初始化项目,由ModelHub XC社区提供模型
Model: fpadovani/tur-latn-100mb-after-ppt-Dp-100mb-ckpt500_seed3407 Source: Original Platform
This commit is contained in:
1202
checkpoint-84000/added_tokens.json
Normal file
1202
checkpoint-84000/added_tokens.json
Normal file
File diff suppressed because it is too large
Load Diff
34
checkpoint-84000/config.json
Normal file
34
checkpoint-84000/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-84000/generation_config.json
Normal file
9
checkpoint-84000/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-84000/model.safetensors
Normal file
3
checkpoint-84000/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:824793a2ff13f6505ee63920d221239399d010092c19da8620bf6d11592634a3
|
||||
size 249556672
|
||||
3
checkpoint-84000/rng_state.pth
Normal file
3
checkpoint-84000/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:8d1ad9f2ca9b7f00669134803e4d3f3bcf978e3e9f66a59e0cff32dd9683f1f9
|
||||
size 14645
|
||||
3
checkpoint-84000/scheduler.pt
Normal file
3
checkpoint-84000/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:b6f5941f1334e45386e6d98bdf9eda6912ddd18c9d777be8bde9c8686932acb8
|
||||
size 1465
|
||||
1249
checkpoint-84000/special_tokens_map.json
Normal file
1249
checkpoint-84000/special_tokens_map.json
Normal file
File diff suppressed because it is too large
Load Diff
3
checkpoint-84000/spiece.model
Normal file
3
checkpoint-84000/spiece.model
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:d95f71157f3785af106d43210eaf3449759831f798d5580ffb139c8e488dabc0
|
||||
size 1200440
|
||||
10825
checkpoint-84000/tokenizer_config.json
Normal file
10825
checkpoint-84000/tokenizer_config.json
Normal file
File diff suppressed because it is too large
Load Diff
168342
checkpoint-84000/trainer_state.json
Normal file
168342
checkpoint-84000/trainer_state.json
Normal file
File diff suppressed because it is too large
Load Diff
3
checkpoint-84000/training_args.bin
Normal file
3
checkpoint-84000/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:7ddb8dfa77c62105a7cb8a108aad8fcbb13a3daa886aa7f6b3fafa67e8723095
|
||||
size 6353
|
||||
Reference in New Issue
Block a user