初始化项目,由ModelHub XC社区提供模型
Model: fpadovani/tur-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
This commit is contained in:
1202
checkpoint-10050/added_tokens.json
Normal file
1202
checkpoint-10050/added_tokens.json
Normal file
File diff suppressed because it is too large
Load Diff
34
checkpoint-10050/config.json
Normal file
34
checkpoint-10050/config.json
Normal file
@@ -0,0 +1,34 @@
|
||||
{
|
||||
"activation_function": "gelu",
|
||||
"architectures": [
|
||||
"GPT2LMHeadModel"
|
||||
],
|
||||
"attn_pdrop": 0.1,
|
||||
"bos_token_id": 50000,
|
||||
"dtype": "bfloat16",
|
||||
"embd_pdrop": 0.1,
|
||||
"eos_token_id": 50001,
|
||||
"initializer_range": 0.02,
|
||||
"layer_norm_epsilon": 1e-05,
|
||||
"model_type": "gpt2",
|
||||
"n_ctx": 512,
|
||||
"n_embd": 768,
|
||||
"n_head": 12,
|
||||
"n_inner": 3072,
|
||||
"n_layer": 12,
|
||||
"n_positions": 512,
|
||||
"pad_token_id": 50002,
|
||||
"prefix": "[CLS]",
|
||||
"reorder_and_upcast_attn": false,
|
||||
"resid_pdrop": 0.1,
|
||||
"scale_attn_by_inverse_layer_idx": false,
|
||||
"scale_attn_weights": true,
|
||||
"summary_activation": null,
|
||||
"summary_first_dropout": 0.1,
|
||||
"summary_proj_to_labels": true,
|
||||
"summary_type": "cls_index",
|
||||
"summary_use_proj": true,
|
||||
"transformers_version": "4.56.2",
|
||||
"use_cache": true,
|
||||
"vocab_size": 51200
|
||||
}
|
||||
9
checkpoint-10050/generation_config.json
Normal file
9
checkpoint-10050/generation_config.json
Normal file
@@ -0,0 +1,9 @@
|
||||
{
|
||||
"_from_model_config": true,
|
||||
"bos_token_id": 50000,
|
||||
"eos_token_id": [
|
||||
50001
|
||||
],
|
||||
"pad_token_id": 50002,
|
||||
"transformers_version": "4.56.2"
|
||||
}
|
||||
3
checkpoint-10050/model.safetensors
Normal file
3
checkpoint-10050/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:58b9304a61a919562f90e6bb6278cd6f6af5fb66bfc9418ed8fe6461e4fa14fc
|
||||
size 249556672
|
||||
3
checkpoint-10050/rng_state.pth
Normal file
3
checkpoint-10050/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:b28e21595a3fd73c3ec2478723eaf8eb9fe738a8117580bfd67535d45a6a6e0d
|
||||
size 14645
|
||||
3
checkpoint-10050/scheduler.pt
Normal file
3
checkpoint-10050/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:b09fb17ddec155ae4d3ff461d531de39b78db1e4d4c4c6135b759166b2815b03
|
||||
size 1465
|
||||
1249
checkpoint-10050/special_tokens_map.json
Normal file
1249
checkpoint-10050/special_tokens_map.json
Normal file
File diff suppressed because it is too large
Load Diff
3
checkpoint-10050/spiece.model
Normal file
3
checkpoint-10050/spiece.model
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:9bc049efcf029ba7a2a2417149c387e629a10bfcfae7158004fe5319936c480f
|
||||
size 1202823
|
||||
10825
checkpoint-10050/tokenizer_config.json
Normal file
10825
checkpoint-10050/tokenizer_config.json
Normal file
File diff suppressed because it is too large
Load Diff
20354
checkpoint-10050/trainer_state.json
Normal file
20354
checkpoint-10050/trainer_state.json
Normal file
File diff suppressed because it is too large
Load Diff
3
checkpoint-10050/training_args.bin
Normal file
3
checkpoint-10050/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:f1c010001d4a3a2740dcf437862bb61eb4b21d50a77ba6ef996d0da99cdfbec1
|
||||
size 6353
|
||||
Reference in New Issue
Block a user