初始化项目,由ModelHub XC社区提供模型
Model: fpadovani/tur_indomain_prepretraining_seed577 Source: Original Platform
This commit is contained in:
35
checkpoint-2500/config.json
Normal file
35
checkpoint-2500/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-2500/generation_config.json
Normal file
7
checkpoint-2500/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-2500/model.safetensors
Normal file
3
checkpoint-2500/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:4deea7f3c0c749a4208cc9ed8eaa8e2a1e6bd59a31521d542f303e1bf57ba01a
|
||||
size 79752272
|
||||
3
checkpoint-2500/optimizer.pt
Normal file
3
checkpoint-2500/optimizer.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:fc4c439cd9df835ef5ffe18ff16eb8ada76c6fe044cf580d7d07567eadfee516
|
||||
size 159538443
|
||||
3
checkpoint-2500/rng_state.pth
Normal file
3
checkpoint-2500/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:8912d33bcb1b75a1de00524c1a07363e0f641629fed837888da0104fda9afe1f
|
||||
size 14645
|
||||
3
checkpoint-2500/scheduler.pt
Normal file
3
checkpoint-2500/scheduler.pt
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:b6b3ad5549d5099e5cee60aa161fa531e77bd60d13a0dc32433f3e6e01a85319
|
||||
size 1465
|
||||
1249
checkpoint-2500/special_tokens_map.json
Normal file
1249
checkpoint-2500/special_tokens_map.json
Normal file
File diff suppressed because it is too large
Load Diff
210940
checkpoint-2500/tokenizer.json
Normal file
210940
checkpoint-2500/tokenizer.json
Normal file
File diff suppressed because one or more lines are too long
10829
checkpoint-2500/tokenizer_config.json
Normal file
10829
checkpoint-2500/tokenizer_config.json
Normal file
File diff suppressed because it is too large
Load Diff
3533
checkpoint-2500/trainer_state.json
Normal file
3533
checkpoint-2500/trainer_state.json
Normal file
File diff suppressed because it is too large
Load Diff
3
checkpoint-2500/training_args.bin
Normal file
3
checkpoint-2500/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:5d844e9bd7269e969751057307f9b24ce1dc61943d4ef008f2ddc415e2d7a0e8
|
||||
size 6161
|
||||
Reference in New Issue
Block a user