Hugging Face's logo

appvoid
/
void.0

Safetensors
GGUF
English
gemma3_text
base
sml
void
pretrained-from-scratch
Model card Files
xet
 Community
2
appvoid commited on
Commit
d1dd948
·
verified ·
1 Parent(s): 549c967

Continued pretraining: FineWeb-Edu + rewrite6 byte hybrid + no-prompt-15k

Browse files
Files changed (7) hide show
  1. all_results.json +18 -0
  2. config.json +62 -0
  3. generation_config.json +12 -0
  4. model.safetensors +3 -0
  5. train_results.json +18 -0
  6. trainer_state.json +0 -0
  7. training_args.bin +3 -0
all_results.json ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "effective_batch_size": 32,
3
+ "epoch": 2.0,
4
+ "final_gradient_accumulation_steps": 1,
5
+ "final_micro_batch_size": 32,
6
+ "num_input_tokens_seen": 726302720,
7
+ "peak_cuda_memory_gib": 213.5380449295044,
8
+ "real_tokens_per_epoch": 363149667,
9
+ "requested_epochs": 2.0,
10
+ "total_flos": 4.372023498453811e+17,
11
+ "total_optimizer_steps_planned": 5542,
12
+ "train_blocks": 88660,
13
+ "train_loss": 2.3499925993530812,
14
+ "train_runtime": 10647.232,
15
+ "train_samples_per_second": 16.654,
16
+ "train_steps_per_second": 0.521,
17
+ "warmup_steps": 111
18
+ }
config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_sliding_window_pattern": 6,
3
+ "architectures": [
4
+ "Gemma3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_logit_softcapping": null,
9
+ "bos_token_id": 2,
10
+ "dtype": "bfloat16",
11
+ "eos_token_id": 1,
12
+ "final_logit_softcapping": null,
13
+ "head_dim": 256,
14
+ "hidden_activation": "gelu_pytorch_tanh",
15
+ "hidden_size": 640,
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 2048,
18
+ "layer_types": [
19
+ "sliding_attention",
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "full_attention",
25
+ "sliding_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "full_attention",
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "full_attention"
37
+ ],
38
+ "max_position_embeddings": 32768,
39
+ "model_type": "gemma3_text",
40
+ "num_attention_heads": 4,
41
+ "num_hidden_layers": 18,
42
+ "num_key_value_heads": 1,
43
+ "pad_token_id": 0,
44
+ "query_pre_attn_scalar": 256,
45
+ "rms_norm_eps": 1e-06,
46
+ "rope_parameters": {
47
+ "full_attention": {
48
+ "rope_theta": 1000000.0,
49
+ "rope_type": "default"
50
+ },
51
+ "sliding_attention": {
52
+ "rope_theta": 10000.0,
53
+ "rope_type": "default"
54
+ }
55
+ },
56
+ "sliding_window": 512,
57
+ "tie_word_embeddings": true,
58
+ "transformers_version": "5.14.1",
59
+ "use_bidirectional_attention": false,
60
+ "use_cache": false,
61
+ "vocab_size": 50176
62
+ }
generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 2,
3
+ "cache_implementation": "hybrid",
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 1
7
+ ],
8
+ "pad_token_id": 0,
9
+ "top_k": 64,
10
+ "top_p": 0.95,
11
+ "transformers_version": "5.14.1"
12
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e1bebf4612c568c5418b5ed4717d04f864f3488381e16ef561154f0a7f52db1
3
+ size 264903928
train_results.json ADDED
@@ -0,0 +1,18 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "effective_batch_size": 32,
3
+ "epoch": 2.0,
4
+ "final_gradient_accumulation_steps": 1,
5
+ "final_micro_batch_size": 32,
6
+ "num_input_tokens_seen": 726302720,
7
+ "peak_cuda_memory_gib": 213.5380449295044,
8
+ "real_tokens_per_epoch": 363149667,
9
+ "requested_epochs": 2.0,
10
+ "total_flos": 4.372023498453811e+17,
11
+ "total_optimizer_steps_planned": 5542,
12
+ "train_blocks": 88660,
13
+ "train_loss": 2.3499925993530812,
14
+ "train_runtime": 10647.232,
15
+ "train_samples_per_second": 16.654,
16
+ "train_steps_per_second": 0.521,
17
+ "warmup_steps": 111
18
+ }
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:f0c7f6b4148dc35718be038102965bec3395f9e66080cac538838495f6be4900
3
+ size 5201