初始化项目，由ModelHub XC社区提供模型
Model: Pageshift-Entertainment/pagestorm-research-preview-24b-first-chapter-only Source: Original Platform
This commit is contained in:
ModelHub XC
38
.gitattributes
vendored
Normal file
38
.gitattributes
vendored
Normal file
@@ -0,0 +1,38 @@
|
||||
*.7z filter=lfs diff=lfs merge=lfs -text
|
||||
*.arrow filter=lfs diff=lfs merge=lfs -text
|
||||
*.bin filter=lfs diff=lfs merge=lfs -text
|
||||
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
||||
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
||||
*.ftz filter=lfs diff=lfs merge=lfs -text
|
||||
*.gz filter=lfs diff=lfs merge=lfs -text
|
||||
*.h5 filter=lfs diff=lfs merge=lfs -text
|
||||
*.joblib filter=lfs diff=lfs merge=lfs -text
|
||||
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
||||
*.model filter=lfs diff=lfs merge=lfs -text
|
||||
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
||||
*.npy filter=lfs diff=lfs merge=lfs -text
|
||||
*.npz filter=lfs diff=lfs merge=lfs -text
|
||||
*.onnx filter=lfs diff=lfs merge=lfs -text
|
||||
*.ot filter=lfs diff=lfs merge=lfs -text
|
||||
*.parquet filter=lfs diff=lfs merge=lfs -text
|
||||
*.pb filter=lfs diff=lfs merge=lfs -text
|
||||
*.pickle filter=lfs diff=lfs merge=lfs -text
|
||||
*.pkl filter=lfs diff=lfs merge=lfs -text
|
||||
*.pt filter=lfs diff=lfs merge=lfs -text
|
||||
*.pth filter=lfs diff=lfs merge=lfs -text
|
||||
*.rar filter=lfs diff=lfs merge=lfs -text
|
||||
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
||||
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
||||
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.tar filter=lfs diff=lfs merge=lfs -text
|
||||
*.tflite filter=lfs diff=lfs merge=lfs -text
|
||||
*.tgz filter=lfs diff=lfs merge=lfs -text
|
||||
*.wasm filter=lfs diff=lfs merge=lfs -text
|
||||
*.xz filter=lfs diff=lfs merge=lfs -text
|
||||
*.zip filter=lfs diff=lfs merge=lfs -text
|
||||
*.zst filter=lfs diff=lfs merge=lfs -text
|
||||
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
||||
tekken.json filter=lfs diff=lfs merge=lfs -text
|
||||
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
||||
cover_image.png filter=lfs diff=lfs merge=lfs -text
|
||||
44
README.md
Normal file
44
README.md
Normal file
@@ -0,0 +1,44 @@
|
||||
---
|
||||
library_name: transformers
|
||||
pipeline_tag: text-generation
|
||||
datasets:
|
||||
- Pageshift-Entertainment/LongPage
|
||||
tags:
|
||||
- text-generation
|
||||
- transformers
|
||||
- vllm
|
||||
- story-generation
|
||||
- staged-generation
|
||||
- first-chapter
|
||||
license: apache-2.0
|
||||
language:
|
||||
- en
|
||||
---
|
||||
|
||||
|
||||
|
||||
# PageStorm Research Preview 24B First Chapter Only
|
||||
|
||||
### Paper: [`2605.17064`](https://huggingface.co/papers/2605.17064)
|
||||
### Collection: [`PageStorm Research Preview`](https://huggingface.co/collections/Pageshift-Entertainment/pagestorm-research-preview)
|
||||
### Twitter / X: [`https://x.com/pageshiftAI`](https://x.com/pageshiftAI)
|
||||
### Job Openings: [`https://pageshift.ai/hiring`](https://pageshift.ai/hiring)
|
||||
|
||||
|
||||
This is an older pipe-cleaning model and is not the one described in the paper. It was trained to only generate until the first chapter, and it uses an older and simpler planning scaffold. It was trained without the LongPage Dataset and was initialized from a Mistral 24B base model. The training was performed on a V5p-64 TPU Pod provided by Google’s TPU Research Cloud.
|
||||
The model is single-turn only and cannot receive additional user prompts after the initial one.
|
||||
|
||||
> [!WARNING]
|
||||
> **Important Safety Note:** This model was trained **without safety** guidelines in place. It **cannot refuse** any requests on its own and should be deployed behind a separate safety classifier.
|
||||
|
||||
## Usage Example
|
||||
|
||||
```bash
|
||||
pip install "git+https://github.com/Pageshift-ai/pagestorm.git"
|
||||
```
|
||||
|
||||
```bash
|
||||
pagestorm generate \
|
||||
--repo-id Pageshift-Entertainment/pagestorm-research-preview-24b-first-chapter-only \
|
||||
--prompt "Thriller in Zurich"
|
||||
```
|
||||
23
chat_template.jinja
Normal file
23
chat_template.jinja
Normal file
@@ -0,0 +1,23 @@
|
||||
{% for message in messages %}{% set role = message['role'] %}{% set content = message['content'] | default('', true) %}{% if role == 'prompt' %}<|start_header_id|>prompt<|stop_header_id|>
|
||||
|
||||
{% elif role == 'book_preview' %}<|start_header_id|>book_preview<|stop_header_id|>
|
||||
|
||||
{% elif role == 'book_plan' %}<|start_header_id|>book_plan<|stop_header_id|>
|
||||
|
||||
{% elif role == 'first_chapter_plan' %}<|start_header_id|>first_chapter_plan<|stop_header_id|>
|
||||
|
||||
{% elif role == 'first_chapter_scene_breakdown' %}<|start_header_id|>first_chapter_scene_breakdown<|stop_header_id|>
|
||||
|
||||
{% elif role == 'first_chapter_text' %}<|start_header_id|>first_chapter_text<|stop_header_id|>
|
||||
|
||||
{% elif role == 'full_book_chapters_plan' %}<|start_header_id|>full_book_chapters_plan<|stop_header_id|>
|
||||
|
||||
{% elif role == 'book_characters_list' %}<|start_header_id|>book_characters_list<|stop_header_id|>
|
||||
|
||||
{% elif role == 'scene_breakdown' %}<|start_header_id|>scene_breakdown<|stop_header_id|>
|
||||
|
||||
{% elif role == 'chapter_text' %}<|start_header_id|>chapter_text<|stop_header_id|>
|
||||
|
||||
{% else %}<|start_header_id|>{{ role }}<|stop_header_id|>
|
||||
|
||||
{% endif %}{{ content }}{% if not loop.last or content %}<|eot_id|>{% endif %}{% endfor %}
|
||||
28
config.json
Normal file
28
config.json
Normal file
@@ -0,0 +1,28 @@
|
||||
{
|
||||
"architectures": [
|
||||
"MistralForCausalLM"
|
||||
],
|
||||
"attention_dropout": 0.0,
|
||||
"bos_token_id": 1,
|
||||
"dtype": "bfloat16",
|
||||
"eos_token_id": 2,
|
||||
"head_dim": 128,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 5120,
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 32768,
|
||||
"max_position_embeddings": 32768,
|
||||
"model_type": "mistral",
|
||||
"num_attention_heads": 32,
|
||||
"num_hidden_layers": 40,
|
||||
"num_key_value_heads": 16,
|
||||
"pad_token_id": null,
|
||||
"rms_norm_eps": 1e-05,
|
||||
"rope_parameters": null,
|
||||
"rope_theta": 100000000.0,
|
||||
"sliding_window": null,
|
||||
"tie_word_embeddings": false,
|
||||
"transformers_version": "5.12.1",
|
||||
"use_cache": true,
|
||||
"vocab_size": 131072
|
||||
}
|
||||
3
cover_image.png
Normal file
3
cover_image.png
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:c18725bf03ea3b952b1f5a9abfcee8afc18c07cd50d4697ba28c477618e7d27f
|
||||
size 10504492
|
||||
6
generation_config.json
Normal file
6
generation_config.json
Normal file
@@ -0,0 +1,6 @@
|
||||
{
|
||||
"_from_model_config": true,
|
||||
"bos_token_id": 1,
|
||||
"eos_token_id": 2,
|
||||
"transformers_version": "4.49.0.dev0"
|
||||
}
|
||||
3
model-00001-of-00032.safetensors
Normal file
3
model-00001-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:8c47db7a088b87d901372b29ca7199875ed7b7e9b53d8f43377a79399005d339
|
||||
size 1342177376
|
||||
3
model-00002-of-00032.safetensors
Normal file
3
model-00002-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:c6f5be7646ca5c679173c794c6d6f5da9a4749611e3f9f03df60966790c37100
|
||||
size 1677732176
|
||||
3
model-00003-of-00032.safetensors
Normal file
3
model-00003-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:b3dc12f2f1746f1eb0d9dfd1123e3fe2ed25659f07940f954da8f1d607e6d7fc
|
||||
size 1468028024
|
||||
3
model-00004-of-00032.safetensors
Normal file
3
model-00004-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:f3f6952590186b067bc2bb22e60833c3b1febe41f2d83656c4ac770ddd01a879
|
||||
size 1593878328
|
||||
3
model-00005-of-00032.safetensors
Normal file
3
model-00005-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:ce10a99144d6c2908bb500e50cb37f1e296cb558cd898895fec5e048a35b3b70
|
||||
size 1468028048
|
||||
3
model-00006-of-00032.safetensors
Normal file
3
model-00006-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:2b5d7c3402f8dd7268d4b2c091c9fa8a6fcccce62e5dac180b122e3aa5611c51
|
||||
size 1468028032
|
||||
3
model-00007-of-00032.safetensors
Normal file
3
model-00007-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:958d97accb4227dda0af18a8ac182a21357fd4ad46df4fc72a8d39e17ce40243
|
||||
size 1593878336
|
||||
3
model-00008-of-00032.safetensors
Normal file
3
model-00008-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:16a700e9abe25d41fd0166366a38deb62b56154855e483091603fa68a3df00ae
|
||||
size 1468028048
|
||||
3
model-00009-of-00032.safetensors
Normal file
3
model-00009-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:1a9dd1a9084c173bcc9e9f6433184c8d67be0de11babc0d1c7f2207dda157255
|
||||
size 1468028032
|
||||
3
model-00010-of-00032.safetensors
Normal file
3
model-00010-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:46384fa3d60f54a2e90a76a20b4f2b9999163d24c83bd347969654fb72cb05d0
|
||||
size 1593878336
|
||||
3
model-00011-of-00032.safetensors
Normal file
3
model-00011-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:61e135bab1a3bef168b39fabcf3cec1fb89bbdaa0a20097bf75facbc07a34a19
|
||||
size 1468028048
|
||||
3
model-00012-of-00032.safetensors
Normal file
3
model-00012-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:e6b895e59b0f13a00d25d89e4ea8ce2335dc9b980b90ab6c4d9925fe772ade0b
|
||||
size 1468028024
|
||||
3
model-00013-of-00032.safetensors
Normal file
3
model-00013-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:5ea6a96255f26ff12ffb53a7510d02119d4066fb13c160b12062bd4dc15b2a20
|
||||
size 1593878336
|
||||
3
model-00014-of-00032.safetensors
Normal file
3
model-00014-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:947edef8e07bdf2ec8d51e38196169fe35d80679eaf9b60900f02a2951868640
|
||||
size 1468028048
|
||||
3
model-00015-of-00032.safetensors
Normal file
3
model-00015-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:7815b69846f0b241efdaf003fa5780eb051cf05865ec85eebb8e4b8e6cc495e7
|
||||
size 1468028032
|
||||
3
model-00016-of-00032.safetensors
Normal file
3
model-00016-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:9f1833c0b69ea3755c61bbf27cd67a93fffab53fc486590fb428128c7706c6b9
|
||||
size 1593878336
|
||||
3
model-00017-of-00032.safetensors
Normal file
3
model-00017-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:62d00a41863f5aa85d7227a668a424433c69d416591ebc38d0b5113a30a90954
|
||||
size 1468028048
|
||||
3
model-00018-of-00032.safetensors
Normal file
3
model-00018-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:716cd100fda86631de424b281c4359c086961c2b3d58b3749be62c2a6c600538
|
||||
size 1468028032
|
||||
3
model-00019-of-00032.safetensors
Normal file
3
model-00019-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:9c323780bce99a3e6f5150d4dfb7aa87ca26de7452f5645e76a032526baf2440
|
||||
size 1593878336
|
||||
3
model-00020-of-00032.safetensors
Normal file
3
model-00020-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:dcaa5e1b1abe44bd435ba5614714fe3ff38c0381906a26d8313cfce0649a6b14
|
||||
size 1468028040
|
||||
3
model-00021-of-00032.safetensors
Normal file
3
model-00021-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:aaaf93fa4b8f1b015f30a635c1c394680c7fe555b8dd3af33ee7e068a4f2bdcb
|
||||
size 1468028032
|
||||
3
model-00022-of-00032.safetensors
Normal file
3
model-00022-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:6daf8368c4e90dfe3e87977e0d4630d1eb7ea929fee1bf20fa051489195e585f
|
||||
size 1593878336
|
||||
3
model-00023-of-00032.safetensors
Normal file
3
model-00023-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:d506b17c6c132caca71e4338556c8d686a3f934ccd876ae453a17fa0f816beb3
|
||||
size 1468028048
|
||||
3
model-00024-of-00032.safetensors
Normal file
3
model-00024-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:cfb7a528da69d28a323238e4401ec97fccc0886183a5d2abeb4cf981459fa693
|
||||
size 1468028032
|
||||
3
model-00025-of-00032.safetensors
Normal file
3
model-00025-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:88250492bd304203e53b07a52ee3856b64e7b665a777d268fe67d28c004292a0
|
||||
size 1593878336
|
||||
3
model-00026-of-00032.safetensors
Normal file
3
model-00026-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:4f9a4b08dd633a4e167351ca6e8a7d7d5c1a0eb369fe4629f22eee74ba5356b6
|
||||
size 1468028048
|
||||
3
model-00027-of-00032.safetensors
Normal file
3
model-00027-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:b8324bb2bad691a180e9a485c64c25915e4959b88d053150fdbb5969375c51ac
|
||||
size 1468028032
|
||||
3
model-00028-of-00032.safetensors
Normal file
3
model-00028-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:df38f8d0e6454a362bafe764033c45385a7a28afad22ab9e0da57fc0103ec5d5
|
||||
size 1593878328
|
||||
3
model-00029-of-00032.safetensors
Normal file
3
model-00029-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:762f3006cbe3a98d8df7df8c13d5592ceb175699d3bf49a4e8701ca1d3118405
|
||||
size 1468028040
|
||||
3
model-00030-of-00032.safetensors
Normal file
3
model-00030-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:e7710f13199307593a7d10d28d5666d857a59c56ec93721a7c0fb41e76d7c0f5
|
||||
size 1468028024
|
||||
3
model-00031-of-00032.safetensors
Normal file
3
model-00031-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:8080bf7226766308fb784858c9500df8789bf2433c9faa417312825f9b569805
|
||||
size 1593878320
|
||||
3
model-00032-of-00032.safetensors
Normal file
3
model-00032-of-00032.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:6e96a6ff68b84f2e747a209a8071206b96c022adc610962647d64904d6ceaa69
|
||||
size 1132483584
|
||||
370
model.safetensors.index.json
Normal file
370
model.safetensors.index.json
Normal file
@@ -0,0 +1,370 @@
|
||||
{
|
||||
"metadata": {
|
||||
"total_size": 47983667200
|
||||
},
|
||||
"weight_map": {
|
||||
"lm_head.weight": "model-00001-of-00032.safetensors",
|
||||
"model.embed_tokens.weight": "model-00002-of-00032.safetensors",
|
||||
"model.layers.0.input_layernorm.weight": "model-00002-of-00032.safetensors",
|
||||
"model.layers.0.mlp.down_proj.weight": "model-00002-of-00032.safetensors",
|
||||
"model.layers.0.mlp.gate_proj.weight": "model-00003-of-00032.safetensors",
|
||||
"model.layers.0.mlp.up_proj.weight": "model-00003-of-00032.safetensors",
|
||||
"model.layers.0.post_attention_layernorm.weight": "model-00003-of-00032.safetensors",
|
||||
"model.layers.0.self_attn.k_proj.weight": "model-00003-of-00032.safetensors",
|
||||
"model.layers.0.self_attn.o_proj.weight": "model-00003-of-00032.safetensors",
|
||||
"model.layers.0.self_attn.q_proj.weight": "model-00003-of-00032.safetensors",
|
||||
"model.layers.0.self_attn.v_proj.weight": "model-00003-of-00032.safetensors",
|
||||
"model.layers.1.input_layernorm.weight": "model-00003-of-00032.safetensors",
|
||||
"model.layers.1.mlp.down_proj.weight": "model-00003-of-00032.safetensors",
|
||||
"model.layers.1.mlp.gate_proj.weight": "model-00003-of-00032.safetensors",
|
||||
"model.layers.1.mlp.up_proj.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.1.post_attention_layernorm.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.1.self_attn.k_proj.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.1.self_attn.o_proj.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.1.self_attn.q_proj.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.1.self_attn.v_proj.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.10.input_layernorm.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.10.mlp.down_proj.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.10.mlp.gate_proj.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.10.mlp.up_proj.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.10.post_attention_layernorm.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.10.self_attn.k_proj.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.10.self_attn.o_proj.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.10.self_attn.q_proj.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.10.self_attn.v_proj.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.11.input_layernorm.weight": "model-00004-of-00032.safetensors",
|
||||
"model.layers.11.mlp.down_proj.weight": "model-00005-of-00032.safetensors",
|
||||
"model.layers.11.mlp.gate_proj.weight": "model-00005-of-00032.safetensors",
|
||||
"model.layers.11.mlp.up_proj.weight": "model-00005-of-00032.safetensors",
|
||||
"model.layers.11.post_attention_layernorm.weight": "model-00005-of-00032.safetensors",
|
||||
"model.layers.11.self_attn.k_proj.weight": "model-00005-of-00032.safetensors",
|
||||
"model.layers.11.self_attn.o_proj.weight": "model-00005-of-00032.safetensors",
|
||||
"model.layers.11.self_attn.q_proj.weight": "model-00005-of-00032.safetensors",
|
||||
"model.layers.11.self_attn.v_proj.weight": "model-00005-of-00032.safetensors",
|
||||
"model.layers.12.input_layernorm.weight": "model-00005-of-00032.safetensors",
|
||||
"model.layers.12.mlp.down_proj.weight": "model-00005-of-00032.safetensors",
|
||||
"model.layers.12.mlp.gate_proj.weight": "model-00006-of-00032.safetensors",
|
||||
"model.layers.12.mlp.up_proj.weight": "model-00006-of-00032.safetensors",
|
||||
"model.layers.12.post_attention_layernorm.weight": "model-00006-of-00032.safetensors",
|
||||
"model.layers.12.self_attn.k_proj.weight": "model-00006-of-00032.safetensors",
|
||||
"model.layers.12.self_attn.o_proj.weight": "model-00006-of-00032.safetensors",
|
||||
"model.layers.12.self_attn.q_proj.weight": "model-00006-of-00032.safetensors",
|
||||
"model.layers.12.self_attn.v_proj.weight": "model-00006-of-00032.safetensors",
|
||||
"model.layers.13.input_layernorm.weight": "model-00006-of-00032.safetensors",
|
||||
"model.layers.13.mlp.down_proj.weight": "model-00006-of-00032.safetensors",
|
||||
"model.layers.13.mlp.gate_proj.weight": "model-00006-of-00032.safetensors",
|
||||
"model.layers.13.mlp.up_proj.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.13.post_attention_layernorm.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.13.self_attn.k_proj.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.13.self_attn.o_proj.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.13.self_attn.q_proj.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.13.self_attn.v_proj.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.14.input_layernorm.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.14.mlp.down_proj.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.14.mlp.gate_proj.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.14.mlp.up_proj.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.14.post_attention_layernorm.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.14.self_attn.k_proj.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.14.self_attn.o_proj.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.14.self_attn.q_proj.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.14.self_attn.v_proj.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.15.input_layernorm.weight": "model-00007-of-00032.safetensors",
|
||||
"model.layers.15.mlp.down_proj.weight": "model-00008-of-00032.safetensors",
|
||||
"model.layers.15.mlp.gate_proj.weight": "model-00008-of-00032.safetensors",
|
||||
"model.layers.15.mlp.up_proj.weight": "model-00008-of-00032.safetensors",
|
||||
"model.layers.15.post_attention_layernorm.weight": "model-00008-of-00032.safetensors",
|
||||
"model.layers.15.self_attn.k_proj.weight": "model-00008-of-00032.safetensors",
|
||||
"model.layers.15.self_attn.o_proj.weight": "model-00008-of-00032.safetensors",
|
||||
"model.layers.15.self_attn.q_proj.weight": "model-00008-of-00032.safetensors",
|
||||
"model.layers.15.self_attn.v_proj.weight": "model-00008-of-00032.safetensors",
|
||||
"model.layers.16.input_layernorm.weight": "model-00008-of-00032.safetensors",
|
||||
"model.layers.16.mlp.down_proj.weight": "model-00008-of-00032.safetensors",
|
||||
"model.layers.16.mlp.gate_proj.weight": "model-00009-of-00032.safetensors",
|
||||
"model.layers.16.mlp.up_proj.weight": "model-00009-of-00032.safetensors",
|
||||
"model.layers.16.post_attention_layernorm.weight": "model-00009-of-00032.safetensors",
|
||||
"model.layers.16.self_attn.k_proj.weight": "model-00009-of-00032.safetensors",
|
||||
"model.layers.16.self_attn.o_proj.weight": "model-00009-of-00032.safetensors",
|
||||
"model.layers.16.self_attn.q_proj.weight": "model-00009-of-00032.safetensors",
|
||||
"model.layers.16.self_attn.v_proj.weight": "model-00009-of-00032.safetensors",
|
||||
"model.layers.17.input_layernorm.weight": "model-00009-of-00032.safetensors",
|
||||
"model.layers.17.mlp.down_proj.weight": "model-00009-of-00032.safetensors",
|
||||
"model.layers.17.mlp.gate_proj.weight": "model-00009-of-00032.safetensors",
|
||||
"model.layers.17.mlp.up_proj.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.17.post_attention_layernorm.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.17.self_attn.k_proj.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.17.self_attn.o_proj.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.17.self_attn.q_proj.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.17.self_attn.v_proj.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.18.input_layernorm.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.18.mlp.down_proj.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.18.mlp.gate_proj.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.18.mlp.up_proj.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.18.post_attention_layernorm.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.18.self_attn.k_proj.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.18.self_attn.o_proj.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.18.self_attn.q_proj.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.18.self_attn.v_proj.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.19.input_layernorm.weight": "model-00010-of-00032.safetensors",
|
||||
"model.layers.19.mlp.down_proj.weight": "model-00011-of-00032.safetensors",
|
||||
"model.layers.19.mlp.gate_proj.weight": "model-00011-of-00032.safetensors",
|
||||
"model.layers.19.mlp.up_proj.weight": "model-00011-of-00032.safetensors",
|
||||
"model.layers.19.post_attention_layernorm.weight": "model-00011-of-00032.safetensors",
|
||||
"model.layers.19.self_attn.k_proj.weight": "model-00011-of-00032.safetensors",
|
||||
"model.layers.19.self_attn.o_proj.weight": "model-00011-of-00032.safetensors",
|
||||
"model.layers.19.self_attn.q_proj.weight": "model-00011-of-00032.safetensors",
|
||||
"model.layers.19.self_attn.v_proj.weight": "model-00011-of-00032.safetensors",
|
||||
"model.layers.2.input_layernorm.weight": "model-00011-of-00032.safetensors",
|
||||
"model.layers.2.mlp.down_proj.weight": "model-00011-of-00032.safetensors",
|
||||
"model.layers.2.mlp.gate_proj.weight": "model-00012-of-00032.safetensors",
|
||||
"model.layers.2.mlp.up_proj.weight": "model-00012-of-00032.safetensors",
|
||||
"model.layers.2.post_attention_layernorm.weight": "model-00012-of-00032.safetensors",
|
||||
"model.layers.2.self_attn.k_proj.weight": "model-00012-of-00032.safetensors",
|
||||
"model.layers.2.self_attn.o_proj.weight": "model-00012-of-00032.safetensors",
|
||||
"model.layers.2.self_attn.q_proj.weight": "model-00012-of-00032.safetensors",
|
||||
"model.layers.2.self_attn.v_proj.weight": "model-00012-of-00032.safetensors",
|
||||
"model.layers.20.input_layernorm.weight": "model-00012-of-00032.safetensors",
|
||||
"model.layers.20.mlp.down_proj.weight": "model-00012-of-00032.safetensors",
|
||||
"model.layers.20.mlp.gate_proj.weight": "model-00012-of-00032.safetensors",
|
||||
"model.layers.20.mlp.up_proj.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.20.post_attention_layernorm.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.20.self_attn.k_proj.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.20.self_attn.o_proj.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.20.self_attn.q_proj.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.20.self_attn.v_proj.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.21.input_layernorm.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.21.mlp.down_proj.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.21.mlp.gate_proj.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.21.mlp.up_proj.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.21.post_attention_layernorm.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.21.self_attn.k_proj.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.21.self_attn.o_proj.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.21.self_attn.q_proj.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.21.self_attn.v_proj.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.22.input_layernorm.weight": "model-00013-of-00032.safetensors",
|
||||
"model.layers.22.mlp.down_proj.weight": "model-00014-of-00032.safetensors",
|
||||
"model.layers.22.mlp.gate_proj.weight": "model-00014-of-00032.safetensors",
|
||||
"model.layers.22.mlp.up_proj.weight": "model-00014-of-00032.safetensors",
|
||||
"model.layers.22.post_attention_layernorm.weight": "model-00014-of-00032.safetensors",
|
||||
"model.layers.22.self_attn.k_proj.weight": "model-00014-of-00032.safetensors",
|
||||
"model.layers.22.self_attn.o_proj.weight": "model-00014-of-00032.safetensors",
|
||||
"model.layers.22.self_attn.q_proj.weight": "model-00014-of-00032.safetensors",
|
||||
"model.layers.22.self_attn.v_proj.weight": "model-00014-of-00032.safetensors",
|
||||
"model.layers.23.input_layernorm.weight": "model-00014-of-00032.safetensors",
|
||||
"model.layers.23.mlp.down_proj.weight": "model-00014-of-00032.safetensors",
|
||||
"model.layers.23.mlp.gate_proj.weight": "model-00015-of-00032.safetensors",
|
||||
"model.layers.23.mlp.up_proj.weight": "model-00015-of-00032.safetensors",
|
||||
"model.layers.23.post_attention_layernorm.weight": "model-00015-of-00032.safetensors",
|
||||
"model.layers.23.self_attn.k_proj.weight": "model-00015-of-00032.safetensors",
|
||||
"model.layers.23.self_attn.o_proj.weight": "model-00015-of-00032.safetensors",
|
||||
"model.layers.23.self_attn.q_proj.weight": "model-00015-of-00032.safetensors",
|
||||
"model.layers.23.self_attn.v_proj.weight": "model-00015-of-00032.safetensors",
|
||||
"model.layers.24.input_layernorm.weight": "model-00015-of-00032.safetensors",
|
||||
"model.layers.24.mlp.down_proj.weight": "model-00015-of-00032.safetensors",
|
||||
"model.layers.24.mlp.gate_proj.weight": "model-00015-of-00032.safetensors",
|
||||
"model.layers.24.mlp.up_proj.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.24.post_attention_layernorm.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.24.self_attn.k_proj.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.24.self_attn.o_proj.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.24.self_attn.q_proj.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.24.self_attn.v_proj.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.25.input_layernorm.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.25.mlp.down_proj.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.25.mlp.gate_proj.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.25.mlp.up_proj.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.25.post_attention_layernorm.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.25.self_attn.k_proj.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.25.self_attn.o_proj.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.25.self_attn.q_proj.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.25.self_attn.v_proj.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.26.input_layernorm.weight": "model-00016-of-00032.safetensors",
|
||||
"model.layers.26.mlp.down_proj.weight": "model-00017-of-00032.safetensors",
|
||||
"model.layers.26.mlp.gate_proj.weight": "model-00017-of-00032.safetensors",
|
||||
"model.layers.26.mlp.up_proj.weight": "model-00017-of-00032.safetensors",
|
||||
"model.layers.26.post_attention_layernorm.weight": "model-00017-of-00032.safetensors",
|
||||
"model.layers.26.self_attn.k_proj.weight": "model-00017-of-00032.safetensors",
|
||||
"model.layers.26.self_attn.o_proj.weight": "model-00017-of-00032.safetensors",
|
||||
"model.layers.26.self_attn.q_proj.weight": "model-00017-of-00032.safetensors",
|
||||
"model.layers.26.self_attn.v_proj.weight": "model-00017-of-00032.safetensors",
|
||||
"model.layers.27.input_layernorm.weight": "model-00017-of-00032.safetensors",
|
||||
"model.layers.27.mlp.down_proj.weight": "model-00017-of-00032.safetensors",
|
||||
"model.layers.27.mlp.gate_proj.weight": "model-00018-of-00032.safetensors",
|
||||
"model.layers.27.mlp.up_proj.weight": "model-00018-of-00032.safetensors",
|
||||
"model.layers.27.post_attention_layernorm.weight": "model-00018-of-00032.safetensors",
|
||||
"model.layers.27.self_attn.k_proj.weight": "model-00018-of-00032.safetensors",
|
||||
"model.layers.27.self_attn.o_proj.weight": "model-00018-of-00032.safetensors",
|
||||
"model.layers.27.self_attn.q_proj.weight": "model-00018-of-00032.safetensors",
|
||||
"model.layers.27.self_attn.v_proj.weight": "model-00018-of-00032.safetensors",
|
||||
"model.layers.28.input_layernorm.weight": "model-00018-of-00032.safetensors",
|
||||
"model.layers.28.mlp.down_proj.weight": "model-00018-of-00032.safetensors",
|
||||
"model.layers.28.mlp.gate_proj.weight": "model-00018-of-00032.safetensors",
|
||||
"model.layers.28.mlp.up_proj.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.28.post_attention_layernorm.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.28.self_attn.k_proj.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.28.self_attn.o_proj.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.28.self_attn.q_proj.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.28.self_attn.v_proj.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.29.input_layernorm.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.29.mlp.down_proj.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.29.mlp.gate_proj.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.29.mlp.up_proj.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.29.post_attention_layernorm.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.29.self_attn.k_proj.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.29.self_attn.o_proj.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.29.self_attn.q_proj.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.29.self_attn.v_proj.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.3.input_layernorm.weight": "model-00019-of-00032.safetensors",
|
||||
"model.layers.3.mlp.down_proj.weight": "model-00020-of-00032.safetensors",
|
||||
"model.layers.3.mlp.gate_proj.weight": "model-00020-of-00032.safetensors",
|
||||
"model.layers.3.mlp.up_proj.weight": "model-00020-of-00032.safetensors",
|
||||
"model.layers.3.post_attention_layernorm.weight": "model-00020-of-00032.safetensors",
|
||||
"model.layers.3.self_attn.k_proj.weight": "model-00020-of-00032.safetensors",
|
||||
"model.layers.3.self_attn.o_proj.weight": "model-00020-of-00032.safetensors",
|
||||
"model.layers.3.self_attn.q_proj.weight": "model-00020-of-00032.safetensors",
|
||||
"model.layers.3.self_attn.v_proj.weight": "model-00020-of-00032.safetensors",
|
||||
"model.layers.30.input_layernorm.weight": "model-00020-of-00032.safetensors",
|
||||
"model.layers.30.mlp.down_proj.weight": "model-00020-of-00032.safetensors",
|
||||
"model.layers.30.mlp.gate_proj.weight": "model-00021-of-00032.safetensors",
|
||||
"model.layers.30.mlp.up_proj.weight": "model-00021-of-00032.safetensors",
|
||||
"model.layers.30.post_attention_layernorm.weight": "model-00021-of-00032.safetensors",
|
||||
"model.layers.30.self_attn.k_proj.weight": "model-00021-of-00032.safetensors",
|
||||
"model.layers.30.self_attn.o_proj.weight": "model-00021-of-00032.safetensors",
|
||||
"model.layers.30.self_attn.q_proj.weight": "model-00021-of-00032.safetensors",
|
||||
"model.layers.30.self_attn.v_proj.weight": "model-00021-of-00032.safetensors",
|
||||
"model.layers.31.input_layernorm.weight": "model-00021-of-00032.safetensors",
|
||||
"model.layers.31.mlp.down_proj.weight": "model-00021-of-00032.safetensors",
|
||||
"model.layers.31.mlp.gate_proj.weight": "model-00021-of-00032.safetensors",
|
||||
"model.layers.31.mlp.up_proj.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.31.post_attention_layernorm.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.31.self_attn.k_proj.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.31.self_attn.o_proj.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.31.self_attn.q_proj.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.31.self_attn.v_proj.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.32.input_layernorm.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.32.mlp.down_proj.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.32.mlp.gate_proj.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.32.mlp.up_proj.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.32.post_attention_layernorm.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.32.self_attn.k_proj.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.32.self_attn.o_proj.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.32.self_attn.q_proj.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.32.self_attn.v_proj.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.33.input_layernorm.weight": "model-00022-of-00032.safetensors",
|
||||
"model.layers.33.mlp.down_proj.weight": "model-00023-of-00032.safetensors",
|
||||
"model.layers.33.mlp.gate_proj.weight": "model-00023-of-00032.safetensors",
|
||||
"model.layers.33.mlp.up_proj.weight": "model-00023-of-00032.safetensors",
|
||||
"model.layers.33.post_attention_layernorm.weight": "model-00023-of-00032.safetensors",
|
||||
"model.layers.33.self_attn.k_proj.weight": "model-00023-of-00032.safetensors",
|
||||
"model.layers.33.self_attn.o_proj.weight": "model-00023-of-00032.safetensors",
|
||||
"model.layers.33.self_attn.q_proj.weight": "model-00023-of-00032.safetensors",
|
||||
"model.layers.33.self_attn.v_proj.weight": "model-00023-of-00032.safetensors",
|
||||
"model.layers.34.input_layernorm.weight": "model-00023-of-00032.safetensors",
|
||||
"model.layers.34.mlp.down_proj.weight": "model-00023-of-00032.safetensors",
|
||||
"model.layers.34.mlp.gate_proj.weight": "model-00024-of-00032.safetensors",
|
||||
"model.layers.34.mlp.up_proj.weight": "model-00024-of-00032.safetensors",
|
||||
"model.layers.34.post_attention_layernorm.weight": "model-00024-of-00032.safetensors",
|
||||
"model.layers.34.self_attn.k_proj.weight": "model-00024-of-00032.safetensors",
|
||||
"model.layers.34.self_attn.o_proj.weight": "model-00024-of-00032.safetensors",
|
||||
"model.layers.34.self_attn.q_proj.weight": "model-00024-of-00032.safetensors",
|
||||
"model.layers.34.self_attn.v_proj.weight": "model-00024-of-00032.safetensors",
|
||||
"model.layers.35.input_layernorm.weight": "model-00024-of-00032.safetensors",
|
||||
"model.layers.35.mlp.down_proj.weight": "model-00024-of-00032.safetensors",
|
||||
"model.layers.35.mlp.gate_proj.weight": "model-00024-of-00032.safetensors",
|
||||
"model.layers.35.mlp.up_proj.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.35.post_attention_layernorm.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.35.self_attn.k_proj.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.35.self_attn.o_proj.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.35.self_attn.q_proj.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.35.self_attn.v_proj.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.36.input_layernorm.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.36.mlp.down_proj.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.36.mlp.gate_proj.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.36.mlp.up_proj.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.36.post_attention_layernorm.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.36.self_attn.k_proj.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.36.self_attn.o_proj.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.36.self_attn.q_proj.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.36.self_attn.v_proj.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.37.input_layernorm.weight": "model-00025-of-00032.safetensors",
|
||||
"model.layers.37.mlp.down_proj.weight": "model-00026-of-00032.safetensors",
|
||||
"model.layers.37.mlp.gate_proj.weight": "model-00026-of-00032.safetensors",
|
||||
"model.layers.37.mlp.up_proj.weight": "model-00026-of-00032.safetensors",
|
||||
"model.layers.37.post_attention_layernorm.weight": "model-00026-of-00032.safetensors",
|
||||
"model.layers.37.self_attn.k_proj.weight": "model-00026-of-00032.safetensors",
|
||||
"model.layers.37.self_attn.o_proj.weight": "model-00026-of-00032.safetensors",
|
||||
"model.layers.37.self_attn.q_proj.weight": "model-00026-of-00032.safetensors",
|
||||
"model.layers.37.self_attn.v_proj.weight": "model-00026-of-00032.safetensors",
|
||||
"model.layers.38.input_layernorm.weight": "model-00026-of-00032.safetensors",
|
||||
"model.layers.38.mlp.down_proj.weight": "model-00026-of-00032.safetensors",
|
||||
"model.layers.38.mlp.gate_proj.weight": "model-00027-of-00032.safetensors",
|
||||
"model.layers.38.mlp.up_proj.weight": "model-00027-of-00032.safetensors",
|
||||
"model.layers.38.post_attention_layernorm.weight": "model-00027-of-00032.safetensors",
|
||||
"model.layers.38.self_attn.k_proj.weight": "model-00027-of-00032.safetensors",
|
||||
"model.layers.38.self_attn.o_proj.weight": "model-00027-of-00032.safetensors",
|
||||
"model.layers.38.self_attn.q_proj.weight": "model-00027-of-00032.safetensors",
|
||||
"model.layers.38.self_attn.v_proj.weight": "model-00027-of-00032.safetensors",
|
||||
"model.layers.39.input_layernorm.weight": "model-00027-of-00032.safetensors",
|
||||
"model.layers.39.mlp.down_proj.weight": "model-00027-of-00032.safetensors",
|
||||
"model.layers.39.mlp.gate_proj.weight": "model-00027-of-00032.safetensors",
|
||||
"model.layers.39.mlp.up_proj.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.39.post_attention_layernorm.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.39.self_attn.k_proj.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.39.self_attn.o_proj.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.39.self_attn.q_proj.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.39.self_attn.v_proj.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.4.input_layernorm.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.4.mlp.down_proj.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.4.mlp.gate_proj.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.4.mlp.up_proj.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.4.post_attention_layernorm.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.4.self_attn.k_proj.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.4.self_attn.o_proj.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.4.self_attn.q_proj.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.4.self_attn.v_proj.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.5.input_layernorm.weight": "model-00028-of-00032.safetensors",
|
||||
"model.layers.5.mlp.down_proj.weight": "model-00029-of-00032.safetensors",
|
||||
"model.layers.5.mlp.gate_proj.weight": "model-00029-of-00032.safetensors",
|
||||
"model.layers.5.mlp.up_proj.weight": "model-00029-of-00032.safetensors",
|
||||
"model.layers.5.post_attention_layernorm.weight": "model-00029-of-00032.safetensors",
|
||||
"model.layers.5.self_attn.k_proj.weight": "model-00029-of-00032.safetensors",
|
||||
"model.layers.5.self_attn.o_proj.weight": "model-00029-of-00032.safetensors",
|
||||
"model.layers.5.self_attn.q_proj.weight": "model-00029-of-00032.safetensors",
|
||||
"model.layers.5.self_attn.v_proj.weight": "model-00029-of-00032.safetensors",
|
||||
"model.layers.6.input_layernorm.weight": "model-00029-of-00032.safetensors",
|
||||
"model.layers.6.mlp.down_proj.weight": "model-00029-of-00032.safetensors",
|
||||
"model.layers.6.mlp.gate_proj.weight": "model-00030-of-00032.safetensors",
|
||||
"model.layers.6.mlp.up_proj.weight": "model-00030-of-00032.safetensors",
|
||||
"model.layers.6.post_attention_layernorm.weight": "model-00030-of-00032.safetensors",
|
||||
"model.layers.6.self_attn.k_proj.weight": "model-00030-of-00032.safetensors",
|
||||
"model.layers.6.self_attn.o_proj.weight": "model-00030-of-00032.safetensors",
|
||||
"model.layers.6.self_attn.q_proj.weight": "model-00030-of-00032.safetensors",
|
||||
"model.layers.6.self_attn.v_proj.weight": "model-00030-of-00032.safetensors",
|
||||
"model.layers.7.input_layernorm.weight": "model-00030-of-00032.safetensors",
|
||||
"model.layers.7.mlp.down_proj.weight": "model-00030-of-00032.safetensors",
|
||||
"model.layers.7.mlp.gate_proj.weight": "model-00030-of-00032.safetensors",
|
||||
"model.layers.7.mlp.up_proj.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.7.post_attention_layernorm.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.7.self_attn.k_proj.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.7.self_attn.o_proj.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.7.self_attn.q_proj.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.7.self_attn.v_proj.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.8.input_layernorm.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.8.mlp.down_proj.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.8.mlp.gate_proj.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.8.mlp.up_proj.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.8.post_attention_layernorm.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.8.self_attn.k_proj.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.8.self_attn.o_proj.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.8.self_attn.q_proj.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.8.self_attn.v_proj.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.9.input_layernorm.weight": "model-00031-of-00032.safetensors",
|
||||
"model.layers.9.mlp.down_proj.weight": "model-00032-of-00032.safetensors",
|
||||
"model.layers.9.mlp.gate_proj.weight": "model-00032-of-00032.safetensors",
|
||||
"model.layers.9.mlp.up_proj.weight": "model-00032-of-00032.safetensors",
|
||||
"model.layers.9.post_attention_layernorm.weight": "model-00032-of-00032.safetensors",
|
||||
"model.layers.9.self_attn.k_proj.weight": "model-00032-of-00032.safetensors",
|
||||
"model.layers.9.self_attn.o_proj.weight": "model-00032-of-00032.safetensors",
|
||||
"model.layers.9.self_attn.q_proj.weight": "model-00032-of-00032.safetensors",
|
||||
"model.layers.9.self_attn.v_proj.weight": "model-00032-of-00032.safetensors",
|
||||
"model.norm.weight": "model-00032-of-00032.safetensors"
|
||||
}
|
||||
}
|
||||
13
params.json
Normal file
13
params.json
Normal file
@@ -0,0 +1,13 @@
|
||||
{
|
||||
"dim": 5120,
|
||||
"n_layers": 40,
|
||||
"head_dim": 128,
|
||||
"hidden_dim": 32768,
|
||||
"n_heads": 32,
|
||||
"n_kv_heads": 8,
|
||||
"norm_eps": 1e-05,
|
||||
"vocab_size": 131072,
|
||||
"rope_theta": 100000000.0,
|
||||
"max_seq_len": 32768,
|
||||
"max_position_embeddings": 32768
|
||||
}
|
||||
7007
special_tokens_map.json
Normal file
7007
special_tokens_map.json
Normal file
File diff suppressed because it is too large
Load Diff
162
story_stage_generation.py
Normal file
162
story_stage_generation.py
Normal file
@@ -0,0 +1,162 @@
|
||||
"""Minimal staged-generation helpers for the exported 24b story model."""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
import re
|
||||
from typing import Any
|
||||
from jinja2 import Environment
|
||||
|
||||
|
||||
EOT_TOKEN = "<|eot_id|>"
|
||||
START_HEADER_TOKEN = "<|start_header_id|>"
|
||||
STOP_HEADER_TOKEN = "<|stop_header_id|>"
|
||||
PROFILE_NAME = "24b"
|
||||
CHAT_TEMPLATE = "{% for message in messages %}{% set role = message['role'] %}{% set content = message['content'] | default('', true) %}{% if role == 'prompt' %}<|start_header_id|>prompt<|stop_header_id|>\n\n{% elif role == 'book_preview' %}<|start_header_id|>book_preview<|stop_header_id|>\n\n{% elif role == 'book_plan' %}<|start_header_id|>book_plan<|stop_header_id|>\n\n{% elif role == 'first_chapter_plan' %}<|start_header_id|>first_chapter_plan<|stop_header_id|>\n\n{% elif role == 'first_chapter_scene_breakdown' %}<|start_header_id|>first_chapter_scene_breakdown<|stop_header_id|>\n\n{% elif role == 'first_chapter_text' %}<|start_header_id|>first_chapter_text<|stop_header_id|>\n\n{% elif role == 'full_book_chapters_plan' %}<|start_header_id|>full_book_chapters_plan<|stop_header_id|>\n\n{% elif role == 'book_characters_list' %}<|start_header_id|>book_characters_list<|stop_header_id|>\n\n{% elif role == 'scene_breakdown' %}<|start_header_id|>scene_breakdown<|stop_header_id|>\n\n{% elif role == 'chapter_text' %}<|start_header_id|>chapter_text<|stop_header_id|>\n\n{% else %}<|start_header_id|>{{ role }}<|stop_header_id|>\n\n{% endif %}{{ content }}{% if not loop.last or content %}<|eot_id|>{% endif %}{% endfor %}"
|
||||
DEFAULT_STAGE_SEQUENCE = ('book_preview', 'book_plan', 'first_chapter_plan', 'first_chapter_scene_breakdown', 'first_chapter_text')
|
||||
STAGE_ROLES = ('prompt', 'book_preview', 'book_plan', 'first_chapter_plan', 'first_chapter_scene_breakdown', 'first_chapter_text')
|
||||
|
||||
|
||||
def stage_header(role: str) -> str:
|
||||
if role not in STAGE_ROLES:
|
||||
raise ValueError(f"Unsupported story stage role: {role}")
|
||||
return f"{START_HEADER_TOKEN}{role}{STOP_HEADER_TOKEN}"
|
||||
|
||||
|
||||
BOOK_PREVIEW_REGEX = '^## Book Highlight\\n(?:\\S+\\s+){0,149}\\S+\\n\\n## Book Title\\n(?:[^\\s\\n]+\\s+){0,7}[^\\s\\n]+\\n\\n## Book Tags\\n(?:- [^\\r\\n]{1,75}\\n){7}## Book Archetype\\n(?:\\S+\\s+){0,199}\\S+\\n?$'
|
||||
BOOK_PLAN_REGEX = '^## World Rules\\n(?:- [^\\n]{1,300}\\n){1,90}\\n## Story Arcs\\n(?:### Arc [1-9][0-9]*: [^\\n]{1,120}\\n(?:- [^\\n]{1,300}\\n){1,20}\\n?){1,12}## Character Archetypes\\n(?:### [^\\n]{1,120}\\n(?:- [^\\n]{1,300}\\n){1,10}\\n?){1,32}## Writing Style\\n(?:- [^\\n]{1,300}\\n){0,39}- [^\\n]{1,300}\\n?$'
|
||||
FIRST_CHAPTER_PLAN_REGEX = '^(?:### [^\\n]{1,120}\\n\\*\\*Word Count:\\*\\* [0-9]{2,5}\\n\\*\\*Embedding Space:\\*\\* erotic: [0-9]{1,2}, action: [0-9]{1,2}, world_building: [0-9]{1,2}, exposition: [0-9]{1,2}, dialog: [0-9]{1,2}, romantic: [0-9]{1,2}, pacing: [0-9]{1,2}\\n\\*\\*Narrative Focuses:\\*\\* \\S(?:[^\\n;]{0,63}\\S)?(?:; \\S(?:[^\\n;]{0,63}\\S)?){0,6}\\n\\*\\*Chapter Summary:\\*\\*\\n(?:- [^\\n]{1,300}\\n){0,31}- [^\\n]{1,300}\\n?\\n?){1,2}$'
|
||||
FULL_BOOK_CHAPTERS_PLAN_REGEX = '^(?:### [^\\n]{1,120}\\n\\*\\*Word Count:\\*\\* [0-9]{2,5}\\n\\*\\*Embedding Space:\\*\\* erotic: [0-9]{1,2}, action: [0-9]{1,2}, world_building: [0-9]{1,2}, exposition: [0-9]{1,2}, dialog: [0-9]{1,2}, romantic: [0-9]{1,2}, pacing: [0-9]{1,2}\\n\\*\\*Narrative Focuses:\\*\\* \\S(?:[^\\n;]{0,63}\\S)?(?:; \\S(?:[^\\n;]{0,63}\\S)?){0,6}\\n\\*\\*Chapter Summary:\\*\\*\\n(?:- [^\\n]{1,300}\\n){0,79}- [^\\n]{1,300}\\n?\\n?){1,24}$'
|
||||
SCENE_BREAKDOWN_REGEX = '^### [^\\n]{1,120}\\n(?:#### Scene (?:[1-9]|1[0-2]): [^\\n]{1,120}\\n\\*\\*Word Count:\\*\\* [0-9]{2,5}\\n\\*\\*Embedding Space:\\*\\* action: [0-9]{1,2}, dialog: [0-9]{1,2}, world_building: [0-9]{1,2}, exposition: [0-9]{1,2}, romantic: [0-9]{1,2}, erotic: [0-9]{1,2}, pacing: [0-9]{1,2}\\n\\*\\*Narrative Focus:\\*\\* \\S(?:[^\\n;]{0,63}\\S)?(?:; \\S(?:[^\\n;]{0,63}\\S)?){0,6}\\n\\*\\*Narrative Perspective:\\*\\* [^\\n]{1,200}\\n\\*\\*Scene Summary:\\*\\*\\n(?:- [^\\n]{1,300}\\n){0,19}- [^\\n]{1,300}\\n?\\n?){1,12}$'
|
||||
BOOK_CHARACTERS_LIST_REGEX = '^### Main Characters\\n(?:#### [^\\n]{1,120}\\n(?:- [^\\n]{1,300}\\n){0,19}- [^\\n]{1,300}\\n?\\n?){1,20}### Side Characters\\n(?:#### [^\\n]{1,120}\\n(?:- [^\\n]{1,300}\\n){0,19}- [^\\n]{1,300}\\n?\\n?){1,20}$'
|
||||
|
||||
|
||||
def _book_plan_section(text: str, section_name: str) -> str:
|
||||
section_match = re.search(
|
||||
rf"(?ms)^## {re.escape(section_name)}\n(?P<body>.*?)(?=^## |\Z)",
|
||||
text,
|
||||
)
|
||||
if section_match is None:
|
||||
raise ValueError(f"Missing book_plan section: {section_name}")
|
||||
return section_match.group("body").strip()
|
||||
|
||||
|
||||
def extract_preloaded_chapters_info(book_plan_text: str) -> list[dict[str, str | int]]:
|
||||
chapter_names_section = _book_plan_section(book_plan_text, "Chapter Names")
|
||||
embedding_section = _book_plan_section(book_plan_text, "Chapters Embedding Space")
|
||||
word_count_section = _book_plan_section(book_plan_text, "Chapters Word Count")
|
||||
|
||||
chapter_names = [
|
||||
chapter_name_match.group(1).strip()
|
||||
for chapter_name_match in re.finditer(r"(?m)^- (.+?)\s*$", chapter_names_section)
|
||||
]
|
||||
if not chapter_names:
|
||||
raise ValueError("Could not extract preloaded chapter names from the 14b book_plan stage.")
|
||||
if len(set(chapter_names)) != len(chapter_names):
|
||||
raise ValueError("Duplicate chapter names are not supported in the 14b book_plan preload.")
|
||||
|
||||
chapter_embedding_space_by_name: dict[str, str] = {}
|
||||
for embedding_match in re.finditer(r"(?ms)^#### (.+?)\s*$\n([^\n]+)", embedding_section):
|
||||
chapter_name = embedding_match.group(1).strip()
|
||||
if chapter_name in chapter_embedding_space_by_name:
|
||||
raise ValueError(f"Duplicate embedding-space preload for chapter {chapter_name!r}.")
|
||||
chapter_embedding_space_by_name[chapter_name] = embedding_match.group(2).strip()
|
||||
|
||||
chapter_word_count_by_name: dict[str, int] = {}
|
||||
for word_count_match in re.finditer(r"(?ms)^#### (.+?)\s*$\n([0-9]{2,5})", word_count_section):
|
||||
chapter_name = word_count_match.group(1).strip()
|
||||
if chapter_name in chapter_word_count_by_name:
|
||||
raise ValueError(f"Duplicate word-count preload for chapter {chapter_name!r}.")
|
||||
chapter_word_count_by_name[chapter_name] = int(word_count_match.group(2))
|
||||
|
||||
chapters_info: list[dict[str, str | int]] = []
|
||||
for chapter_name in chapter_names:
|
||||
if chapter_name not in chapter_embedding_space_by_name:
|
||||
raise ValueError(f"Missing embedding-space preload for chapter {chapter_name!r}.")
|
||||
if chapter_name not in chapter_word_count_by_name:
|
||||
raise ValueError(f"Missing word-count preload for chapter {chapter_name!r}.")
|
||||
chapters_info.append(
|
||||
{
|
||||
"chapter_name": chapter_name,
|
||||
"chapter_word_count": chapter_word_count_by_name[chapter_name],
|
||||
"chapter_embedding_space": chapter_embedding_space_by_name[chapter_name],
|
||||
}
|
||||
)
|
||||
return chapters_info
|
||||
|
||||
|
||||
def build_full_book_chapters_plan_regex_from_book_plan(book_plan_text: str) -> str:
|
||||
chapters_info = extract_preloaded_chapters_info(book_plan_text)
|
||||
if len(chapters_info) < 2:
|
||||
raise ValueError("14b book_plan must preload the first chapter and at least one later chapter.")
|
||||
|
||||
later_chapter_patterns: list[str] = []
|
||||
for chapter_info in chapters_info[1:]:
|
||||
later_chapter_patterns.append(
|
||||
r"### "
|
||||
+ re.escape(str(chapter_info["chapter_name"]))
|
||||
+ r"\n"
|
||||
r"\*\*Word Count:\*\* "
|
||||
+ re.escape(str(chapter_info["chapter_word_count"]))
|
||||
+ r"\n"
|
||||
r"\*\*Embedding Space:\*\* "
|
||||
+ re.escape(str(chapter_info["chapter_embedding_space"]))
|
||||
+ r"\n"
|
||||
r"\*\*Narrative Focuses:\*\* "
|
||||
+ '\\S(?:[^\\n;]{0,63}\\S)?(?:; \\S(?:[^\\n;]{0,63}\\S)?){0,6}'
|
||||
+ r"\n"
|
||||
r"\*\*Chapter Summary:\*\*\n"
|
||||
+ '(?:- [^\\n]{1,300}\\n){0,79}- [^\\n]{1,300}\\n?'
|
||||
)
|
||||
return r"^" + r"\n\n".join(later_chapter_patterns) + r"$"
|
||||
|
||||
STAGE_STRUCTURED_OUTPUTS = {
|
||||
"book_preview": {"regex": BOOK_PREVIEW_REGEX},
|
||||
"book_plan": {"regex": BOOK_PLAN_REGEX},
|
||||
"first_chapter_plan": {"regex": FIRST_CHAPTER_PLAN_REGEX},
|
||||
"first_chapter_scene_breakdown": {"regex": SCENE_BREAKDOWN_REGEX},
|
||||
}
|
||||
|
||||
|
||||
def stage_structured_outputs(role: str) -> dict[str, Any] | None:
|
||||
structured_outputs = STAGE_STRUCTURED_OUTPUTS.get(role)
|
||||
if structured_outputs is None:
|
||||
return None
|
||||
return dict(structured_outputs)
|
||||
|
||||
|
||||
def build_stage_messages(
|
||||
prompt: str,
|
||||
completed_stages: list[tuple[str, str]] | None,
|
||||
next_stage_role: str,
|
||||
) -> list[dict[str, str]]:
|
||||
if next_stage_role == "prompt":
|
||||
raise ValueError("prompt cannot be used as a generation stage")
|
||||
if next_stage_role not in STAGE_ROLES:
|
||||
raise ValueError(f"Unsupported story stage role: {next_stage_role}")
|
||||
messages = [{"role": "prompt", "content": prompt}]
|
||||
for role, content in completed_stages or []:
|
||||
if role == "prompt":
|
||||
raise ValueError("completed_stages must not include prompt")
|
||||
if role not in STAGE_ROLES:
|
||||
raise ValueError(f"Unsupported completed stage role: {role}")
|
||||
messages.append({"role": role, "content": content})
|
||||
messages.append({"role": next_stage_role, "content": ""})
|
||||
return messages
|
||||
|
||||
|
||||
def render_stage_prompt(
|
||||
tokenizer: Any,
|
||||
prompt: str,
|
||||
completed_stages: list[tuple[str, str]] | None,
|
||||
next_stage_role: str,
|
||||
) -> str:
|
||||
messages = build_stage_messages(prompt, completed_stages, next_stage_role)
|
||||
rendered = Environment().from_string(CHAT_TEMPLATE).render(messages=messages)
|
||||
if not isinstance(rendered, str):
|
||||
raise TypeError("Stage chat template rendering must return a string")
|
||||
return rendered
|
||||
|
||||
|
||||
def trim_generated_stage_text(text: str, generated_stage_role: str) -> str:
|
||||
del generated_stage_role
|
||||
return text.split(EOT_TOKEN, 1)[0].rstrip()
|
||||
3
tekken.json
Normal file
3
tekken.json
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:c4b90a968dbc67ef3975129d0b78a2e3cbb6bea340ab9205f22e8a0308b1ffc5
|
||||
size 14801223
|
||||
3
tokenizer.json
Normal file
3
tokenizer.json
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:57b42b226e2726c538edc62e5687ac5cb7518c7d880c6b8b590217c99574147c
|
||||
size 29906843
|
||||
9020
tokenizer_config.json
Normal file
9020
tokenizer_config.json
Normal file
File diff suppressed because it is too large
Load Diff
Reference in New Issue
Block a user