos_token": "<|end_of_text|>",
39
+ "extra_special_tokens": {},
40
+ "model_max_length": 1000000000000000019884624838656,
41
+ "pad_token": "<|end_of_text|>",
42
+ "padding_side": "left",
43
+ "return_token_type_ids": false,
44
+ "tokenizer_class": "PreTrainedTokenizer",
45
+ "unk_token": "[UNK]",
46
+ "use_token_type_ids": false,
47
+ "vocab_size": 65536
48
+ }
checkpoint-1000/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3dace8ce1a5e09ab309937787e5aeea778b0c13c6aea3d4d0402aead8da30b6b
3
+ size 5880
checkpoint-1250/config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "PleIAs/Pleias-350m-Preview",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "head_dim": 64,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 1024,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 2560,
15
+ "max_position_embeddings": 2048,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "num_attention_heads": 16,
19
+ "num_hidden_layers": 26,
20
+ "num_key_value_heads": 8,
21
+ "pretraining_tp": 1,
22
+ "rms_norm_eps": 1e-05,
23
+ "rope_scaling": null,
24
+ "rope_theta": 10000,
25
+ "tie_word_embeddings": true,
26
+ "torch_dtype": "bfloat16",
27
+ "transformers_version": "4.49.0",
28
+ "use_cache": true,
29
+ "vocab_size": 65536
30
+ }
checkpoint-1250/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "_padding_side": "left",
4
+ "bos_token_id": 1,
5
+ "eos_token_id": 2,
6
+ "transformers_version": "4.49.0"
7
+ }
checkpoint-1250/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c884a74ea906511756e5576fde66e9697b9239bc6db21905815300296c32ac6
3
+ size 706875632
checkpoint-1250/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8fd4958f56bbc4adc2f9afe2c3667168f4fa18c105b1a5680bc0de5f81dfa5ab
3
+ size 1413896442
checkpoint-1250/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e87723368a46b43369a3e6b14a6ee8e34c835eddb72e056859bef62d5c2a22db
3
+ size 14180
checkpoint-1250/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9197244c5ee8fd84c23cb5387dd7cd4b0d34bb7720142963e9ea404ddb17646d
3
+ size 1064
checkpoint-1250/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|end_of_text|>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|end_of_text|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<|end_of_text|>",
17
+ "unk_token": {
18
+ "content": "[UNK]",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
checkpoint-1250/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1250/tokenizer_config.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[UNK]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<|begin_of_text|>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "<|end_of_text|>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[PAD]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ }
35
+ },
36
+ "bos_token": "<|end_of_text|>",
37
+ "clean_up_tokenization_spaces": true,
38
+ "eos_token": "<|end_of_text|>",
39
+ "extra_special_tokens": {},
40
+ "model_max_length": 1000000000000000019884624838656,
41
+ "pad_token": "<|end_of_text|>",
42
+ "padding_side": "left",
43
+ "return_token_type_ids": false,
44
+ "tokenizer_class": "PreTrainedTokenizer",
45
+ "unk_token": "[UNK]",
46
+ "use_token_type_ids": false,
47
+ "vocab_size": 65536
48
+ }
checkpoint-1250/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1250/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3dace8ce1a5e09ab309937787e5aeea778b0c13c6aea3d4d0402aead8da30b6b
3
+ size 5880
checkpoint-1500/config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "PleIAs/Pleias-350m-Preview",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "head_dim": 64,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 1024,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 2560,
15
+ "max_position_embeddings": 2048,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "num_attention_heads": 16,
19
+ "num_hidden_layers": 26,
20
+ "num_key_value_heads": 8,
21
+ "pretraining_tp": 1,
22
+ "rms_norm_eps": 1e-05,
23
+ "rope_scaling": null,
24
+ "rope_theta": 10000,
25
+ "tie_word_embeddings": true,
26
+ "torch_dtype": "bfloat16",
27
+ "transformers_version": "4.49.0",
28
+ "use_cache": true,
29
+ "vocab_size": 65536
30
+ }
checkpoint-1500/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "_padding_side": "left",
4
+ "bos_token_id": 1,
5
+ "eos_token_id": 2,
6
+ "transformers_version": "4.49.0"
7
+ }
checkpoint-1500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59120a1064af2219dd54d88cc56cc5d00ef4f4de1d64d77bfbac111863c31e77
3
+ size 706875632
checkpoint-1500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d1fafd7579f493aa6f25b736efccf2f516513adca1e385de2f6701bbd993ba9
3
+ size 1413896442
checkpoint-1500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e35710e5e17312d93e4215cb44671c05347760215f22c2ecbaf25e73bc98e470
3
+ size 14180
checkpoint-1500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:716b0cc2a83a19d3701a38cb850a1b7b7616b3a6df57a76c7551804368d34520
3
+ size 1064
checkpoint-1500/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|end_of_text|>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|end_of_text|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<|end_of_text|>",
17
+ "unk_token": {
18
+ "content": "[UNK]",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
checkpoint-1500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1500/tokenizer_config.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[UNK]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<|begin_of_text|>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "<|end_of_text|>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[PAD]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ }
35
+ },
36
+ "bos_token": "<|end_of_text|>",
37
+ "clean_up_tokenization_spaces": true,
38
+ "eos_token": "<|end_of_text|>",
39
+ "extra_special_tokens": {},
40
+ "model_max_length": 1000000000000000019884624838656,
41
+ "pad_token": "<|end_of_text|>",
42
+ "padding_side": "left",
43
+ "return_token_type_ids": false,
44
+ "tokenizer_class": "PreTrainedTokenizer",
45
+ "unk_token": "[UNK]",
46
+ "use_token_type_ids": false,
47
+ "vocab_size": 65536
48
+ }
checkpoint-1500/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1559899b9b88fe3d4f3dabf6d3282e5e364bcb1e111be269752af2ed005eec0b
3
+ size 5880
checkpoint-2000/config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "PleIAs/Pleias-350m-Preview",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "head_dim": 64,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 1024,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 2560,
15
+ "max_position_embeddings": 2048,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "num_attention_heads": 16,
19
+ "num_hidden_layers": 26,
20
+ "num_key_value_heads": 8,
21
+ "pretraining_tp": 1,
22
+ "rms_norm_eps": 1e-05,
23
+ "rope_scaling": null,
24
+ "rope_theta": 10000,
25
+ "tie_word_embeddings": true,
26
+ "torch_dtype": "bfloat16",
27
+ "transformers_version": "4.49.0",
28
+ "use_cache": true,
29
+ "vocab_size": 65536
30
+ }
checkpoint-2000/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "_padding_side": "left",
4
+ "bos_token_id": 1,
5
+ "eos_token_id": 2,
6
+ "transformers_version": "4.49.0"
7
+ }
checkpoint-2000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce882410a504acd57ed1ed27cc13f0b5ef0fb2886a9a28244bc71e60b0251939
3
+ size 706875632
checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4aaddf897a002f7a3a2ebbb6ca42d71148eb6ed3781637ece24922bce7aba7d0
3
+ size 1413896442
checkpoint-2000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f255457753ea35f0069238317250b39ef37861895782ee428f627e50ae33db3c
3
+ size 14180
checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c922bb438fdf675d859401330ee0258b252c6818f3938691cd76bcac69d9e029
3
+ size 1064
checkpoint-2000/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|end_of_text|>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|end_of_text|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<|end_of_text|>",
17
+ "unk_token": {
18
+ "content": "[UNK]",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
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 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[UNK]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<|begin_of_text|>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "<|end_of_text|>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[PAD]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ }
35
+ },
36
+ "bos_token": "<|end_of_text|>",
37
+ "clean_up_tokenization_spaces": true,
38
+ "eos_token": "<|end_of_text|>",
39
+ "extra_special_tokens": {},
40
+ "model_max_length": 1000000000000000019884624838656,
41
+ "pad_token": "<|end_of_text|>",
42
+ "padding_side": "left",
43
+ "return_token_type_ids": false,
44
+ "tokenizer_class": "PreTrainedTokenizer",
45
+ "unk_token": "[UNK]",
46
+ "use_token_type_ids": false,
47
+ "vocab_size": 65536
48
+ }
checkpoint-2000/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f201212167739e155fa7b22956dd37b24a655942dd7b4770636ef85f6b04e79a
3
+ size 5880
checkpoint-500/config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "PleIAs/Pleias-350m-Preview",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "head_dim": 64,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 1024,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 2560,
15
+ "max_position_embeddings": 2048,
16
+ "mlp_bias": false,
17
+ "model_type": "llama",
18
+ "num_attention_heads": 16,
19
+ "num_hidden_layers": 26,
20
+ "num_key_value_heads": 8,
21
+ "pretraining_tp": 1,
22
+ "rms_norm_eps": 1e-05,
23
+ "rope_scaling": null,
24
+ "rope_theta": 10000,
25
+ "tie_word_embeddings": true,
26
+ "torch_dtype": "bfloat16",
27
+ "transformers_version": "4.49.0",
28
+ "use_cache": true,
29
+ "vocab_size": 65536
30
+ }
checkpoint-500/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "_padding_side": "left",
4
+ "bos_token_id": 1,
5
+ "eos_token_id": 2,
6
+ "transformers_version": "4.49.0"
7
+ }
checkpoint-500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1aab5d62a8011bab50d352269cd6cd509637f90fad702fa520c19d0a188e7bad
3
+ size 706875632
checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0fe839021fcbd0ec5cff605de5c54a254be8670393b1d184611b049accd1f736
3
+ size 1413896442
checkpoint-500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:91061687b1770a5a2d6c0979161d48c72e463899f99f39d8f0fcf4b974bcd8d5
3
+ size 14180