Upload folder using huggingface_hub
Browse files- exp5.ethereal-galaxy-3/checkpoint-125500/config.json +32 -0
- exp5.ethereal-galaxy-3/checkpoint-125500/generation_config.json +10 -0
- exp5.ethereal-galaxy-3/checkpoint-125500/model.safetensors +3 -0
- exp5.ethereal-galaxy-3/checkpoint-125500/optimizer.pt +3 -0
- exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_0.pth +3 -0
- exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_1.pth +3 -0
- exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_2.pth +3 -0
- exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_3.pth +3 -0
- exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_4.pth +3 -0
- exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_5.pth +3 -0
- exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_6.pth +3 -0
- exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_7.pth +3 -0
- exp5.ethereal-galaxy-3/checkpoint-125500/scheduler.pt +3 -0
- exp5.ethereal-galaxy-3/checkpoint-125500/trainer_state.json +0 -0
- exp5.ethereal-galaxy-3/checkpoint-125500/training_args.bin +3 -0
exp5.ethereal-galaxy-3/checkpoint-125500/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 1,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 8192,
|
| 15 |
+
"max_position_embeddings": 8192,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 32,
|
| 19 |
+
"num_hidden_layers": 16,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"pad_token_id": 0,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"rms_norm_eps": 1e-05,
|
| 24 |
+
"rope_parameters": {
|
| 25 |
+
"rope_theta": 10000.0,
|
| 26 |
+
"rope_type": "default"
|
| 27 |
+
},
|
| 28 |
+
"tie_word_embeddings": false,
|
| 29 |
+
"transformers_version": "5.0.0",
|
| 30 |
+
"use_cache": false,
|
| 31 |
+
"vocab_size": 2779
|
| 32 |
+
}
|
exp5.ethereal-galaxy-3/checkpoint-125500/generation_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 1,
|
| 5 |
+
"output_attentions": false,
|
| 6 |
+
"output_hidden_states": false,
|
| 7 |
+
"pad_token_id": 0,
|
| 8 |
+
"transformers_version": "5.0.0",
|
| 9 |
+
"use_cache": true
|
| 10 |
+
}
|
exp5.ethereal-galaxy-3/checkpoint-125500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b6e0cd05cd782d278a36e2aabcf3c0ab32d222917927c24eeef4edfeb7a46fa2
|
| 3 |
+
size 3938132328
|
exp5.ethereal-galaxy-3/checkpoint-125500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0eb58cf3860e165842475e3a240bd59e732f7b1d622ab42b4dd3f85cc7f7cfc1
|
| 3 |
+
size 7876360739
|
exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5dddc629c0bf68f9e2f8b2e4129cbb1c2b10baf4e5ced2ca3c8f4e7466f26f4a
|
| 3 |
+
size 16389
|
exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:62b6f927092e187145a929f85f72710546b141d92258a4f504c9315b645e2f5f
|
| 3 |
+
size 16389
|
exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:69b4d822c85e1fc408e49968cc1cacd131e5f749328c08e1d71d5d0a908a5bdc
|
| 3 |
+
size 16389
|
exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bd29105983242012d55d70b99528b4d06e122f1d24e6cca8bea63ac5eb404e64
|
| 3 |
+
size 16389
|
exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_4.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d24424a038ad2f54a5463356a45c4634915c5611300d8ae4c9282b1a9b99328e
|
| 3 |
+
size 16389
|
exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_5.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd1ca3d1df61b5844141f424cadd37c15d100864c0a0a2a7cd3809f4f0f3b702
|
| 3 |
+
size 16389
|
exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_6.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7da0ae1e97644704b082717153dbd03d1f9b6431abed721acb8d50ad856b5f13
|
| 3 |
+
size 16389
|
exp5.ethereal-galaxy-3/checkpoint-125500/rng_state_7.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ba7751406af3a294ddd4a25a8b63a0a72b3696b16683d64efbaab400289a27dc
|
| 3 |
+
size 16389
|
exp5.ethereal-galaxy-3/checkpoint-125500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a1e11561668c2527e75fbae0595c5d8094de5e69a9823d8db963742e35831154
|
| 3 |
+
size 1465
|
exp5.ethereal-galaxy-3/checkpoint-125500/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
exp5.ethereal-galaxy-3/checkpoint-125500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:943ac99d00594726002a0fa2924e2687ab75e8fa425a956b0e51b21bf84e0ffd
|
| 3 |
+
size 5201
|