初始化项目,由ModelHub XC社区提供模型
Model: fpadovani/tur-latn-100mb-10mb_seed3407 Source: Original Platform
This commit is contained in:
1202
checkpoint-3500/added_tokens.json
Normal file
1202
checkpoint-3500/added_tokens.json
Normal file
File diff suppressed because it is too large
Load Diff
34
checkpoint-3500/config.json
Normal file
34
checkpoint-3500/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-3500/generation_config.json
Normal file
9
checkpoint-3500/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-3500/model.safetensors
Normal file
3
checkpoint-3500/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:64db35cc4e93e2bffeaaa50d7b70b5012fa0b3782f3acc5325376abe1073ae80
|
||||
size 249556672
|
||||
3
checkpoint-3500/optimizer.pt
Normal file
3
checkpoint-3500/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:df6b380fa42d64e940de1e4d8355729155371463d41abc540664098ec373f6bf
|
||||
size 499209675
|
||||
3
checkpoint-3500/rng_state.pth
Normal file
3
checkpoint-3500/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:e0530e81e840098f09bc5f74af40d665ac1236f80d0b71e5cec24420c54fe472
|
||||
size 14244
|
||||
3
checkpoint-3500/scheduler.pt
Normal file
3
checkpoint-3500/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:237798a28cf8f82c9aaa7ed420d37ddfb124fb729e9152f32895784c5e4c7608
|
||||
size 1465
|
||||
1249
checkpoint-3500/special_tokens_map.json
Normal file
1249
checkpoint-3500/special_tokens_map.json
Normal file
File diff suppressed because it is too large
Load Diff
3
checkpoint-3500/spiece.model
Normal file
3
checkpoint-3500/spiece.model
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:9bc049efcf029ba7a2a2417149c387e629a10bfcfae7158004fe5319936c480f
|
||||
size 1202823
|
||||
10825
checkpoint-3500/tokenizer_config.json
Normal file
10825
checkpoint-3500/tokenizer_config.json
Normal file
File diff suppressed because it is too large
Load Diff
7111
checkpoint-3500/trainer_state.json
Normal file
7111
checkpoint-3500/trainer_state.json
Normal file
File diff suppressed because it is too large
Load Diff
3
checkpoint-3500/training_args.bin
Normal file
3
checkpoint-3500/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:f0c9f63c2c1822415062037349e594a4ec827e89f15ac4a20fb772165310e396
|
||||
size 5880
|
||||
Reference in New Issue
Block a user