初始化项目,由ModelHub XC社区提供模型
Model: fpadovani/heb-hebr-100mb-100mb_seed3407 Source: Original Platform
This commit is contained in:
1202
checkpoint-87000/added_tokens.json
Normal file
1202
checkpoint-87000/added_tokens.json
Normal file
File diff suppressed because it is too large
Load Diff
34
checkpoint-87000/config.json
Normal file
34
checkpoint-87000/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-87000/generation_config.json
Normal file
9
checkpoint-87000/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-87000/model.safetensors
Normal file
3
checkpoint-87000/model.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:1d4780d281161248fa726ed3d614adcd07d39c198200660f9f8fae1e2fa58b8c
|
||||
size 249556672
|
||||
3
checkpoint-87000/rng_state.pth
Normal file
3
checkpoint-87000/rng_state.pth
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:aa2de3329593b58d404154dee13dd58473b45251d3db260065072d9bb23b107f
|
||||
size 14645
|
||||
1249
checkpoint-87000/special_tokens_map.json
Normal file
1249
checkpoint-87000/special_tokens_map.json
Normal file
File diff suppressed because it is too large
Load Diff
3
checkpoint-87000/spiece.model
Normal file
3
checkpoint-87000/spiece.model
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:38b155328359cdd2d3d2776654f2d30a8fa8923fa48c2edf841f5c870e4b14c7
|
||||
size 1284754
|
||||
10825
checkpoint-87000/tokenizer_config.json
Normal file
10825
checkpoint-87000/tokenizer_config.json
Normal file
File diff suppressed because it is too large
Load Diff
174353
checkpoint-87000/trainer_state.json
Normal file
174353
checkpoint-87000/trainer_state.json
Normal file
File diff suppressed because it is too large
Load Diff
3
checkpoint-87000/training_args.bin
Normal file
3
checkpoint-87000/training_args.bin
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:e4c8cea7be5572185e565f6a8b3d32c9fa11075b9b51d0ec405be082f54d0fd6
|
||||
size 6289
|
||||
Reference in New Issue
Block a user