初始化项目,由ModelHub XC社区提供模型
Model: fpadovani/tur_indomain_prepretraining_seed443 Source: Original Platform
This commit is contained in:
35
checkpoint-1500/config.json
Normal file
35
checkpoint-1500/config.json
Normal file
@@ -0,0 +1,35 @@
|
||||
{
|
||||
"_name_or_path": "goldfish-models/tur_latn_10mb",
|
||||
"activation_function": "gelu",
|
||||
"architectures": [
|
||||
"GPT2LMHeadModel"
|
||||
],
|
||||
"attn_pdrop": 0.1,
|
||||
"bos_token_id": 50000,
|
||||
"embd_pdrop": 0.1,
|
||||
"eos_token_id": 50001,
|
||||
"initializer_range": 0.02,
|
||||
"layer_norm_epsilon": 1e-05,
|
||||
"model_type": "gpt2",
|
||||
"n_ctx": 2048,
|
||||
"n_embd": 512,
|
||||
"n_head": 8,
|
||||
"n_inner": 2048,
|
||||
"n_layer": 4,
|
||||
"n_positions": 2048,
|
||||
"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,
|
||||
"torch_dtype": "bfloat16",
|
||||
"transformers_version": "4.47.0",
|
||||
"use_cache": true,
|
||||
"vocab_size": 51200
|
||||
}
|
||||
7
checkpoint-1500/generation_config.json
Normal file
7
checkpoint-1500/generation_config.json
Normal file
@@ -0,0 +1,7 @@
|
||||
{
|
||||
"_from_model_config": true,
|
||||
"bos_token_id": 50000,
|
||||
"eos_token_id": 50001,
|
||||
"pad_token_id": 50002,
|
||||
"transformers_version": "4.47.0"
|
||||
}
|
||||
3
checkpoint-1500/model.safetensors
Normal file
3
checkpoint-1500/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:3fc62649876a24d8523dea8a5a4f38e38fb265d13fedb0e000c4d93e8e7f3401
|
||||
size 79752272
|
||||
3
checkpoint-1500/optimizer.pt
Normal file
3
checkpoint-1500/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:10a3e995bc0a215d0fc53c5364682b1aee2d53bc4e1073e48a62926d52f64c71
|
||||
size 159538443
|
||||
3
checkpoint-1500/rng_state.pth
Normal file
3
checkpoint-1500/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:765b0d7858091d7ba87c8b5b857d0c75a24b9ba3e1217688224015760db749aa
|
||||
size 14645
|
||||
3
checkpoint-1500/scheduler.pt
Normal file
3
checkpoint-1500/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:cdf706bbb7bb6e14789b8a64f9da210a0ebd06388777b0fcde0c23247b8fa573
|
||||
size 1465
|
||||
1249
checkpoint-1500/special_tokens_map.json
Normal file
1249
checkpoint-1500/special_tokens_map.json
Normal file
File diff suppressed because it is too large
Load Diff
210940
checkpoint-1500/tokenizer.json
Normal file
210940
checkpoint-1500/tokenizer.json
Normal file
File diff suppressed because one or more lines are too long
10829
checkpoint-1500/tokenizer_config.json
Normal file
10829
checkpoint-1500/tokenizer_config.json
Normal file
File diff suppressed because it is too large
Load Diff
2133
checkpoint-1500/trainer_state.json
Normal file
2133
checkpoint-1500/trainer_state.json
Normal file
File diff suppressed because it is too large
Load Diff
3
checkpoint-1500/training_args.bin
Normal file
3
checkpoint-1500/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:6dc7942391c97f5c71d525af067d82cf56a6d2234bb905634d507c9b596fea7c
|
||||
size 6161
|
||||
Reference in New Issue
Block a user