Add MKL layer3 S0 and S2 and temporal-merge baseline S2 checkpoints
Browse files- .gitattributes +2 -0
- ckpts/S0/qwen3_5_2b/v14_1_mkl_layer3/step005258/config.json +21 -0
- ckpts/S0/qwen3_5_2b/v14_1_mkl_layer3/step005258/export_manifest.json +70 -0
- ckpts/S0/qwen3_5_2b/v14_1_mkl_layer3/step005258/mp_rank_00_model_states.pt +3 -0
- ckpts/S0/qwen3_5_2b/v14_1_mkl_layer3/step005258/trainer_state.json +6 -0
- ckpts/S0/qwen3_5_2b/v14_1_mkl_layer3/step005258/v12_vision_model_states.pt +3 -0
- ckpts/S0/qwen3_5_2b/v14_1_mkl_layer3/step005258/vtc_compressor_model_states.pt +3 -0
- ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/chat_template.jinja +61 -0
- ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/chat_template.json +3 -0
- ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/config.json +286 -0
- ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/generation_config.json +12 -0
- ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/model-00001-of-00002.safetensors +3 -0
- ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/model-00002-of-00002.safetensors +3 -0
- ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/model.safetensors.index.json +705 -0
- ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/modeling_qwen3_5vit_qwen3.py +0 -0
- ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/processing_qwen3_5vit_qwen3.py +302 -0
- ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/processor_config.json +209 -0
- ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/tokenizer.json +3 -0
- ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/tokenizer_config.json +19 -0
- ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/trainer_state.json +0 -0
- ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/chat_template.jinja +61 -0
- ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/chat_template.json +3 -0
- ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/config.json +290 -0
- ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/generation_config.json +12 -0
- ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/model-00001-of-00002.safetensors +3 -0
- ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/model-00002-of-00002.safetensors +3 -0
- ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/model.safetensors.index.json +720 -0
- ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/modeling_qwen3_5vit_qwen3.py +0 -0
- ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/processing_qwen3_5vit_qwen3.py +302 -0
- ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/processor_config.json +209 -0
- ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/tokenizer.json +3 -0
- ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/tokenizer_config.json +19 -0
- ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/trainer_state.json +2490 -0
.gitattributes
CHANGED
|
@@ -91,3 +91,5 @@ ckpts/S2/4b/qwen3_5_2b/v13_2/v1-20260817-033406/checkpoint-1530/tokenizer.json f
|
|
| 91 |
ckpts/S2/4b/qwen3_5_2b/v13_3/v0-20260817-190428/checkpoint-1530/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 92 |
ckpts/S2/4b/qwen3_5_2b/v13_5/v0-20260821-172306/checkpoint-1530/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 93 |
ckpts/S2/4b/qwen3_5_2b/v13_6/v0-20260820-192228/checkpoint-1400/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
| 91 |
ckpts/S2/4b/qwen3_5_2b/v13_3/v0-20260817-190428/checkpoint-1530/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 92 |
ckpts/S2/4b/qwen3_5_2b/v13_5/v0-20260821-172306/checkpoint-1530/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 93 |
ckpts/S2/4b/qwen3_5_2b/v13_6/v0-20260820-192228/checkpoint-1400/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 94 |
+
ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
| 95 |
+
ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
ckpts/S0/qwen3_5_2b/v14_1_mkl_layer3/step005258/config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"deepstack_visual_indexes": [],
|
| 3 |
+
"depth": 24,
|
| 4 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 5 |
+
"hidden_size": 1024,
|
| 6 |
+
"in_channels": 3,
|
| 7 |
+
"initializer_range": 0.02,
|
| 8 |
+
"intermediate_size": 4096,
|
| 9 |
+
"model_type": "qwen3_5",
|
| 10 |
+
"num_heads": 16,
|
| 11 |
+
"num_position_embeddings": 2304,
|
| 12 |
+
"out_hidden_size": 2048,
|
| 13 |
+
"patch_size": 16,
|
| 14 |
+
"spatial_merge_size": 2,
|
| 15 |
+
"temporal_patch_size": 2,
|
| 16 |
+
"transformers_version": "5.5.4",
|
| 17 |
+
"vtc_compressor_type": "vtc_x2",
|
| 18 |
+
"vtc_compressor_variant": "baseline",
|
| 19 |
+
"vtc_feature_fusion_layers": [],
|
| 20 |
+
"vtc_temporal_attention_window": 2
|
| 21 |
+
}
|
ckpts/S0/qwen3_5_2b/v14_1_mkl_layer3/step005258/export_manifest.json
ADDED
|
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architecture": {
|
| 3 |
+
"compressor": "vtc_x2",
|
| 4 |
+
"decoder_bottleneck_size": 704,
|
| 5 |
+
"decoder_temporal_attention_window": 2,
|
| 6 |
+
"image_route": "duplicate_two_then_compress_one",
|
| 7 |
+
"placement": "post_final_layernorm_pre_projector",
|
| 8 |
+
"spatial_attention": true,
|
| 9 |
+
"temporal_attention": true,
|
| 10 |
+
"temporal_attention_window": 2,
|
| 11 |
+
"temporal_downsample": "conv1d_stride2",
|
| 12 |
+
"temporal_group_size": 2,
|
| 13 |
+
"temporal_position": "learned_local_two_slot",
|
| 14 |
+
"variant": "baseline",
|
| 15 |
+
"vtc_feature_fusion_layers": []
|
| 16 |
+
},
|
| 17 |
+
"downstream": {
|
| 18 |
+
"native_merger_mlp_discarded": true,
|
| 19 |
+
"native_merger_norm_retained_as_final_layernorm": true,
|
| 20 |
+
"preferred_model_type": "llava_qwen3_5vit_qwen3",
|
| 21 |
+
"projector": "standard_llava_ov_mlp_reinitialized_in_s1",
|
| 22 |
+
"reconstruction_decoder_discarded": true,
|
| 23 |
+
"train_frames": 64,
|
| 24 |
+
"video_mode": "vtc_compression",
|
| 25 |
+
"visual_units": 32
|
| 26 |
+
},
|
| 27 |
+
"export": {
|
| 28 |
+
"dropped": {
|
| 29 |
+
"merger_mlp": 4,
|
| 30 |
+
"non_visual": 355
|
| 31 |
+
},
|
| 32 |
+
"exported_keys": 293,
|
| 33 |
+
"final_layernorm_from": "student.model.visual.merger.norm",
|
| 34 |
+
"retained": {
|
| 35 |
+
"merger_norm_as_final_layernorm": 2
|
| 36 |
+
}
|
| 37 |
+
},
|
| 38 |
+
"freeze": {
|
| 39 |
+
"language_model": true,
|
| 40 |
+
"lm_head": true,
|
| 41 |
+
"native_merger": true,
|
| 42 |
+
"patch_embed": true,
|
| 43 |
+
"pos_embed": true,
|
| 44 |
+
"trainable": [
|
| 45 |
+
"model.visual.blocks.*",
|
| 46 |
+
"vtc_compressor.*",
|
| 47 |
+
"reconstruction_decoder.*"
|
| 48 |
+
]
|
| 49 |
+
},
|
| 50 |
+
"loss": {
|
| 51 |
+
"image": "v12_4_llm_semantic_change_hidden_fusion_forward_kl",
|
| 52 |
+
"image_weight": 0.5,
|
| 53 |
+
"video": "vtc_compress_reconstruct_full_vocab_forward_kl",
|
| 54 |
+
"video_compressed_units": 16,
|
| 55 |
+
"video_direct_target": "teacher_vit_temporal_mean_llm",
|
| 56 |
+
"video_direct_weight": 0.0,
|
| 57 |
+
"video_input_mode": "repeat2",
|
| 58 |
+
"video_llm_relation": "full_token_cosine_matrix_mse",
|
| 59 |
+
"video_llm_relation_weight": 0.0,
|
| 60 |
+
"video_reconstructed_units": 32,
|
| 61 |
+
"video_teacher_frames": 32,
|
| 62 |
+
"video_weight": 0.5
|
| 63 |
+
},
|
| 64 |
+
"resolution": 384,
|
| 65 |
+
"schema": "visionencoder.v14_1_mkl",
|
| 66 |
+
"source_model": "/share/m2v_intern_v3/wangjunjie09/model_cache/huggingface/Qwen/Qwen3.5-2B",
|
| 67 |
+
"step": 5258,
|
| 68 |
+
"version": 1,
|
| 69 |
+
"vision_source_id": "qwen3_5_2b"
|
| 70 |
+
}
|
ckpts/S0/qwen3_5_2b/v14_1_mkl_layer3/step005258/mp_rank_00_model_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2ba66fc267c59f2212d6ec8c311047e8fde7d75f8ae92165998ac1c0392ac42a
|
| 3 |
+
size 4460548929
|
ckpts/S0/qwen3_5_2b/v14_1_mkl_layer3/step005258/trainer_state.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"batch_in_epoch": 0,
|
| 3 |
+
"epoch": 1,
|
| 4 |
+
"global_step": 5258,
|
| 5 |
+
"max_steps": 5258
|
| 6 |
+
}
|
ckpts/S0/qwen3_5_2b/v14_1_mkl_layer3/step005258/v12_vision_model_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2b7a2e8c5139162cc0dc5a37e51af74f7c0fc769fcbfca91590cfb24fbb2a650
|
| 3 |
+
size 612522145
|
ckpts/S0/qwen3_5_2b/v14_1_mkl_layer3/step005258/vtc_compressor_model_states.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0bde1f903958ce742a72bf376007c4388c5387323f038a27d126a21c362c35a9
|
| 3 |
+
size 33826053
|
ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/chat_template.jinja
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0].role == 'system' %}
|
| 4 |
+
{{- messages[0].content + '\n\n' }}
|
| 5 |
+
{%- endif %}
|
| 6 |
+
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
| 7 |
+
{%- for tool in tools %}
|
| 8 |
+
{{- "\n" }}
|
| 9 |
+
{{- tool | tojson }}
|
| 10 |
+
{%- endfor %}
|
| 11 |
+
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
| 12 |
+
{%- else %}
|
| 13 |
+
{%- if messages[0].role == 'system' %}
|
| 14 |
+
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
| 15 |
+
{%- endif %}
|
| 16 |
+
{%- endif %}
|
| 17 |
+
{%- for message in messages %}
|
| 18 |
+
{%- if message.content is string %}
|
| 19 |
+
{%- set content = message.content %}
|
| 20 |
+
{%- else %}
|
| 21 |
+
{%- set content = '' %}
|
| 22 |
+
{%- endif %}
|
| 23 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
| 24 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 25 |
+
{%- elif message.role == "assistant" %}
|
| 26 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 27 |
+
{%- if message.tool_calls %}
|
| 28 |
+
{%- for tool_call in message.tool_calls %}
|
| 29 |
+
{%- if (loop.first and content) or (not loop.first) %}
|
| 30 |
+
{{- '\n' }}
|
| 31 |
+
{%- endif %}
|
| 32 |
+
{%- if tool_call.function %}
|
| 33 |
+
{%- set tool_call = tool_call.function %}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{{- '<tool_call>\n{"name": "' }}
|
| 36 |
+
{{- tool_call.name }}
|
| 37 |
+
{{- '", "arguments": ' }}
|
| 38 |
+
{%- if tool_call.arguments is string %}
|
| 39 |
+
{{- tool_call.arguments }}
|
| 40 |
+
{%- else %}
|
| 41 |
+
{{- tool_call.arguments | tojson }}
|
| 42 |
+
{%- endif %}
|
| 43 |
+
{{- '}\n</tool_call>' }}
|
| 44 |
+
{%- endfor %}
|
| 45 |
+
{%- endif %}
|
| 46 |
+
{{- '<|im_end|>\n' }}
|
| 47 |
+
{%- elif message.role == "tool" %}
|
| 48 |
+
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
| 49 |
+
{{- '<|im_start|>user' }}
|
| 50 |
+
{%- endif %}
|
| 51 |
+
{{- '\n<tool_response>\n' }}
|
| 52 |
+
{{- content }}
|
| 53 |
+
{{- '\n</tool_response>' }}
|
| 54 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 55 |
+
{{- '<|im_end|>\n' }}
|
| 56 |
+
{%- endif %}
|
| 57 |
+
{%- endif %}
|
| 58 |
+
{%- endfor %}
|
| 59 |
+
{%- if add_generation_prompt %}
|
| 60 |
+
{{- '<|im_start|>assistant\n' }}
|
| 61 |
+
{%- endif %}
|
ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/chat_template.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + ' '}}{# Render all images first #}{% for content in message['content'] | selectattr('type', 'equalto', 'image') %}{{ '<image>' }}{% endfor %}{# Render all video then #}{% for content in message['content'] | selectattr('type', 'equalto', 'video') %}{{ '<video>' }}{% endfor %}{# Render all text next #}{% if message['role'] != 'assistant' %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{{ '\n' + content['text'] }}{% endfor %}{% else %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{% generation %}{{ '\n' + content['text'] }}{% endgeneration %}{% endfor %}{% endif %}{{'<|im_end|>'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}"
|
| 3 |
+
}
|
ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/config.json
ADDED
|
@@ -0,0 +1,286 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlavaQwen3_5ViTForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"bos_token_id": null,
|
| 6 |
+
"dtype": "bfloat16",
|
| 7 |
+
"eos_token_id": 151645,
|
| 8 |
+
"eval_num_frames": 64,
|
| 9 |
+
"hidden_size": 2560,
|
| 10 |
+
"image_grid_pinpoints": [
|
| 11 |
+
[
|
| 12 |
+
384,
|
| 13 |
+
384
|
| 14 |
+
],
|
| 15 |
+
[
|
| 16 |
+
384,
|
| 17 |
+
768
|
| 18 |
+
],
|
| 19 |
+
[
|
| 20 |
+
384,
|
| 21 |
+
1152
|
| 22 |
+
],
|
| 23 |
+
[
|
| 24 |
+
384,
|
| 25 |
+
1536
|
| 26 |
+
],
|
| 27 |
+
[
|
| 28 |
+
384,
|
| 29 |
+
1920
|
| 30 |
+
],
|
| 31 |
+
[
|
| 32 |
+
384,
|
| 33 |
+
2304
|
| 34 |
+
],
|
| 35 |
+
[
|
| 36 |
+
768,
|
| 37 |
+
384
|
| 38 |
+
],
|
| 39 |
+
[
|
| 40 |
+
768,
|
| 41 |
+
768
|
| 42 |
+
],
|
| 43 |
+
[
|
| 44 |
+
768,
|
| 45 |
+
1152
|
| 46 |
+
],
|
| 47 |
+
[
|
| 48 |
+
768,
|
| 49 |
+
1536
|
| 50 |
+
],
|
| 51 |
+
[
|
| 52 |
+
768,
|
| 53 |
+
1920
|
| 54 |
+
],
|
| 55 |
+
[
|
| 56 |
+
768,
|
| 57 |
+
2304
|
| 58 |
+
],
|
| 59 |
+
[
|
| 60 |
+
1152,
|
| 61 |
+
384
|
| 62 |
+
],
|
| 63 |
+
[
|
| 64 |
+
1152,
|
| 65 |
+
768
|
| 66 |
+
],
|
| 67 |
+
[
|
| 68 |
+
1152,
|
| 69 |
+
1152
|
| 70 |
+
],
|
| 71 |
+
[
|
| 72 |
+
1152,
|
| 73 |
+
1536
|
| 74 |
+
],
|
| 75 |
+
[
|
| 76 |
+
1152,
|
| 77 |
+
1920
|
| 78 |
+
],
|
| 79 |
+
[
|
| 80 |
+
1152,
|
| 81 |
+
2304
|
| 82 |
+
],
|
| 83 |
+
[
|
| 84 |
+
1536,
|
| 85 |
+
384
|
| 86 |
+
],
|
| 87 |
+
[
|
| 88 |
+
1536,
|
| 89 |
+
768
|
| 90 |
+
],
|
| 91 |
+
[
|
| 92 |
+
1536,
|
| 93 |
+
1152
|
| 94 |
+
],
|
| 95 |
+
[
|
| 96 |
+
1536,
|
| 97 |
+
1536
|
| 98 |
+
],
|
| 99 |
+
[
|
| 100 |
+
1536,
|
| 101 |
+
1920
|
| 102 |
+
],
|
| 103 |
+
[
|
| 104 |
+
1536,
|
| 105 |
+
2304
|
| 106 |
+
],
|
| 107 |
+
[
|
| 108 |
+
1920,
|
| 109 |
+
384
|
| 110 |
+
],
|
| 111 |
+
[
|
| 112 |
+
1920,
|
| 113 |
+
768
|
| 114 |
+
],
|
| 115 |
+
[
|
| 116 |
+
1920,
|
| 117 |
+
1152
|
| 118 |
+
],
|
| 119 |
+
[
|
| 120 |
+
1920,
|
| 121 |
+
1536
|
| 122 |
+
],
|
| 123 |
+
[
|
| 124 |
+
1920,
|
| 125 |
+
1920
|
| 126 |
+
],
|
| 127 |
+
[
|
| 128 |
+
1920,
|
| 129 |
+
2304
|
| 130 |
+
],
|
| 131 |
+
[
|
| 132 |
+
2304,
|
| 133 |
+
384
|
| 134 |
+
],
|
| 135 |
+
[
|
| 136 |
+
2304,
|
| 137 |
+
768
|
| 138 |
+
],
|
| 139 |
+
[
|
| 140 |
+
2304,
|
| 141 |
+
1152
|
| 142 |
+
],
|
| 143 |
+
[
|
| 144 |
+
2304,
|
| 145 |
+
1536
|
| 146 |
+
],
|
| 147 |
+
[
|
| 148 |
+
2304,
|
| 149 |
+
1920
|
| 150 |
+
],
|
| 151 |
+
[
|
| 152 |
+
2304,
|
| 153 |
+
2304
|
| 154 |
+
]
|
| 155 |
+
],
|
| 156 |
+
"image_token_index": 151669,
|
| 157 |
+
"model_type": "llava_qwen3_5vit_qwen3",
|
| 158 |
+
"multimodal_projector_bias": true,
|
| 159 |
+
"pad_token_id": 151643,
|
| 160 |
+
"pair_fusion_clip_selection": "",
|
| 161 |
+
"pair_fusion_lr": 0.0001,
|
| 162 |
+
"pair_fusion_motion_cache_format": "",
|
| 163 |
+
"pair_fusion_motion_cache_version": 0,
|
| 164 |
+
"pair_fusion_motion_record_count": 0,
|
| 165 |
+
"pair_fusion_pretraining": "",
|
| 166 |
+
"pair_fusion_pretraining_step": 0,
|
| 167 |
+
"projector_hidden_act": "gelu",
|
| 168 |
+
"shared_image_temporal_refiner": false,
|
| 169 |
+
"temporal_refiner_lr": 0.0001,
|
| 170 |
+
"temporal_refiner_variant": "attention_only",
|
| 171 |
+
"temporal_refiner_window_groups": 4,
|
| 172 |
+
"text_config": {
|
| 173 |
+
"_name_or_path": "/share/m2v_intern_v3/wangjunjie09/model_cache/huggingface/Qwen/Qwen3-4B-Instruct-2507",
|
| 174 |
+
"architectures": [
|
| 175 |
+
"Qwen3ForCausalLM"
|
| 176 |
+
],
|
| 177 |
+
"attention_bias": false,
|
| 178 |
+
"attention_dropout": 0.0,
|
| 179 |
+
"bos_token_id": 151643,
|
| 180 |
+
"dtype": "bfloat16",
|
| 181 |
+
"eos_token_id": 151645,
|
| 182 |
+
"head_dim": 128,
|
| 183 |
+
"hidden_act": "silu",
|
| 184 |
+
"hidden_size": 2560,
|
| 185 |
+
"initializer_range": 0.02,
|
| 186 |
+
"intermediate_size": 9728,
|
| 187 |
+
"layer_types": [
|
| 188 |
+
"full_attention",
|
| 189 |
+
"full_attention",
|
| 190 |
+
"full_attention",
|
| 191 |
+
"full_attention",
|
| 192 |
+
"full_attention",
|
| 193 |
+
"full_attention",
|
| 194 |
+
"full_attention",
|
| 195 |
+
"full_attention",
|
| 196 |
+
"full_attention",
|
| 197 |
+
"full_attention",
|
| 198 |
+
"full_attention",
|
| 199 |
+
"full_attention",
|
| 200 |
+
"full_attention",
|
| 201 |
+
"full_attention",
|
| 202 |
+
"full_attention",
|
| 203 |
+
"full_attention",
|
| 204 |
+
"full_attention",
|
| 205 |
+
"full_attention",
|
| 206 |
+
"full_attention",
|
| 207 |
+
"full_attention",
|
| 208 |
+
"full_attention",
|
| 209 |
+
"full_attention",
|
| 210 |
+
"full_attention",
|
| 211 |
+
"full_attention",
|
| 212 |
+
"full_attention",
|
| 213 |
+
"full_attention",
|
| 214 |
+
"full_attention",
|
| 215 |
+
"full_attention",
|
| 216 |
+
"full_attention",
|
| 217 |
+
"full_attention",
|
| 218 |
+
"full_attention",
|
| 219 |
+
"full_attention",
|
| 220 |
+
"full_attention",
|
| 221 |
+
"full_attention",
|
| 222 |
+
"full_attention",
|
| 223 |
+
"full_attention"
|
| 224 |
+
],
|
| 225 |
+
"max_position_embeddings": 262144,
|
| 226 |
+
"max_window_layers": 36,
|
| 227 |
+
"model_type": "qwen3",
|
| 228 |
+
"num_attention_heads": 32,
|
| 229 |
+
"num_hidden_layers": 36,
|
| 230 |
+
"num_key_value_heads": 8,
|
| 231 |
+
"pad_token_id": 151643,
|
| 232 |
+
"rms_norm_eps": 1e-06,
|
| 233 |
+
"rope_parameters": {
|
| 234 |
+
"rope_theta": 5000000,
|
| 235 |
+
"rope_type": "default"
|
| 236 |
+
},
|
| 237 |
+
"sliding_window": null,
|
| 238 |
+
"tie_word_embeddings": true,
|
| 239 |
+
"use_cache": false,
|
| 240 |
+
"use_sliding_window": false,
|
| 241 |
+
"vocab_size": 151936
|
| 242 |
+
},
|
| 243 |
+
"tie_word_embeddings": true,
|
| 244 |
+
"train_num_frames": 64,
|
| 245 |
+
"transformers_version": "5.5.4",
|
| 246 |
+
"use_cache": false,
|
| 247 |
+
"video_mode": "temporal_merge",
|
| 248 |
+
"video_temporal_refiner": 0,
|
| 249 |
+
"video_token_index": 151670,
|
| 250 |
+
"vision_aspect_ratio": "anyres_max_9",
|
| 251 |
+
"vision_config": {
|
| 252 |
+
"deepstack_visual_indexes": [],
|
| 253 |
+
"depth": 24,
|
| 254 |
+
"dtype": "bfloat16",
|
| 255 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 256 |
+
"hidden_size": 1024,
|
| 257 |
+
"image_size": 384,
|
| 258 |
+
"in_channels": 3,
|
| 259 |
+
"initializer_range": 0.02,
|
| 260 |
+
"intermediate_size": 4096,
|
| 261 |
+
"model_type": "qwen3_5",
|
| 262 |
+
"num_heads": 16,
|
| 263 |
+
"num_position_embeddings": 2304,
|
| 264 |
+
"out_hidden_size": 2048,
|
| 265 |
+
"patch_size": 16,
|
| 266 |
+
"spatial_merge_size": 2,
|
| 267 |
+
"temporal_patch_size": 2
|
| 268 |
+
},
|
| 269 |
+
"vision_feature_layer": -1,
|
| 270 |
+
"vision_feature_select_strategy": "full",
|
| 271 |
+
"vra_layer": 18,
|
| 272 |
+
"vra_loss": 0,
|
| 273 |
+
"vra_scope": "image_video",
|
| 274 |
+
"vra_target_path": "",
|
| 275 |
+
"vra_teacher_kind": "dino",
|
| 276 |
+
"vra_teacher_res": 384,
|
| 277 |
+
"vra_video_loss": "direct",
|
| 278 |
+
"vra_weight": 0.5,
|
| 279 |
+
"vra_z_dim": 1024,
|
| 280 |
+
"auto_map": {
|
| 281 |
+
"AutoConfig": "modeling_qwen3_5vit_qwen3.LlavaQwen3_5ViTConfig",
|
| 282 |
+
"AutoModel": "modeling_qwen3_5vit_qwen3.LlavaQwen3_5ViTForConditionalGeneration",
|
| 283 |
+
"AutoModelForCausalLM": "modeling_qwen3_5vit_qwen3.LlavaQwen3_5ViTForConditionalGeneration",
|
| 284 |
+
"AutoModelForImageTextToText": "modeling_qwen3_5vit_qwen3.LlavaQwen3_5ViTForConditionalGeneration"
|
| 285 |
+
}
|
| 286 |
+
}
|
ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 151643,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"output_attentions": false,
|
| 9 |
+
"output_hidden_states": false,
|
| 10 |
+
"transformers_version": "5.5.4",
|
| 11 |
+
"use_cache": true
|
| 12 |
+
}
|
ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/model-00001-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d33c95f5c94fe5ae2fa9ee4e77fb27ecc53de29bbb0a10c703ff396ce77ae97d
|
| 3 |
+
size 4987507408
|
ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/model-00002-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70e72bb9e7f7876576f759a0f14942cd65cb0fb36f641aff7f41984e25145f3e
|
| 3 |
+
size 4466285248
|
ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/model.safetensors.index.json
ADDED
|
@@ -0,0 +1,705 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 4337895424,
|
| 4 |
+
"total_size": 9453703168
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"lm_head.weight": "model-00001-of-00002.safetensors",
|
| 8 |
+
"model.image_newline": "model-00001-of-00002.safetensors",
|
| 9 |
+
"model.language_model.embed_tokens.weight": "model-00001-of-00002.safetensors",
|
| 10 |
+
"model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 11 |
+
"model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 12 |
+
"model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 13 |
+
"model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 14 |
+
"model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 15 |
+
"model.language_model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 16 |
+
"model.language_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 17 |
+
"model.language_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 18 |
+
"model.language_model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 19 |
+
"model.language_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 20 |
+
"model.language_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 21 |
+
"model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 22 |
+
"model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 23 |
+
"model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 24 |
+
"model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 25 |
+
"model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 26 |
+
"model.language_model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 27 |
+
"model.language_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 28 |
+
"model.language_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 29 |
+
"model.language_model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 30 |
+
"model.language_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 31 |
+
"model.language_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 32 |
+
"model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 33 |
+
"model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 34 |
+
"model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 35 |
+
"model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 36 |
+
"model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 37 |
+
"model.language_model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 38 |
+
"model.language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 39 |
+
"model.language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 40 |
+
"model.language_model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 41 |
+
"model.language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 42 |
+
"model.language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 43 |
+
"model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 44 |
+
"model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 45 |
+
"model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 46 |
+
"model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 47 |
+
"model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 48 |
+
"model.language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 49 |
+
"model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 50 |
+
"model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 51 |
+
"model.language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 52 |
+
"model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 53 |
+
"model.language_model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 54 |
+
"model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 55 |
+
"model.language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 56 |
+
"model.language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 57 |
+
"model.language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 58 |
+
"model.language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 59 |
+
"model.language_model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 60 |
+
"model.language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 61 |
+
"model.language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 62 |
+
"model.language_model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 63 |
+
"model.language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 64 |
+
"model.language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 65 |
+
"model.language_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 66 |
+
"model.language_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 67 |
+
"model.language_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 68 |
+
"model.language_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 69 |
+
"model.language_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 70 |
+
"model.language_model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 71 |
+
"model.language_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 72 |
+
"model.language_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 73 |
+
"model.language_model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 74 |
+
"model.language_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 75 |
+
"model.language_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 76 |
+
"model.language_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 77 |
+
"model.language_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 78 |
+
"model.language_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 79 |
+
"model.language_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 80 |
+
"model.language_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 81 |
+
"model.language_model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 82 |
+
"model.language_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 83 |
+
"model.language_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 84 |
+
"model.language_model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 85 |
+
"model.language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 86 |
+
"model.language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 87 |
+
"model.language_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 88 |
+
"model.language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 89 |
+
"model.language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 90 |
+
"model.language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 91 |
+
"model.language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 92 |
+
"model.language_model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 93 |
+
"model.language_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 94 |
+
"model.language_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 95 |
+
"model.language_model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 96 |
+
"model.language_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 97 |
+
"model.language_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 98 |
+
"model.language_model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 99 |
+
"model.language_model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 100 |
+
"model.language_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 101 |
+
"model.language_model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 102 |
+
"model.language_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 103 |
+
"model.language_model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 104 |
+
"model.language_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 105 |
+
"model.language_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 106 |
+
"model.language_model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 107 |
+
"model.language_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 108 |
+
"model.language_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 109 |
+
"model.language_model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 110 |
+
"model.language_model.layers.17.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 111 |
+
"model.language_model.layers.17.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 112 |
+
"model.language_model.layers.17.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 113 |
+
"model.language_model.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 114 |
+
"model.language_model.layers.17.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 115 |
+
"model.language_model.layers.17.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 116 |
+
"model.language_model.layers.17.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 117 |
+
"model.language_model.layers.17.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 118 |
+
"model.language_model.layers.17.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 119 |
+
"model.language_model.layers.17.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 120 |
+
"model.language_model.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 121 |
+
"model.language_model.layers.18.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 122 |
+
"model.language_model.layers.18.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 123 |
+
"model.language_model.layers.18.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 124 |
+
"model.language_model.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 125 |
+
"model.language_model.layers.18.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 126 |
+
"model.language_model.layers.18.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 127 |
+
"model.language_model.layers.18.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 128 |
+
"model.language_model.layers.18.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 129 |
+
"model.language_model.layers.18.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 130 |
+
"model.language_model.layers.18.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 131 |
+
"model.language_model.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 132 |
+
"model.language_model.layers.19.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 133 |
+
"model.language_model.layers.19.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 134 |
+
"model.language_model.layers.19.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 135 |
+
"model.language_model.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 136 |
+
"model.language_model.layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 137 |
+
"model.language_model.layers.19.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 138 |
+
"model.language_model.layers.19.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 139 |
+
"model.language_model.layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 140 |
+
"model.language_model.layers.19.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 141 |
+
"model.language_model.layers.19.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 142 |
+
"model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 143 |
+
"model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 144 |
+
"model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 145 |
+
"model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 146 |
+
"model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 147 |
+
"model.language_model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 148 |
+
"model.language_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 149 |
+
"model.language_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 150 |
+
"model.language_model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 151 |
+
"model.language_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 152 |
+
"model.language_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 153 |
+
"model.language_model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 154 |
+
"model.language_model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 155 |
+
"model.language_model.layers.20.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 156 |
+
"model.language_model.layers.20.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 157 |
+
"model.language_model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 158 |
+
"model.language_model.layers.20.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 159 |
+
"model.language_model.layers.20.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 160 |
+
"model.language_model.layers.20.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 161 |
+
"model.language_model.layers.20.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 162 |
+
"model.language_model.layers.20.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 163 |
+
"model.language_model.layers.20.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 164 |
+
"model.language_model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 165 |
+
"model.language_model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 166 |
+
"model.language_model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 167 |
+
"model.language_model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 168 |
+
"model.language_model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 169 |
+
"model.language_model.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 170 |
+
"model.language_model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 171 |
+
"model.language_model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 172 |
+
"model.language_model.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 173 |
+
"model.language_model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 174 |
+
"model.language_model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 175 |
+
"model.language_model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 176 |
+
"model.language_model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 177 |
+
"model.language_model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 178 |
+
"model.language_model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 179 |
+
"model.language_model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 180 |
+
"model.language_model.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 181 |
+
"model.language_model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 182 |
+
"model.language_model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 183 |
+
"model.language_model.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 184 |
+
"model.language_model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 185 |
+
"model.language_model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 186 |
+
"model.language_model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 187 |
+
"model.language_model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 188 |
+
"model.language_model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 189 |
+
"model.language_model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 190 |
+
"model.language_model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 191 |
+
"model.language_model.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 192 |
+
"model.language_model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 193 |
+
"model.language_model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 194 |
+
"model.language_model.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 195 |
+
"model.language_model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 196 |
+
"model.language_model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 197 |
+
"model.language_model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 198 |
+
"model.language_model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 199 |
+
"model.language_model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 200 |
+
"model.language_model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 201 |
+
"model.language_model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 202 |
+
"model.language_model.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 203 |
+
"model.language_model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 204 |
+
"model.language_model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 205 |
+
"model.language_model.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 206 |
+
"model.language_model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 207 |
+
"model.language_model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 208 |
+
"model.language_model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 209 |
+
"model.language_model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 210 |
+
"model.language_model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 211 |
+
"model.language_model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 212 |
+
"model.language_model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 213 |
+
"model.language_model.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 214 |
+
"model.language_model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 215 |
+
"model.language_model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 216 |
+
"model.language_model.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 217 |
+
"model.language_model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 218 |
+
"model.language_model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 219 |
+
"model.language_model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 220 |
+
"model.language_model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 221 |
+
"model.language_model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 222 |
+
"model.language_model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 223 |
+
"model.language_model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 224 |
+
"model.language_model.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 225 |
+
"model.language_model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 226 |
+
"model.language_model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 227 |
+
"model.language_model.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 228 |
+
"model.language_model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 229 |
+
"model.language_model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 230 |
+
"model.language_model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 231 |
+
"model.language_model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 232 |
+
"model.language_model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 233 |
+
"model.language_model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 234 |
+
"model.language_model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 235 |
+
"model.language_model.layers.27.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 236 |
+
"model.language_model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 237 |
+
"model.language_model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 238 |
+
"model.language_model.layers.27.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 239 |
+
"model.language_model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 240 |
+
"model.language_model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 241 |
+
"model.language_model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 242 |
+
"model.language_model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 243 |
+
"model.language_model.layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 244 |
+
"model.language_model.layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 245 |
+
"model.language_model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 246 |
+
"model.language_model.layers.28.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 247 |
+
"model.language_model.layers.28.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 248 |
+
"model.language_model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 249 |
+
"model.language_model.layers.28.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 250 |
+
"model.language_model.layers.28.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 251 |
+
"model.language_model.layers.28.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 252 |
+
"model.language_model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 253 |
+
"model.language_model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 254 |
+
"model.language_model.layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 255 |
+
"model.language_model.layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 256 |
+
"model.language_model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 257 |
+
"model.language_model.layers.29.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 258 |
+
"model.language_model.layers.29.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 259 |
+
"model.language_model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 260 |
+
"model.language_model.layers.29.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 261 |
+
"model.language_model.layers.29.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 262 |
+
"model.language_model.layers.29.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 263 |
+
"model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 264 |
+
"model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 265 |
+
"model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 266 |
+
"model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 267 |
+
"model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 268 |
+
"model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 269 |
+
"model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 270 |
+
"model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 271 |
+
"model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 272 |
+
"model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 273 |
+
"model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 274 |
+
"model.language_model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 275 |
+
"model.language_model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 276 |
+
"model.language_model.layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 277 |
+
"model.language_model.layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 278 |
+
"model.language_model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 279 |
+
"model.language_model.layers.30.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 280 |
+
"model.language_model.layers.30.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 281 |
+
"model.language_model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 282 |
+
"model.language_model.layers.30.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 283 |
+
"model.language_model.layers.30.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 284 |
+
"model.language_model.layers.30.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 285 |
+
"model.language_model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 286 |
+
"model.language_model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 287 |
+
"model.language_model.layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 288 |
+
"model.language_model.layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 289 |
+
"model.language_model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 290 |
+
"model.language_model.layers.31.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 291 |
+
"model.language_model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 292 |
+
"model.language_model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 293 |
+
"model.language_model.layers.31.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 294 |
+
"model.language_model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 295 |
+
"model.language_model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 296 |
+
"model.language_model.layers.32.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 297 |
+
"model.language_model.layers.32.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 298 |
+
"model.language_model.layers.32.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 299 |
+
"model.language_model.layers.32.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 300 |
+
"model.language_model.layers.32.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 301 |
+
"model.language_model.layers.32.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 302 |
+
"model.language_model.layers.32.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 303 |
+
"model.language_model.layers.32.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 304 |
+
"model.language_model.layers.32.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 305 |
+
"model.language_model.layers.32.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 306 |
+
"model.language_model.layers.32.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 307 |
+
"model.language_model.layers.33.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 308 |
+
"model.language_model.layers.33.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 309 |
+
"model.language_model.layers.33.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 310 |
+
"model.language_model.layers.33.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 311 |
+
"model.language_model.layers.33.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 312 |
+
"model.language_model.layers.33.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 313 |
+
"model.language_model.layers.33.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 314 |
+
"model.language_model.layers.33.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 315 |
+
"model.language_model.layers.33.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 316 |
+
"model.language_model.layers.33.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 317 |
+
"model.language_model.layers.33.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 318 |
+
"model.language_model.layers.34.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 319 |
+
"model.language_model.layers.34.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 320 |
+
"model.language_model.layers.34.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 321 |
+
"model.language_model.layers.34.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 322 |
+
"model.language_model.layers.34.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 323 |
+
"model.language_model.layers.34.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 324 |
+
"model.language_model.layers.34.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 325 |
+
"model.language_model.layers.34.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 326 |
+
"model.language_model.layers.34.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 327 |
+
"model.language_model.layers.34.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 328 |
+
"model.language_model.layers.34.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 329 |
+
"model.language_model.layers.35.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 330 |
+
"model.language_model.layers.35.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 331 |
+
"model.language_model.layers.35.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 332 |
+
"model.language_model.layers.35.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 333 |
+
"model.language_model.layers.35.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 334 |
+
"model.language_model.layers.35.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 335 |
+
"model.language_model.layers.35.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 336 |
+
"model.language_model.layers.35.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 337 |
+
"model.language_model.layers.35.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 338 |
+
"model.language_model.layers.35.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 339 |
+
"model.language_model.layers.35.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 340 |
+
"model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 341 |
+
"model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 342 |
+
"model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 343 |
+
"model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 344 |
+
"model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 345 |
+
"model.language_model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 346 |
+
"model.language_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 347 |
+
"model.language_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 348 |
+
"model.language_model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 349 |
+
"model.language_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 350 |
+
"model.language_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 351 |
+
"model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 352 |
+
"model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 353 |
+
"model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 354 |
+
"model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 355 |
+
"model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 356 |
+
"model.language_model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 357 |
+
"model.language_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 358 |
+
"model.language_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 359 |
+
"model.language_model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 360 |
+
"model.language_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 361 |
+
"model.language_model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 362 |
+
"model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 363 |
+
"model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 364 |
+
"model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 365 |
+
"model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 366 |
+
"model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 367 |
+
"model.language_model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 368 |
+
"model.language_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 369 |
+
"model.language_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 370 |
+
"model.language_model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 371 |
+
"model.language_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 372 |
+
"model.language_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 373 |
+
"model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 374 |
+
"model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 375 |
+
"model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 376 |
+
"model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 377 |
+
"model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 378 |
+
"model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 379 |
+
"model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 380 |
+
"model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 381 |
+
"model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 382 |
+
"model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 383 |
+
"model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 384 |
+
"model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 385 |
+
"model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 386 |
+
"model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 387 |
+
"model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 388 |
+
"model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 389 |
+
"model.language_model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 390 |
+
"model.language_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 391 |
+
"model.language_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 392 |
+
"model.language_model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 393 |
+
"model.language_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 394 |
+
"model.language_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 395 |
+
"model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 396 |
+
"model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 397 |
+
"model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 398 |
+
"model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 399 |
+
"model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 400 |
+
"model.language_model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 401 |
+
"model.language_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 402 |
+
"model.language_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 403 |
+
"model.language_model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 404 |
+
"model.language_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 405 |
+
"model.language_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 406 |
+
"model.language_model.norm.weight": "model-00002-of-00002.safetensors",
|
| 407 |
+
"model.multi_modal_projector.linear_1.bias": "model-00002-of-00002.safetensors",
|
| 408 |
+
"model.multi_modal_projector.linear_1.weight": "model-00002-of-00002.safetensors",
|
| 409 |
+
"model.multi_modal_projector.linear_2.bias": "model-00002-of-00002.safetensors",
|
| 410 |
+
"model.multi_modal_projector.linear_2.weight": "model-00002-of-00002.safetensors",
|
| 411 |
+
"model.vision_tower.vision.blocks.0.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 412 |
+
"model.vision_tower.vision.blocks.0.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 413 |
+
"model.vision_tower.vision.blocks.0.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 414 |
+
"model.vision_tower.vision.blocks.0.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 415 |
+
"model.vision_tower.vision.blocks.0.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 416 |
+
"model.vision_tower.vision.blocks.0.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 417 |
+
"model.vision_tower.vision.blocks.0.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 418 |
+
"model.vision_tower.vision.blocks.0.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 419 |
+
"model.vision_tower.vision.blocks.0.norm1.bias": "model-00002-of-00002.safetensors",
|
| 420 |
+
"model.vision_tower.vision.blocks.0.norm1.weight": "model-00002-of-00002.safetensors",
|
| 421 |
+
"model.vision_tower.vision.blocks.0.norm2.bias": "model-00002-of-00002.safetensors",
|
| 422 |
+
"model.vision_tower.vision.blocks.0.norm2.weight": "model-00002-of-00002.safetensors",
|
| 423 |
+
"model.vision_tower.vision.blocks.1.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 424 |
+
"model.vision_tower.vision.blocks.1.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 425 |
+
"model.vision_tower.vision.blocks.1.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 426 |
+
"model.vision_tower.vision.blocks.1.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 427 |
+
"model.vision_tower.vision.blocks.1.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 428 |
+
"model.vision_tower.vision.blocks.1.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 429 |
+
"model.vision_tower.vision.blocks.1.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 430 |
+
"model.vision_tower.vision.blocks.1.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 431 |
+
"model.vision_tower.vision.blocks.1.norm1.bias": "model-00002-of-00002.safetensors",
|
| 432 |
+
"model.vision_tower.vision.blocks.1.norm1.weight": "model-00002-of-00002.safetensors",
|
| 433 |
+
"model.vision_tower.vision.blocks.1.norm2.bias": "model-00002-of-00002.safetensors",
|
| 434 |
+
"model.vision_tower.vision.blocks.1.norm2.weight": "model-00002-of-00002.safetensors",
|
| 435 |
+
"model.vision_tower.vision.blocks.10.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 436 |
+
"model.vision_tower.vision.blocks.10.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 437 |
+
"model.vision_tower.vision.blocks.10.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 438 |
+
"model.vision_tower.vision.blocks.10.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 439 |
+
"model.vision_tower.vision.blocks.10.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 440 |
+
"model.vision_tower.vision.blocks.10.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 441 |
+
"model.vision_tower.vision.blocks.10.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 442 |
+
"model.vision_tower.vision.blocks.10.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 443 |
+
"model.vision_tower.vision.blocks.10.norm1.bias": "model-00002-of-00002.safetensors",
|
| 444 |
+
"model.vision_tower.vision.blocks.10.norm1.weight": "model-00002-of-00002.safetensors",
|
| 445 |
+
"model.vision_tower.vision.blocks.10.norm2.bias": "model-00002-of-00002.safetensors",
|
| 446 |
+
"model.vision_tower.vision.blocks.10.norm2.weight": "model-00002-of-00002.safetensors",
|
| 447 |
+
"model.vision_tower.vision.blocks.11.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 448 |
+
"model.vision_tower.vision.blocks.11.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 449 |
+
"model.vision_tower.vision.blocks.11.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 450 |
+
"model.vision_tower.vision.blocks.11.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 451 |
+
"model.vision_tower.vision.blocks.11.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 452 |
+
"model.vision_tower.vision.blocks.11.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 453 |
+
"model.vision_tower.vision.blocks.11.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 454 |
+
"model.vision_tower.vision.blocks.11.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 455 |
+
"model.vision_tower.vision.blocks.11.norm1.bias": "model-00002-of-00002.safetensors",
|
| 456 |
+
"model.vision_tower.vision.blocks.11.norm1.weight": "model-00002-of-00002.safetensors",
|
| 457 |
+
"model.vision_tower.vision.blocks.11.norm2.bias": "model-00002-of-00002.safetensors",
|
| 458 |
+
"model.vision_tower.vision.blocks.11.norm2.weight": "model-00002-of-00002.safetensors",
|
| 459 |
+
"model.vision_tower.vision.blocks.12.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 460 |
+
"model.vision_tower.vision.blocks.12.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 461 |
+
"model.vision_tower.vision.blocks.12.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 462 |
+
"model.vision_tower.vision.blocks.12.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 463 |
+
"model.vision_tower.vision.blocks.12.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 464 |
+
"model.vision_tower.vision.blocks.12.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 465 |
+
"model.vision_tower.vision.blocks.12.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 466 |
+
"model.vision_tower.vision.blocks.12.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 467 |
+
"model.vision_tower.vision.blocks.12.norm1.bias": "model-00002-of-00002.safetensors",
|
| 468 |
+
"model.vision_tower.vision.blocks.12.norm1.weight": "model-00002-of-00002.safetensors",
|
| 469 |
+
"model.vision_tower.vision.blocks.12.norm2.bias": "model-00002-of-00002.safetensors",
|
| 470 |
+
"model.vision_tower.vision.blocks.12.norm2.weight": "model-00002-of-00002.safetensors",
|
| 471 |
+
"model.vision_tower.vision.blocks.13.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 472 |
+
"model.vision_tower.vision.blocks.13.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 473 |
+
"model.vision_tower.vision.blocks.13.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 474 |
+
"model.vision_tower.vision.blocks.13.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 475 |
+
"model.vision_tower.vision.blocks.13.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 476 |
+
"model.vision_tower.vision.blocks.13.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 477 |
+
"model.vision_tower.vision.blocks.13.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 478 |
+
"model.vision_tower.vision.blocks.13.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 479 |
+
"model.vision_tower.vision.blocks.13.norm1.bias": "model-00002-of-00002.safetensors",
|
| 480 |
+
"model.vision_tower.vision.blocks.13.norm1.weight": "model-00002-of-00002.safetensors",
|
| 481 |
+
"model.vision_tower.vision.blocks.13.norm2.bias": "model-00002-of-00002.safetensors",
|
| 482 |
+
"model.vision_tower.vision.blocks.13.norm2.weight": "model-00002-of-00002.safetensors",
|
| 483 |
+
"model.vision_tower.vision.blocks.14.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 484 |
+
"model.vision_tower.vision.blocks.14.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 485 |
+
"model.vision_tower.vision.blocks.14.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 486 |
+
"model.vision_tower.vision.blocks.14.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 487 |
+
"model.vision_tower.vision.blocks.14.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 488 |
+
"model.vision_tower.vision.blocks.14.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 489 |
+
"model.vision_tower.vision.blocks.14.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 490 |
+
"model.vision_tower.vision.blocks.14.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 491 |
+
"model.vision_tower.vision.blocks.14.norm1.bias": "model-00002-of-00002.safetensors",
|
| 492 |
+
"model.vision_tower.vision.blocks.14.norm1.weight": "model-00002-of-00002.safetensors",
|
| 493 |
+
"model.vision_tower.vision.blocks.14.norm2.bias": "model-00002-of-00002.safetensors",
|
| 494 |
+
"model.vision_tower.vision.blocks.14.norm2.weight": "model-00002-of-00002.safetensors",
|
| 495 |
+
"model.vision_tower.vision.blocks.15.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 496 |
+
"model.vision_tower.vision.blocks.15.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 497 |
+
"model.vision_tower.vision.blocks.15.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 498 |
+
"model.vision_tower.vision.blocks.15.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 499 |
+
"model.vision_tower.vision.blocks.15.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 500 |
+
"model.vision_tower.vision.blocks.15.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 501 |
+
"model.vision_tower.vision.blocks.15.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 502 |
+
"model.vision_tower.vision.blocks.15.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 503 |
+
"model.vision_tower.vision.blocks.15.norm1.bias": "model-00002-of-00002.safetensors",
|
| 504 |
+
"model.vision_tower.vision.blocks.15.norm1.weight": "model-00002-of-00002.safetensors",
|
| 505 |
+
"model.vision_tower.vision.blocks.15.norm2.bias": "model-00002-of-00002.safetensors",
|
| 506 |
+
"model.vision_tower.vision.blocks.15.norm2.weight": "model-00002-of-00002.safetensors",
|
| 507 |
+
"model.vision_tower.vision.blocks.16.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 508 |
+
"model.vision_tower.vision.blocks.16.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 509 |
+
"model.vision_tower.vision.blocks.16.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 510 |
+
"model.vision_tower.vision.blocks.16.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 511 |
+
"model.vision_tower.vision.blocks.16.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 512 |
+
"model.vision_tower.vision.blocks.16.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 513 |
+
"model.vision_tower.vision.blocks.16.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 514 |
+
"model.vision_tower.vision.blocks.16.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 515 |
+
"model.vision_tower.vision.blocks.16.norm1.bias": "model-00002-of-00002.safetensors",
|
| 516 |
+
"model.vision_tower.vision.blocks.16.norm1.weight": "model-00002-of-00002.safetensors",
|
| 517 |
+
"model.vision_tower.vision.blocks.16.norm2.bias": "model-00002-of-00002.safetensors",
|
| 518 |
+
"model.vision_tower.vision.blocks.16.norm2.weight": "model-00002-of-00002.safetensors",
|
| 519 |
+
"model.vision_tower.vision.blocks.17.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 520 |
+
"model.vision_tower.vision.blocks.17.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 521 |
+
"model.vision_tower.vision.blocks.17.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 522 |
+
"model.vision_tower.vision.blocks.17.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 523 |
+
"model.vision_tower.vision.blocks.17.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 524 |
+
"model.vision_tower.vision.blocks.17.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 525 |
+
"model.vision_tower.vision.blocks.17.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 526 |
+
"model.vision_tower.vision.blocks.17.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 527 |
+
"model.vision_tower.vision.blocks.17.norm1.bias": "model-00002-of-00002.safetensors",
|
| 528 |
+
"model.vision_tower.vision.blocks.17.norm1.weight": "model-00002-of-00002.safetensors",
|
| 529 |
+
"model.vision_tower.vision.blocks.17.norm2.bias": "model-00002-of-00002.safetensors",
|
| 530 |
+
"model.vision_tower.vision.blocks.17.norm2.weight": "model-00002-of-00002.safetensors",
|
| 531 |
+
"model.vision_tower.vision.blocks.18.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 532 |
+
"model.vision_tower.vision.blocks.18.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 533 |
+
"model.vision_tower.vision.blocks.18.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 534 |
+
"model.vision_tower.vision.blocks.18.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 535 |
+
"model.vision_tower.vision.blocks.18.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 536 |
+
"model.vision_tower.vision.blocks.18.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 537 |
+
"model.vision_tower.vision.blocks.18.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 538 |
+
"model.vision_tower.vision.blocks.18.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 539 |
+
"model.vision_tower.vision.blocks.18.norm1.bias": "model-00002-of-00002.safetensors",
|
| 540 |
+
"model.vision_tower.vision.blocks.18.norm1.weight": "model-00002-of-00002.safetensors",
|
| 541 |
+
"model.vision_tower.vision.blocks.18.norm2.bias": "model-00002-of-00002.safetensors",
|
| 542 |
+
"model.vision_tower.vision.blocks.18.norm2.weight": "model-00002-of-00002.safetensors",
|
| 543 |
+
"model.vision_tower.vision.blocks.19.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 544 |
+
"model.vision_tower.vision.blocks.19.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 545 |
+
"model.vision_tower.vision.blocks.19.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 546 |
+
"model.vision_tower.vision.blocks.19.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 547 |
+
"model.vision_tower.vision.blocks.19.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 548 |
+
"model.vision_tower.vision.blocks.19.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 549 |
+
"model.vision_tower.vision.blocks.19.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 550 |
+
"model.vision_tower.vision.blocks.19.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 551 |
+
"model.vision_tower.vision.blocks.19.norm1.bias": "model-00002-of-00002.safetensors",
|
| 552 |
+
"model.vision_tower.vision.blocks.19.norm1.weight": "model-00002-of-00002.safetensors",
|
| 553 |
+
"model.vision_tower.vision.blocks.19.norm2.bias": "model-00002-of-00002.safetensors",
|
| 554 |
+
"model.vision_tower.vision.blocks.19.norm2.weight": "model-00002-of-00002.safetensors",
|
| 555 |
+
"model.vision_tower.vision.blocks.2.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 556 |
+
"model.vision_tower.vision.blocks.2.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 557 |
+
"model.vision_tower.vision.blocks.2.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 558 |
+
"model.vision_tower.vision.blocks.2.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 559 |
+
"model.vision_tower.vision.blocks.2.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 560 |
+
"model.vision_tower.vision.blocks.2.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 561 |
+
"model.vision_tower.vision.blocks.2.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 562 |
+
"model.vision_tower.vision.blocks.2.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 563 |
+
"model.vision_tower.vision.blocks.2.norm1.bias": "model-00002-of-00002.safetensors",
|
| 564 |
+
"model.vision_tower.vision.blocks.2.norm1.weight": "model-00002-of-00002.safetensors",
|
| 565 |
+
"model.vision_tower.vision.blocks.2.norm2.bias": "model-00002-of-00002.safetensors",
|
| 566 |
+
"model.vision_tower.vision.blocks.2.norm2.weight": "model-00002-of-00002.safetensors",
|
| 567 |
+
"model.vision_tower.vision.blocks.20.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 568 |
+
"model.vision_tower.vision.blocks.20.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 569 |
+
"model.vision_tower.vision.blocks.20.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 570 |
+
"model.vision_tower.vision.blocks.20.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 571 |
+
"model.vision_tower.vision.blocks.20.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 572 |
+
"model.vision_tower.vision.blocks.20.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 573 |
+
"model.vision_tower.vision.blocks.20.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 574 |
+
"model.vision_tower.vision.blocks.20.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 575 |
+
"model.vision_tower.vision.blocks.20.norm1.bias": "model-00002-of-00002.safetensors",
|
| 576 |
+
"model.vision_tower.vision.blocks.20.norm1.weight": "model-00002-of-00002.safetensors",
|
| 577 |
+
"model.vision_tower.vision.blocks.20.norm2.bias": "model-00002-of-00002.safetensors",
|
| 578 |
+
"model.vision_tower.vision.blocks.20.norm2.weight": "model-00002-of-00002.safetensors",
|
| 579 |
+
"model.vision_tower.vision.blocks.21.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 580 |
+
"model.vision_tower.vision.blocks.21.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 581 |
+
"model.vision_tower.vision.blocks.21.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 582 |
+
"model.vision_tower.vision.blocks.21.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 583 |
+
"model.vision_tower.vision.blocks.21.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 584 |
+
"model.vision_tower.vision.blocks.21.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 585 |
+
"model.vision_tower.vision.blocks.21.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 586 |
+
"model.vision_tower.vision.blocks.21.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 587 |
+
"model.vision_tower.vision.blocks.21.norm1.bias": "model-00002-of-00002.safetensors",
|
| 588 |
+
"model.vision_tower.vision.blocks.21.norm1.weight": "model-00002-of-00002.safetensors",
|
| 589 |
+
"model.vision_tower.vision.blocks.21.norm2.bias": "model-00002-of-00002.safetensors",
|
| 590 |
+
"model.vision_tower.vision.blocks.21.norm2.weight": "model-00002-of-00002.safetensors",
|
| 591 |
+
"model.vision_tower.vision.blocks.22.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 592 |
+
"model.vision_tower.vision.blocks.22.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 593 |
+
"model.vision_tower.vision.blocks.22.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 594 |
+
"model.vision_tower.vision.blocks.22.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 595 |
+
"model.vision_tower.vision.blocks.22.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 596 |
+
"model.vision_tower.vision.blocks.22.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 597 |
+
"model.vision_tower.vision.blocks.22.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 598 |
+
"model.vision_tower.vision.blocks.22.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 599 |
+
"model.vision_tower.vision.blocks.22.norm1.bias": "model-00002-of-00002.safetensors",
|
| 600 |
+
"model.vision_tower.vision.blocks.22.norm1.weight": "model-00002-of-00002.safetensors",
|
| 601 |
+
"model.vision_tower.vision.blocks.22.norm2.bias": "model-00002-of-00002.safetensors",
|
| 602 |
+
"model.vision_tower.vision.blocks.22.norm2.weight": "model-00002-of-00002.safetensors",
|
| 603 |
+
"model.vision_tower.vision.blocks.23.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 604 |
+
"model.vision_tower.vision.blocks.23.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 605 |
+
"model.vision_tower.vision.blocks.23.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 606 |
+
"model.vision_tower.vision.blocks.23.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 607 |
+
"model.vision_tower.vision.blocks.23.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 608 |
+
"model.vision_tower.vision.blocks.23.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 609 |
+
"model.vision_tower.vision.blocks.23.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 610 |
+
"model.vision_tower.vision.blocks.23.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 611 |
+
"model.vision_tower.vision.blocks.23.norm1.bias": "model-00002-of-00002.safetensors",
|
| 612 |
+
"model.vision_tower.vision.blocks.23.norm1.weight": "model-00002-of-00002.safetensors",
|
| 613 |
+
"model.vision_tower.vision.blocks.23.norm2.bias": "model-00002-of-00002.safetensors",
|
| 614 |
+
"model.vision_tower.vision.blocks.23.norm2.weight": "model-00002-of-00002.safetensors",
|
| 615 |
+
"model.vision_tower.vision.blocks.3.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 616 |
+
"model.vision_tower.vision.blocks.3.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 617 |
+
"model.vision_tower.vision.blocks.3.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 618 |
+
"model.vision_tower.vision.blocks.3.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 619 |
+
"model.vision_tower.vision.blocks.3.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 620 |
+
"model.vision_tower.vision.blocks.3.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 621 |
+
"model.vision_tower.vision.blocks.3.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 622 |
+
"model.vision_tower.vision.blocks.3.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 623 |
+
"model.vision_tower.vision.blocks.3.norm1.bias": "model-00002-of-00002.safetensors",
|
| 624 |
+
"model.vision_tower.vision.blocks.3.norm1.weight": "model-00002-of-00002.safetensors",
|
| 625 |
+
"model.vision_tower.vision.blocks.3.norm2.bias": "model-00002-of-00002.safetensors",
|
| 626 |
+
"model.vision_tower.vision.blocks.3.norm2.weight": "model-00002-of-00002.safetensors",
|
| 627 |
+
"model.vision_tower.vision.blocks.4.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 628 |
+
"model.vision_tower.vision.blocks.4.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 629 |
+
"model.vision_tower.vision.blocks.4.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 630 |
+
"model.vision_tower.vision.blocks.4.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 631 |
+
"model.vision_tower.vision.blocks.4.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 632 |
+
"model.vision_tower.vision.blocks.4.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 633 |
+
"model.vision_tower.vision.blocks.4.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 634 |
+
"model.vision_tower.vision.blocks.4.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 635 |
+
"model.vision_tower.vision.blocks.4.norm1.bias": "model-00002-of-00002.safetensors",
|
| 636 |
+
"model.vision_tower.vision.blocks.4.norm1.weight": "model-00002-of-00002.safetensors",
|
| 637 |
+
"model.vision_tower.vision.blocks.4.norm2.bias": "model-00002-of-00002.safetensors",
|
| 638 |
+
"model.vision_tower.vision.blocks.4.norm2.weight": "model-00002-of-00002.safetensors",
|
| 639 |
+
"model.vision_tower.vision.blocks.5.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 640 |
+
"model.vision_tower.vision.blocks.5.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 641 |
+
"model.vision_tower.vision.blocks.5.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 642 |
+
"model.vision_tower.vision.blocks.5.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 643 |
+
"model.vision_tower.vision.blocks.5.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 644 |
+
"model.vision_tower.vision.blocks.5.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 645 |
+
"model.vision_tower.vision.blocks.5.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 646 |
+
"model.vision_tower.vision.blocks.5.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 647 |
+
"model.vision_tower.vision.blocks.5.norm1.bias": "model-00002-of-00002.safetensors",
|
| 648 |
+
"model.vision_tower.vision.blocks.5.norm1.weight": "model-00002-of-00002.safetensors",
|
| 649 |
+
"model.vision_tower.vision.blocks.5.norm2.bias": "model-00002-of-00002.safetensors",
|
| 650 |
+
"model.vision_tower.vision.blocks.5.norm2.weight": "model-00002-of-00002.safetensors",
|
| 651 |
+
"model.vision_tower.vision.blocks.6.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 652 |
+
"model.vision_tower.vision.blocks.6.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 653 |
+
"model.vision_tower.vision.blocks.6.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 654 |
+
"model.vision_tower.vision.blocks.6.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 655 |
+
"model.vision_tower.vision.blocks.6.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 656 |
+
"model.vision_tower.vision.blocks.6.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 657 |
+
"model.vision_tower.vision.blocks.6.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 658 |
+
"model.vision_tower.vision.blocks.6.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 659 |
+
"model.vision_tower.vision.blocks.6.norm1.bias": "model-00002-of-00002.safetensors",
|
| 660 |
+
"model.vision_tower.vision.blocks.6.norm1.weight": "model-00002-of-00002.safetensors",
|
| 661 |
+
"model.vision_tower.vision.blocks.6.norm2.bias": "model-00002-of-00002.safetensors",
|
| 662 |
+
"model.vision_tower.vision.blocks.6.norm2.weight": "model-00002-of-00002.safetensors",
|
| 663 |
+
"model.vision_tower.vision.blocks.7.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 664 |
+
"model.vision_tower.vision.blocks.7.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 665 |
+
"model.vision_tower.vision.blocks.7.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 666 |
+
"model.vision_tower.vision.blocks.7.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 667 |
+
"model.vision_tower.vision.blocks.7.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 668 |
+
"model.vision_tower.vision.blocks.7.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 669 |
+
"model.vision_tower.vision.blocks.7.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 670 |
+
"model.vision_tower.vision.blocks.7.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 671 |
+
"model.vision_tower.vision.blocks.7.norm1.bias": "model-00002-of-00002.safetensors",
|
| 672 |
+
"model.vision_tower.vision.blocks.7.norm1.weight": "model-00002-of-00002.safetensors",
|
| 673 |
+
"model.vision_tower.vision.blocks.7.norm2.bias": "model-00002-of-00002.safetensors",
|
| 674 |
+
"model.vision_tower.vision.blocks.7.norm2.weight": "model-00002-of-00002.safetensors",
|
| 675 |
+
"model.vision_tower.vision.blocks.8.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 676 |
+
"model.vision_tower.vision.blocks.8.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 677 |
+
"model.vision_tower.vision.blocks.8.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 678 |
+
"model.vision_tower.vision.blocks.8.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 679 |
+
"model.vision_tower.vision.blocks.8.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 680 |
+
"model.vision_tower.vision.blocks.8.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 681 |
+
"model.vision_tower.vision.blocks.8.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 682 |
+
"model.vision_tower.vision.blocks.8.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 683 |
+
"model.vision_tower.vision.blocks.8.norm1.bias": "model-00002-of-00002.safetensors",
|
| 684 |
+
"model.vision_tower.vision.blocks.8.norm1.weight": "model-00002-of-00002.safetensors",
|
| 685 |
+
"model.vision_tower.vision.blocks.8.norm2.bias": "model-00002-of-00002.safetensors",
|
| 686 |
+
"model.vision_tower.vision.blocks.8.norm2.weight": "model-00002-of-00002.safetensors",
|
| 687 |
+
"model.vision_tower.vision.blocks.9.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 688 |
+
"model.vision_tower.vision.blocks.9.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 689 |
+
"model.vision_tower.vision.blocks.9.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 690 |
+
"model.vision_tower.vision.blocks.9.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 691 |
+
"model.vision_tower.vision.blocks.9.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 692 |
+
"model.vision_tower.vision.blocks.9.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 693 |
+
"model.vision_tower.vision.blocks.9.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 694 |
+
"model.vision_tower.vision.blocks.9.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 695 |
+
"model.vision_tower.vision.blocks.9.norm1.bias": "model-00002-of-00002.safetensors",
|
| 696 |
+
"model.vision_tower.vision.blocks.9.norm1.weight": "model-00002-of-00002.safetensors",
|
| 697 |
+
"model.vision_tower.vision.blocks.9.norm2.bias": "model-00002-of-00002.safetensors",
|
| 698 |
+
"model.vision_tower.vision.blocks.9.norm2.weight": "model-00002-of-00002.safetensors",
|
| 699 |
+
"model.vision_tower.vision.final_layernorm.bias": "model-00002-of-00002.safetensors",
|
| 700 |
+
"model.vision_tower.vision.final_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 701 |
+
"model.vision_tower.vision.patch_embed.proj.bias": "model-00002-of-00002.safetensors",
|
| 702 |
+
"model.vision_tower.vision.patch_embed.proj.weight": "model-00002-of-00002.safetensors",
|
| 703 |
+
"model.vision_tower.vision.pos_embed.weight": "model-00002-of-00002.safetensors"
|
| 704 |
+
}
|
| 705 |
+
}
|
ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/modeling_qwen3_5vit_qwen3.py
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/processing_qwen3_5vit_qwen3.py
ADDED
|
@@ -0,0 +1,302 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"""processing_qwen3_5vit_qwen3.py — LlavaOnevision processor with optional
|
| 2 |
+
video temporal-merge (tmerge) token accounting.
|
| 3 |
+
|
| 4 |
+
Why a subclass: at eval the model is loaded via HF (lmms-eval `llava_hf` →
|
| 5 |
+
`AutoProcessor.from_pretrained(ckpt, trust_remote_code=True)`), NOT the swift
|
| 6 |
+
template. The `LlavaOnevisionProcessor.__call__` expands `<video>` by
|
| 7 |
+
`num_frames*144+1`. Under `video_temporal_merge=1` the model emits
|
| 8 |
+
`(num_frames//2)*144+1` features (grid_t=T/2 temporal groups), so the default
|
| 9 |
+
count over-expands → `masked_scatter` token/feature mismatch crash.
|
| 10 |
+
|
| 11 |
+
This subclass is byte-identical to the upstream processor EXCEPT the single video
|
| 12 |
+
token-count line, which routes through `qwen3_5_video_token_count` — the same
|
| 13 |
+
SoT helper the swift training template uses. The mllm_eval adapter
|
| 14 |
+
(`mllm_eval/checkpoint/__init__.py::qwen3_5vit`) declares this file as
|
| 15 |
+
`processor_file`, so prep cp's it into the ckpt and injects `auto_map.AutoProcessor`.
|
| 16 |
+
|
| 17 |
+
`video_temporal_merge` is a persisted init kwarg (lands in processor_config.json),
|
| 18 |
+
set by the plugin `get_processor` from env/config so eval inherits it with no env.
|
| 19 |
+
"""
|
| 20 |
+
|
| 21 |
+
import math
|
| 22 |
+
|
| 23 |
+
import numpy as np
|
| 24 |
+
|
| 25 |
+
from transformers.feature_extraction_utils import BatchFeature
|
| 26 |
+
from transformers.image_processing_utils import select_best_resolution
|
| 27 |
+
from transformers.image_utils import ImageInput, get_image_size, to_numpy_array
|
| 28 |
+
from transformers.processing_utils import Unpack
|
| 29 |
+
from transformers.tokenization_utils_base import PreTokenizedInput, TextInput
|
| 30 |
+
from transformers.video_utils import VideoInput
|
| 31 |
+
from transformers.models.llava_onevision.processing_llava_onevision import (
|
| 32 |
+
LlavaOnevisionProcessor,
|
| 33 |
+
LlavaOnevisionProcessorKwargs,
|
| 34 |
+
)
|
| 35 |
+
|
| 36 |
+
from typing import Optional
|
| 37 |
+
|
| 38 |
+
# Flat-name sibling import: works both in-repo (model_plugin adds video_mllm/model/ to
|
| 39 |
+
# sys.path) and at eval (HF dynamic_module loads modeling_qwen3_5vit_qwen3.py as a
|
| 40 |
+
# ckpt-dir sibling — the qwen3_5vit adapter ships it as modeling_file).
|
| 41 |
+
from modeling_qwen3_5vit_qwen3 import (
|
| 42 |
+
build_faithful_video_token_string,
|
| 43 |
+
vtc_compression_factor,
|
| 44 |
+
vtc_temporal_patch_size,
|
| 45 |
+
qwen3_5_video_token_count,
|
| 46 |
+
video_is_vtc_compression,
|
| 47 |
+
video_is_spatial_catp,
|
| 48 |
+
video_is_image_mrope,
|
| 49 |
+
video_is_pair_fusion,
|
| 50 |
+
video_is_temporal_compression,
|
| 51 |
+
video_is_temporal_merge,
|
| 52 |
+
video_mrope_active,
|
| 53 |
+
)
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
class LlavaQwen3_5ViTProcessor(LlavaOnevisionProcessor):
|
| 57 |
+
"""LlavaOnevisionProcessor whose `<video>` expansion is tmerge-aware."""
|
| 58 |
+
|
| 59 |
+
def __init__(
|
| 60 |
+
self,
|
| 61 |
+
image_processor=None,
|
| 62 |
+
tokenizer=None,
|
| 63 |
+
video_processor=None,
|
| 64 |
+
num_image_tokens=None,
|
| 65 |
+
vision_feature_select_strategy=None,
|
| 66 |
+
chat_template=None,
|
| 67 |
+
image_token="<image>",
|
| 68 |
+
video_token="<video>",
|
| 69 |
+
vision_aspect_ratio="anyres_max_9",
|
| 70 |
+
video_mode: Optional[str] = None,
|
| 71 |
+
vtc_compression_factor: int = 2,
|
| 72 |
+
vtc_temporal_patch_size: int = 1,
|
| 73 |
+
# Legacy int flags kept in the signature ONLY so a pre-2026-06-22 processor_config.json
|
| 74 |
+
# still reaches __init__ on reload (ProcessorMixin filters keys NOT in the signature BEFORE
|
| 75 |
+
# calling __init__). Folded into video_mode and NEVER set as attributes -> to_dict (which
|
| 76 |
+
# dumps __dict__) serializes ONLY video_mode (single SoT; old ckpts migrate on re-save).
|
| 77 |
+
video_temporal_merge: int = 0,
|
| 78 |
+
image_mrope: int = 0,
|
| 79 |
+
video_mrope: int = 0,
|
| 80 |
+
**kwargs,
|
| 81 |
+
):
|
| 82 |
+
# The parent's sub-processors AND scalar kwargs MUST be explicit in THIS signature -- NOT
|
| 83 |
+
# swallowed by a bare (*args, **kwargs). ProcessorMixin serialization enumerates what to
|
| 84 |
+
# persist via get_attributes() = inspect.signature(cls.__init__).parameters; a (*args,
|
| 85 |
+
# **kwargs) signature returns [] -> save_pretrained writes neither the sub-processor files
|
| 86 |
+
# nor video_mode into the ckpt. So video_mode is an explicit, serialized param here.
|
| 87 |
+
# Backward-compat: an old processor_config.json carries the legacy int flags (received via
|
| 88 |
+
# the explicit params above so reload reaches here) and no video_mode. Fold them into
|
| 89 |
+
# video_mode; do NOT set them as attributes, so to_dict writes only video_mode (single SoT;
|
| 90 |
+
# old ckpts migrate on re-save). New ckpts pass video_mode directly (fold skipped).
|
| 91 |
+
if video_mode is None:
|
| 92 |
+
if int(image_mrope or 0):
|
| 93 |
+
video_mode = "image_mrope"
|
| 94 |
+
elif int(video_mrope or 0):
|
| 95 |
+
video_mode = "video_mrope"
|
| 96 |
+
elif int(video_temporal_merge or 0):
|
| 97 |
+
video_mode = "temporal_merge"
|
| 98 |
+
self.video_mode = video_mode
|
| 99 |
+
self.vtc_compression_factor = vtc_compression_factor
|
| 100 |
+
self.vtc_temporal_patch_size = vtc_temporal_patch_size
|
| 101 |
+
super().__init__(
|
| 102 |
+
image_processor=image_processor,
|
| 103 |
+
tokenizer=tokenizer,
|
| 104 |
+
video_processor=video_processor,
|
| 105 |
+
num_image_tokens=num_image_tokens,
|
| 106 |
+
vision_feature_select_strategy=vision_feature_select_strategy,
|
| 107 |
+
chat_template=chat_template,
|
| 108 |
+
image_token=image_token,
|
| 109 |
+
video_token=video_token,
|
| 110 |
+
vision_aspect_ratio=vision_aspect_ratio,
|
| 111 |
+
**kwargs,
|
| 112 |
+
)
|
| 113 |
+
|
| 114 |
+
def __call__(
|
| 115 |
+
self,
|
| 116 |
+
images: ImageInput = None,
|
| 117 |
+
text: "TextInput | PreTokenizedInput | list[TextInput] | list[PreTokenizedInput]" = None,
|
| 118 |
+
videos: VideoInput = None,
|
| 119 |
+
video_fps: float = None,
|
| 120 |
+
video_sample_indices=None,
|
| 121 |
+
**kwargs: Unpack[LlavaOnevisionProcessorKwargs],
|
| 122 |
+
) -> BatchFeature:
|
| 123 |
+
# ── verbatim from LlavaOnevisionProcessor.__call__ (transformers) ──
|
| 124 |
+
output_kwargs = self._merge_kwargs(
|
| 125 |
+
LlavaOnevisionProcessorKwargs,
|
| 126 |
+
tokenizer_init_kwargs=self.tokenizer.init_kwargs,
|
| 127 |
+
**kwargs,
|
| 128 |
+
)
|
| 129 |
+
|
| 130 |
+
if isinstance(text, str):
|
| 131 |
+
text = [text]
|
| 132 |
+
elif not isinstance(text, list) and not isinstance(text[0], str):
|
| 133 |
+
raise TypeError("Invalid input text. Please provide a string, or a list of strings")
|
| 134 |
+
|
| 135 |
+
image_inputs = video_inputs = {}
|
| 136 |
+
|
| 137 |
+
if images is not None:
|
| 138 |
+
image_inputs = self.image_processor(images, **output_kwargs["images_kwargs"])
|
| 139 |
+
|
| 140 |
+
batch_num_images = iter(image_inputs["batch_num_images"])
|
| 141 |
+
image_sizes = iter(image_inputs["image_sizes"])
|
| 142 |
+
height, width = get_image_size(
|
| 143 |
+
to_numpy_array(image_inputs["pixel_values"][0][0]),
|
| 144 |
+
channel_dim=output_kwargs["images_kwargs"].get("data_format"),
|
| 145 |
+
)
|
| 146 |
+
text, num_image_tokens = self._expand_image_tokens(
|
| 147 |
+
text, image_sizes, height, width, self.image_token, batch_num_images
|
| 148 |
+
)
|
| 149 |
+
|
| 150 |
+
if videos is not None:
|
| 151 |
+
video_inputs = self.video_processor(videos, **output_kwargs["videos_kwargs"])
|
| 152 |
+
|
| 153 |
+
one_video = video_inputs.get("pixel_values_videos")[0]
|
| 154 |
+
if isinstance(video_inputs.get("pixel_values_videos")[0], (list, tuple)):
|
| 155 |
+
one_video = np.array(one_video)
|
| 156 |
+
else:
|
| 157 |
+
one_video = to_numpy_array(one_video)
|
| 158 |
+
num_frames = one_video.shape[0] # frame dim is always after batch dim
|
| 159 |
+
# ── ONLY deviation from upstream __call__: the video token count.
|
| 160 |
+
# Upstream derives it inline as `get_image_size(one_video[0])` (unused
|
| 161 |
+
# for count) + `num_frames * ceil(sqrt(num_image_tokens)/2)**2 + 1`.
|
| 162 |
+
# We drop the get_image_size line ON PURPOSE — the SoT helper hardcodes
|
| 163 |
+
# 144 tokens/readout-unit (= ceil(24/2)**2 for Qwen3.5 384/16 grid) and
|
| 164 |
+
# folds in the grid_t=num_frames//2 tubelet grouping when merge is on.
|
| 165 |
+
# Do NOT "restore" get_image_size when re-syncing against the default.
|
| 166 |
+
if video_mrope_active(self):
|
| 167 |
+
# FAIL-FAST: the faithful eval path builds ONE timestamp stream from a single video's
|
| 168 |
+
# (fps, sampled_indices) and replaces ONE <video> placeholder. A multi-video sample
|
| 169 |
+
# or batched (B>1) call would mis-apply the first video's timestamps to the rest.
|
| 170 |
+
# lmms-eval llava_hf runs generate at bs=1 single-video, so this never fires; it
|
| 171 |
+
# guards a future multi-video / batched entry point from silent mis-scoring.
|
| 172 |
+
n_video_ph = sum(sample.count(self.video_token) for sample in text)
|
| 173 |
+
if n_video_ph != 1:
|
| 174 |
+
raise RuntimeError(
|
| 175 |
+
f"LlavaQwen3_5ViTProcessor: faithful video_mrope eval supports exactly ONE "
|
| 176 |
+
f"<video> per call (single video, batch_size=1); found {n_video_ph}. "
|
| 177 |
+
f"Multi-video / batched faithful eval is unsupported."
|
| 178 |
+
)
|
| 179 |
+
if video_fps is None:
|
| 180 |
+
# FAIL-FAST (not a silent blob fallback): video_mrope=1 means the ckpt was trained
|
| 181 |
+
# on the faithful per-frame timestamp stream, so evaluating it on the contiguous
|
| 182 |
+
# blob is a train/eval mismatch that produces silently-wrong numbers. Only the
|
| 183 |
+
# lmms-eval simple (--force_simple) path threads fps today; the chat / VLMEvalKit
|
| 184 |
+
# video paths do not. Current 6-bench video tasks all use --force_simple, so this
|
| 185 |
+
# never fires; it guards a future entry-point swap from silently degrading.
|
| 186 |
+
# (Image-only eval never reaches here — `videos is None` skips this whole block.)
|
| 187 |
+
raise RuntimeError(
|
| 188 |
+
"LlavaQwen3_5ViTProcessor: video_mrope=1 but no video_fps/video_sample_indices "
|
| 189 |
+
"were provided — cannot build the faithful per-frame timestamp stream. Route "
|
| 190 |
+
"video eval through the lmms-eval simple (--force_simple) path, which threads fps."
|
| 191 |
+
)
|
| 192 |
+
# Faithful video-mRoPE (video_mrope=1): build the per-frame timestamp + vision-marker
|
| 193 |
+
# stream via the SAME shared SoT the training template _encode uses, so eval and
|
| 194 |
+
# train produce byte-identical video token ids (zero drift). Needs fps + sampled
|
| 195 |
+
# frame indices (threaded from the eval video loader). One video per eval request.
|
| 196 |
+
faithful_str = build_faithful_video_token_string(
|
| 197 |
+
num_frames, video_fps, video_sample_indices,
|
| 198 |
+
video_is_temporal_merge(self), video_token=self.video_token,
|
| 199 |
+
)
|
| 200 |
+
text = [sample.replace(self.video_token, faithful_str, 1) for sample in text]
|
| 201 |
+
else:
|
| 202 |
+
# Default / non-mrope: byte-identical contiguous blob.
|
| 203 |
+
num_video_tokens = qwen3_5_video_token_count(
|
| 204 |
+
num_frames,
|
| 205 |
+
temporal_merge=video_is_temporal_merge(self),
|
| 206 |
+
temporal_compression=video_is_temporal_compression(self),
|
| 207 |
+
pair_fusion=video_is_pair_fusion(self),
|
| 208 |
+
vtc_compression=video_is_vtc_compression(self),
|
| 209 |
+
spatial_catp=video_is_spatial_catp(self),
|
| 210 |
+
vtc_compression_factor=vtc_compression_factor(self),
|
| 211 |
+
vtc_temporal_patch_size=vtc_temporal_patch_size(self),
|
| 212 |
+
)
|
| 213 |
+
text = [sample.replace(self.video_token, self.video_token * num_video_tokens) for sample in text]
|
| 214 |
+
|
| 215 |
+
return_tensors = output_kwargs["text_kwargs"].pop("return_tensors", None)
|
| 216 |
+
return_mm_token_type_ids = output_kwargs["text_kwargs"].pop("return_mm_token_type_ids", None)
|
| 217 |
+
text_inputs = self.tokenizer(text, **output_kwargs["text_kwargs"])
|
| 218 |
+
self._check_special_mm_tokens(text, text_inputs, modalities=["image"])
|
| 219 |
+
|
| 220 |
+
if return_mm_token_type_ids:
|
| 221 |
+
# Byte-identical to LlavaOnevisionProcessor.__call__ (transformers 5.5.4):
|
| 222 |
+
# delegate to ProcessorMixin.create_mm_token_type_ids, which marks image=1 / video=2
|
| 223 |
+
# / audio=3 over self.image_ids/video_ids/audio_ids (set by the base __init__, so
|
| 224 |
+
# present on the fresh-build AND the __class__-retype reload path). An earlier inline
|
| 225 |
+
# copy here came from an OLDER transformers that lacked this method and marked
|
| 226 |
+
# image-only (mis-typing video tokens); 5.5.4 HAS the method (processing_utils.py) and
|
| 227 |
+
# the default path calls it, so tracking the default keeps the <video> token COUNT the only intentional
|
| 228 |
+
# deviation in this __call__. (This path is dormant: return_mm_token_type_ids defaults
|
| 229 |
+
# False for LlavaOnevision.)
|
| 230 |
+
text_inputs["mm_token_type_ids"] = self.create_mm_token_type_ids(text_inputs["input_ids"])
|
| 231 |
+
return BatchFeature(data={**text_inputs, **image_inputs, **video_inputs}, tensor_type=return_tensors)
|
| 232 |
+
|
| 233 |
+
def _get_number_of_features(self, orig_height: int, orig_width: int, height: int, width: int) -> int:
|
| 234 |
+
"""image_mrope=1: drop the per-row image_newline from the count (the model calls
|
| 235 |
+
pack_image_features with image_newline=None), so the image token count is
|
| 236 |
+
base + unpadded (clean H x W grid). Default count = base + unpadded + newline; here we
|
| 237 |
+
subtract exactly `newline_features`. flag 0 (default / video_mrope) → byte-identical to parent.
|
| 238 |
+
Used by BOTH the swift training _encode and eval __call__/_expand_image_tokens, so the
|
| 239 |
+
model's pack feature count and the input_ids placeholder count stay matched."""
|
| 240 |
+
n = super()._get_number_of_features(orig_height, orig_width, height, width)
|
| 241 |
+
if video_is_image_mrope(self):
|
| 242 |
+
# Guard the one config that would desync the position-id block sum from the emitted
|
| 243 |
+
# token count: vision_feature_select_strategy="default" makes _expand_image_tokens
|
| 244 |
+
# subtract 1 (a CLS-token drop), but image_mrope_grid_blocks emits the full base+
|
| 245 |
+
# unpadded grid (no CLS to drop). Qwen3.5 ViT has NO CLS token, so the strategy is
|
| 246 |
+
# always 'full'/None here; fail loud instead of silently misaligning if that ever changes.
|
| 247 |
+
if getattr(self, "vision_feature_select_strategy", None) == "default":
|
| 248 |
+
raise ValueError(
|
| 249 |
+
"image_mrope is incompatible with vision_feature_select_strategy='default' "
|
| 250 |
+
"(no CLS token in the clean H x W grid; the -1 would desync the position-id "
|
| 251 |
+
"block sum from the token count). Use 'full' / None."
|
| 252 |
+
)
|
| 253 |
+
image_grid_pinpoints = self.image_processor.image_grid_pinpoints
|
| 254 |
+
height_best, width_best = select_best_resolution([orig_height, orig_width], image_grid_pinpoints)
|
| 255 |
+
scale_height, scale_width = height_best // height, width_best // width
|
| 256 |
+
patches = int(math.sqrt(self.num_image_tokens))
|
| 257 |
+
_, newline_features = self._get_unpadded_features(
|
| 258 |
+
orig_height, orig_width, patches, patches, scale_height, scale_width
|
| 259 |
+
)
|
| 260 |
+
n -= newline_features
|
| 261 |
+
return n
|
| 262 |
+
|
| 263 |
+
def _expand_image_tokens(self, text, image_sizes, height, width, special_token, batch_num_images):
|
| 264 |
+
"""image_mrope=1 multi-image fix: LlavaOV expands each image of a MULTI-image sample as
|
| 265 |
+
`num_image_tokens + 1` (base 576 + one image_newline), bypassing `_get_number_of_features`
|
| 266 |
+
(HF processing_llava_onevision.py). But under image_mrope the model strips image_newline
|
| 267 |
+
(pack_image_features(image_newline=None)) → 576 features/image, so the default +1 over-counts
|
| 268 |
+
by one per image → masked_scatter token/feature mismatch at multi-image eval (e.g. MMMU).
|
| 269 |
+
This override drops the +1 ONLY for image_mrope multi-image; single-image still routes
|
| 270 |
+
through `_get_number_of_features` (AnyRes, newline already stripped there). flag 0 →
|
| 271 |
+
byte-identical to the parent. Verbatim from LlavaOnevisionProcessor._expand_image_tokens
|
| 272 |
+
EXCEPT the gated multi-image count line."""
|
| 273 |
+
image_mrope = video_is_image_mrope(self)
|
| 274 |
+
prompt_strings = []
|
| 275 |
+
max_num_vision_tokens = 0
|
| 276 |
+
for sample in text:
|
| 277 |
+
if special_token in sample:
|
| 278 |
+
num_images = next(batch_num_images) # should consume iterable
|
| 279 |
+
is_multi_image = num_images != 1
|
| 280 |
+
else:
|
| 281 |
+
is_multi_image = False
|
| 282 |
+
while special_token in sample:
|
| 283 |
+
original_size = next(image_sizes) # should consume iterable
|
| 284 |
+
if is_multi_image:
|
| 285 |
+
# ── ONLY deviation: drop the +1 newline under image_mrope (base-only 576).
|
| 286 |
+
num_image_tokens = self.num_image_tokens if image_mrope else self.num_image_tokens + 1
|
| 287 |
+
else:
|
| 288 |
+
if not isinstance(original_size, (list, tuple)):
|
| 289 |
+
# cast to list to avoid numerical precision errors when calculating unpadding
|
| 290 |
+
original_size = original_size.tolist()
|
| 291 |
+
orig_height, orig_width = original_size
|
| 292 |
+
num_image_tokens = self._get_number_of_features(orig_height, orig_width, height, width)
|
| 293 |
+
max_num_vision_tokens = max(max_num_vision_tokens, num_image_tokens)
|
| 294 |
+
if self.vision_feature_select_strategy == "default":
|
| 295 |
+
num_image_tokens -= 1
|
| 296 |
+
sample = sample.replace(special_token, "<placeholder>" * num_image_tokens, 1)
|
| 297 |
+
prompt_strings.append(sample)
|
| 298 |
+
text = [sample.replace("<placeholder>", special_token) for sample in prompt_strings]
|
| 299 |
+
return text, max_num_vision_tokens
|
| 300 |
+
|
| 301 |
+
|
| 302 |
+
__all__ = ["LlavaQwen3_5ViTProcessor"]
|
ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/processor_config.json
ADDED
|
@@ -0,0 +1,209 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"image_processor": {
|
| 3 |
+
"do_convert_rgb": true,
|
| 4 |
+
"do_normalize": true,
|
| 5 |
+
"do_pad": true,
|
| 6 |
+
"do_rescale": true,
|
| 7 |
+
"do_resize": true,
|
| 8 |
+
"image_grid_pinpoints": [
|
| 9 |
+
[
|
| 10 |
+
384,
|
| 11 |
+
384
|
| 12 |
+
],
|
| 13 |
+
[
|
| 14 |
+
384,
|
| 15 |
+
768
|
| 16 |
+
],
|
| 17 |
+
[
|
| 18 |
+
384,
|
| 19 |
+
1152
|
| 20 |
+
],
|
| 21 |
+
[
|
| 22 |
+
384,
|
| 23 |
+
1536
|
| 24 |
+
],
|
| 25 |
+
[
|
| 26 |
+
384,
|
| 27 |
+
1920
|
| 28 |
+
],
|
| 29 |
+
[
|
| 30 |
+
384,
|
| 31 |
+
2304
|
| 32 |
+
],
|
| 33 |
+
[
|
| 34 |
+
768,
|
| 35 |
+
384
|
| 36 |
+
],
|
| 37 |
+
[
|
| 38 |
+
768,
|
| 39 |
+
768
|
| 40 |
+
],
|
| 41 |
+
[
|
| 42 |
+
768,
|
| 43 |
+
1152
|
| 44 |
+
],
|
| 45 |
+
[
|
| 46 |
+
768,
|
| 47 |
+
1536
|
| 48 |
+
],
|
| 49 |
+
[
|
| 50 |
+
768,
|
| 51 |
+
1920
|
| 52 |
+
],
|
| 53 |
+
[
|
| 54 |
+
768,
|
| 55 |
+
2304
|
| 56 |
+
],
|
| 57 |
+
[
|
| 58 |
+
1152,
|
| 59 |
+
384
|
| 60 |
+
],
|
| 61 |
+
[
|
| 62 |
+
1152,
|
| 63 |
+
768
|
| 64 |
+
],
|
| 65 |
+
[
|
| 66 |
+
1152,
|
| 67 |
+
1152
|
| 68 |
+
],
|
| 69 |
+
[
|
| 70 |
+
1152,
|
| 71 |
+
1536
|
| 72 |
+
],
|
| 73 |
+
[
|
| 74 |
+
1152,
|
| 75 |
+
1920
|
| 76 |
+
],
|
| 77 |
+
[
|
| 78 |
+
1152,
|
| 79 |
+
2304
|
| 80 |
+
],
|
| 81 |
+
[
|
| 82 |
+
1536,
|
| 83 |
+
384
|
| 84 |
+
],
|
| 85 |
+
[
|
| 86 |
+
1536,
|
| 87 |
+
768
|
| 88 |
+
],
|
| 89 |
+
[
|
| 90 |
+
1536,
|
| 91 |
+
1152
|
| 92 |
+
],
|
| 93 |
+
[
|
| 94 |
+
1536,
|
| 95 |
+
1536
|
| 96 |
+
],
|
| 97 |
+
[
|
| 98 |
+
1536,
|
| 99 |
+
1920
|
| 100 |
+
],
|
| 101 |
+
[
|
| 102 |
+
1536,
|
| 103 |
+
2304
|
| 104 |
+
],
|
| 105 |
+
[
|
| 106 |
+
1920,
|
| 107 |
+
384
|
| 108 |
+
],
|
| 109 |
+
[
|
| 110 |
+
1920,
|
| 111 |
+
768
|
| 112 |
+
],
|
| 113 |
+
[
|
| 114 |
+
1920,
|
| 115 |
+
1152
|
| 116 |
+
],
|
| 117 |
+
[
|
| 118 |
+
1920,
|
| 119 |
+
1536
|
| 120 |
+
],
|
| 121 |
+
[
|
| 122 |
+
1920,
|
| 123 |
+
1920
|
| 124 |
+
],
|
| 125 |
+
[
|
| 126 |
+
1920,
|
| 127 |
+
2304
|
| 128 |
+
],
|
| 129 |
+
[
|
| 130 |
+
2304,
|
| 131 |
+
384
|
| 132 |
+
],
|
| 133 |
+
[
|
| 134 |
+
2304,
|
| 135 |
+
768
|
| 136 |
+
],
|
| 137 |
+
[
|
| 138 |
+
2304,
|
| 139 |
+
1152
|
| 140 |
+
],
|
| 141 |
+
[
|
| 142 |
+
2304,
|
| 143 |
+
1536
|
| 144 |
+
],
|
| 145 |
+
[
|
| 146 |
+
2304,
|
| 147 |
+
1920
|
| 148 |
+
],
|
| 149 |
+
[
|
| 150 |
+
2304,
|
| 151 |
+
2304
|
| 152 |
+
]
|
| 153 |
+
],
|
| 154 |
+
"image_mean": [
|
| 155 |
+
0.5,
|
| 156 |
+
0.5,
|
| 157 |
+
0.5
|
| 158 |
+
],
|
| 159 |
+
"image_processor_type": "LlavaOnevisionImageProcessor",
|
| 160 |
+
"image_std": [
|
| 161 |
+
0.5,
|
| 162 |
+
0.5,
|
| 163 |
+
0.5
|
| 164 |
+
],
|
| 165 |
+
"resample": 3,
|
| 166 |
+
"rescale_factor": 0.00392156862745098,
|
| 167 |
+
"size": {
|
| 168 |
+
"height": 384,
|
| 169 |
+
"width": 384
|
| 170 |
+
}
|
| 171 |
+
},
|
| 172 |
+
"image_token": "<image>",
|
| 173 |
+
"num_image_tokens": 576,
|
| 174 |
+
"processor_class": "LlavaQwen3_5ViTProcessor",
|
| 175 |
+
"video_mode": "temporal_merge",
|
| 176 |
+
"video_processor": {
|
| 177 |
+
"do_convert_rgb": true,
|
| 178 |
+
"do_normalize": true,
|
| 179 |
+
"do_rescale": true,
|
| 180 |
+
"do_resize": true,
|
| 181 |
+
"do_sample_frames": false,
|
| 182 |
+
"image_mean": [
|
| 183 |
+
0.5,
|
| 184 |
+
0.5,
|
| 185 |
+
0.5
|
| 186 |
+
],
|
| 187 |
+
"image_std": [
|
| 188 |
+
0.5,
|
| 189 |
+
0.5,
|
| 190 |
+
0.5
|
| 191 |
+
],
|
| 192 |
+
"resample": 3,
|
| 193 |
+
"rescale_factor": 0.00392156862745098,
|
| 194 |
+
"return_metadata": false,
|
| 195 |
+
"size": {
|
| 196 |
+
"height": 384,
|
| 197 |
+
"width": 384
|
| 198 |
+
},
|
| 199 |
+
"video_processor_type": "LlavaOnevisionVideoProcessor"
|
| 200 |
+
},
|
| 201 |
+
"video_token": "<video>",
|
| 202 |
+
"vision_aspect_ratio": "anyres_max_9",
|
| 203 |
+
"vision_feature_select_strategy": null,
|
| 204 |
+
"vtc_compression_factor": 2,
|
| 205 |
+
"vtc_temporal_patch_size": 1,
|
| 206 |
+
"auto_map": {
|
| 207 |
+
"AutoProcessor": "processing_qwen3_5vit_qwen3.LlavaQwen3_5ViTProcessor"
|
| 208 |
+
}
|
| 209 |
+
}
|
ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9024318c850eaadf26be79389d21b07a7afd8f1af749b89f9109b06c0d12173c
|
| 3 |
+
size 11423018
|
ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/tokenizer_config.json
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": null,
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"extra_special_tokens": [
|
| 9 |
+
"<image>",
|
| 10 |
+
"<video>"
|
| 11 |
+
],
|
| 12 |
+
"is_local": true,
|
| 13 |
+
"model_max_length": 1010000,
|
| 14 |
+
"pad_token": "<|endoftext|>",
|
| 15 |
+
"processor_class": "LlavaQwen3_5ViTProcessor",
|
| 16 |
+
"split_special_tokens": false,
|
| 17 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 18 |
+
"unk_token": null
|
| 19 |
+
}
|
ckpts/S2/4b/qwen3_5_2b/baseline_tmerge_64f_20pct/v0-20261012-171609/checkpoint-1530/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/chat_template.jinja
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{%- if tools %}
|
| 2 |
+
{{- '<|im_start|>system\n' }}
|
| 3 |
+
{%- if messages[0].role == 'system' %}
|
| 4 |
+
{{- messages[0].content + '\n\n' }}
|
| 5 |
+
{%- endif %}
|
| 6 |
+
{{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
|
| 7 |
+
{%- for tool in tools %}
|
| 8 |
+
{{- "\n" }}
|
| 9 |
+
{{- tool | tojson }}
|
| 10 |
+
{%- endfor %}
|
| 11 |
+
{{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
|
| 12 |
+
{%- else %}
|
| 13 |
+
{%- if messages[0].role == 'system' %}
|
| 14 |
+
{{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
|
| 15 |
+
{%- endif %}
|
| 16 |
+
{%- endif %}
|
| 17 |
+
{%- for message in messages %}
|
| 18 |
+
{%- if message.content is string %}
|
| 19 |
+
{%- set content = message.content %}
|
| 20 |
+
{%- else %}
|
| 21 |
+
{%- set content = '' %}
|
| 22 |
+
{%- endif %}
|
| 23 |
+
{%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
|
| 24 |
+
{{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
|
| 25 |
+
{%- elif message.role == "assistant" %}
|
| 26 |
+
{{- '<|im_start|>' + message.role + '\n' + content }}
|
| 27 |
+
{%- if message.tool_calls %}
|
| 28 |
+
{%- for tool_call in message.tool_calls %}
|
| 29 |
+
{%- if (loop.first and content) or (not loop.first) %}
|
| 30 |
+
{{- '\n' }}
|
| 31 |
+
{%- endif %}
|
| 32 |
+
{%- if tool_call.function %}
|
| 33 |
+
{%- set tool_call = tool_call.function %}
|
| 34 |
+
{%- endif %}
|
| 35 |
+
{{- '<tool_call>\n{"name": "' }}
|
| 36 |
+
{{- tool_call.name }}
|
| 37 |
+
{{- '", "arguments": ' }}
|
| 38 |
+
{%- if tool_call.arguments is string %}
|
| 39 |
+
{{- tool_call.arguments }}
|
| 40 |
+
{%- else %}
|
| 41 |
+
{{- tool_call.arguments | tojson }}
|
| 42 |
+
{%- endif %}
|
| 43 |
+
{{- '}\n</tool_call>' }}
|
| 44 |
+
{%- endfor %}
|
| 45 |
+
{%- endif %}
|
| 46 |
+
{{- '<|im_end|>\n' }}
|
| 47 |
+
{%- elif message.role == "tool" %}
|
| 48 |
+
{%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
|
| 49 |
+
{{- '<|im_start|>user' }}
|
| 50 |
+
{%- endif %}
|
| 51 |
+
{{- '\n<tool_response>\n' }}
|
| 52 |
+
{{- content }}
|
| 53 |
+
{{- '\n</tool_response>' }}
|
| 54 |
+
{%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
|
| 55 |
+
{{- '<|im_end|>\n' }}
|
| 56 |
+
{%- endif %}
|
| 57 |
+
{%- endif %}
|
| 58 |
+
{%- endfor %}
|
| 59 |
+
{%- if add_generation_prompt %}
|
| 60 |
+
{{- '<|im_start|>assistant\n' }}
|
| 61 |
+
{%- endif %}
|
ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/chat_template.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + ' '}}{# Render all images first #}{% for content in message['content'] | selectattr('type', 'equalto', 'image') %}{{ '<image>' }}{% endfor %}{# Render all video then #}{% for content in message['content'] | selectattr('type', 'equalto', 'video') %}{{ '<video>' }}{% endfor %}{# Render all text next #}{% if message['role'] != 'assistant' %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{{ '\n' + content['text'] }}{% endfor %}{% else %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{% generation %}{{ '\n' + content['text'] }}{% endgeneration %}{% endfor %}{% endif %}{{'<|im_end|>'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}"
|
| 3 |
+
}
|
ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/config.json
ADDED
|
@@ -0,0 +1,290 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlavaQwen3_5ViTForConditionalGeneration"
|
| 4 |
+
],
|
| 5 |
+
"bos_token_id": null,
|
| 6 |
+
"dtype": "bfloat16",
|
| 7 |
+
"eos_token_id": 151645,
|
| 8 |
+
"eval_num_frames": 64,
|
| 9 |
+
"hidden_size": 2560,
|
| 10 |
+
"image_grid_pinpoints": [
|
| 11 |
+
[
|
| 12 |
+
384,
|
| 13 |
+
384
|
| 14 |
+
],
|
| 15 |
+
[
|
| 16 |
+
384,
|
| 17 |
+
768
|
| 18 |
+
],
|
| 19 |
+
[
|
| 20 |
+
384,
|
| 21 |
+
1152
|
| 22 |
+
],
|
| 23 |
+
[
|
| 24 |
+
384,
|
| 25 |
+
1536
|
| 26 |
+
],
|
| 27 |
+
[
|
| 28 |
+
384,
|
| 29 |
+
1920
|
| 30 |
+
],
|
| 31 |
+
[
|
| 32 |
+
384,
|
| 33 |
+
2304
|
| 34 |
+
],
|
| 35 |
+
[
|
| 36 |
+
768,
|
| 37 |
+
384
|
| 38 |
+
],
|
| 39 |
+
[
|
| 40 |
+
768,
|
| 41 |
+
768
|
| 42 |
+
],
|
| 43 |
+
[
|
| 44 |
+
768,
|
| 45 |
+
1152
|
| 46 |
+
],
|
| 47 |
+
[
|
| 48 |
+
768,
|
| 49 |
+
1536
|
| 50 |
+
],
|
| 51 |
+
[
|
| 52 |
+
768,
|
| 53 |
+
1920
|
| 54 |
+
],
|
| 55 |
+
[
|
| 56 |
+
768,
|
| 57 |
+
2304
|
| 58 |
+
],
|
| 59 |
+
[
|
| 60 |
+
1152,
|
| 61 |
+
384
|
| 62 |
+
],
|
| 63 |
+
[
|
| 64 |
+
1152,
|
| 65 |
+
768
|
| 66 |
+
],
|
| 67 |
+
[
|
| 68 |
+
1152,
|
| 69 |
+
1152
|
| 70 |
+
],
|
| 71 |
+
[
|
| 72 |
+
1152,
|
| 73 |
+
1536
|
| 74 |
+
],
|
| 75 |
+
[
|
| 76 |
+
1152,
|
| 77 |
+
1920
|
| 78 |
+
],
|
| 79 |
+
[
|
| 80 |
+
1152,
|
| 81 |
+
2304
|
| 82 |
+
],
|
| 83 |
+
[
|
| 84 |
+
1536,
|
| 85 |
+
384
|
| 86 |
+
],
|
| 87 |
+
[
|
| 88 |
+
1536,
|
| 89 |
+
768
|
| 90 |
+
],
|
| 91 |
+
[
|
| 92 |
+
1536,
|
| 93 |
+
1152
|
| 94 |
+
],
|
| 95 |
+
[
|
| 96 |
+
1536,
|
| 97 |
+
1536
|
| 98 |
+
],
|
| 99 |
+
[
|
| 100 |
+
1536,
|
| 101 |
+
1920
|
| 102 |
+
],
|
| 103 |
+
[
|
| 104 |
+
1536,
|
| 105 |
+
2304
|
| 106 |
+
],
|
| 107 |
+
[
|
| 108 |
+
1920,
|
| 109 |
+
384
|
| 110 |
+
],
|
| 111 |
+
[
|
| 112 |
+
1920,
|
| 113 |
+
768
|
| 114 |
+
],
|
| 115 |
+
[
|
| 116 |
+
1920,
|
| 117 |
+
1152
|
| 118 |
+
],
|
| 119 |
+
[
|
| 120 |
+
1920,
|
| 121 |
+
1536
|
| 122 |
+
],
|
| 123 |
+
[
|
| 124 |
+
1920,
|
| 125 |
+
1920
|
| 126 |
+
],
|
| 127 |
+
[
|
| 128 |
+
1920,
|
| 129 |
+
2304
|
| 130 |
+
],
|
| 131 |
+
[
|
| 132 |
+
2304,
|
| 133 |
+
384
|
| 134 |
+
],
|
| 135 |
+
[
|
| 136 |
+
2304,
|
| 137 |
+
768
|
| 138 |
+
],
|
| 139 |
+
[
|
| 140 |
+
2304,
|
| 141 |
+
1152
|
| 142 |
+
],
|
| 143 |
+
[
|
| 144 |
+
2304,
|
| 145 |
+
1536
|
| 146 |
+
],
|
| 147 |
+
[
|
| 148 |
+
2304,
|
| 149 |
+
1920
|
| 150 |
+
],
|
| 151 |
+
[
|
| 152 |
+
2304,
|
| 153 |
+
2304
|
| 154 |
+
]
|
| 155 |
+
],
|
| 156 |
+
"image_token_index": 151669,
|
| 157 |
+
"model_type": "llava_qwen3_5vit_qwen3",
|
| 158 |
+
"multimodal_projector_bias": true,
|
| 159 |
+
"pad_token_id": 151643,
|
| 160 |
+
"pair_fusion_clip_selection": "",
|
| 161 |
+
"pair_fusion_lr": 0.0001,
|
| 162 |
+
"pair_fusion_motion_cache_format": "",
|
| 163 |
+
"pair_fusion_motion_cache_version": 0,
|
| 164 |
+
"pair_fusion_motion_record_count": 0,
|
| 165 |
+
"pair_fusion_pretraining": "",
|
| 166 |
+
"pair_fusion_pretraining_step": 0,
|
| 167 |
+
"projector_hidden_act": "gelu",
|
| 168 |
+
"shared_image_temporal_refiner": false,
|
| 169 |
+
"temporal_refiner_lr": 0.0001,
|
| 170 |
+
"temporal_refiner_variant": "attention_only",
|
| 171 |
+
"temporal_refiner_window_groups": 4,
|
| 172 |
+
"text_config": {
|
| 173 |
+
"_name_or_path": "/share/m2v_intern_v3/wangjunjie09/model_cache/huggingface/Qwen/Qwen3-4B-Instruct-2507",
|
| 174 |
+
"architectures": [
|
| 175 |
+
"Qwen3ForCausalLM"
|
| 176 |
+
],
|
| 177 |
+
"attention_bias": false,
|
| 178 |
+
"attention_dropout": 0.0,
|
| 179 |
+
"bos_token_id": 151643,
|
| 180 |
+
"dtype": "bfloat16",
|
| 181 |
+
"eos_token_id": 151645,
|
| 182 |
+
"head_dim": 128,
|
| 183 |
+
"hidden_act": "silu",
|
| 184 |
+
"hidden_size": 2560,
|
| 185 |
+
"initializer_range": 0.02,
|
| 186 |
+
"intermediate_size": 9728,
|
| 187 |
+
"layer_types": [
|
| 188 |
+
"full_attention",
|
| 189 |
+
"full_attention",
|
| 190 |
+
"full_attention",
|
| 191 |
+
"full_attention",
|
| 192 |
+
"full_attention",
|
| 193 |
+
"full_attention",
|
| 194 |
+
"full_attention",
|
| 195 |
+
"full_attention",
|
| 196 |
+
"full_attention",
|
| 197 |
+
"full_attention",
|
| 198 |
+
"full_attention",
|
| 199 |
+
"full_attention",
|
| 200 |
+
"full_attention",
|
| 201 |
+
"full_attention",
|
| 202 |
+
"full_attention",
|
| 203 |
+
"full_attention",
|
| 204 |
+
"full_attention",
|
| 205 |
+
"full_attention",
|
| 206 |
+
"full_attention",
|
| 207 |
+
"full_attention",
|
| 208 |
+
"full_attention",
|
| 209 |
+
"full_attention",
|
| 210 |
+
"full_attention",
|
| 211 |
+
"full_attention",
|
| 212 |
+
"full_attention",
|
| 213 |
+
"full_attention",
|
| 214 |
+
"full_attention",
|
| 215 |
+
"full_attention",
|
| 216 |
+
"full_attention",
|
| 217 |
+
"full_attention",
|
| 218 |
+
"full_attention",
|
| 219 |
+
"full_attention",
|
| 220 |
+
"full_attention",
|
| 221 |
+
"full_attention",
|
| 222 |
+
"full_attention",
|
| 223 |
+
"full_attention"
|
| 224 |
+
],
|
| 225 |
+
"max_position_embeddings": 262144,
|
| 226 |
+
"max_window_layers": 36,
|
| 227 |
+
"model_type": "qwen3",
|
| 228 |
+
"num_attention_heads": 32,
|
| 229 |
+
"num_hidden_layers": 36,
|
| 230 |
+
"num_key_value_heads": 8,
|
| 231 |
+
"pad_token_id": 151643,
|
| 232 |
+
"rms_norm_eps": 1e-06,
|
| 233 |
+
"rope_parameters": {
|
| 234 |
+
"rope_theta": 5000000,
|
| 235 |
+
"rope_type": "default"
|
| 236 |
+
},
|
| 237 |
+
"sliding_window": null,
|
| 238 |
+
"tie_word_embeddings": true,
|
| 239 |
+
"use_cache": false,
|
| 240 |
+
"use_sliding_window": false,
|
| 241 |
+
"vocab_size": 151936
|
| 242 |
+
},
|
| 243 |
+
"tie_word_embeddings": true,
|
| 244 |
+
"train_num_frames": 64,
|
| 245 |
+
"transformers_version": "5.5.4",
|
| 246 |
+
"use_cache": false,
|
| 247 |
+
"video_mode": "vtc_compression",
|
| 248 |
+
"video_temporal_refiner": 0,
|
| 249 |
+
"video_token_index": 151670,
|
| 250 |
+
"vision_aspect_ratio": "anyres_max_9",
|
| 251 |
+
"vision_config": {
|
| 252 |
+
"deepstack_visual_indexes": [],
|
| 253 |
+
"depth": 24,
|
| 254 |
+
"dtype": "bfloat16",
|
| 255 |
+
"hidden_act": "gelu_pytorch_tanh",
|
| 256 |
+
"hidden_size": 1024,
|
| 257 |
+
"image_size": 384,
|
| 258 |
+
"in_channels": 3,
|
| 259 |
+
"initializer_range": 0.02,
|
| 260 |
+
"intermediate_size": 4096,
|
| 261 |
+
"model_type": "qwen3_5",
|
| 262 |
+
"num_heads": 16,
|
| 263 |
+
"num_position_embeddings": 2304,
|
| 264 |
+
"out_hidden_size": 2048,
|
| 265 |
+
"patch_size": 16,
|
| 266 |
+
"spatial_merge_size": 2,
|
| 267 |
+
"temporal_patch_size": 2,
|
| 268 |
+
"vtc_compressor_type": "vtc_x2",
|
| 269 |
+
"vtc_compressor_variant": "baseline",
|
| 270 |
+
"vtc_feature_fusion_layers": [],
|
| 271 |
+
"vtc_temporal_attention_window": 2
|
| 272 |
+
},
|
| 273 |
+
"vision_feature_layer": -1,
|
| 274 |
+
"vision_feature_select_strategy": "full",
|
| 275 |
+
"vra_layer": 18,
|
| 276 |
+
"vra_loss": 0,
|
| 277 |
+
"vra_scope": "image_video",
|
| 278 |
+
"vra_target_path": "/share/m2v_intern_v3/wangjunjie09/model_cache/huggingface/facebook/dinov3-vitl16-pretrain-lvd1689m",
|
| 279 |
+
"vra_teacher_kind": "dino",
|
| 280 |
+
"vra_teacher_res": 384,
|
| 281 |
+
"vra_video_loss": "direct",
|
| 282 |
+
"vra_weight": 0.5,
|
| 283 |
+
"vra_z_dim": 1024,
|
| 284 |
+
"auto_map": {
|
| 285 |
+
"AutoConfig": "modeling_qwen3_5vit_qwen3.LlavaQwen3_5ViTConfig",
|
| 286 |
+
"AutoModel": "modeling_qwen3_5vit_qwen3.LlavaQwen3_5ViTForConditionalGeneration",
|
| 287 |
+
"AutoModelForCausalLM": "modeling_qwen3_5vit_qwen3.LlavaQwen3_5ViTForConditionalGeneration",
|
| 288 |
+
"AutoModelForImageTextToText": "modeling_qwen3_5vit_qwen3.LlavaQwen3_5ViTForConditionalGeneration"
|
| 289 |
+
}
|
| 290 |
+
}
|
ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 151643,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"output_attentions": false,
|
| 9 |
+
"output_hidden_states": false,
|
| 10 |
+
"transformers_version": "5.5.4",
|
| 11 |
+
"use_cache": true
|
| 12 |
+
}
|
ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/model-00001-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:809a8b737804e078596061cd0c5e9ab4200c232304951f7e0810b6ec7629e066
|
| 3 |
+
size 4987507408
|
ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/model-00002-of-00002.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:71e920d3075a38be3570357e191230633daa8dac5a2b4015ee44660b673d5d33
|
| 3 |
+
size 4487289528
|
ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/model.safetensors.index.json
ADDED
|
@@ -0,0 +1,720 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_parameters": 4348396544,
|
| 4 |
+
"total_size": 9474705408
|
| 5 |
+
},
|
| 6 |
+
"weight_map": {
|
| 7 |
+
"lm_head.weight": "model-00001-of-00002.safetensors",
|
| 8 |
+
"model.image_newline": "model-00001-of-00002.safetensors",
|
| 9 |
+
"model.language_model.embed_tokens.weight": "model-00001-of-00002.safetensors",
|
| 10 |
+
"model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 11 |
+
"model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 12 |
+
"model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 13 |
+
"model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 14 |
+
"model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 15 |
+
"model.language_model.layers.0.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 16 |
+
"model.language_model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 17 |
+
"model.language_model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 18 |
+
"model.language_model.layers.0.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 19 |
+
"model.language_model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 20 |
+
"model.language_model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 21 |
+
"model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 22 |
+
"model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 23 |
+
"model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 24 |
+
"model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 25 |
+
"model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 26 |
+
"model.language_model.layers.1.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 27 |
+
"model.language_model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 28 |
+
"model.language_model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 29 |
+
"model.language_model.layers.1.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 30 |
+
"model.language_model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 31 |
+
"model.language_model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 32 |
+
"model.language_model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 33 |
+
"model.language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 34 |
+
"model.language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 35 |
+
"model.language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 36 |
+
"model.language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 37 |
+
"model.language_model.layers.10.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 38 |
+
"model.language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 39 |
+
"model.language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 40 |
+
"model.language_model.layers.10.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 41 |
+
"model.language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 42 |
+
"model.language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 43 |
+
"model.language_model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 44 |
+
"model.language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 45 |
+
"model.language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 46 |
+
"model.language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 47 |
+
"model.language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 48 |
+
"model.language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 49 |
+
"model.language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 50 |
+
"model.language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 51 |
+
"model.language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 52 |
+
"model.language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 53 |
+
"model.language_model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 54 |
+
"model.language_model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 55 |
+
"model.language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 56 |
+
"model.language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 57 |
+
"model.language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 58 |
+
"model.language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 59 |
+
"model.language_model.layers.12.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 60 |
+
"model.language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 61 |
+
"model.language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 62 |
+
"model.language_model.layers.12.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 63 |
+
"model.language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 64 |
+
"model.language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 65 |
+
"model.language_model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 66 |
+
"model.language_model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 67 |
+
"model.language_model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 68 |
+
"model.language_model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 69 |
+
"model.language_model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 70 |
+
"model.language_model.layers.13.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 71 |
+
"model.language_model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 72 |
+
"model.language_model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 73 |
+
"model.language_model.layers.13.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 74 |
+
"model.language_model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 75 |
+
"model.language_model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 76 |
+
"model.language_model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 77 |
+
"model.language_model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 78 |
+
"model.language_model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 79 |
+
"model.language_model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 80 |
+
"model.language_model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 81 |
+
"model.language_model.layers.14.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 82 |
+
"model.language_model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 83 |
+
"model.language_model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 84 |
+
"model.language_model.layers.14.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 85 |
+
"model.language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 86 |
+
"model.language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 87 |
+
"model.language_model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 88 |
+
"model.language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 89 |
+
"model.language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 90 |
+
"model.language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 91 |
+
"model.language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 92 |
+
"model.language_model.layers.15.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 93 |
+
"model.language_model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 94 |
+
"model.language_model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 95 |
+
"model.language_model.layers.15.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 96 |
+
"model.language_model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 97 |
+
"model.language_model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 98 |
+
"model.language_model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 99 |
+
"model.language_model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 100 |
+
"model.language_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 101 |
+
"model.language_model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 102 |
+
"model.language_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 103 |
+
"model.language_model.layers.16.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 104 |
+
"model.language_model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 105 |
+
"model.language_model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 106 |
+
"model.language_model.layers.16.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 107 |
+
"model.language_model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 108 |
+
"model.language_model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 109 |
+
"model.language_model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 110 |
+
"model.language_model.layers.17.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 111 |
+
"model.language_model.layers.17.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 112 |
+
"model.language_model.layers.17.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 113 |
+
"model.language_model.layers.17.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 114 |
+
"model.language_model.layers.17.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 115 |
+
"model.language_model.layers.17.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 116 |
+
"model.language_model.layers.17.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 117 |
+
"model.language_model.layers.17.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 118 |
+
"model.language_model.layers.17.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 119 |
+
"model.language_model.layers.17.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 120 |
+
"model.language_model.layers.18.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 121 |
+
"model.language_model.layers.18.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 122 |
+
"model.language_model.layers.18.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 123 |
+
"model.language_model.layers.18.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 124 |
+
"model.language_model.layers.18.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 125 |
+
"model.language_model.layers.18.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 126 |
+
"model.language_model.layers.18.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 127 |
+
"model.language_model.layers.18.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 128 |
+
"model.language_model.layers.18.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 129 |
+
"model.language_model.layers.18.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 130 |
+
"model.language_model.layers.18.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 131 |
+
"model.language_model.layers.19.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 132 |
+
"model.language_model.layers.19.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 133 |
+
"model.language_model.layers.19.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 134 |
+
"model.language_model.layers.19.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 135 |
+
"model.language_model.layers.19.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 136 |
+
"model.language_model.layers.19.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 137 |
+
"model.language_model.layers.19.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 138 |
+
"model.language_model.layers.19.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 139 |
+
"model.language_model.layers.19.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 140 |
+
"model.language_model.layers.19.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 141 |
+
"model.language_model.layers.19.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 142 |
+
"model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 143 |
+
"model.language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 144 |
+
"model.language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 145 |
+
"model.language_model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 146 |
+
"model.language_model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 147 |
+
"model.language_model.layers.2.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 148 |
+
"model.language_model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 149 |
+
"model.language_model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 150 |
+
"model.language_model.layers.2.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 151 |
+
"model.language_model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 152 |
+
"model.language_model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 153 |
+
"model.language_model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 154 |
+
"model.language_model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 155 |
+
"model.language_model.layers.20.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 156 |
+
"model.language_model.layers.20.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 157 |
+
"model.language_model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 158 |
+
"model.language_model.layers.20.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 159 |
+
"model.language_model.layers.20.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 160 |
+
"model.language_model.layers.20.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 161 |
+
"model.language_model.layers.20.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 162 |
+
"model.language_model.layers.20.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 163 |
+
"model.language_model.layers.20.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 164 |
+
"model.language_model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 165 |
+
"model.language_model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 166 |
+
"model.language_model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 167 |
+
"model.language_model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 168 |
+
"model.language_model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 169 |
+
"model.language_model.layers.21.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 170 |
+
"model.language_model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 171 |
+
"model.language_model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 172 |
+
"model.language_model.layers.21.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 173 |
+
"model.language_model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 174 |
+
"model.language_model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 175 |
+
"model.language_model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 176 |
+
"model.language_model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 177 |
+
"model.language_model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 178 |
+
"model.language_model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 179 |
+
"model.language_model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 180 |
+
"model.language_model.layers.22.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 181 |
+
"model.language_model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 182 |
+
"model.language_model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 183 |
+
"model.language_model.layers.22.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 184 |
+
"model.language_model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 185 |
+
"model.language_model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 186 |
+
"model.language_model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 187 |
+
"model.language_model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 188 |
+
"model.language_model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 189 |
+
"model.language_model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 190 |
+
"model.language_model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 191 |
+
"model.language_model.layers.23.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 192 |
+
"model.language_model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 193 |
+
"model.language_model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 194 |
+
"model.language_model.layers.23.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 195 |
+
"model.language_model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 196 |
+
"model.language_model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 197 |
+
"model.language_model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 198 |
+
"model.language_model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 199 |
+
"model.language_model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 200 |
+
"model.language_model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 201 |
+
"model.language_model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 202 |
+
"model.language_model.layers.24.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 203 |
+
"model.language_model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 204 |
+
"model.language_model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 205 |
+
"model.language_model.layers.24.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 206 |
+
"model.language_model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 207 |
+
"model.language_model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 208 |
+
"model.language_model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 209 |
+
"model.language_model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 210 |
+
"model.language_model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 211 |
+
"model.language_model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 212 |
+
"model.language_model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 213 |
+
"model.language_model.layers.25.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 214 |
+
"model.language_model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 215 |
+
"model.language_model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 216 |
+
"model.language_model.layers.25.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 217 |
+
"model.language_model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 218 |
+
"model.language_model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 219 |
+
"model.language_model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 220 |
+
"model.language_model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 221 |
+
"model.language_model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 222 |
+
"model.language_model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 223 |
+
"model.language_model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 224 |
+
"model.language_model.layers.26.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 225 |
+
"model.language_model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 226 |
+
"model.language_model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 227 |
+
"model.language_model.layers.26.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 228 |
+
"model.language_model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 229 |
+
"model.language_model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 230 |
+
"model.language_model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 231 |
+
"model.language_model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 232 |
+
"model.language_model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 233 |
+
"model.language_model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 234 |
+
"model.language_model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 235 |
+
"model.language_model.layers.27.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 236 |
+
"model.language_model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 237 |
+
"model.language_model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 238 |
+
"model.language_model.layers.27.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 239 |
+
"model.language_model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 240 |
+
"model.language_model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 241 |
+
"model.language_model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 242 |
+
"model.language_model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 243 |
+
"model.language_model.layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 244 |
+
"model.language_model.layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 245 |
+
"model.language_model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 246 |
+
"model.language_model.layers.28.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 247 |
+
"model.language_model.layers.28.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 248 |
+
"model.language_model.layers.28.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 249 |
+
"model.language_model.layers.28.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 250 |
+
"model.language_model.layers.28.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 251 |
+
"model.language_model.layers.28.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 252 |
+
"model.language_model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 253 |
+
"model.language_model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 254 |
+
"model.language_model.layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 255 |
+
"model.language_model.layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 256 |
+
"model.language_model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 257 |
+
"model.language_model.layers.29.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 258 |
+
"model.language_model.layers.29.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 259 |
+
"model.language_model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 260 |
+
"model.language_model.layers.29.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 261 |
+
"model.language_model.layers.29.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 262 |
+
"model.language_model.layers.29.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 263 |
+
"model.language_model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 264 |
+
"model.language_model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 265 |
+
"model.language_model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 266 |
+
"model.language_model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 267 |
+
"model.language_model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 268 |
+
"model.language_model.layers.3.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 269 |
+
"model.language_model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 270 |
+
"model.language_model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 271 |
+
"model.language_model.layers.3.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 272 |
+
"model.language_model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 273 |
+
"model.language_model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 274 |
+
"model.language_model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 275 |
+
"model.language_model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 276 |
+
"model.language_model.layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 277 |
+
"model.language_model.layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 278 |
+
"model.language_model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 279 |
+
"model.language_model.layers.30.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 280 |
+
"model.language_model.layers.30.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 281 |
+
"model.language_model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 282 |
+
"model.language_model.layers.30.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 283 |
+
"model.language_model.layers.30.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 284 |
+
"model.language_model.layers.30.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 285 |
+
"model.language_model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 286 |
+
"model.language_model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 287 |
+
"model.language_model.layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 288 |
+
"model.language_model.layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 289 |
+
"model.language_model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 290 |
+
"model.language_model.layers.31.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 291 |
+
"model.language_model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 292 |
+
"model.language_model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 293 |
+
"model.language_model.layers.31.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 294 |
+
"model.language_model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 295 |
+
"model.language_model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 296 |
+
"model.language_model.layers.32.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 297 |
+
"model.language_model.layers.32.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 298 |
+
"model.language_model.layers.32.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 299 |
+
"model.language_model.layers.32.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 300 |
+
"model.language_model.layers.32.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 301 |
+
"model.language_model.layers.32.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 302 |
+
"model.language_model.layers.32.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 303 |
+
"model.language_model.layers.32.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 304 |
+
"model.language_model.layers.32.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 305 |
+
"model.language_model.layers.32.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 306 |
+
"model.language_model.layers.32.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 307 |
+
"model.language_model.layers.33.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 308 |
+
"model.language_model.layers.33.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 309 |
+
"model.language_model.layers.33.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 310 |
+
"model.language_model.layers.33.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 311 |
+
"model.language_model.layers.33.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 312 |
+
"model.language_model.layers.33.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 313 |
+
"model.language_model.layers.33.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 314 |
+
"model.language_model.layers.33.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 315 |
+
"model.language_model.layers.33.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 316 |
+
"model.language_model.layers.33.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 317 |
+
"model.language_model.layers.33.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 318 |
+
"model.language_model.layers.34.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 319 |
+
"model.language_model.layers.34.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 320 |
+
"model.language_model.layers.34.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 321 |
+
"model.language_model.layers.34.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 322 |
+
"model.language_model.layers.34.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 323 |
+
"model.language_model.layers.34.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 324 |
+
"model.language_model.layers.34.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 325 |
+
"model.language_model.layers.34.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 326 |
+
"model.language_model.layers.34.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 327 |
+
"model.language_model.layers.34.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 328 |
+
"model.language_model.layers.34.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 329 |
+
"model.language_model.layers.35.input_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 330 |
+
"model.language_model.layers.35.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
|
| 331 |
+
"model.language_model.layers.35.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
|
| 332 |
+
"model.language_model.layers.35.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
|
| 333 |
+
"model.language_model.layers.35.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 334 |
+
"model.language_model.layers.35.self_attn.k_norm.weight": "model-00002-of-00002.safetensors",
|
| 335 |
+
"model.language_model.layers.35.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
|
| 336 |
+
"model.language_model.layers.35.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
|
| 337 |
+
"model.language_model.layers.35.self_attn.q_norm.weight": "model-00002-of-00002.safetensors",
|
| 338 |
+
"model.language_model.layers.35.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
|
| 339 |
+
"model.language_model.layers.35.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
|
| 340 |
+
"model.language_model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 341 |
+
"model.language_model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 342 |
+
"model.language_model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 343 |
+
"model.language_model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 344 |
+
"model.language_model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 345 |
+
"model.language_model.layers.4.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 346 |
+
"model.language_model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 347 |
+
"model.language_model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 348 |
+
"model.language_model.layers.4.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 349 |
+
"model.language_model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 350 |
+
"model.language_model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 351 |
+
"model.language_model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 352 |
+
"model.language_model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 353 |
+
"model.language_model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 354 |
+
"model.language_model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 355 |
+
"model.language_model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 356 |
+
"model.language_model.layers.5.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 357 |
+
"model.language_model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 358 |
+
"model.language_model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 359 |
+
"model.language_model.layers.5.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 360 |
+
"model.language_model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 361 |
+
"model.language_model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 362 |
+
"model.language_model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 363 |
+
"model.language_model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 364 |
+
"model.language_model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 365 |
+
"model.language_model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 366 |
+
"model.language_model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 367 |
+
"model.language_model.layers.6.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 368 |
+
"model.language_model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 369 |
+
"model.language_model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 370 |
+
"model.language_model.layers.6.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 371 |
+
"model.language_model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 372 |
+
"model.language_model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 373 |
+
"model.language_model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 374 |
+
"model.language_model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 375 |
+
"model.language_model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 376 |
+
"model.language_model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 377 |
+
"model.language_model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 378 |
+
"model.language_model.layers.7.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 379 |
+
"model.language_model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 380 |
+
"model.language_model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 381 |
+
"model.language_model.layers.7.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 382 |
+
"model.language_model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 383 |
+
"model.language_model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 384 |
+
"model.language_model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 385 |
+
"model.language_model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 386 |
+
"model.language_model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 387 |
+
"model.language_model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 388 |
+
"model.language_model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 389 |
+
"model.language_model.layers.8.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 390 |
+
"model.language_model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 391 |
+
"model.language_model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 392 |
+
"model.language_model.layers.8.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 393 |
+
"model.language_model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 394 |
+
"model.language_model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 395 |
+
"model.language_model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 396 |
+
"model.language_model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
|
| 397 |
+
"model.language_model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
|
| 398 |
+
"model.language_model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
|
| 399 |
+
"model.language_model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
|
| 400 |
+
"model.language_model.layers.9.self_attn.k_norm.weight": "model-00001-of-00002.safetensors",
|
| 401 |
+
"model.language_model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
|
| 402 |
+
"model.language_model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
|
| 403 |
+
"model.language_model.layers.9.self_attn.q_norm.weight": "model-00001-of-00002.safetensors",
|
| 404 |
+
"model.language_model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
|
| 405 |
+
"model.language_model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
|
| 406 |
+
"model.language_model.norm.weight": "model-00002-of-00002.safetensors",
|
| 407 |
+
"model.multi_modal_projector.linear_1.bias": "model-00002-of-00002.safetensors",
|
| 408 |
+
"model.multi_modal_projector.linear_1.weight": "model-00002-of-00002.safetensors",
|
| 409 |
+
"model.multi_modal_projector.linear_2.bias": "model-00002-of-00002.safetensors",
|
| 410 |
+
"model.multi_modal_projector.linear_2.weight": "model-00002-of-00002.safetensors",
|
| 411 |
+
"model.vision_tower.vision.blocks.0.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 412 |
+
"model.vision_tower.vision.blocks.0.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 413 |
+
"model.vision_tower.vision.blocks.0.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 414 |
+
"model.vision_tower.vision.blocks.0.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 415 |
+
"model.vision_tower.vision.blocks.0.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 416 |
+
"model.vision_tower.vision.blocks.0.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 417 |
+
"model.vision_tower.vision.blocks.0.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 418 |
+
"model.vision_tower.vision.blocks.0.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 419 |
+
"model.vision_tower.vision.blocks.0.norm1.bias": "model-00002-of-00002.safetensors",
|
| 420 |
+
"model.vision_tower.vision.blocks.0.norm1.weight": "model-00002-of-00002.safetensors",
|
| 421 |
+
"model.vision_tower.vision.blocks.0.norm2.bias": "model-00002-of-00002.safetensors",
|
| 422 |
+
"model.vision_tower.vision.blocks.0.norm2.weight": "model-00002-of-00002.safetensors",
|
| 423 |
+
"model.vision_tower.vision.blocks.1.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 424 |
+
"model.vision_tower.vision.blocks.1.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 425 |
+
"model.vision_tower.vision.blocks.1.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 426 |
+
"model.vision_tower.vision.blocks.1.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 427 |
+
"model.vision_tower.vision.blocks.1.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 428 |
+
"model.vision_tower.vision.blocks.1.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 429 |
+
"model.vision_tower.vision.blocks.1.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 430 |
+
"model.vision_tower.vision.blocks.1.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 431 |
+
"model.vision_tower.vision.blocks.1.norm1.bias": "model-00002-of-00002.safetensors",
|
| 432 |
+
"model.vision_tower.vision.blocks.1.norm1.weight": "model-00002-of-00002.safetensors",
|
| 433 |
+
"model.vision_tower.vision.blocks.1.norm2.bias": "model-00002-of-00002.safetensors",
|
| 434 |
+
"model.vision_tower.vision.blocks.1.norm2.weight": "model-00002-of-00002.safetensors",
|
| 435 |
+
"model.vision_tower.vision.blocks.10.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 436 |
+
"model.vision_tower.vision.blocks.10.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 437 |
+
"model.vision_tower.vision.blocks.10.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 438 |
+
"model.vision_tower.vision.blocks.10.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 439 |
+
"model.vision_tower.vision.blocks.10.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 440 |
+
"model.vision_tower.vision.blocks.10.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 441 |
+
"model.vision_tower.vision.blocks.10.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 442 |
+
"model.vision_tower.vision.blocks.10.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 443 |
+
"model.vision_tower.vision.blocks.10.norm1.bias": "model-00002-of-00002.safetensors",
|
| 444 |
+
"model.vision_tower.vision.blocks.10.norm1.weight": "model-00002-of-00002.safetensors",
|
| 445 |
+
"model.vision_tower.vision.blocks.10.norm2.bias": "model-00002-of-00002.safetensors",
|
| 446 |
+
"model.vision_tower.vision.blocks.10.norm2.weight": "model-00002-of-00002.safetensors",
|
| 447 |
+
"model.vision_tower.vision.blocks.11.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 448 |
+
"model.vision_tower.vision.blocks.11.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 449 |
+
"model.vision_tower.vision.blocks.11.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 450 |
+
"model.vision_tower.vision.blocks.11.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 451 |
+
"model.vision_tower.vision.blocks.11.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 452 |
+
"model.vision_tower.vision.blocks.11.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 453 |
+
"model.vision_tower.vision.blocks.11.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 454 |
+
"model.vision_tower.vision.blocks.11.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 455 |
+
"model.vision_tower.vision.blocks.11.norm1.bias": "model-00002-of-00002.safetensors",
|
| 456 |
+
"model.vision_tower.vision.blocks.11.norm1.weight": "model-00002-of-00002.safetensors",
|
| 457 |
+
"model.vision_tower.vision.blocks.11.norm2.bias": "model-00002-of-00002.safetensors",
|
| 458 |
+
"model.vision_tower.vision.blocks.11.norm2.weight": "model-00002-of-00002.safetensors",
|
| 459 |
+
"model.vision_tower.vision.blocks.12.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 460 |
+
"model.vision_tower.vision.blocks.12.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 461 |
+
"model.vision_tower.vision.blocks.12.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 462 |
+
"model.vision_tower.vision.blocks.12.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 463 |
+
"model.vision_tower.vision.blocks.12.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 464 |
+
"model.vision_tower.vision.blocks.12.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 465 |
+
"model.vision_tower.vision.blocks.12.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 466 |
+
"model.vision_tower.vision.blocks.12.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 467 |
+
"model.vision_tower.vision.blocks.12.norm1.bias": "model-00002-of-00002.safetensors",
|
| 468 |
+
"model.vision_tower.vision.blocks.12.norm1.weight": "model-00002-of-00002.safetensors",
|
| 469 |
+
"model.vision_tower.vision.blocks.12.norm2.bias": "model-00002-of-00002.safetensors",
|
| 470 |
+
"model.vision_tower.vision.blocks.12.norm2.weight": "model-00002-of-00002.safetensors",
|
| 471 |
+
"model.vision_tower.vision.blocks.13.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 472 |
+
"model.vision_tower.vision.blocks.13.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 473 |
+
"model.vision_tower.vision.blocks.13.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 474 |
+
"model.vision_tower.vision.blocks.13.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 475 |
+
"model.vision_tower.vision.blocks.13.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 476 |
+
"model.vision_tower.vision.blocks.13.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 477 |
+
"model.vision_tower.vision.blocks.13.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 478 |
+
"model.vision_tower.vision.blocks.13.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 479 |
+
"model.vision_tower.vision.blocks.13.norm1.bias": "model-00002-of-00002.safetensors",
|
| 480 |
+
"model.vision_tower.vision.blocks.13.norm1.weight": "model-00002-of-00002.safetensors",
|
| 481 |
+
"model.vision_tower.vision.blocks.13.norm2.bias": "model-00002-of-00002.safetensors",
|
| 482 |
+
"model.vision_tower.vision.blocks.13.norm2.weight": "model-00002-of-00002.safetensors",
|
| 483 |
+
"model.vision_tower.vision.blocks.14.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 484 |
+
"model.vision_tower.vision.blocks.14.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 485 |
+
"model.vision_tower.vision.blocks.14.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 486 |
+
"model.vision_tower.vision.blocks.14.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 487 |
+
"model.vision_tower.vision.blocks.14.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 488 |
+
"model.vision_tower.vision.blocks.14.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 489 |
+
"model.vision_tower.vision.blocks.14.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 490 |
+
"model.vision_tower.vision.blocks.14.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 491 |
+
"model.vision_tower.vision.blocks.14.norm1.bias": "model-00002-of-00002.safetensors",
|
| 492 |
+
"model.vision_tower.vision.blocks.14.norm1.weight": "model-00002-of-00002.safetensors",
|
| 493 |
+
"model.vision_tower.vision.blocks.14.norm2.bias": "model-00002-of-00002.safetensors",
|
| 494 |
+
"model.vision_tower.vision.blocks.14.norm2.weight": "model-00002-of-00002.safetensors",
|
| 495 |
+
"model.vision_tower.vision.blocks.15.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 496 |
+
"model.vision_tower.vision.blocks.15.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 497 |
+
"model.vision_tower.vision.blocks.15.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 498 |
+
"model.vision_tower.vision.blocks.15.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 499 |
+
"model.vision_tower.vision.blocks.15.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 500 |
+
"model.vision_tower.vision.blocks.15.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 501 |
+
"model.vision_tower.vision.blocks.15.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 502 |
+
"model.vision_tower.vision.blocks.15.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 503 |
+
"model.vision_tower.vision.blocks.15.norm1.bias": "model-00002-of-00002.safetensors",
|
| 504 |
+
"model.vision_tower.vision.blocks.15.norm1.weight": "model-00002-of-00002.safetensors",
|
| 505 |
+
"model.vision_tower.vision.blocks.15.norm2.bias": "model-00002-of-00002.safetensors",
|
| 506 |
+
"model.vision_tower.vision.blocks.15.norm2.weight": "model-00002-of-00002.safetensors",
|
| 507 |
+
"model.vision_tower.vision.blocks.16.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 508 |
+
"model.vision_tower.vision.blocks.16.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 509 |
+
"model.vision_tower.vision.blocks.16.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 510 |
+
"model.vision_tower.vision.blocks.16.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 511 |
+
"model.vision_tower.vision.blocks.16.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 512 |
+
"model.vision_tower.vision.blocks.16.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 513 |
+
"model.vision_tower.vision.blocks.16.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 514 |
+
"model.vision_tower.vision.blocks.16.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 515 |
+
"model.vision_tower.vision.blocks.16.norm1.bias": "model-00002-of-00002.safetensors",
|
| 516 |
+
"model.vision_tower.vision.blocks.16.norm1.weight": "model-00002-of-00002.safetensors",
|
| 517 |
+
"model.vision_tower.vision.blocks.16.norm2.bias": "model-00002-of-00002.safetensors",
|
| 518 |
+
"model.vision_tower.vision.blocks.16.norm2.weight": "model-00002-of-00002.safetensors",
|
| 519 |
+
"model.vision_tower.vision.blocks.17.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 520 |
+
"model.vision_tower.vision.blocks.17.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 521 |
+
"model.vision_tower.vision.blocks.17.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 522 |
+
"model.vision_tower.vision.blocks.17.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 523 |
+
"model.vision_tower.vision.blocks.17.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 524 |
+
"model.vision_tower.vision.blocks.17.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 525 |
+
"model.vision_tower.vision.blocks.17.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 526 |
+
"model.vision_tower.vision.blocks.17.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 527 |
+
"model.vision_tower.vision.blocks.17.norm1.bias": "model-00002-of-00002.safetensors",
|
| 528 |
+
"model.vision_tower.vision.blocks.17.norm1.weight": "model-00002-of-00002.safetensors",
|
| 529 |
+
"model.vision_tower.vision.blocks.17.norm2.bias": "model-00002-of-00002.safetensors",
|
| 530 |
+
"model.vision_tower.vision.blocks.17.norm2.weight": "model-00002-of-00002.safetensors",
|
| 531 |
+
"model.vision_tower.vision.blocks.18.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 532 |
+
"model.vision_tower.vision.blocks.18.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 533 |
+
"model.vision_tower.vision.blocks.18.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 534 |
+
"model.vision_tower.vision.blocks.18.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 535 |
+
"model.vision_tower.vision.blocks.18.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 536 |
+
"model.vision_tower.vision.blocks.18.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 537 |
+
"model.vision_tower.vision.blocks.18.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 538 |
+
"model.vision_tower.vision.blocks.18.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 539 |
+
"model.vision_tower.vision.blocks.18.norm1.bias": "model-00002-of-00002.safetensors",
|
| 540 |
+
"model.vision_tower.vision.blocks.18.norm1.weight": "model-00002-of-00002.safetensors",
|
| 541 |
+
"model.vision_tower.vision.blocks.18.norm2.bias": "model-00002-of-00002.safetensors",
|
| 542 |
+
"model.vision_tower.vision.blocks.18.norm2.weight": "model-00002-of-00002.safetensors",
|
| 543 |
+
"model.vision_tower.vision.blocks.19.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 544 |
+
"model.vision_tower.vision.blocks.19.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 545 |
+
"model.vision_tower.vision.blocks.19.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 546 |
+
"model.vision_tower.vision.blocks.19.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 547 |
+
"model.vision_tower.vision.blocks.19.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 548 |
+
"model.vision_tower.vision.blocks.19.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 549 |
+
"model.vision_tower.vision.blocks.19.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 550 |
+
"model.vision_tower.vision.blocks.19.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 551 |
+
"model.vision_tower.vision.blocks.19.norm1.bias": "model-00002-of-00002.safetensors",
|
| 552 |
+
"model.vision_tower.vision.blocks.19.norm1.weight": "model-00002-of-00002.safetensors",
|
| 553 |
+
"model.vision_tower.vision.blocks.19.norm2.bias": "model-00002-of-00002.safetensors",
|
| 554 |
+
"model.vision_tower.vision.blocks.19.norm2.weight": "model-00002-of-00002.safetensors",
|
| 555 |
+
"model.vision_tower.vision.blocks.2.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 556 |
+
"model.vision_tower.vision.blocks.2.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 557 |
+
"model.vision_tower.vision.blocks.2.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 558 |
+
"model.vision_tower.vision.blocks.2.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 559 |
+
"model.vision_tower.vision.blocks.2.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 560 |
+
"model.vision_tower.vision.blocks.2.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 561 |
+
"model.vision_tower.vision.blocks.2.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 562 |
+
"model.vision_tower.vision.blocks.2.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 563 |
+
"model.vision_tower.vision.blocks.2.norm1.bias": "model-00002-of-00002.safetensors",
|
| 564 |
+
"model.vision_tower.vision.blocks.2.norm1.weight": "model-00002-of-00002.safetensors",
|
| 565 |
+
"model.vision_tower.vision.blocks.2.norm2.bias": "model-00002-of-00002.safetensors",
|
| 566 |
+
"model.vision_tower.vision.blocks.2.norm2.weight": "model-00002-of-00002.safetensors",
|
| 567 |
+
"model.vision_tower.vision.blocks.20.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 568 |
+
"model.vision_tower.vision.blocks.20.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 569 |
+
"model.vision_tower.vision.blocks.20.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 570 |
+
"model.vision_tower.vision.blocks.20.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 571 |
+
"model.vision_tower.vision.blocks.20.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 572 |
+
"model.vision_tower.vision.blocks.20.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 573 |
+
"model.vision_tower.vision.blocks.20.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 574 |
+
"model.vision_tower.vision.blocks.20.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 575 |
+
"model.vision_tower.vision.blocks.20.norm1.bias": "model-00002-of-00002.safetensors",
|
| 576 |
+
"model.vision_tower.vision.blocks.20.norm1.weight": "model-00002-of-00002.safetensors",
|
| 577 |
+
"model.vision_tower.vision.blocks.20.norm2.bias": "model-00002-of-00002.safetensors",
|
| 578 |
+
"model.vision_tower.vision.blocks.20.norm2.weight": "model-00002-of-00002.safetensors",
|
| 579 |
+
"model.vision_tower.vision.blocks.21.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 580 |
+
"model.vision_tower.vision.blocks.21.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 581 |
+
"model.vision_tower.vision.blocks.21.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 582 |
+
"model.vision_tower.vision.blocks.21.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 583 |
+
"model.vision_tower.vision.blocks.21.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 584 |
+
"model.vision_tower.vision.blocks.21.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 585 |
+
"model.vision_tower.vision.blocks.21.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 586 |
+
"model.vision_tower.vision.blocks.21.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 587 |
+
"model.vision_tower.vision.blocks.21.norm1.bias": "model-00002-of-00002.safetensors",
|
| 588 |
+
"model.vision_tower.vision.blocks.21.norm1.weight": "model-00002-of-00002.safetensors",
|
| 589 |
+
"model.vision_tower.vision.blocks.21.norm2.bias": "model-00002-of-00002.safetensors",
|
| 590 |
+
"model.vision_tower.vision.blocks.21.norm2.weight": "model-00002-of-00002.safetensors",
|
| 591 |
+
"model.vision_tower.vision.blocks.22.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 592 |
+
"model.vision_tower.vision.blocks.22.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 593 |
+
"model.vision_tower.vision.blocks.22.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 594 |
+
"model.vision_tower.vision.blocks.22.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 595 |
+
"model.vision_tower.vision.blocks.22.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 596 |
+
"model.vision_tower.vision.blocks.22.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 597 |
+
"model.vision_tower.vision.blocks.22.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 598 |
+
"model.vision_tower.vision.blocks.22.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 599 |
+
"model.vision_tower.vision.blocks.22.norm1.bias": "model-00002-of-00002.safetensors",
|
| 600 |
+
"model.vision_tower.vision.blocks.22.norm1.weight": "model-00002-of-00002.safetensors",
|
| 601 |
+
"model.vision_tower.vision.blocks.22.norm2.bias": "model-00002-of-00002.safetensors",
|
| 602 |
+
"model.vision_tower.vision.blocks.22.norm2.weight": "model-00002-of-00002.safetensors",
|
| 603 |
+
"model.vision_tower.vision.blocks.23.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 604 |
+
"model.vision_tower.vision.blocks.23.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 605 |
+
"model.vision_tower.vision.blocks.23.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 606 |
+
"model.vision_tower.vision.blocks.23.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 607 |
+
"model.vision_tower.vision.blocks.23.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 608 |
+
"model.vision_tower.vision.blocks.23.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 609 |
+
"model.vision_tower.vision.blocks.23.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 610 |
+
"model.vision_tower.vision.blocks.23.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 611 |
+
"model.vision_tower.vision.blocks.23.norm1.bias": "model-00002-of-00002.safetensors",
|
| 612 |
+
"model.vision_tower.vision.blocks.23.norm1.weight": "model-00002-of-00002.safetensors",
|
| 613 |
+
"model.vision_tower.vision.blocks.23.norm2.bias": "model-00002-of-00002.safetensors",
|
| 614 |
+
"model.vision_tower.vision.blocks.23.norm2.weight": "model-00002-of-00002.safetensors",
|
| 615 |
+
"model.vision_tower.vision.blocks.3.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 616 |
+
"model.vision_tower.vision.blocks.3.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 617 |
+
"model.vision_tower.vision.blocks.3.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 618 |
+
"model.vision_tower.vision.blocks.3.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 619 |
+
"model.vision_tower.vision.blocks.3.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 620 |
+
"model.vision_tower.vision.blocks.3.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 621 |
+
"model.vision_tower.vision.blocks.3.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 622 |
+
"model.vision_tower.vision.blocks.3.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 623 |
+
"model.vision_tower.vision.blocks.3.norm1.bias": "model-00002-of-00002.safetensors",
|
| 624 |
+
"model.vision_tower.vision.blocks.3.norm1.weight": "model-00002-of-00002.safetensors",
|
| 625 |
+
"model.vision_tower.vision.blocks.3.norm2.bias": "model-00002-of-00002.safetensors",
|
| 626 |
+
"model.vision_tower.vision.blocks.3.norm2.weight": "model-00002-of-00002.safetensors",
|
| 627 |
+
"model.vision_tower.vision.blocks.4.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 628 |
+
"model.vision_tower.vision.blocks.4.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 629 |
+
"model.vision_tower.vision.blocks.4.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 630 |
+
"model.vision_tower.vision.blocks.4.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 631 |
+
"model.vision_tower.vision.blocks.4.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 632 |
+
"model.vision_tower.vision.blocks.4.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 633 |
+
"model.vision_tower.vision.blocks.4.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 634 |
+
"model.vision_tower.vision.blocks.4.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 635 |
+
"model.vision_tower.vision.blocks.4.norm1.bias": "model-00002-of-00002.safetensors",
|
| 636 |
+
"model.vision_tower.vision.blocks.4.norm1.weight": "model-00002-of-00002.safetensors",
|
| 637 |
+
"model.vision_tower.vision.blocks.4.norm2.bias": "model-00002-of-00002.safetensors",
|
| 638 |
+
"model.vision_tower.vision.blocks.4.norm2.weight": "model-00002-of-00002.safetensors",
|
| 639 |
+
"model.vision_tower.vision.blocks.5.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 640 |
+
"model.vision_tower.vision.blocks.5.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 641 |
+
"model.vision_tower.vision.blocks.5.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 642 |
+
"model.vision_tower.vision.blocks.5.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 643 |
+
"model.vision_tower.vision.blocks.5.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 644 |
+
"model.vision_tower.vision.blocks.5.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 645 |
+
"model.vision_tower.vision.blocks.5.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 646 |
+
"model.vision_tower.vision.blocks.5.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 647 |
+
"model.vision_tower.vision.blocks.5.norm1.bias": "model-00002-of-00002.safetensors",
|
| 648 |
+
"model.vision_tower.vision.blocks.5.norm1.weight": "model-00002-of-00002.safetensors",
|
| 649 |
+
"model.vision_tower.vision.blocks.5.norm2.bias": "model-00002-of-00002.safetensors",
|
| 650 |
+
"model.vision_tower.vision.blocks.5.norm2.weight": "model-00002-of-00002.safetensors",
|
| 651 |
+
"model.vision_tower.vision.blocks.6.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 652 |
+
"model.vision_tower.vision.blocks.6.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 653 |
+
"model.vision_tower.vision.blocks.6.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 654 |
+
"model.vision_tower.vision.blocks.6.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 655 |
+
"model.vision_tower.vision.blocks.6.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 656 |
+
"model.vision_tower.vision.blocks.6.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 657 |
+
"model.vision_tower.vision.blocks.6.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 658 |
+
"model.vision_tower.vision.blocks.6.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 659 |
+
"model.vision_tower.vision.blocks.6.norm1.bias": "model-00002-of-00002.safetensors",
|
| 660 |
+
"model.vision_tower.vision.blocks.6.norm1.weight": "model-00002-of-00002.safetensors",
|
| 661 |
+
"model.vision_tower.vision.blocks.6.norm2.bias": "model-00002-of-00002.safetensors",
|
| 662 |
+
"model.vision_tower.vision.blocks.6.norm2.weight": "model-00002-of-00002.safetensors",
|
| 663 |
+
"model.vision_tower.vision.blocks.7.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 664 |
+
"model.vision_tower.vision.blocks.7.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 665 |
+
"model.vision_tower.vision.blocks.7.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 666 |
+
"model.vision_tower.vision.blocks.7.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 667 |
+
"model.vision_tower.vision.blocks.7.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 668 |
+
"model.vision_tower.vision.blocks.7.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 669 |
+
"model.vision_tower.vision.blocks.7.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 670 |
+
"model.vision_tower.vision.blocks.7.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 671 |
+
"model.vision_tower.vision.blocks.7.norm1.bias": "model-00002-of-00002.safetensors",
|
| 672 |
+
"model.vision_tower.vision.blocks.7.norm1.weight": "model-00002-of-00002.safetensors",
|
| 673 |
+
"model.vision_tower.vision.blocks.7.norm2.bias": "model-00002-of-00002.safetensors",
|
| 674 |
+
"model.vision_tower.vision.blocks.7.norm2.weight": "model-00002-of-00002.safetensors",
|
| 675 |
+
"model.vision_tower.vision.blocks.8.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 676 |
+
"model.vision_tower.vision.blocks.8.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 677 |
+
"model.vision_tower.vision.blocks.8.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 678 |
+
"model.vision_tower.vision.blocks.8.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 679 |
+
"model.vision_tower.vision.blocks.8.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 680 |
+
"model.vision_tower.vision.blocks.8.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 681 |
+
"model.vision_tower.vision.blocks.8.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 682 |
+
"model.vision_tower.vision.blocks.8.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 683 |
+
"model.vision_tower.vision.blocks.8.norm1.bias": "model-00002-of-00002.safetensors",
|
| 684 |
+
"model.vision_tower.vision.blocks.8.norm1.weight": "model-00002-of-00002.safetensors",
|
| 685 |
+
"model.vision_tower.vision.blocks.8.norm2.bias": "model-00002-of-00002.safetensors",
|
| 686 |
+
"model.vision_tower.vision.blocks.8.norm2.weight": "model-00002-of-00002.safetensors",
|
| 687 |
+
"model.vision_tower.vision.blocks.9.attn.proj.bias": "model-00002-of-00002.safetensors",
|
| 688 |
+
"model.vision_tower.vision.blocks.9.attn.proj.weight": "model-00002-of-00002.safetensors",
|
| 689 |
+
"model.vision_tower.vision.blocks.9.attn.qkv.bias": "model-00002-of-00002.safetensors",
|
| 690 |
+
"model.vision_tower.vision.blocks.9.attn.qkv.weight": "model-00002-of-00002.safetensors",
|
| 691 |
+
"model.vision_tower.vision.blocks.9.mlp.linear_fc1.bias": "model-00002-of-00002.safetensors",
|
| 692 |
+
"model.vision_tower.vision.blocks.9.mlp.linear_fc1.weight": "model-00002-of-00002.safetensors",
|
| 693 |
+
"model.vision_tower.vision.blocks.9.mlp.linear_fc2.bias": "model-00002-of-00002.safetensors",
|
| 694 |
+
"model.vision_tower.vision.blocks.9.mlp.linear_fc2.weight": "model-00002-of-00002.safetensors",
|
| 695 |
+
"model.vision_tower.vision.blocks.9.norm1.bias": "model-00002-of-00002.safetensors",
|
| 696 |
+
"model.vision_tower.vision.blocks.9.norm1.weight": "model-00002-of-00002.safetensors",
|
| 697 |
+
"model.vision_tower.vision.blocks.9.norm2.bias": "model-00002-of-00002.safetensors",
|
| 698 |
+
"model.vision_tower.vision.blocks.9.norm2.weight": "model-00002-of-00002.safetensors",
|
| 699 |
+
"model.vision_tower.vision.final_layernorm.bias": "model-00002-of-00002.safetensors",
|
| 700 |
+
"model.vision_tower.vision.final_layernorm.weight": "model-00002-of-00002.safetensors",
|
| 701 |
+
"model.vision_tower.vision.patch_embed.proj.bias": "model-00002-of-00002.safetensors",
|
| 702 |
+
"model.vision_tower.vision.patch_embed.proj.weight": "model-00002-of-00002.safetensors",
|
| 703 |
+
"model.vision_tower.vision.pos_embed.weight": "model-00002-of-00002.safetensors",
|
| 704 |
+
"model.vision_tower.vtc_compressor.downsample.bias": "model-00002-of-00002.safetensors",
|
| 705 |
+
"model.vision_tower.vtc_compressor.downsample.weight": "model-00002-of-00002.safetensors",
|
| 706 |
+
"model.vision_tower.vtc_compressor.spatial_attention.norm.bias": "model-00002-of-00002.safetensors",
|
| 707 |
+
"model.vision_tower.vtc_compressor.spatial_attention.norm.weight": "model-00002-of-00002.safetensors",
|
| 708 |
+
"model.vision_tower.vtc_compressor.spatial_attention.proj.bias": "model-00002-of-00002.safetensors",
|
| 709 |
+
"model.vision_tower.vtc_compressor.spatial_attention.proj.weight": "model-00002-of-00002.safetensors",
|
| 710 |
+
"model.vision_tower.vtc_compressor.spatial_attention.qkv.bias": "model-00002-of-00002.safetensors",
|
| 711 |
+
"model.vision_tower.vtc_compressor.spatial_attention.qkv.weight": "model-00002-of-00002.safetensors",
|
| 712 |
+
"model.vision_tower.vtc_compressor.temporal_attention.norm.bias": "model-00002-of-00002.safetensors",
|
| 713 |
+
"model.vision_tower.vtc_compressor.temporal_attention.norm.weight": "model-00002-of-00002.safetensors",
|
| 714 |
+
"model.vision_tower.vtc_compressor.temporal_attention.proj.bias": "model-00002-of-00002.safetensors",
|
| 715 |
+
"model.vision_tower.vtc_compressor.temporal_attention.proj.weight": "model-00002-of-00002.safetensors",
|
| 716 |
+
"model.vision_tower.vtc_compressor.temporal_attention.qkv.bias": "model-00002-of-00002.safetensors",
|
| 717 |
+
"model.vision_tower.vtc_compressor.temporal_attention.qkv.weight": "model-00002-of-00002.safetensors",
|
| 718 |
+
"model.vision_tower.vtc_compressor.temporal_position": "model-00002-of-00002.safetensors"
|
| 719 |
+
}
|
| 720 |
+
}
|
ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/modeling_qwen3_5vit_qwen3.py
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/processing_qwen3_5vit_qwen3.py
ADDED
|
@@ -0,0 +1,302 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"""processing_qwen3_5vit_qwen3.py — LlavaOnevision processor with optional
|
| 2 |
+
video temporal-merge (tmerge) token accounting.
|
| 3 |
+
|
| 4 |
+
Why a subclass: at eval the model is loaded via HF (lmms-eval `llava_hf` →
|
| 5 |
+
`AutoProcessor.from_pretrained(ckpt, trust_remote_code=True)`), NOT the swift
|
| 6 |
+
template. The `LlavaOnevisionProcessor.__call__` expands `<video>` by
|
| 7 |
+
`num_frames*144+1`. Under `video_temporal_merge=1` the model emits
|
| 8 |
+
`(num_frames//2)*144+1` features (grid_t=T/2 temporal groups), so the default
|
| 9 |
+
count over-expands → `masked_scatter` token/feature mismatch crash.
|
| 10 |
+
|
| 11 |
+
This subclass is byte-identical to the upstream processor EXCEPT the single video
|
| 12 |
+
token-count line, which routes through `qwen3_5_video_token_count` — the same
|
| 13 |
+
SoT helper the swift training template uses. The mllm_eval adapter
|
| 14 |
+
(`mllm_eval/checkpoint/__init__.py::qwen3_5vit`) declares this file as
|
| 15 |
+
`processor_file`, so prep cp's it into the ckpt and injects `auto_map.AutoProcessor`.
|
| 16 |
+
|
| 17 |
+
`video_temporal_merge` is a persisted init kwarg (lands in processor_config.json),
|
| 18 |
+
set by the plugin `get_processor` from env/config so eval inherits it with no env.
|
| 19 |
+
"""
|
| 20 |
+
|
| 21 |
+
import math
|
| 22 |
+
|
| 23 |
+
import numpy as np
|
| 24 |
+
|
| 25 |
+
from transformers.feature_extraction_utils import BatchFeature
|
| 26 |
+
from transformers.image_processing_utils import select_best_resolution
|
| 27 |
+
from transformers.image_utils import ImageInput, get_image_size, to_numpy_array
|
| 28 |
+
from transformers.processing_utils import Unpack
|
| 29 |
+
from transformers.tokenization_utils_base import PreTokenizedInput, TextInput
|
| 30 |
+
from transformers.video_utils import VideoInput
|
| 31 |
+
from transformers.models.llava_onevision.processing_llava_onevision import (
|
| 32 |
+
LlavaOnevisionProcessor,
|
| 33 |
+
LlavaOnevisionProcessorKwargs,
|
| 34 |
+
)
|
| 35 |
+
|
| 36 |
+
from typing import Optional
|
| 37 |
+
|
| 38 |
+
# Flat-name sibling import: works both in-repo (model_plugin adds video_mllm/model/ to
|
| 39 |
+
# sys.path) and at eval (HF dynamic_module loads modeling_qwen3_5vit_qwen3.py as a
|
| 40 |
+
# ckpt-dir sibling — the qwen3_5vit adapter ships it as modeling_file).
|
| 41 |
+
from modeling_qwen3_5vit_qwen3 import (
|
| 42 |
+
build_faithful_video_token_string,
|
| 43 |
+
vtc_compression_factor,
|
| 44 |
+
vtc_temporal_patch_size,
|
| 45 |
+
qwen3_5_video_token_count,
|
| 46 |
+
video_is_vtc_compression,
|
| 47 |
+
video_is_spatial_catp,
|
| 48 |
+
video_is_image_mrope,
|
| 49 |
+
video_is_pair_fusion,
|
| 50 |
+
video_is_temporal_compression,
|
| 51 |
+
video_is_temporal_merge,
|
| 52 |
+
video_mrope_active,
|
| 53 |
+
)
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
class LlavaQwen3_5ViTProcessor(LlavaOnevisionProcessor):
|
| 57 |
+
"""LlavaOnevisionProcessor whose `<video>` expansion is tmerge-aware."""
|
| 58 |
+
|
| 59 |
+
def __init__(
|
| 60 |
+
self,
|
| 61 |
+
image_processor=None,
|
| 62 |
+
tokenizer=None,
|
| 63 |
+
video_processor=None,
|
| 64 |
+
num_image_tokens=None,
|
| 65 |
+
vision_feature_select_strategy=None,
|
| 66 |
+
chat_template=None,
|
| 67 |
+
image_token="<image>",
|
| 68 |
+
video_token="<video>",
|
| 69 |
+
vision_aspect_ratio="anyres_max_9",
|
| 70 |
+
video_mode: Optional[str] = None,
|
| 71 |
+
vtc_compression_factor: int = 2,
|
| 72 |
+
vtc_temporal_patch_size: int = 1,
|
| 73 |
+
# Legacy int flags kept in the signature ONLY so a pre-2026-06-22 processor_config.json
|
| 74 |
+
# still reaches __init__ on reload (ProcessorMixin filters keys NOT in the signature BEFORE
|
| 75 |
+
# calling __init__). Folded into video_mode and NEVER set as attributes -> to_dict (which
|
| 76 |
+
# dumps __dict__) serializes ONLY video_mode (single SoT; old ckpts migrate on re-save).
|
| 77 |
+
video_temporal_merge: int = 0,
|
| 78 |
+
image_mrope: int = 0,
|
| 79 |
+
video_mrope: int = 0,
|
| 80 |
+
**kwargs,
|
| 81 |
+
):
|
| 82 |
+
# The parent's sub-processors AND scalar kwargs MUST be explicit in THIS signature -- NOT
|
| 83 |
+
# swallowed by a bare (*args, **kwargs). ProcessorMixin serialization enumerates what to
|
| 84 |
+
# persist via get_attributes() = inspect.signature(cls.__init__).parameters; a (*args,
|
| 85 |
+
# **kwargs) signature returns [] -> save_pretrained writes neither the sub-processor files
|
| 86 |
+
# nor video_mode into the ckpt. So video_mode is an explicit, serialized param here.
|
| 87 |
+
# Backward-compat: an old processor_config.json carries the legacy int flags (received via
|
| 88 |
+
# the explicit params above so reload reaches here) and no video_mode. Fold them into
|
| 89 |
+
# video_mode; do NOT set them as attributes, so to_dict writes only video_mode (single SoT;
|
| 90 |
+
# old ckpts migrate on re-save). New ckpts pass video_mode directly (fold skipped).
|
| 91 |
+
if video_mode is None:
|
| 92 |
+
if int(image_mrope or 0):
|
| 93 |
+
video_mode = "image_mrope"
|
| 94 |
+
elif int(video_mrope or 0):
|
| 95 |
+
video_mode = "video_mrope"
|
| 96 |
+
elif int(video_temporal_merge or 0):
|
| 97 |
+
video_mode = "temporal_merge"
|
| 98 |
+
self.video_mode = video_mode
|
| 99 |
+
self.vtc_compression_factor = vtc_compression_factor
|
| 100 |
+
self.vtc_temporal_patch_size = vtc_temporal_patch_size
|
| 101 |
+
super().__init__(
|
| 102 |
+
image_processor=image_processor,
|
| 103 |
+
tokenizer=tokenizer,
|
| 104 |
+
video_processor=video_processor,
|
| 105 |
+
num_image_tokens=num_image_tokens,
|
| 106 |
+
vision_feature_select_strategy=vision_feature_select_strategy,
|
| 107 |
+
chat_template=chat_template,
|
| 108 |
+
image_token=image_token,
|
| 109 |
+
video_token=video_token,
|
| 110 |
+
vision_aspect_ratio=vision_aspect_ratio,
|
| 111 |
+
**kwargs,
|
| 112 |
+
)
|
| 113 |
+
|
| 114 |
+
def __call__(
|
| 115 |
+
self,
|
| 116 |
+
images: ImageInput = None,
|
| 117 |
+
text: "TextInput | PreTokenizedInput | list[TextInput] | list[PreTokenizedInput]" = None,
|
| 118 |
+
videos: VideoInput = None,
|
| 119 |
+
video_fps: float = None,
|
| 120 |
+
video_sample_indices=None,
|
| 121 |
+
**kwargs: Unpack[LlavaOnevisionProcessorKwargs],
|
| 122 |
+
) -> BatchFeature:
|
| 123 |
+
# ── verbatim from LlavaOnevisionProcessor.__call__ (transformers) ──
|
| 124 |
+
output_kwargs = self._merge_kwargs(
|
| 125 |
+
LlavaOnevisionProcessorKwargs,
|
| 126 |
+
tokenizer_init_kwargs=self.tokenizer.init_kwargs,
|
| 127 |
+
**kwargs,
|
| 128 |
+
)
|
| 129 |
+
|
| 130 |
+
if isinstance(text, str):
|
| 131 |
+
text = [text]
|
| 132 |
+
elif not isinstance(text, list) and not isinstance(text[0], str):
|
| 133 |
+
raise TypeError("Invalid input text. Please provide a string, or a list of strings")
|
| 134 |
+
|
| 135 |
+
image_inputs = video_inputs = {}
|
| 136 |
+
|
| 137 |
+
if images is not None:
|
| 138 |
+
image_inputs = self.image_processor(images, **output_kwargs["images_kwargs"])
|
| 139 |
+
|
| 140 |
+
batch_num_images = iter(image_inputs["batch_num_images"])
|
| 141 |
+
image_sizes = iter(image_inputs["image_sizes"])
|
| 142 |
+
height, width = get_image_size(
|
| 143 |
+
to_numpy_array(image_inputs["pixel_values"][0][0]),
|
| 144 |
+
channel_dim=output_kwargs["images_kwargs"].get("data_format"),
|
| 145 |
+
)
|
| 146 |
+
text, num_image_tokens = self._expand_image_tokens(
|
| 147 |
+
text, image_sizes, height, width, self.image_token, batch_num_images
|
| 148 |
+
)
|
| 149 |
+
|
| 150 |
+
if videos is not None:
|
| 151 |
+
video_inputs = self.video_processor(videos, **output_kwargs["videos_kwargs"])
|
| 152 |
+
|
| 153 |
+
one_video = video_inputs.get("pixel_values_videos")[0]
|
| 154 |
+
if isinstance(video_inputs.get("pixel_values_videos")[0], (list, tuple)):
|
| 155 |
+
one_video = np.array(one_video)
|
| 156 |
+
else:
|
| 157 |
+
one_video = to_numpy_array(one_video)
|
| 158 |
+
num_frames = one_video.shape[0] # frame dim is always after batch dim
|
| 159 |
+
# ── ONLY deviation from upstream __call__: the video token count.
|
| 160 |
+
# Upstream derives it inline as `get_image_size(one_video[0])` (unused
|
| 161 |
+
# for count) + `num_frames * ceil(sqrt(num_image_tokens)/2)**2 + 1`.
|
| 162 |
+
# We drop the get_image_size line ON PURPOSE — the SoT helper hardcodes
|
| 163 |
+
# 144 tokens/readout-unit (= ceil(24/2)**2 for Qwen3.5 384/16 grid) and
|
| 164 |
+
# folds in the grid_t=num_frames//2 tubelet grouping when merge is on.
|
| 165 |
+
# Do NOT "restore" get_image_size when re-syncing against the default.
|
| 166 |
+
if video_mrope_active(self):
|
| 167 |
+
# FAIL-FAST: the faithful eval path builds ONE timestamp stream from a single video's
|
| 168 |
+
# (fps, sampled_indices) and replaces ONE <video> placeholder. A multi-video sample
|
| 169 |
+
# or batched (B>1) call would mis-apply the first video's timestamps to the rest.
|
| 170 |
+
# lmms-eval llava_hf runs generate at bs=1 single-video, so this never fires; it
|
| 171 |
+
# guards a future multi-video / batched entry point from silent mis-scoring.
|
| 172 |
+
n_video_ph = sum(sample.count(self.video_token) for sample in text)
|
| 173 |
+
if n_video_ph != 1:
|
| 174 |
+
raise RuntimeError(
|
| 175 |
+
f"LlavaQwen3_5ViTProcessor: faithful video_mrope eval supports exactly ONE "
|
| 176 |
+
f"<video> per call (single video, batch_size=1); found {n_video_ph}. "
|
| 177 |
+
f"Multi-video / batched faithful eval is unsupported."
|
| 178 |
+
)
|
| 179 |
+
if video_fps is None:
|
| 180 |
+
# FAIL-FAST (not a silent blob fallback): video_mrope=1 means the ckpt was trained
|
| 181 |
+
# on the faithful per-frame timestamp stream, so evaluating it on the contiguous
|
| 182 |
+
# blob is a train/eval mismatch that produces silently-wrong numbers. Only the
|
| 183 |
+
# lmms-eval simple (--force_simple) path threads fps today; the chat / VLMEvalKit
|
| 184 |
+
# video paths do not. Current 6-bench video tasks all use --force_simple, so this
|
| 185 |
+
# never fires; it guards a future entry-point swap from silently degrading.
|
| 186 |
+
# (Image-only eval never reaches here — `videos is None` skips this whole block.)
|
| 187 |
+
raise RuntimeError(
|
| 188 |
+
"LlavaQwen3_5ViTProcessor: video_mrope=1 but no video_fps/video_sample_indices "
|
| 189 |
+
"were provided — cannot build the faithful per-frame timestamp stream. Route "
|
| 190 |
+
"video eval through the lmms-eval simple (--force_simple) path, which threads fps."
|
| 191 |
+
)
|
| 192 |
+
# Faithful video-mRoPE (video_mrope=1): build the per-frame timestamp + vision-marker
|
| 193 |
+
# stream via the SAME shared SoT the training template _encode uses, so eval and
|
| 194 |
+
# train produce byte-identical video token ids (zero drift). Needs fps + sampled
|
| 195 |
+
# frame indices (threaded from the eval video loader). One video per eval request.
|
| 196 |
+
faithful_str = build_faithful_video_token_string(
|
| 197 |
+
num_frames, video_fps, video_sample_indices,
|
| 198 |
+
video_is_temporal_merge(self), video_token=self.video_token,
|
| 199 |
+
)
|
| 200 |
+
text = [sample.replace(self.video_token, faithful_str, 1) for sample in text]
|
| 201 |
+
else:
|
| 202 |
+
# Default / non-mrope: byte-identical contiguous blob.
|
| 203 |
+
num_video_tokens = qwen3_5_video_token_count(
|
| 204 |
+
num_frames,
|
| 205 |
+
temporal_merge=video_is_temporal_merge(self),
|
| 206 |
+
temporal_compression=video_is_temporal_compression(self),
|
| 207 |
+
pair_fusion=video_is_pair_fusion(self),
|
| 208 |
+
vtc_compression=video_is_vtc_compression(self),
|
| 209 |
+
spatial_catp=video_is_spatial_catp(self),
|
| 210 |
+
vtc_compression_factor=vtc_compression_factor(self),
|
| 211 |
+
vtc_temporal_patch_size=vtc_temporal_patch_size(self),
|
| 212 |
+
)
|
| 213 |
+
text = [sample.replace(self.video_token, self.video_token * num_video_tokens) for sample in text]
|
| 214 |
+
|
| 215 |
+
return_tensors = output_kwargs["text_kwargs"].pop("return_tensors", None)
|
| 216 |
+
return_mm_token_type_ids = output_kwargs["text_kwargs"].pop("return_mm_token_type_ids", None)
|
| 217 |
+
text_inputs = self.tokenizer(text, **output_kwargs["text_kwargs"])
|
| 218 |
+
self._check_special_mm_tokens(text, text_inputs, modalities=["image"])
|
| 219 |
+
|
| 220 |
+
if return_mm_token_type_ids:
|
| 221 |
+
# Byte-identical to LlavaOnevisionProcessor.__call__ (transformers 5.5.4):
|
| 222 |
+
# delegate to ProcessorMixin.create_mm_token_type_ids, which marks image=1 / video=2
|
| 223 |
+
# / audio=3 over self.image_ids/video_ids/audio_ids (set by the base __init__, so
|
| 224 |
+
# present on the fresh-build AND the __class__-retype reload path). An earlier inline
|
| 225 |
+
# copy here came from an OLDER transformers that lacked this method and marked
|
| 226 |
+
# image-only (mis-typing video tokens); 5.5.4 HAS the method (processing_utils.py) and
|
| 227 |
+
# the default path calls it, so tracking the default keeps the <video> token COUNT the only intentional
|
| 228 |
+
# deviation in this __call__. (This path is dormant: return_mm_token_type_ids defaults
|
| 229 |
+
# False for LlavaOnevision.)
|
| 230 |
+
text_inputs["mm_token_type_ids"] = self.create_mm_token_type_ids(text_inputs["input_ids"])
|
| 231 |
+
return BatchFeature(data={**text_inputs, **image_inputs, **video_inputs}, tensor_type=return_tensors)
|
| 232 |
+
|
| 233 |
+
def _get_number_of_features(self, orig_height: int, orig_width: int, height: int, width: int) -> int:
|
| 234 |
+
"""image_mrope=1: drop the per-row image_newline from the count (the model calls
|
| 235 |
+
pack_image_features with image_newline=None), so the image token count is
|
| 236 |
+
base + unpadded (clean H x W grid). Default count = base + unpadded + newline; here we
|
| 237 |
+
subtract exactly `newline_features`. flag 0 (default / video_mrope) → byte-identical to parent.
|
| 238 |
+
Used by BOTH the swift training _encode and eval __call__/_expand_image_tokens, so the
|
| 239 |
+
model's pack feature count and the input_ids placeholder count stay matched."""
|
| 240 |
+
n = super()._get_number_of_features(orig_height, orig_width, height, width)
|
| 241 |
+
if video_is_image_mrope(self):
|
| 242 |
+
# Guard the one config that would desync the position-id block sum from the emitted
|
| 243 |
+
# token count: vision_feature_select_strategy="default" makes _expand_image_tokens
|
| 244 |
+
# subtract 1 (a CLS-token drop), but image_mrope_grid_blocks emits the full base+
|
| 245 |
+
# unpadded grid (no CLS to drop). Qwen3.5 ViT has NO CLS token, so the strategy is
|
| 246 |
+
# always 'full'/None here; fail loud instead of silently misaligning if that ever changes.
|
| 247 |
+
if getattr(self, "vision_feature_select_strategy", None) == "default":
|
| 248 |
+
raise ValueError(
|
| 249 |
+
"image_mrope is incompatible with vision_feature_select_strategy='default' "
|
| 250 |
+
"(no CLS token in the clean H x W grid; the -1 would desync the position-id "
|
| 251 |
+
"block sum from the token count). Use 'full' / None."
|
| 252 |
+
)
|
| 253 |
+
image_grid_pinpoints = self.image_processor.image_grid_pinpoints
|
| 254 |
+
height_best, width_best = select_best_resolution([orig_height, orig_width], image_grid_pinpoints)
|
| 255 |
+
scale_height, scale_width = height_best // height, width_best // width
|
| 256 |
+
patches = int(math.sqrt(self.num_image_tokens))
|
| 257 |
+
_, newline_features = self._get_unpadded_features(
|
| 258 |
+
orig_height, orig_width, patches, patches, scale_height, scale_width
|
| 259 |
+
)
|
| 260 |
+
n -= newline_features
|
| 261 |
+
return n
|
| 262 |
+
|
| 263 |
+
def _expand_image_tokens(self, text, image_sizes, height, width, special_token, batch_num_images):
|
| 264 |
+
"""image_mrope=1 multi-image fix: LlavaOV expands each image of a MULTI-image sample as
|
| 265 |
+
`num_image_tokens + 1` (base 576 + one image_newline), bypassing `_get_number_of_features`
|
| 266 |
+
(HF processing_llava_onevision.py). But under image_mrope the model strips image_newline
|
| 267 |
+
(pack_image_features(image_newline=None)) → 576 features/image, so the default +1 over-counts
|
| 268 |
+
by one per image → masked_scatter token/feature mismatch at multi-image eval (e.g. MMMU).
|
| 269 |
+
This override drops the +1 ONLY for image_mrope multi-image; single-image still routes
|
| 270 |
+
through `_get_number_of_features` (AnyRes, newline already stripped there). flag 0 →
|
| 271 |
+
byte-identical to the parent. Verbatim from LlavaOnevisionProcessor._expand_image_tokens
|
| 272 |
+
EXCEPT the gated multi-image count line."""
|
| 273 |
+
image_mrope = video_is_image_mrope(self)
|
| 274 |
+
prompt_strings = []
|
| 275 |
+
max_num_vision_tokens = 0
|
| 276 |
+
for sample in text:
|
| 277 |
+
if special_token in sample:
|
| 278 |
+
num_images = next(batch_num_images) # should consume iterable
|
| 279 |
+
is_multi_image = num_images != 1
|
| 280 |
+
else:
|
| 281 |
+
is_multi_image = False
|
| 282 |
+
while special_token in sample:
|
| 283 |
+
original_size = next(image_sizes) # should consume iterable
|
| 284 |
+
if is_multi_image:
|
| 285 |
+
# ── ONLY deviation: drop the +1 newline under image_mrope (base-only 576).
|
| 286 |
+
num_image_tokens = self.num_image_tokens if image_mrope else self.num_image_tokens + 1
|
| 287 |
+
else:
|
| 288 |
+
if not isinstance(original_size, (list, tuple)):
|
| 289 |
+
# cast to list to avoid numerical precision errors when calculating unpadding
|
| 290 |
+
original_size = original_size.tolist()
|
| 291 |
+
orig_height, orig_width = original_size
|
| 292 |
+
num_image_tokens = self._get_number_of_features(orig_height, orig_width, height, width)
|
| 293 |
+
max_num_vision_tokens = max(max_num_vision_tokens, num_image_tokens)
|
| 294 |
+
if self.vision_feature_select_strategy == "default":
|
| 295 |
+
num_image_tokens -= 1
|
| 296 |
+
sample = sample.replace(special_token, "<placeholder>" * num_image_tokens, 1)
|
| 297 |
+
prompt_strings.append(sample)
|
| 298 |
+
text = [sample.replace("<placeholder>", special_token) for sample in prompt_strings]
|
| 299 |
+
return text, max_num_vision_tokens
|
| 300 |
+
|
| 301 |
+
|
| 302 |
+
__all__ = ["LlavaQwen3_5ViTProcessor"]
|
ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/processor_config.json
ADDED
|
@@ -0,0 +1,209 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"image_processor": {
|
| 3 |
+
"do_convert_rgb": true,
|
| 4 |
+
"do_normalize": true,
|
| 5 |
+
"do_pad": true,
|
| 6 |
+
"do_rescale": true,
|
| 7 |
+
"do_resize": true,
|
| 8 |
+
"image_grid_pinpoints": [
|
| 9 |
+
[
|
| 10 |
+
384,
|
| 11 |
+
384
|
| 12 |
+
],
|
| 13 |
+
[
|
| 14 |
+
384,
|
| 15 |
+
768
|
| 16 |
+
],
|
| 17 |
+
[
|
| 18 |
+
384,
|
| 19 |
+
1152
|
| 20 |
+
],
|
| 21 |
+
[
|
| 22 |
+
384,
|
| 23 |
+
1536
|
| 24 |
+
],
|
| 25 |
+
[
|
| 26 |
+
384,
|
| 27 |
+
1920
|
| 28 |
+
],
|
| 29 |
+
[
|
| 30 |
+
384,
|
| 31 |
+
2304
|
| 32 |
+
],
|
| 33 |
+
[
|
| 34 |
+
768,
|
| 35 |
+
384
|
| 36 |
+
],
|
| 37 |
+
[
|
| 38 |
+
768,
|
| 39 |
+
768
|
| 40 |
+
],
|
| 41 |
+
[
|
| 42 |
+
768,
|
| 43 |
+
1152
|
| 44 |
+
],
|
| 45 |
+
[
|
| 46 |
+
768,
|
| 47 |
+
1536
|
| 48 |
+
],
|
| 49 |
+
[
|
| 50 |
+
768,
|
| 51 |
+
1920
|
| 52 |
+
],
|
| 53 |
+
[
|
| 54 |
+
768,
|
| 55 |
+
2304
|
| 56 |
+
],
|
| 57 |
+
[
|
| 58 |
+
1152,
|
| 59 |
+
384
|
| 60 |
+
],
|
| 61 |
+
[
|
| 62 |
+
1152,
|
| 63 |
+
768
|
| 64 |
+
],
|
| 65 |
+
[
|
| 66 |
+
1152,
|
| 67 |
+
1152
|
| 68 |
+
],
|
| 69 |
+
[
|
| 70 |
+
1152,
|
| 71 |
+
1536
|
| 72 |
+
],
|
| 73 |
+
[
|
| 74 |
+
1152,
|
| 75 |
+
1920
|
| 76 |
+
],
|
| 77 |
+
[
|
| 78 |
+
1152,
|
| 79 |
+
2304
|
| 80 |
+
],
|
| 81 |
+
[
|
| 82 |
+
1536,
|
| 83 |
+
384
|
| 84 |
+
],
|
| 85 |
+
[
|
| 86 |
+
1536,
|
| 87 |
+
768
|
| 88 |
+
],
|
| 89 |
+
[
|
| 90 |
+
1536,
|
| 91 |
+
1152
|
| 92 |
+
],
|
| 93 |
+
[
|
| 94 |
+
1536,
|
| 95 |
+
1536
|
| 96 |
+
],
|
| 97 |
+
[
|
| 98 |
+
1536,
|
| 99 |
+
1920
|
| 100 |
+
],
|
| 101 |
+
[
|
| 102 |
+
1536,
|
| 103 |
+
2304
|
| 104 |
+
],
|
| 105 |
+
[
|
| 106 |
+
1920,
|
| 107 |
+
384
|
| 108 |
+
],
|
| 109 |
+
[
|
| 110 |
+
1920,
|
| 111 |
+
768
|
| 112 |
+
],
|
| 113 |
+
[
|
| 114 |
+
1920,
|
| 115 |
+
1152
|
| 116 |
+
],
|
| 117 |
+
[
|
| 118 |
+
1920,
|
| 119 |
+
1536
|
| 120 |
+
],
|
| 121 |
+
[
|
| 122 |
+
1920,
|
| 123 |
+
1920
|
| 124 |
+
],
|
| 125 |
+
[
|
| 126 |
+
1920,
|
| 127 |
+
2304
|
| 128 |
+
],
|
| 129 |
+
[
|
| 130 |
+
2304,
|
| 131 |
+
384
|
| 132 |
+
],
|
| 133 |
+
[
|
| 134 |
+
2304,
|
| 135 |
+
768
|
| 136 |
+
],
|
| 137 |
+
[
|
| 138 |
+
2304,
|
| 139 |
+
1152
|
| 140 |
+
],
|
| 141 |
+
[
|
| 142 |
+
2304,
|
| 143 |
+
1536
|
| 144 |
+
],
|
| 145 |
+
[
|
| 146 |
+
2304,
|
| 147 |
+
1920
|
| 148 |
+
],
|
| 149 |
+
[
|
| 150 |
+
2304,
|
| 151 |
+
2304
|
| 152 |
+
]
|
| 153 |
+
],
|
| 154 |
+
"image_mean": [
|
| 155 |
+
0.5,
|
| 156 |
+
0.5,
|
| 157 |
+
0.5
|
| 158 |
+
],
|
| 159 |
+
"image_processor_type": "LlavaOnevisionImageProcessor",
|
| 160 |
+
"image_std": [
|
| 161 |
+
0.5,
|
| 162 |
+
0.5,
|
| 163 |
+
0.5
|
| 164 |
+
],
|
| 165 |
+
"resample": 3,
|
| 166 |
+
"rescale_factor": 0.00392156862745098,
|
| 167 |
+
"size": {
|
| 168 |
+
"height": 384,
|
| 169 |
+
"width": 384
|
| 170 |
+
}
|
| 171 |
+
},
|
| 172 |
+
"image_token": "<image>",
|
| 173 |
+
"num_image_tokens": 576,
|
| 174 |
+
"processor_class": "LlavaQwen3_5ViTProcessor",
|
| 175 |
+
"video_mode": "vtc_compression",
|
| 176 |
+
"video_processor": {
|
| 177 |
+
"do_convert_rgb": true,
|
| 178 |
+
"do_normalize": true,
|
| 179 |
+
"do_rescale": true,
|
| 180 |
+
"do_resize": true,
|
| 181 |
+
"do_sample_frames": false,
|
| 182 |
+
"image_mean": [
|
| 183 |
+
0.5,
|
| 184 |
+
0.5,
|
| 185 |
+
0.5
|
| 186 |
+
],
|
| 187 |
+
"image_std": [
|
| 188 |
+
0.5,
|
| 189 |
+
0.5,
|
| 190 |
+
0.5
|
| 191 |
+
],
|
| 192 |
+
"resample": 3,
|
| 193 |
+
"rescale_factor": 0.00392156862745098,
|
| 194 |
+
"return_metadata": false,
|
| 195 |
+
"size": {
|
| 196 |
+
"height": 384,
|
| 197 |
+
"width": 384
|
| 198 |
+
},
|
| 199 |
+
"video_processor_type": "LlavaOnevisionVideoProcessor"
|
| 200 |
+
},
|
| 201 |
+
"video_token": "<video>",
|
| 202 |
+
"vision_aspect_ratio": "anyres_max_9",
|
| 203 |
+
"vision_feature_select_strategy": null,
|
| 204 |
+
"vtc_compression_factor": 2,
|
| 205 |
+
"vtc_temporal_patch_size": 1,
|
| 206 |
+
"auto_map": {
|
| 207 |
+
"AutoProcessor": "processing_qwen3_5vit_qwen3.LlavaQwen3_5ViTProcessor"
|
| 208 |
+
}
|
| 209 |
+
}
|
ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9024318c850eaadf26be79389d21b07a7afd8f1af749b89f9109b06c0d12173c
|
| 3 |
+
size 11423018
|
ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/tokenizer_config.json
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": null,
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|im_end|>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"extra_special_tokens": [
|
| 9 |
+
"<image>",
|
| 10 |
+
"<video>"
|
| 11 |
+
],
|
| 12 |
+
"is_local": true,
|
| 13 |
+
"model_max_length": 1010000,
|
| 14 |
+
"pad_token": "<|endoftext|>",
|
| 15 |
+
"processor_class": "LlavaQwen3_5ViTProcessor",
|
| 16 |
+
"split_special_tokens": false,
|
| 17 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 18 |
+
"unk_token": null
|
| 19 |
+
}
|
ckpts/S2/4b/qwen3_5_2b/v14_1_mkl_layer3/v0-20261012-043050/checkpoint-1530/trainer_state.json
ADDED
|
@@ -0,0 +1,2490 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 100.0,
|
| 7 |
+
"global_step": 1530,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.00065359477124183,
|
| 14 |
+
"grad_norm": 19.855953216552734,
|
| 15 |
+
"learning_rate": 1.2987012987012988e-08,
|
| 16 |
+
"loss": 1.8338881731033325,
|
| 17 |
+
"step": 1,
|
| 18 |
+
"token_acc": 0.6095767575322812
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 0.0032679738562091504,
|
| 22 |
+
"grad_norm": 20.445348739624023,
|
| 23 |
+
"learning_rate": 6.493506493506492e-08,
|
| 24 |
+
"loss": 1.796581745147705,
|
| 25 |
+
"step": 5,
|
| 26 |
+
"token_acc": 0.610710261529642
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.006535947712418301,
|
| 30 |
+
"grad_norm": 17.406198501586914,
|
| 31 |
+
"learning_rate": 1.2987012987012984e-07,
|
| 32 |
+
"loss": 1.758521270751953,
|
| 33 |
+
"step": 10,
|
| 34 |
+
"token_acc": 0.6177650302630192
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.00980392156862745,
|
| 38 |
+
"grad_norm": 14.155068397521973,
|
| 39 |
+
"learning_rate": 1.948051948051948e-07,
|
| 40 |
+
"loss": 1.6537511825561524,
|
| 41 |
+
"step": 15,
|
| 42 |
+
"token_acc": 0.6256829540222871
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.013071895424836602,
|
| 46 |
+
"grad_norm": 8.195589065551758,
|
| 47 |
+
"learning_rate": 2.597402597402597e-07,
|
| 48 |
+
"loss": 1.4368864059448243,
|
| 49 |
+
"step": 20,
|
| 50 |
+
"token_acc": 0.6549158217840201
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.016339869281045753,
|
| 54 |
+
"grad_norm": 6.412547588348389,
|
| 55 |
+
"learning_rate": 3.2467532467532465e-07,
|
| 56 |
+
"loss": 1.331705093383789,
|
| 57 |
+
"step": 25,
|
| 58 |
+
"token_acc": 0.6687547322432126
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.0196078431372549,
|
| 62 |
+
"grad_norm": 7.091923236846924,
|
| 63 |
+
"learning_rate": 3.896103896103896e-07,
|
| 64 |
+
"loss": 1.2620681762695312,
|
| 65 |
+
"step": 30,
|
| 66 |
+
"token_acc": 0.6841354200003148
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.02287581699346405,
|
| 70 |
+
"grad_norm": 7.774296760559082,
|
| 71 |
+
"learning_rate": 4.545454545454545e-07,
|
| 72 |
+
"loss": 1.211052703857422,
|
| 73 |
+
"step": 35,
|
| 74 |
+
"token_acc": 0.6901409162589462
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.026143790849673203,
|
| 78 |
+
"grad_norm": 5.958282470703125,
|
| 79 |
+
"learning_rate": 5.194805194805194e-07,
|
| 80 |
+
"loss": 1.1563541412353515,
|
| 81 |
+
"step": 40,
|
| 82 |
+
"token_acc": 0.6971471971936419
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.029411764705882353,
|
| 86 |
+
"grad_norm": 6.463438987731934,
|
| 87 |
+
"learning_rate": 5.844155844155844e-07,
|
| 88 |
+
"loss": 1.0958675384521483,
|
| 89 |
+
"step": 45,
|
| 90 |
+
"token_acc": 0.7100328999904175
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.032679738562091505,
|
| 94 |
+
"grad_norm": 6.096618175506592,
|
| 95 |
+
"learning_rate": 6.493506493506493e-07,
|
| 96 |
+
"loss": 1.0740789413452148,
|
| 97 |
+
"step": 50,
|
| 98 |
+
"token_acc": 0.7153402011391955
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.03594771241830065,
|
| 102 |
+
"grad_norm": 11.768878936767578,
|
| 103 |
+
"learning_rate": 7.142857142857143e-07,
|
| 104 |
+
"loss": 1.0851158142089843,
|
| 105 |
+
"step": 55,
|
| 106 |
+
"token_acc": 0.7121884670749384
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.0392156862745098,
|
| 110 |
+
"grad_norm": 6.751704692840576,
|
| 111 |
+
"learning_rate": 7.792207792207792e-07,
|
| 112 |
+
"loss": 1.0402315139770508,
|
| 113 |
+
"step": 60,
|
| 114 |
+
"token_acc": 0.7256622158276748
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.042483660130718956,
|
| 118 |
+
"grad_norm": 6.28439474105835,
|
| 119 |
+
"learning_rate": 8.44155844155844e-07,
|
| 120 |
+
"loss": 1.0236156463623047,
|
| 121 |
+
"step": 65,
|
| 122 |
+
"token_acc": 0.7255372353336306
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.0457516339869281,
|
| 126 |
+
"grad_norm": 5.036556720733643,
|
| 127 |
+
"learning_rate": 9.09090909090909e-07,
|
| 128 |
+
"loss": 1.031804084777832,
|
| 129 |
+
"step": 70,
|
| 130 |
+
"token_acc": 0.7243787004386745
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.049019607843137254,
|
| 134 |
+
"grad_norm": 10.287496566772461,
|
| 135 |
+
"learning_rate": 9.74025974025974e-07,
|
| 136 |
+
"loss": 1.0415892601013184,
|
| 137 |
+
"step": 75,
|
| 138 |
+
"token_acc": 0.7174946815167063
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.05228758169934641,
|
| 142 |
+
"grad_norm": 11.181119918823242,
|
| 143 |
+
"learning_rate": 9.999894816041221e-07,
|
| 144 |
+
"loss": 0.9901890754699707,
|
| 145 |
+
"step": 80,
|
| 146 |
+
"token_acc": 0.7292383377907213
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.05555555555555555,
|
| 150 |
+
"grad_norm": 8.260233879089355,
|
| 151 |
+
"learning_rate": 9.999252041208325e-07,
|
| 152 |
+
"loss": 1.0289884567260743,
|
| 153 |
+
"step": 85,
|
| 154 |
+
"token_acc": 0.7209439206496266
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.058823529411764705,
|
| 158 |
+
"grad_norm": 9.87868595123291,
|
| 159 |
+
"learning_rate": 9.99802500210544e-07,
|
| 160 |
+
"loss": 0.9935386657714844,
|
| 161 |
+
"step": 90,
|
| 162 |
+
"token_acc": 0.7332437426507643
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 0.06209150326797386,
|
| 166 |
+
"grad_norm": 7.0180535316467285,
|
| 167 |
+
"learning_rate": 9.99621384213704e-07,
|
| 168 |
+
"loss": 0.9806125640869141,
|
| 169 |
+
"step": 95,
|
| 170 |
+
"token_acc": 0.7324640886240542
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.06535947712418301,
|
| 174 |
+
"grad_norm": 8.04842758178711,
|
| 175 |
+
"learning_rate": 9.993818772974002e-07,
|
| 176 |
+
"loss": 0.9608866691589355,
|
| 177 |
+
"step": 100,
|
| 178 |
+
"token_acc": 0.7395402285163111
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.06862745098039216,
|
| 182 |
+
"grad_norm": 5.766630172729492,
|
| 183 |
+
"learning_rate": 9.99084007452886e-07,
|
| 184 |
+
"loss": 0.9635747909545899,
|
| 185 |
+
"step": 105,
|
| 186 |
+
"token_acc": 0.7366975835977545
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.0718954248366013,
|
| 190 |
+
"grad_norm": 10.151355743408203,
|
| 191 |
+
"learning_rate": 9.98727809492311e-07,
|
| 192 |
+
"loss": 0.9714326858520508,
|
| 193 |
+
"step": 110,
|
| 194 |
+
"token_acc": 0.7360298323579882
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.07516339869281045,
|
| 198 |
+
"grad_norm": 10.803625106811523,
|
| 199 |
+
"learning_rate": 9.983133250446507e-07,
|
| 200 |
+
"loss": 0.9860940933227539,
|
| 201 |
+
"step": 115,
|
| 202 |
+
"token_acc": 0.7282516443878634
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.0784313725490196,
|
| 206 |
+
"grad_norm": 15.469793319702148,
|
| 207 |
+
"learning_rate": 9.978406025508422e-07,
|
| 208 |
+
"loss": 0.9767313957214355,
|
| 209 |
+
"step": 120,
|
| 210 |
+
"token_acc": 0.7322962514979449
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 0.08169934640522876,
|
| 214 |
+
"grad_norm": 21.7315616607666,
|
| 215 |
+
"learning_rate": 9.973096972581224e-07,
|
| 216 |
+
"loss": 0.9655673027038574,
|
| 217 |
+
"step": 125,
|
| 218 |
+
"token_acc": 0.734423684238321
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.08496732026143791,
|
| 222 |
+
"grad_norm": 10.19537353515625,
|
| 223 |
+
"learning_rate": 9.967206712135716e-07,
|
| 224 |
+
"loss": 0.9567217826843262,
|
| 225 |
+
"step": 130,
|
| 226 |
+
"token_acc": 0.7379129168183947
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.08823529411764706,
|
| 230 |
+
"grad_norm": 19.695636749267578,
|
| 231 |
+
"learning_rate": 9.960735932568622e-07,
|
| 232 |
+
"loss": 0.9229909896850585,
|
| 233 |
+
"step": 135,
|
| 234 |
+
"token_acc": 0.7451511905255122
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.0915032679738562,
|
| 238 |
+
"grad_norm": 10.887438774108887,
|
| 239 |
+
"learning_rate": 9.95368539012213e-07,
|
| 240 |
+
"loss": 0.9508550643920899,
|
| 241 |
+
"step": 140,
|
| 242 |
+
"token_acc": 0.7377114375099422
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.09477124183006536,
|
| 246 |
+
"grad_norm": 12.469518661499023,
|
| 247 |
+
"learning_rate": 9.94605590879551e-07,
|
| 248 |
+
"loss": 0.9250885963439941,
|
| 249 |
+
"step": 145,
|
| 250 |
+
"token_acc": 0.7460594727987855
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.09803921568627451,
|
| 254 |
+
"grad_norm": 9.998051643371582,
|
| 255 |
+
"learning_rate": 9.937848380248816e-07,
|
| 256 |
+
"loss": 0.9237943649291992,
|
| 257 |
+
"step": 150,
|
| 258 |
+
"token_acc": 0.7431540721276627
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 0.10130718954248366,
|
| 262 |
+
"grad_norm": 12.615669250488281,
|
| 263 |
+
"learning_rate": 9.929063763698672e-07,
|
| 264 |
+
"loss": 0.926536750793457,
|
| 265 |
+
"step": 155,
|
| 266 |
+
"token_acc": 0.74386002551971
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.10457516339869281,
|
| 270 |
+
"grad_norm": 7.4902472496032715,
|
| 271 |
+
"learning_rate": 9.919703085806174e-07,
|
| 272 |
+
"loss": 0.9132516860961915,
|
| 273 |
+
"step": 160,
|
| 274 |
+
"token_acc": 0.7511147603864503
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 0.10784313725490197,
|
| 278 |
+
"grad_norm": 7.313841819763184,
|
| 279 |
+
"learning_rate": 9.9097674405569e-07,
|
| 280 |
+
"loss": 0.947848892211914,
|
| 281 |
+
"step": 165,
|
| 282 |
+
"token_acc": 0.7378941112527747
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.1111111111111111,
|
| 286 |
+
"grad_norm": 11.811566352844238,
|
| 287 |
+
"learning_rate": 9.899257989133057e-07,
|
| 288 |
+
"loss": 0.9072443962097168,
|
| 289 |
+
"step": 170,
|
| 290 |
+
"token_acc": 0.7499969591183876
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.11437908496732026,
|
| 294 |
+
"grad_norm": 10.67428970336914,
|
| 295 |
+
"learning_rate": 9.88817595977777e-07,
|
| 296 |
+
"loss": 0.9251197814941406,
|
| 297 |
+
"step": 175,
|
| 298 |
+
"token_acc": 0.7391460928150348
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 0.11764705882352941,
|
| 302 |
+
"grad_norm": 9.573467254638672,
|
| 303 |
+
"learning_rate": 9.87652264765155e-07,
|
| 304 |
+
"loss": 0.9112161636352539,
|
| 305 |
+
"step": 180,
|
| 306 |
+
"token_acc": 0.7452073249566772
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.12091503267973856,
|
| 310 |
+
"grad_norm": 12.068974494934082,
|
| 311 |
+
"learning_rate": 9.864299414680904e-07,
|
| 312 |
+
"loss": 0.9501218795776367,
|
| 313 |
+
"step": 185,
|
| 314 |
+
"token_acc": 0.7399626493645837
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.12418300653594772,
|
| 318 |
+
"grad_norm": 11.75418758392334,
|
| 319 |
+
"learning_rate": 9.851507689399188e-07,
|
| 320 |
+
"loss": 0.8926944732666016,
|
| 321 |
+
"step": 190,
|
| 322 |
+
"token_acc": 0.7470107190564624
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 0.12745098039215685,
|
| 326 |
+
"grad_norm": 7.0023369789123535,
|
| 327 |
+
"learning_rate": 9.838148966779644e-07,
|
| 328 |
+
"loss": 0.9030766487121582,
|
| 329 |
+
"step": 195,
|
| 330 |
+
"token_acc": 0.7473020846131794
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 0.13071895424836602,
|
| 334 |
+
"grad_norm": 9.188921928405762,
|
| 335 |
+
"learning_rate": 9.824224808060689e-07,
|
| 336 |
+
"loss": 0.8898355484008789,
|
| 337 |
+
"step": 200,
|
| 338 |
+
"token_acc": 0.7526864566121588
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.13398692810457516,
|
| 342 |
+
"grad_norm": 5.432031631469727,
|
| 343 |
+
"learning_rate": 9.809736840563434e-07,
|
| 344 |
+
"loss": 0.908603286743164,
|
| 345 |
+
"step": 205,
|
| 346 |
+
"token_acc": 0.7480688923989008
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.13725490196078433,
|
| 350 |
+
"grad_norm": 8.152050018310547,
|
| 351 |
+
"learning_rate": 9.794686757501526e-07,
|
| 352 |
+
"loss": 0.8831151962280274,
|
| 353 |
+
"step": 210,
|
| 354 |
+
"token_acc": 0.754504370577392
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 0.14052287581699346,
|
| 358 |
+
"grad_norm": 8.81791877746582,
|
| 359 |
+
"learning_rate": 9.779076317783244e-07,
|
| 360 |
+
"loss": 0.8753261566162109,
|
| 361 |
+
"step": 215,
|
| 362 |
+
"token_acc": 0.7566091182872648
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.1437908496732026,
|
| 366 |
+
"grad_norm": 10.696077346801758,
|
| 367 |
+
"learning_rate": 9.762907345805932e-07,
|
| 368 |
+
"loss": 0.9344842910766602,
|
| 369 |
+
"step": 220,
|
| 370 |
+
"token_acc": 0.7421422490777679
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.14705882352941177,
|
| 374 |
+
"grad_norm": 11.021129608154297,
|
| 375 |
+
"learning_rate": 9.746181731242791e-07,
|
| 376 |
+
"loss": 0.900820541381836,
|
| 377 |
+
"step": 225,
|
| 378 |
+
"token_acc": 0.750125657978315
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 0.1503267973856209,
|
| 382 |
+
"grad_norm": 10.349108695983887,
|
| 383 |
+
"learning_rate": 9.72890142882203e-07,
|
| 384 |
+
"loss": 0.8732796669006347,
|
| 385 |
+
"step": 230,
|
| 386 |
+
"token_acc": 0.7578910104147634
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.15359477124183007,
|
| 390 |
+
"grad_norm": 10.01156997680664,
|
| 391 |
+
"learning_rate": 9.711068458098417e-07,
|
| 392 |
+
"loss": 0.8918936729431153,
|
| 393 |
+
"step": 235,
|
| 394 |
+
"token_acc": 0.7505512974222831
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.1568627450980392,
|
| 398 |
+
"grad_norm": 9.233423233032227,
|
| 399 |
+
"learning_rate": 9.69268490321724e-07,
|
| 400 |
+
"loss": 0.9119072914123535,
|
| 401 |
+
"step": 240,
|
| 402 |
+
"token_acc": 0.7467435324791896
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.16013071895424835,
|
| 406 |
+
"grad_norm": 9.706135749816895,
|
| 407 |
+
"learning_rate": 9.67375291267076e-07,
|
| 408 |
+
"loss": 0.8936971664428711,
|
| 409 |
+
"step": 245,
|
| 410 |
+
"token_acc": 0.7504708280749441
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.16339869281045752,
|
| 414 |
+
"grad_norm": 8.238842010498047,
|
| 415 |
+
"learning_rate": 9.654274699047078e-07,
|
| 416 |
+
"loss": 0.9122947692871094,
|
| 417 |
+
"step": 250,
|
| 418 |
+
"token_acc": 0.7478949131874831
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 0.16666666666666666,
|
| 422 |
+
"grad_norm": 23.064035415649414,
|
| 423 |
+
"learning_rate": 9.634252538771587e-07,
|
| 424 |
+
"loss": 0.8869501113891601,
|
| 425 |
+
"step": 255,
|
| 426 |
+
"token_acc": 0.7499183354833708
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 0.16993464052287582,
|
| 430 |
+
"grad_norm": 9.165678024291992,
|
| 431 |
+
"learning_rate": 9.613688771840906e-07,
|
| 432 |
+
"loss": 0.8962480545043945,
|
| 433 |
+
"step": 260,
|
| 434 |
+
"token_acc": 0.7451406421339862
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.17320261437908496,
|
| 438 |
+
"grad_norm": 21.120023727416992,
|
| 439 |
+
"learning_rate": 9.592585801549396e-07,
|
| 440 |
+
"loss": 0.8865818023681641,
|
| 441 |
+
"step": 265,
|
| 442 |
+
"token_acc": 0.7520313141918108
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 0.17647058823529413,
|
| 446 |
+
"grad_norm": 10.798661231994629,
|
| 447 |
+
"learning_rate": 9.570946094208307e-07,
|
| 448 |
+
"loss": 0.8834690093994141,
|
| 449 |
+
"step": 270,
|
| 450 |
+
"token_acc": 0.7549796136621884
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.17973856209150327,
|
| 454 |
+
"grad_norm": 9.285658836364746,
|
| 455 |
+
"learning_rate": 9.548772178857525e-07,
|
| 456 |
+
"loss": 0.8783238410949707,
|
| 457 |
+
"step": 275,
|
| 458 |
+
"token_acc": 0.7535333435379357
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.1830065359477124,
|
| 462 |
+
"grad_norm": 14.306886672973633,
|
| 463 |
+
"learning_rate": 9.526066646970005e-07,
|
| 464 |
+
"loss": 0.8692483901977539,
|
| 465 |
+
"step": 280,
|
| 466 |
+
"token_acc": 0.7575995630212904
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 0.18627450980392157,
|
| 470 |
+
"grad_norm": 10.609308242797852,
|
| 471 |
+
"learning_rate": 9.502832152148906e-07,
|
| 472 |
+
"loss": 0.9016864776611329,
|
| 473 |
+
"step": 285,
|
| 474 |
+
"token_acc": 0.747729183048182
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 0.1895424836601307,
|
| 478 |
+
"grad_norm": 13.762345314025879,
|
| 479 |
+
"learning_rate": 9.479071409817466e-07,
|
| 480 |
+
"loss": 0.8667617797851562,
|
| 481 |
+
"step": 290,
|
| 482 |
+
"token_acc": 0.7545149479887847
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.19281045751633988,
|
| 486 |
+
"grad_norm": 26.821304321289062,
|
| 487 |
+
"learning_rate": 9.454787196901644e-07,
|
| 488 |
+
"loss": 0.882562255859375,
|
| 489 |
+
"step": 295,
|
| 490 |
+
"token_acc": 0.7533141679323063
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 0.19607843137254902,
|
| 494 |
+
"grad_norm": 21.474590301513672,
|
| 495 |
+
"learning_rate": 9.429982351505583e-07,
|
| 496 |
+
"loss": 0.881015396118164,
|
| 497 |
+
"step": 300,
|
| 498 |
+
"token_acc": 0.7526286986414185
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.19934640522875818,
|
| 502 |
+
"grad_norm": 17.721975326538086,
|
| 503 |
+
"learning_rate": 9.40465977257992e-07,
|
| 504 |
+
"loss": 0.8866046905517578,
|
| 505 |
+
"step": 305,
|
| 506 |
+
"token_acc": 0.7502130300766908
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.20261437908496732,
|
| 510 |
+
"grad_norm": 12.351274490356445,
|
| 511 |
+
"learning_rate": 9.378822419582984e-07,
|
| 512 |
+
"loss": 0.8542521476745606,
|
| 513 |
+
"step": 310,
|
| 514 |
+
"token_acc": 0.7567264063614428
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.20588235294117646,
|
| 518 |
+
"grad_norm": 10.410445213317871,
|
| 519 |
+
"learning_rate": 9.352473312134921e-07,
|
| 520 |
+
"loss": 0.8521980285644531,
|
| 521 |
+
"step": 315,
|
| 522 |
+
"token_acc": 0.7616740918124851
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.20915032679738563,
|
| 526 |
+
"grad_norm": 13.164457321166992,
|
| 527 |
+
"learning_rate": 9.325615529664795e-07,
|
| 528 |
+
"loss": 0.8816440582275391,
|
| 529 |
+
"step": 320,
|
| 530 |
+
"token_acc": 0.7543240287525349
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.21241830065359477,
|
| 534 |
+
"grad_norm": 9.843915939331055,
|
| 535 |
+
"learning_rate": 9.29825221105069e-07,
|
| 536 |
+
"loss": 0.8506572723388672,
|
| 537 |
+
"step": 325,
|
| 538 |
+
"token_acc": 0.7611864380654874
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 0.21568627450980393,
|
| 542 |
+
"grad_norm": 10.565733909606934,
|
| 543 |
+
"learning_rate": 9.270386554252869e-07,
|
| 544 |
+
"loss": 0.8983824729919434,
|
| 545 |
+
"step": 330,
|
| 546 |
+
"token_acc": 0.7484428678338038
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 0.21895424836601307,
|
| 550 |
+
"grad_norm": 10.865199089050293,
|
| 551 |
+
"learning_rate": 9.24202181594003e-07,
|
| 552 |
+
"loss": 0.8470457077026368,
|
| 553 |
+
"step": 335,
|
| 554 |
+
"token_acc": 0.7619183896067431
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.2222222222222222,
|
| 558 |
+
"grad_norm": 11.11334228515625,
|
| 559 |
+
"learning_rate": 9.21316131110869e-07,
|
| 560 |
+
"loss": 0.8703369140625,
|
| 561 |
+
"step": 340,
|
| 562 |
+
"token_acc": 0.7520899607901161
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.22549019607843138,
|
| 566 |
+
"grad_norm": 11.682340621948242,
|
| 567 |
+
"learning_rate": 9.183808412695775e-07,
|
| 568 |
+
"loss": 0.8469175338745117,
|
| 569 |
+
"step": 345,
|
| 570 |
+
"token_acc": 0.7620305956559453
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.22875816993464052,
|
| 574 |
+
"grad_norm": 17.082984924316406,
|
| 575 |
+
"learning_rate": 9.153966551184404e-07,
|
| 576 |
+
"loss": 0.8685555458068848,
|
| 577 |
+
"step": 350,
|
| 578 |
+
"token_acc": 0.7556311982827465
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.23202614379084968,
|
| 582 |
+
"grad_norm": 21.304964065551758,
|
| 583 |
+
"learning_rate": 9.12363921420299e-07,
|
| 584 |
+
"loss": 0.8746360778808594,
|
| 585 |
+
"step": 355,
|
| 586 |
+
"token_acc": 0.7553185041517668
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 0.23529411764705882,
|
| 590 |
+
"grad_norm": 17.049705505371094,
|
| 591 |
+
"learning_rate": 9.092829946117615e-07,
|
| 592 |
+
"loss": 0.8922021865844727,
|
| 593 |
+
"step": 360,
|
| 594 |
+
"token_acc": 0.7508404020938327
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 0.238562091503268,
|
| 598 |
+
"grad_norm": 12.26673412322998,
|
| 599 |
+
"learning_rate": 9.061542347617824e-07,
|
| 600 |
+
"loss": 0.8763910293579101,
|
| 601 |
+
"step": 365,
|
| 602 |
+
"token_acc": 0.7520791863223625
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.24183006535947713,
|
| 606 |
+
"grad_norm": 10.861994743347168,
|
| 607 |
+
"learning_rate": 9.029780075295787e-07,
|
| 608 |
+
"loss": 0.8442388534545898,
|
| 609 |
+
"step": 370,
|
| 610 |
+
"token_acc": 0.7629595850721794
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 0.24509803921568626,
|
| 614 |
+
"grad_norm": 9.749073028564453,
|
| 615 |
+
"learning_rate": 8.99754684121897e-07,
|
| 616 |
+
"loss": 0.87369384765625,
|
| 617 |
+
"step": 375,
|
| 618 |
+
"token_acc": 0.7515194799516098
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.24836601307189543,
|
| 622 |
+
"grad_norm": 9.403931617736816,
|
| 623 |
+
"learning_rate": 8.964846412496301e-07,
|
| 624 |
+
"loss": 0.8741621017456055,
|
| 625 |
+
"step": 380,
|
| 626 |
+
"token_acc": 0.7517588448018613
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.25163398692810457,
|
| 630 |
+
"grad_norm": 14.506192207336426,
|
| 631 |
+
"learning_rate": 8.931682610837896e-07,
|
| 632 |
+
"loss": 0.8476000785827636,
|
| 633 |
+
"step": 385,
|
| 634 |
+
"token_acc": 0.760447963392239
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.2549019607843137,
|
| 638 |
+
"grad_norm": 13.438100814819336,
|
| 639 |
+
"learning_rate": 8.898059312108425e-07,
|
| 640 |
+
"loss": 0.8526718139648437,
|
| 641 |
+
"step": 390,
|
| 642 |
+
"token_acc": 0.7612079221347601
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 0.2581699346405229,
|
| 646 |
+
"grad_norm": 9.219259262084961,
|
| 647 |
+
"learning_rate": 8.86398044587414e-07,
|
| 648 |
+
"loss": 0.8688133239746094,
|
| 649 |
+
"step": 395,
|
| 650 |
+
"token_acc": 0.7554470808818775
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.26143790849673204,
|
| 654 |
+
"grad_norm": 10.211725234985352,
|
| 655 |
+
"learning_rate": 8.829449994943612e-07,
|
| 656 |
+
"loss": 0.8704776763916016,
|
| 657 |
+
"step": 400,
|
| 658 |
+
"token_acc": 0.75205035073613
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 0.2647058823529412,
|
| 662 |
+
"grad_norm": 11.363197326660156,
|
| 663 |
+
"learning_rate": 8.794471994902276e-07,
|
| 664 |
+
"loss": 0.848613166809082,
|
| 665 |
+
"step": 405,
|
| 666 |
+
"token_acc": 0.7580784854970022
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 0.2679738562091503,
|
| 670 |
+
"grad_norm": 14.390101432800293,
|
| 671 |
+
"learning_rate": 8.759050533640777e-07,
|
| 672 |
+
"loss": 0.8752598762512207,
|
| 673 |
+
"step": 410,
|
| 674 |
+
"token_acc": 0.7534999311141491
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.27124183006535946,
|
| 678 |
+
"grad_norm": 9.734460830688477,
|
| 679 |
+
"learning_rate": 8.723189750877224e-07,
|
| 680 |
+
"loss": 0.8558663368225098,
|
| 681 |
+
"step": 415,
|
| 682 |
+
"token_acc": 0.7604358532874189
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.27450980392156865,
|
| 686 |
+
"grad_norm": 22.253286361694336,
|
| 687 |
+
"learning_rate": 8.686893837673379e-07,
|
| 688 |
+
"loss": 0.8463081359863281,
|
| 689 |
+
"step": 420,
|
| 690 |
+
"token_acc": 0.7608954284281454
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.2777777777777778,
|
| 694 |
+
"grad_norm": 14.327561378479004,
|
| 695 |
+
"learning_rate": 8.650167035944841e-07,
|
| 696 |
+
"loss": 0.8459568023681641,
|
| 697 |
+
"step": 425,
|
| 698 |
+
"token_acc": 0.7599806450108435
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.28104575163398693,
|
| 702 |
+
"grad_norm": 11.866144180297852,
|
| 703 |
+
"learning_rate": 8.613013637965303e-07,
|
| 704 |
+
"loss": 0.8597783088684082,
|
| 705 |
+
"step": 430,
|
| 706 |
+
"token_acc": 0.75692995190999
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 0.28431372549019607,
|
| 710 |
+
"grad_norm": 8.261825561523438,
|
| 711 |
+
"learning_rate": 8.5754379858649e-07,
|
| 712 |
+
"loss": 0.8413874626159668,
|
| 713 |
+
"step": 435,
|
| 714 |
+
"token_acc": 0.763401763459174
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 0.2875816993464052,
|
| 718 |
+
"grad_norm": 8.602568626403809,
|
| 719 |
+
"learning_rate": 8.537444471122747e-07,
|
| 720 |
+
"loss": 0.8405612945556641,
|
| 721 |
+
"step": 440,
|
| 722 |
+
"token_acc": 0.7626938602082005
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.2908496732026144,
|
| 726 |
+
"grad_norm": 10.16855239868164,
|
| 727 |
+
"learning_rate": 8.499037534053712e-07,
|
| 728 |
+
"loss": 0.83404541015625,
|
| 729 |
+
"step": 445,
|
| 730 |
+
"token_acc": 0.7660154774821217
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.29411764705882354,
|
| 734 |
+
"grad_norm": 9.481566429138184,
|
| 735 |
+
"learning_rate": 8.460221663289468e-07,
|
| 736 |
+
"loss": 0.822332763671875,
|
| 737 |
+
"step": 450,
|
| 738 |
+
"token_acc": 0.7673596517994392
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.2973856209150327,
|
| 742 |
+
"grad_norm": 11.954257011413574,
|
| 743 |
+
"learning_rate": 8.421001395253906e-07,
|
| 744 |
+
"loss": 0.8428705215454102,
|
| 745 |
+
"step": 455,
|
| 746 |
+
"token_acc": 0.7629577134757934
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.3006535947712418,
|
| 750 |
+
"grad_norm": 14.700212478637695,
|
| 751 |
+
"learning_rate": 8.38138131363296e-07,
|
| 752 |
+
"loss": 0.8308051109313965,
|
| 753 |
+
"step": 460,
|
| 754 |
+
"token_acc": 0.7647826887860696
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 0.30392156862745096,
|
| 758 |
+
"grad_norm": 11.61778736114502,
|
| 759 |
+
"learning_rate": 8.341366048838915e-07,
|
| 760 |
+
"loss": 0.8619503974914551,
|
| 761 |
+
"step": 465,
|
| 762 |
+
"token_acc": 0.7568585376321013
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 0.30718954248366015,
|
| 766 |
+
"grad_norm": 10.546567916870117,
|
| 767 |
+
"learning_rate": 8.300960277469247e-07,
|
| 768 |
+
"loss": 0.8505170822143555,
|
| 769 |
+
"step": 470,
|
| 770 |
+
"token_acc": 0.7596334874691328
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.3104575163398693,
|
| 774 |
+
"grad_norm": 13.461160659790039,
|
| 775 |
+
"learning_rate": 8.260168721760066e-07,
|
| 776 |
+
"loss": 0.8434528350830078,
|
| 777 |
+
"step": 475,
|
| 778 |
+
"token_acc": 0.7591824849220685
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 0.3137254901960784,
|
| 782 |
+
"grad_norm": 9.887128829956055,
|
| 783 |
+
"learning_rate": 8.218996149034228e-07,
|
| 784 |
+
"loss": 0.8470413208007812,
|
| 785 |
+
"step": 480,
|
| 786 |
+
"token_acc": 0.7591767159107208
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.31699346405228757,
|
| 790 |
+
"grad_norm": 11.56584644317627,
|
| 791 |
+
"learning_rate": 8.177447371144174e-07,
|
| 792 |
+
"loss": 0.8538051605224609,
|
| 793 |
+
"step": 485,
|
| 794 |
+
"token_acc": 0.7558111015730388
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.3202614379084967,
|
| 798 |
+
"grad_norm": 12.656098365783691,
|
| 799 |
+
"learning_rate": 8.135527243909572e-07,
|
| 800 |
+
"loss": 0.8436816215515137,
|
| 801 |
+
"step": 490,
|
| 802 |
+
"token_acc": 0.7604942116390075
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 0.3235294117647059,
|
| 806 |
+
"grad_norm": 11.942681312561035,
|
| 807 |
+
"learning_rate": 8.093240666549815e-07,
|
| 808 |
+
"loss": 0.8648993492126464,
|
| 809 |
+
"step": 495,
|
| 810 |
+
"token_acc": 0.7569572858999387
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 0.32679738562091504,
|
| 814 |
+
"grad_norm": 15.744888305664062,
|
| 815 |
+
"learning_rate": 8.050592581111439e-07,
|
| 816 |
+
"loss": 0.8473841667175293,
|
| 817 |
+
"step": 500,
|
| 818 |
+
"token_acc": 0.7625280745326853
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 0.3300653594771242,
|
| 822 |
+
"grad_norm": 18.36495018005371,
|
| 823 |
+
"learning_rate": 8.007587971890558e-07,
|
| 824 |
+
"loss": 0.8407678604125977,
|
| 825 |
+
"step": 505,
|
| 826 |
+
"token_acc": 0.7635814445718317
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 0.3333333333333333,
|
| 830 |
+
"grad_norm": 12.46922779083252,
|
| 831 |
+
"learning_rate": 7.964231864850335e-07,
|
| 832 |
+
"loss": 0.8600395202636719,
|
| 833 |
+
"step": 510,
|
| 834 |
+
"token_acc": 0.7575112788369374
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 0.3366013071895425,
|
| 838 |
+
"grad_norm": 40.35283279418945,
|
| 839 |
+
"learning_rate": 7.920529327033603e-07,
|
| 840 |
+
"loss": 0.839472770690918,
|
| 841 |
+
"step": 515,
|
| 842 |
+
"token_acc": 0.7620206747770853
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.33986928104575165,
|
| 846 |
+
"grad_norm": 13.720703125,
|
| 847 |
+
"learning_rate": 7.876485465970679e-07,
|
| 848 |
+
"loss": 0.8494855880737304,
|
| 849 |
+
"step": 520,
|
| 850 |
+
"token_acc": 0.7581153564424281
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 0.3431372549019608,
|
| 854 |
+
"grad_norm": 13.580601692199707,
|
| 855 |
+
"learning_rate": 7.832105429082441e-07,
|
| 856 |
+
"loss": 0.8347602844238281,
|
| 857 |
+
"step": 525,
|
| 858 |
+
"token_acc": 0.7627168182380251
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 0.3464052287581699,
|
| 862 |
+
"grad_norm": 16.54994773864746,
|
| 863 |
+
"learning_rate": 7.787394403078746e-07,
|
| 864 |
+
"loss": 0.8405153274536132,
|
| 865 |
+
"step": 530,
|
| 866 |
+
"token_acc": 0.7606589147286822
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.34967320261437906,
|
| 870 |
+
"grad_norm": 16.629894256591797,
|
| 871 |
+
"learning_rate": 7.742357613352268e-07,
|
| 872 |
+
"loss": 0.8442140579223633,
|
| 873 |
+
"step": 535,
|
| 874 |
+
"token_acc": 0.7593705080261373
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 0.35294117647058826,
|
| 878 |
+
"grad_norm": 11.812152862548828,
|
| 879 |
+
"learning_rate": 7.69700032336779e-07,
|
| 880 |
+
"loss": 0.8355265617370605,
|
| 881 |
+
"step": 540,
|
| 882 |
+
"token_acc": 0.7614466910847967
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 0.3562091503267974,
|
| 886 |
+
"grad_norm": 13.99330997467041,
|
| 887 |
+
"learning_rate": 7.65132783404707e-07,
|
| 888 |
+
"loss": 0.8402891159057617,
|
| 889 |
+
"step": 545,
|
| 890 |
+
"token_acc": 0.76289203646244
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.35947712418300654,
|
| 894 |
+
"grad_norm": 17.06981086730957,
|
| 895 |
+
"learning_rate": 7.605345483149322e-07,
|
| 896 |
+
"loss": 0.8363126754760742,
|
| 897 |
+
"step": 550,
|
| 898 |
+
"token_acc": 0.7636697957998645
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.3627450980392157,
|
| 902 |
+
"grad_norm": 11.00927448272705,
|
| 903 |
+
"learning_rate": 7.559058644647378e-07,
|
| 904 |
+
"loss": 0.8396165847778321,
|
| 905 |
+
"step": 555,
|
| 906 |
+
"token_acc": 0.7614336156707792
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 0.3660130718954248,
|
| 910 |
+
"grad_norm": 41.30654525756836,
|
| 911 |
+
"learning_rate": 7.51247272809965e-07,
|
| 912 |
+
"loss": 0.8121014595031738,
|
| 913 |
+
"step": 560,
|
| 914 |
+
"token_acc": 0.7668876131857726
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 0.369281045751634,
|
| 918 |
+
"grad_norm": 15.057770729064941,
|
| 919 |
+
"learning_rate": 7.465593178017896e-07,
|
| 920 |
+
"loss": 0.8397438049316406,
|
| 921 |
+
"step": 565,
|
| 922 |
+
"token_acc": 0.7622377622377622
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 0.37254901960784315,
|
| 926 |
+
"grad_norm": 10.871357917785645,
|
| 927 |
+
"learning_rate": 7.418425473230923e-07,
|
| 928 |
+
"loss": 0.8423866271972656,
|
| 929 |
+
"step": 570,
|
| 930 |
+
"token_acc": 0.7609689785203858
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 0.3758169934640523,
|
| 934 |
+
"grad_norm": 13.613224983215332,
|
| 935 |
+
"learning_rate": 7.370975126244279e-07,
|
| 936 |
+
"loss": 0.8342062950134277,
|
| 937 |
+
"step": 575,
|
| 938 |
+
"token_acc": 0.7610806711328252
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 0.3790849673202614,
|
| 942 |
+
"grad_norm": 12.01113224029541,
|
| 943 |
+
"learning_rate": 7.323247682596e-07,
|
| 944 |
+
"loss": 0.8046956062316895,
|
| 945 |
+
"step": 580,
|
| 946 |
+
"token_acc": 0.7730626214140806
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 0.38235294117647056,
|
| 950 |
+
"grad_norm": 13.857526779174805,
|
| 951 |
+
"learning_rate": 7.275248720208499e-07,
|
| 952 |
+
"loss": 0.8374439239501953,
|
| 953 |
+
"step": 585,
|
| 954 |
+
"token_acc": 0.7629432436561946
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.38562091503267976,
|
| 958 |
+
"grad_norm": 11.863401412963867,
|
| 959 |
+
"learning_rate": 7.226983848736678e-07,
|
| 960 |
+
"loss": 0.8024310111999512,
|
| 961 |
+
"step": 590,
|
| 962 |
+
"token_acc": 0.772104691402269
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.3888888888888889,
|
| 966 |
+
"grad_norm": 12.410170555114746,
|
| 967 |
+
"learning_rate": 7.178458708912327e-07,
|
| 968 |
+
"loss": 0.8280567169189453,
|
| 969 |
+
"step": 595,
|
| 970 |
+
"token_acc": 0.7643324462186738
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 0.39215686274509803,
|
| 974 |
+
"grad_norm": 17.5835018157959,
|
| 975 |
+
"learning_rate": 7.12967897188488e-07,
|
| 976 |
+
"loss": 0.8388355255126954,
|
| 977 |
+
"step": 600,
|
| 978 |
+
"token_acc": 0.7613294700405282
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 0.3954248366013072,
|
| 982 |
+
"grad_norm": 16.445173263549805,
|
| 983 |
+
"learning_rate": 7.080650338558633e-07,
|
| 984 |
+
"loss": 0.8303842544555664,
|
| 985 |
+
"step": 605,
|
| 986 |
+
"token_acc": 0.7617286152323862
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 0.39869281045751637,
|
| 990 |
+
"grad_norm": 16.148969650268555,
|
| 991 |
+
"learning_rate": 7.03137853892648e-07,
|
| 992 |
+
"loss": 0.8134881973266601,
|
| 993 |
+
"step": 610,
|
| 994 |
+
"token_acc": 0.7701163229592012
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 0.4019607843137255,
|
| 998 |
+
"grad_norm": 20.40843963623047,
|
| 999 |
+
"learning_rate": 6.981869331400237e-07,
|
| 1000 |
+
"loss": 0.8203607559204101,
|
| 1001 |
+
"step": 615,
|
| 1002 |
+
"token_acc": 0.7647343072982942
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 0.40522875816993464,
|
| 1006 |
+
"grad_norm": 18.9699649810791,
|
| 1007 |
+
"learning_rate": 6.93212850213767e-07,
|
| 1008 |
+
"loss": 0.8277750968933105,
|
| 1009 |
+
"step": 620,
|
| 1010 |
+
"token_acc": 0.7626589656634167
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.4084967320261438,
|
| 1014 |
+
"grad_norm": 12.926063537597656,
|
| 1015 |
+
"learning_rate": 6.882161864366243e-07,
|
| 1016 |
+
"loss": 0.8561537742614747,
|
| 1017 |
+
"step": 625,
|
| 1018 |
+
"token_acc": 0.7584628082581394
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 0.4117647058823529,
|
| 1022 |
+
"grad_norm": 123.43069458007812,
|
| 1023 |
+
"learning_rate": 6.831975257703747e-07,
|
| 1024 |
+
"loss": 0.8185005187988281,
|
| 1025 |
+
"step": 630,
|
| 1026 |
+
"token_acc": 0.7683428166943453
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 0.4150326797385621,
|
| 1030 |
+
"grad_norm": 16.874107360839844,
|
| 1031 |
+
"learning_rate": 6.781574547475805e-07,
|
| 1032 |
+
"loss": 0.8033872604370117,
|
| 1033 |
+
"step": 635,
|
| 1034 |
+
"token_acc": 0.772021244085693
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 0.41830065359477125,
|
| 1038 |
+
"grad_norm": 35.09786605834961,
|
| 1039 |
+
"learning_rate": 6.730965624030401e-07,
|
| 1040 |
+
"loss": 0.8158361434936523,
|
| 1041 |
+
"step": 640,
|
| 1042 |
+
"token_acc": 0.7684266243218209
|
| 1043 |
+
},
|
| 1044 |
+
{
|
| 1045 |
+
"epoch": 0.4215686274509804,
|
| 1046 |
+
"grad_norm": 13.792229652404785,
|
| 1047 |
+
"learning_rate": 6.680154402049463e-07,
|
| 1048 |
+
"loss": 0.8064857482910156,
|
| 1049 |
+
"step": 645,
|
| 1050 |
+
"token_acc": 0.7695724083180806
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 0.42483660130718953,
|
| 1054 |
+
"grad_norm": 16.574562072753906,
|
| 1055 |
+
"learning_rate": 6.629146819857618e-07,
|
| 1056 |
+
"loss": 0.8209724426269531,
|
| 1057 |
+
"step": 650,
|
| 1058 |
+
"token_acc": 0.7657072096554584
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.42810457516339867,
|
| 1062 |
+
"grad_norm": 17.467939376831055,
|
| 1063 |
+
"learning_rate": 6.577948838728183e-07,
|
| 1064 |
+
"loss": 0.8391878128051757,
|
| 1065 |
+
"step": 655,
|
| 1066 |
+
"token_acc": 0.7623057477457853
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.43137254901960786,
|
| 1070 |
+
"grad_norm": 18.3452205657959,
|
| 1071 |
+
"learning_rate": 6.52656644218645e-07,
|
| 1072 |
+
"loss": 0.849530029296875,
|
| 1073 |
+
"step": 660,
|
| 1074 |
+
"token_acc": 0.7606905888666634
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 0.434640522875817,
|
| 1078 |
+
"grad_norm": 11.495086669921875,
|
| 1079 |
+
"learning_rate": 6.475005635310417e-07,
|
| 1080 |
+
"loss": 0.818548583984375,
|
| 1081 |
+
"step": 665,
|
| 1082 |
+
"token_acc": 0.7667387092800592
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 0.43790849673202614,
|
| 1086 |
+
"grad_norm": 14.218070983886719,
|
| 1087 |
+
"learning_rate": 6.423272444028947e-07,
|
| 1088 |
+
"loss": 0.8181242942810059,
|
| 1089 |
+
"step": 670,
|
| 1090 |
+
"token_acc": 0.7692475405477267
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 0.4411764705882353,
|
| 1094 |
+
"grad_norm": 20.244361877441406,
|
| 1095 |
+
"learning_rate": 6.371372914417535e-07,
|
| 1096 |
+
"loss": 0.8055562019348145,
|
| 1097 |
+
"step": 675,
|
| 1098 |
+
"token_acc": 0.7723559649837134
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 0.4444444444444444,
|
| 1102 |
+
"grad_norm": 20.16501808166504,
|
| 1103 |
+
"learning_rate": 6.319313111991684e-07,
|
| 1104 |
+
"loss": 0.8287031173706054,
|
| 1105 |
+
"step": 680,
|
| 1106 |
+
"token_acc": 0.7660015369531229
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 0.4477124183006536,
|
| 1110 |
+
"grad_norm": 21.247982025146484,
|
| 1111 |
+
"learning_rate": 6.267099120998041e-07,
|
| 1112 |
+
"loss": 0.813420295715332,
|
| 1113 |
+
"step": 685,
|
| 1114 |
+
"token_acc": 0.7699741107609185
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 0.45098039215686275,
|
| 1118 |
+
"grad_norm": 8.700778007507324,
|
| 1119 |
+
"learning_rate": 6.214737043703322e-07,
|
| 1120 |
+
"loss": 0.8061553001403808,
|
| 1121 |
+
"step": 690,
|
| 1122 |
+
"token_acc": 0.7697841726618705
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.4542483660130719,
|
| 1126 |
+
"grad_norm": 9.28950023651123,
|
| 1127 |
+
"learning_rate": 6.162232999681138e-07,
|
| 1128 |
+
"loss": 0.8056695938110352,
|
| 1129 |
+
"step": 695,
|
| 1130 |
+
"token_acc": 0.7695348439138157
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.45751633986928103,
|
| 1134 |
+
"grad_norm": 10.441835403442383,
|
| 1135 |
+
"learning_rate": 6.109593125096798e-07,
|
| 1136 |
+
"loss": 0.8270071029663086,
|
| 1137 |
+
"step": 700,
|
| 1138 |
+
"token_acc": 0.7612057232395091
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 0.46078431372549017,
|
| 1142 |
+
"grad_norm": 10.06371021270752,
|
| 1143 |
+
"learning_rate": 6.056823571990176e-07,
|
| 1144 |
+
"loss": 0.8276515007019043,
|
| 1145 |
+
"step": 705,
|
| 1146 |
+
"token_acc": 0.7636677721024963
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 0.46405228758169936,
|
| 1150 |
+
"grad_norm": 7.716113090515137,
|
| 1151 |
+
"learning_rate": 6.003930507556718e-07,
|
| 1152 |
+
"loss": 0.8147080421447754,
|
| 1153 |
+
"step": 710,
|
| 1154 |
+
"token_acc": 0.7682322895847213
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 0.4673202614379085,
|
| 1158 |
+
"grad_norm": 15.472831726074219,
|
| 1159 |
+
"learning_rate": 5.95092011342668e-07,
|
| 1160 |
+
"loss": 0.816691780090332,
|
| 1161 |
+
"step": 715,
|
| 1162 |
+
"token_acc": 0.7655702665795879
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 0.47058823529411764,
|
| 1166 |
+
"grad_norm": 12.411829948425293,
|
| 1167 |
+
"learning_rate": 5.89779858494268e-07,
|
| 1168 |
+
"loss": 0.8144680023193359,
|
| 1169 |
+
"step": 720,
|
| 1170 |
+
"token_acc": 0.7690880067166364
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 0.4738562091503268,
|
| 1174 |
+
"grad_norm": 10.708953857421875,
|
| 1175 |
+
"learning_rate": 5.844572130435648e-07,
|
| 1176 |
+
"loss": 0.8147815704345703,
|
| 1177 |
+
"step": 725,
|
| 1178 |
+
"token_acc": 0.767069484777139
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.477124183006536,
|
| 1182 |
+
"grad_norm": 14.9004487991333,
|
| 1183 |
+
"learning_rate": 5.79124697049925e-07,
|
| 1184 |
+
"loss": 0.8133182525634766,
|
| 1185 |
+
"step": 730,
|
| 1186 |
+
"token_acc": 0.7692034514330106
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 0.4803921568627451,
|
| 1190 |
+
"grad_norm": 19.81015396118164,
|
| 1191 |
+
"learning_rate": 5.737829337262887e-07,
|
| 1192 |
+
"loss": 0.8090297698974609,
|
| 1193 |
+
"step": 735,
|
| 1194 |
+
"token_acc": 0.7716722811138239
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 0.48366013071895425,
|
| 1198 |
+
"grad_norm": 18.869447708129883,
|
| 1199 |
+
"learning_rate": 5.684325473663348e-07,
|
| 1200 |
+
"loss": 0.8260050773620605,
|
| 1201 |
+
"step": 740,
|
| 1202 |
+
"token_acc": 0.7651222962759748
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 0.4869281045751634,
|
| 1206 |
+
"grad_norm": 10.783040046691895,
|
| 1207 |
+
"learning_rate": 5.630741632715197e-07,
|
| 1208 |
+
"loss": 0.8219170570373535,
|
| 1209 |
+
"step": 745,
|
| 1210 |
+
"token_acc": 0.7662494168869538
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 0.49019607843137253,
|
| 1214 |
+
"grad_norm": 14.318591117858887,
|
| 1215 |
+
"learning_rate": 5.577084076779975e-07,
|
| 1216 |
+
"loss": 0.8044301986694335,
|
| 1217 |
+
"step": 750,
|
| 1218 |
+
"token_acc": 0.7715717924176305
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 0.4934640522875817,
|
| 1222 |
+
"grad_norm": 16.11661720275879,
|
| 1223 |
+
"learning_rate": 5.523359076834315e-07,
|
| 1224 |
+
"loss": 0.7847557544708252,
|
| 1225 |
+
"step": 755,
|
| 1226 |
+
"token_acc": 0.776726990363275
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 0.49673202614379086,
|
| 1230 |
+
"grad_norm": 13.084589958190918,
|
| 1231 |
+
"learning_rate": 5.469572911737059e-07,
|
| 1232 |
+
"loss": 0.8072048187255859,
|
| 1233 |
+
"step": 760,
|
| 1234 |
+
"token_acc": 0.7706756866710837
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.5,
|
| 1238 |
+
"grad_norm": 74.14442443847656,
|
| 1239 |
+
"learning_rate": 5.415731867495439e-07,
|
| 1240 |
+
"loss": 0.7891597270965576,
|
| 1241 |
+
"step": 765,
|
| 1242 |
+
"token_acc": 0.7766007913661308
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 0.5032679738562091,
|
| 1246 |
+
"grad_norm": 71.82379150390625,
|
| 1247 |
+
"learning_rate": 5.361842236530431e-07,
|
| 1248 |
+
"loss": 0.808492374420166,
|
| 1249 |
+
"step": 770,
|
| 1250 |
+
"token_acc": 0.7684806201550387
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 0.5065359477124183,
|
| 1254 |
+
"grad_norm": 15.662697792053223,
|
| 1255 |
+
"learning_rate": 5.307910316941358e-07,
|
| 1256 |
+
"loss": 0.7847880363464356,
|
| 1257 |
+
"step": 775,
|
| 1258 |
+
"token_acc": 0.7714996285649688
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 0.5098039215686274,
|
| 1262 |
+
"grad_norm": 10.902754783630371,
|
| 1263 |
+
"learning_rate": 5.253942411769828e-07,
|
| 1264 |
+
"loss": 0.8251161575317383,
|
| 1265 |
+
"step": 780,
|
| 1266 |
+
"token_acc": 0.767316236694041
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 0.5130718954248366,
|
| 1270 |
+
"grad_norm": 16.743911743164062,
|
| 1271 |
+
"learning_rate": 5.199944828263099e-07,
|
| 1272 |
+
"loss": 0.8166553497314453,
|
| 1273 |
+
"step": 785,
|
| 1274 |
+
"token_acc": 0.7665164246297607
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 0.5163398692810458,
|
| 1278 |
+
"grad_norm": 15.667343139648438,
|
| 1279 |
+
"learning_rate": 5.145923877136939e-07,
|
| 1280 |
+
"loss": 0.8361772537231446,
|
| 1281 |
+
"step": 790,
|
| 1282 |
+
"token_acc": 0.7632328585624379
|
| 1283 |
+
},
|
| 1284 |
+
{
|
| 1285 |
+
"epoch": 0.5196078431372549,
|
| 1286 |
+
"grad_norm": 51.58697509765625,
|
| 1287 |
+
"learning_rate": 5.091885871838107e-07,
|
| 1288 |
+
"loss": 0.7939982891082764,
|
| 1289 |
+
"step": 795,
|
| 1290 |
+
"token_acc": 0.7730233833916204
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.5228758169934641,
|
| 1294 |
+
"grad_norm": 9.2147855758667,
|
| 1295 |
+
"learning_rate": 5.037837127806485e-07,
|
| 1296 |
+
"loss": 0.805908489227295,
|
| 1297 |
+
"step": 800,
|
| 1298 |
+
"token_acc": 0.7726681647287982
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.5261437908496732,
|
| 1302 |
+
"grad_norm": 10.623895645141602,
|
| 1303 |
+
"learning_rate": 4.983783961736989e-07,
|
| 1304 |
+
"loss": 0.8057775497436523,
|
| 1305 |
+
"step": 805,
|
| 1306 |
+
"token_acc": 0.7718298957833689
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 0.5294117647058824,
|
| 1310 |
+
"grad_norm": 11.640973091125488,
|
| 1311 |
+
"learning_rate": 4.929732690841345e-07,
|
| 1312 |
+
"loss": 0.8148061752319335,
|
| 1313 |
+
"step": 810,
|
| 1314 |
+
"token_acc": 0.770098414434117
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 0.5326797385620915,
|
| 1318 |
+
"grad_norm": 8.896126747131348,
|
| 1319 |
+
"learning_rate": 4.875689632109795e-07,
|
| 1320 |
+
"loss": 0.8047978401184082,
|
| 1321 |
+
"step": 815,
|
| 1322 |
+
"token_acc": 0.7708750523913752
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 0.5359477124183006,
|
| 1326 |
+
"grad_norm": 16.067562103271484,
|
| 1327 |
+
"learning_rate": 4.821661101572809e-07,
|
| 1328 |
+
"loss": 0.8254037857055664,
|
| 1329 |
+
"step": 820,
|
| 1330 |
+
"token_acc": 0.7648166316016309
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 0.5392156862745098,
|
| 1334 |
+
"grad_norm": 8.182540893554688,
|
| 1335 |
+
"learning_rate": 4.767653413562956e-07,
|
| 1336 |
+
"loss": 0.8067924499511718,
|
| 1337 |
+
"step": 825,
|
| 1338 |
+
"token_acc": 0.7700773824096052
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 0.5424836601307189,
|
| 1342 |
+
"grad_norm": 21.027469635009766,
|
| 1343 |
+
"learning_rate": 4.713672879976923e-07,
|
| 1344 |
+
"loss": 0.8266950607299804,
|
| 1345 |
+
"step": 830,
|
| 1346 |
+
"token_acc": 0.7643826263160213
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.545751633986928,
|
| 1350 |
+
"grad_norm": 13.580276489257812,
|
| 1351 |
+
"learning_rate": 4.65972580953785e-07,
|
| 1352 |
+
"loss": 0.8016036987304688,
|
| 1353 |
+
"step": 835,
|
| 1354 |
+
"token_acc": 0.7704610326865622
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 0.5490196078431373,
|
| 1358 |
+
"grad_norm": 10.898624420166016,
|
| 1359 |
+
"learning_rate": 4.605818507058036e-07,
|
| 1360 |
+
"loss": 0.8136810302734375,
|
| 1361 |
+
"step": 840,
|
| 1362 |
+
"token_acc": 0.769013163358525
|
| 1363 |
+
},
|
| 1364 |
+
{
|
| 1365 |
+
"epoch": 0.5522875816993464,
|
| 1366 |
+
"grad_norm": 11.562179565429688,
|
| 1367 |
+
"learning_rate": 4.551957272702075e-07,
|
| 1368 |
+
"loss": 0.81622314453125,
|
| 1369 |
+
"step": 845,
|
| 1370 |
+
"token_acc": 0.7676546284765463
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 0.5555555555555556,
|
| 1374 |
+
"grad_norm": 11.148411750793457,
|
| 1375 |
+
"learning_rate": 4.4981484012505717e-07,
|
| 1376 |
+
"loss": 0.8086343765258789,
|
| 1377 |
+
"step": 850,
|
| 1378 |
+
"token_acc": 0.7677541734308541
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 0.5588235294117647,
|
| 1382 |
+
"grad_norm": 22.669891357421875,
|
| 1383 |
+
"learning_rate": 4.444398181364455e-07,
|
| 1384 |
+
"loss": 0.7971403121948242,
|
| 1385 |
+
"step": 855,
|
| 1386 |
+
"token_acc": 0.7708641916081483
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 0.5620915032679739,
|
| 1390 |
+
"grad_norm": 12.720222473144531,
|
| 1391 |
+
"learning_rate": 4.390712894850032e-07,
|
| 1392 |
+
"loss": 0.8296154975891114,
|
| 1393 |
+
"step": 860,
|
| 1394 |
+
"token_acc": 0.7647365466356564
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 0.565359477124183,
|
| 1398 |
+
"grad_norm": 16.401975631713867,
|
| 1399 |
+
"learning_rate": 4.337098815924818e-07,
|
| 1400 |
+
"loss": 0.7979914665222168,
|
| 1401 |
+
"step": 865,
|
| 1402 |
+
"token_acc": 0.770620347394541
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.5686274509803921,
|
| 1406 |
+
"grad_norm": 9.437167167663574,
|
| 1407 |
+
"learning_rate": 4.2835622104842737e-07,
|
| 1408 |
+
"loss": 0.8097938537597656,
|
| 1409 |
+
"step": 870,
|
| 1410 |
+
"token_acc": 0.7671590709481474
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 0.5718954248366013,
|
| 1414 |
+
"grad_norm": 12.183428764343262,
|
| 1415 |
+
"learning_rate": 4.2301093353695157e-07,
|
| 1416 |
+
"loss": 0.7752901077270508,
|
| 1417 |
+
"step": 875,
|
| 1418 |
+
"token_acc": 0.7773678010180266
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 0.5751633986928104,
|
| 1422 |
+
"grad_norm": 57.63277053833008,
|
| 1423 |
+
"learning_rate": 4.1767464376360583e-07,
|
| 1424 |
+
"loss": 0.7992642402648926,
|
| 1425 |
+
"step": 880,
|
| 1426 |
+
"token_acc": 0.7730707893309073
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 0.5784313725490197,
|
| 1430 |
+
"grad_norm": 11.309215545654297,
|
| 1431 |
+
"learning_rate": 4.123479753823739e-07,
|
| 1432 |
+
"loss": 0.8178821563720703,
|
| 1433 |
+
"step": 885,
|
| 1434 |
+
"token_acc": 0.7657045458380914
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 0.5816993464052288,
|
| 1438 |
+
"grad_norm": 9.985339164733887,
|
| 1439 |
+
"learning_rate": 4.070315509227832e-07,
|
| 1440 |
+
"loss": 0.8014745712280273,
|
| 1441 |
+
"step": 890,
|
| 1442 |
+
"token_acc": 0.7731357939120892
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 0.5849673202614379,
|
| 1446 |
+
"grad_norm": 8.616693496704102,
|
| 1447 |
+
"learning_rate": 4.017259917171516e-07,
|
| 1448 |
+
"loss": 0.8079689025878907,
|
| 1449 |
+
"step": 895,
|
| 1450 |
+
"token_acc": 0.7677836465907532
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 0.5882352941176471,
|
| 1454 |
+
"grad_norm": 12.550949096679688,
|
| 1455 |
+
"learning_rate": 3.964319178279703e-07,
|
| 1456 |
+
"loss": 0.7903870582580567,
|
| 1457 |
+
"step": 900,
|
| 1458 |
+
"token_acc": 0.7721659602463304
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.5915032679738562,
|
| 1462 |
+
"grad_norm": 8.316669464111328,
|
| 1463 |
+
"learning_rate": 3.911499479754379e-07,
|
| 1464 |
+
"loss": 0.8000009536743165,
|
| 1465 |
+
"step": 905,
|
| 1466 |
+
"token_acc": 0.773729905599319
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.5947712418300654,
|
| 1470 |
+
"grad_norm": 15.755565643310547,
|
| 1471 |
+
"learning_rate": 3.8588069946515053e-07,
|
| 1472 |
+
"loss": 0.7649762153625488,
|
| 1473 |
+
"step": 910,
|
| 1474 |
+
"token_acc": 0.7819265047002808
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 0.5980392156862745,
|
| 1478 |
+
"grad_norm": 10.586187362670898,
|
| 1479 |
+
"learning_rate": 3.80624788115956e-07,
|
| 1480 |
+
"loss": 0.8030073165893554,
|
| 1481 |
+
"step": 915,
|
| 1482 |
+
"token_acc": 0.7693005012501629
|
| 1483 |
+
},
|
| 1484 |
+
{
|
| 1485 |
+
"epoch": 0.6013071895424836,
|
| 1486 |
+
"grad_norm": 7.650852680206299,
|
| 1487 |
+
"learning_rate": 3.753828281879845e-07,
|
| 1488 |
+
"loss": 0.806195068359375,
|
| 1489 |
+
"step": 920,
|
| 1490 |
+
"token_acc": 0.7688406947738297
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 0.6045751633986928,
|
| 1494 |
+
"grad_norm": 9.451173782348633,
|
| 1495 |
+
"learning_rate": 3.701554323108582e-07,
|
| 1496 |
+
"loss": 0.7724335193634033,
|
| 1497 |
+
"step": 925,
|
| 1498 |
+
"token_acc": 0.7795763547915225
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 0.6078431372549019,
|
| 1502 |
+
"grad_norm": 15.554048538208008,
|
| 1503 |
+
"learning_rate": 3.649432114120943e-07,
|
| 1504 |
+
"loss": 0.8206496238708496,
|
| 1505 |
+
"step": 930,
|
| 1506 |
+
"token_acc": 0.7666409765275244
|
| 1507 |
+
},
|
| 1508 |
+
{
|
| 1509 |
+
"epoch": 0.6111111111111112,
|
| 1510 |
+
"grad_norm": 12.310651779174805,
|
| 1511 |
+
"learning_rate": 3.597467746457049e-07,
|
| 1512 |
+
"loss": 0.8173147201538086,
|
| 1513 |
+
"step": 935,
|
| 1514 |
+
"token_acc": 0.7691580556470823
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.6143790849673203,
|
| 1518 |
+
"grad_norm": 10.026083946228027,
|
| 1519 |
+
"learning_rate": 3.5456672932100486e-07,
|
| 1520 |
+
"loss": 0.8049489974975585,
|
| 1521 |
+
"step": 940,
|
| 1522 |
+
"token_acc": 0.7708023560960594
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 0.6176470588235294,
|
| 1526 |
+
"grad_norm": 10.743084907531738,
|
| 1527 |
+
"learning_rate": 3.494036808316366e-07,
|
| 1528 |
+
"loss": 0.7841720581054688,
|
| 1529 |
+
"step": 945,
|
| 1530 |
+
"token_acc": 0.777067921990585
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 0.6209150326797386,
|
| 1534 |
+
"grad_norm": 16.597537994384766,
|
| 1535 |
+
"learning_rate": 3.4425823258481577e-07,
|
| 1536 |
+
"loss": 0.7723310470581055,
|
| 1537 |
+
"step": 950,
|
| 1538 |
+
"token_acc": 0.7776802821203455
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 0.6241830065359477,
|
| 1542 |
+
"grad_norm": 12.68069839477539,
|
| 1543 |
+
"learning_rate": 3.3913098593081257e-07,
|
| 1544 |
+
"loss": 0.7836595058441163,
|
| 1545 |
+
"step": 955,
|
| 1546 |
+
"token_acc": 0.7766279442548136
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 0.6274509803921569,
|
| 1550 |
+
"grad_norm": 25.00177764892578,
|
| 1551 |
+
"learning_rate": 3.3402254009266993e-07,
|
| 1552 |
+
"loss": 0.8226211547851563,
|
| 1553 |
+
"step": 960,
|
| 1554 |
+
"token_acc": 0.7657885161591641
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 0.630718954248366,
|
| 1558 |
+
"grad_norm": 43.81291198730469,
|
| 1559 |
+
"learning_rate": 3.289334920961743e-07,
|
| 1560 |
+
"loss": 0.7897493362426757,
|
| 1561 |
+
"step": 965,
|
| 1562 |
+
"token_acc": 0.775090884610206
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 0.6339869281045751,
|
| 1566 |
+
"grad_norm": 12.591288566589355,
|
| 1567 |
+
"learning_rate": 3.238644367000787e-07,
|
| 1568 |
+
"loss": 0.7990244388580322,
|
| 1569 |
+
"step": 970,
|
| 1570 |
+
"token_acc": 0.7713068338006488
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.6372549019607843,
|
| 1574 |
+
"grad_norm": 10.87752914428711,
|
| 1575 |
+
"learning_rate": 3.1881596632659507e-07,
|
| 1576 |
+
"loss": 0.7952980995178223,
|
| 1577 |
+
"step": 975,
|
| 1578 |
+
"token_acc": 0.7722554008020353
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.6405228758169934,
|
| 1582 |
+
"grad_norm": 8.995305061340332,
|
| 1583 |
+
"learning_rate": 3.1378867099215637e-07,
|
| 1584 |
+
"loss": 0.8093055725097656,
|
| 1585 |
+
"step": 980,
|
| 1586 |
+
"token_acc": 0.7718945744604929
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 0.6437908496732027,
|
| 1590 |
+
"grad_norm": 33.82859420776367,
|
| 1591 |
+
"learning_rate": 3.0878313823846147e-07,
|
| 1592 |
+
"loss": 0.8125652313232422,
|
| 1593 |
+
"step": 985,
|
| 1594 |
+
"token_acc": 0.7680579775065193
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 0.6470588235294118,
|
| 1598 |
+
"grad_norm": 10.589845657348633,
|
| 1599 |
+
"learning_rate": 3.037999530638098e-07,
|
| 1600 |
+
"loss": 0.7739749908447265,
|
| 1601 |
+
"step": 990,
|
| 1602 |
+
"token_acc": 0.7794414184725924
|
| 1603 |
+
},
|
| 1604 |
+
{
|
| 1605 |
+
"epoch": 0.6503267973856209,
|
| 1606 |
+
"grad_norm": 10.574102401733398,
|
| 1607 |
+
"learning_rate": 2.988396978547308e-07,
|
| 1608 |
+
"loss": 0.8024740219116211,
|
| 1609 |
+
"step": 995,
|
| 1610 |
+
"token_acc": 0.7719405974829644
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 0.6535947712418301,
|
| 1614 |
+
"grad_norm": 13.721783638000488,
|
| 1615 |
+
"learning_rate": 2.9390295231792177e-07,
|
| 1616 |
+
"loss": 0.7813970565795898,
|
| 1617 |
+
"step": 1000,
|
| 1618 |
+
"token_acc": 0.7793375983562539
|
| 1619 |
+
},
|
| 1620 |
+
{
|
| 1621 |
+
"epoch": 0.6568627450980392,
|
| 1622 |
+
"grad_norm": 26.136316299438477,
|
| 1623 |
+
"learning_rate": 2.8899029341249647e-07,
|
| 1624 |
+
"loss": 0.8029732704162598,
|
| 1625 |
+
"step": 1005,
|
| 1626 |
+
"token_acc": 0.7684636736356922
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.6601307189542484,
|
| 1630 |
+
"grad_norm": 12.889707565307617,
|
| 1631 |
+
"learning_rate": 2.8410229528255603e-07,
|
| 1632 |
+
"loss": 0.8104598999023438,
|
| 1633 |
+
"step": 1010,
|
| 1634 |
+
"token_acc": 0.7674994680183548
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.6633986928104575,
|
| 1638 |
+
"grad_norm": 13.532292366027832,
|
| 1639 |
+
"learning_rate": 2.792395291900882e-07,
|
| 1640 |
+
"loss": 0.7894612789154053,
|
| 1641 |
+
"step": 1015,
|
| 1642 |
+
"token_acc": 0.7747884800881907
|
| 1643 |
+
},
|
| 1644 |
+
{
|
| 1645 |
+
"epoch": 0.6666666666666666,
|
| 1646 |
+
"grad_norm": 14.569544792175293,
|
| 1647 |
+
"learning_rate": 2.74402563448204e-07,
|
| 1648 |
+
"loss": 0.7679402351379394,
|
| 1649 |
+
"step": 1020,
|
| 1650 |
+
"token_acc": 0.7773644006614033
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 0.6699346405228758,
|
| 1654 |
+
"grad_norm": 12.618599891662598,
|
| 1655 |
+
"learning_rate": 2.6959196335471934e-07,
|
| 1656 |
+
"loss": 0.80099515914917,
|
| 1657 |
+
"step": 1025,
|
| 1658 |
+
"token_acc": 0.7711608918046714
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 0.673202614379085,
|
| 1662 |
+
"grad_norm": 16.700902938842773,
|
| 1663 |
+
"learning_rate": 2.6480829112608725e-07,
|
| 1664 |
+
"loss": 0.7949865341186524,
|
| 1665 |
+
"step": 1030,
|
| 1666 |
+
"token_acc": 0.7741975523173548
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 0.6764705882352942,
|
| 1670 |
+
"grad_norm": 12.214275360107422,
|
| 1671 |
+
"learning_rate": 2.6005210583169267e-07,
|
| 1672 |
+
"loss": 0.7988276481628418,
|
| 1673 |
+
"step": 1035,
|
| 1674 |
+
"token_acc": 0.7708385046914561
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 0.6797385620915033,
|
| 1678 |
+
"grad_norm": 11.995553970336914,
|
| 1679 |
+
"learning_rate": 2.5532396332851264e-07,
|
| 1680 |
+
"loss": 0.7960112571716309,
|
| 1681 |
+
"step": 1040,
|
| 1682 |
+
"token_acc": 0.7720841449597303
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.6830065359477124,
|
| 1686 |
+
"grad_norm": 8.099811553955078,
|
| 1687 |
+
"learning_rate": 2.5062441619615455e-07,
|
| 1688 |
+
"loss": 0.794053316116333,
|
| 1689 |
+
"step": 1045,
|
| 1690 |
+
"token_acc": 0.7732998336332462
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 0.6862745098039216,
|
| 1694 |
+
"grad_norm": 15.760690689086914,
|
| 1695 |
+
"learning_rate": 2.4595401367227454e-07,
|
| 1696 |
+
"loss": 0.7991122245788574,
|
| 1697 |
+
"step": 1050,
|
| 1698 |
+
"token_acc": 0.7714444629845931
|
| 1699 |
+
},
|
| 1700 |
+
{
|
| 1701 |
+
"epoch": 0.6895424836601307,
|
| 1702 |
+
"grad_norm": 19.17706871032715,
|
| 1703 |
+
"learning_rate": 2.413133015883885e-07,
|
| 1704 |
+
"loss": 0.8024096488952637,
|
| 1705 |
+
"step": 1055,
|
| 1706 |
+
"token_acc": 0.7732000941566812
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 0.6928104575163399,
|
| 1710 |
+
"grad_norm": 11.341483116149902,
|
| 1711 |
+
"learning_rate": 2.367028223060813e-07,
|
| 1712 |
+
"loss": 0.7807505607604981,
|
| 1713 |
+
"step": 1060,
|
| 1714 |
+
"token_acc": 0.7752502170833143
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 0.696078431372549,
|
| 1718 |
+
"grad_norm": 12.454402923583984,
|
| 1719 |
+
"learning_rate": 2.3212311465361918e-07,
|
| 1720 |
+
"loss": 0.7818704605102539,
|
| 1721 |
+
"step": 1065,
|
| 1722 |
+
"token_acc": 0.7762884070448313
|
| 1723 |
+
},
|
| 1724 |
+
{
|
| 1725 |
+
"epoch": 0.6993464052287581,
|
| 1726 |
+
"grad_norm": 12.52795124053955,
|
| 1727 |
+
"learning_rate": 2.2757471386297854e-07,
|
| 1728 |
+
"loss": 0.7916586875915528,
|
| 1729 |
+
"step": 1070,
|
| 1730 |
+
"token_acc": 0.774573363119228
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 0.7026143790849673,
|
| 1734 |
+
"grad_norm": 11.589825630187988,
|
| 1735 |
+
"learning_rate": 2.2305815150729164e-07,
|
| 1736 |
+
"loss": 0.8006845474243164,
|
| 1737 |
+
"step": 1075,
|
| 1738 |
+
"token_acc": 0.7745477857667884
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.7058823529411765,
|
| 1742 |
+
"grad_norm": 15.58310604095459,
|
| 1743 |
+
"learning_rate": 2.185739554387223e-07,
|
| 1744 |
+
"loss": 0.7996684074401855,
|
| 1745 |
+
"step": 1080,
|
| 1746 |
+
"token_acc": 0.7711560223042617
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 0.7091503267973857,
|
| 1750 |
+
"grad_norm": 9.493857383728027,
|
| 1751 |
+
"learning_rate": 2.1412264972677535e-07,
|
| 1752 |
+
"loss": 0.8115836143493652,
|
| 1753 |
+
"step": 1085,
|
| 1754 |
+
"token_acc": 0.7708925236289547
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 0.7124183006535948,
|
| 1758 |
+
"grad_norm": 7.889930248260498,
|
| 1759 |
+
"learning_rate": 2.0970475459704906e-07,
|
| 1760 |
+
"loss": 0.7898194789886475,
|
| 1761 |
+
"step": 1090,
|
| 1762 |
+
"token_acc": 0.772780275594524
|
| 1763 |
+
},
|
| 1764 |
+
{
|
| 1765 |
+
"epoch": 0.7156862745098039,
|
| 1766 |
+
"grad_norm": 18.62689971923828,
|
| 1767 |
+
"learning_rate": 2.053207863704352e-07,
|
| 1768 |
+
"loss": 0.7930598258972168,
|
| 1769 |
+
"step": 1095,
|
| 1770 |
+
"token_acc": 0.7711671543356091
|
| 1771 |
+
},
|
| 1772 |
+
{
|
| 1773 |
+
"epoch": 0.7189542483660131,
|
| 1774 |
+
"grad_norm": 12.486854553222656,
|
| 1775 |
+
"learning_rate": 2.0097125740277694e-07,
|
| 1776 |
+
"loss": 0.7758184909820557,
|
| 1777 |
+
"step": 1100,
|
| 1778 |
+
"token_acc": 0.7760485163128061
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 0.7222222222222222,
|
| 1782 |
+
"grad_norm": 13.689062118530273,
|
| 1783 |
+
"learning_rate": 1.966566760249901e-07,
|
| 1784 |
+
"loss": 0.7781660079956054,
|
| 1785 |
+
"step": 1105,
|
| 1786 |
+
"token_acc": 0.7775909431449264
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 0.7254901960784313,
|
| 1790 |
+
"grad_norm": 10.267345428466797,
|
| 1791 |
+
"learning_rate": 1.9237754648365324e-07,
|
| 1792 |
+
"loss": 0.8121429443359375,
|
| 1793 |
+
"step": 1110,
|
| 1794 |
+
"token_acc": 0.7692980842655608
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.7287581699346405,
|
| 1798 |
+
"grad_norm": 14.282358169555664,
|
| 1799 |
+
"learning_rate": 1.8813436888207769e-07,
|
| 1800 |
+
"loss": 0.7903863906860351,
|
| 1801 |
+
"step": 1115,
|
| 1802 |
+
"token_acc": 0.7759478338586011
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.7320261437908496,
|
| 1806 |
+
"grad_norm": 18.916223526000977,
|
| 1807 |
+
"learning_rate": 1.839276391218585e-07,
|
| 1808 |
+
"loss": 0.780515480041504,
|
| 1809 |
+
"step": 1120,
|
| 1810 |
+
"token_acc": 0.7761903516799582
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 0.7352941176470589,
|
| 1814 |
+
"grad_norm": 12.354140281677246,
|
| 1815 |
+
"learning_rate": 1.7975784884492023e-07,
|
| 1816 |
+
"loss": 0.7762207984924316,
|
| 1817 |
+
"step": 1125,
|
| 1818 |
+
"token_acc": 0.7771318682931841
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"epoch": 0.738562091503268,
|
| 1822 |
+
"grad_norm": 13.276935577392578,
|
| 1823 |
+
"learning_rate": 1.7562548537605682e-07,
|
| 1824 |
+
"loss": 0.7915905952453614,
|
| 1825 |
+
"step": 1130,
|
| 1826 |
+
"token_acc": 0.7723898929173292
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 0.7418300653594772,
|
| 1830 |
+
"grad_norm": 19.15022087097168,
|
| 1831 |
+
"learning_rate": 1.7153103166597866e-07,
|
| 1832 |
+
"loss": 0.7942383766174317,
|
| 1833 |
+
"step": 1135,
|
| 1834 |
+
"token_acc": 0.7734970343430398
|
| 1835 |
+
},
|
| 1836 |
+
{
|
| 1837 |
+
"epoch": 0.7450980392156863,
|
| 1838 |
+
"grad_norm": 11.207582473754883,
|
| 1839 |
+
"learning_rate": 1.674749662348704e-07,
|
| 1840 |
+
"loss": 0.7850894451141357,
|
| 1841 |
+
"step": 1140,
|
| 1842 |
+
"token_acc": 0.7736343036786021
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 0.7483660130718954,
|
| 1846 |
+
"grad_norm": 11.658065795898438,
|
| 1847 |
+
"learning_rate": 1.6345776311646475e-07,
|
| 1848 |
+
"loss": 0.7763532638549805,
|
| 1849 |
+
"step": 1145,
|
| 1850 |
+
"token_acc": 0.7733978738680316
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.7516339869281046,
|
| 1854 |
+
"grad_norm": 12.42931842803955,
|
| 1855 |
+
"learning_rate": 1.5947989180264376e-07,
|
| 1856 |
+
"loss": 0.7802921295166015,
|
| 1857 |
+
"step": 1150,
|
| 1858 |
+
"token_acc": 0.7749346665151687
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 0.7549019607843137,
|
| 1862 |
+
"grad_norm": 17.95322036743164,
|
| 1863 |
+
"learning_rate": 1.5554181718856735e-07,
|
| 1864 |
+
"loss": 0.7752510070800781,
|
| 1865 |
+
"step": 1155,
|
| 1866 |
+
"token_acc": 0.7759107659039255
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 0.7581699346405228,
|
| 1870 |
+
"grad_norm": 8.104009628295898,
|
| 1871 |
+
"learning_rate": 1.5164399951834255e-07,
|
| 1872 |
+
"loss": 0.7809647560119629,
|
| 1873 |
+
"step": 1160,
|
| 1874 |
+
"token_acc": 0.7783235765760544
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"epoch": 0.761437908496732,
|
| 1878 |
+
"grad_norm": 13.01050853729248,
|
| 1879 |
+
"learning_rate": 1.477868943312332e-07,
|
| 1880 |
+
"loss": 0.7745242595672608,
|
| 1881 |
+
"step": 1165,
|
| 1882 |
+
"token_acc": 0.7764719406557887
|
| 1883 |
+
},
|
| 1884 |
+
{
|
| 1885 |
+
"epoch": 0.7647058823529411,
|
| 1886 |
+
"grad_norm": 15.886526107788086,
|
| 1887 |
+
"learning_rate": 1.439709524084216e-07,
|
| 1888 |
+
"loss": 0.793631649017334,
|
| 1889 |
+
"step": 1170,
|
| 1890 |
+
"token_acc": 0.7713033076708243
|
| 1891 |
+
},
|
| 1892 |
+
{
|
| 1893 |
+
"epoch": 0.7679738562091504,
|
| 1894 |
+
"grad_norm": 10.532743453979492,
|
| 1895 |
+
"learning_rate": 1.401966197203259e-07,
|
| 1896 |
+
"loss": 0.8169536590576172,
|
| 1897 |
+
"step": 1175,
|
| 1898 |
+
"token_acc": 0.7648686721349145
|
| 1899 |
+
},
|
| 1900 |
+
{
|
| 1901 |
+
"epoch": 0.7712418300653595,
|
| 1902 |
+
"grad_norm": 13.834940910339355,
|
| 1903 |
+
"learning_rate": 1.364643373744786e-07,
|
| 1904 |
+
"loss": 0.769162368774414,
|
| 1905 |
+
"step": 1180,
|
| 1906 |
+
"token_acc": 0.7768779105082
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.7745098039215687,
|
| 1910 |
+
"grad_norm": 11.915435791015625,
|
| 1911 |
+
"learning_rate": 1.3277454156397455e-07,
|
| 1912 |
+
"loss": 0.7994132995605469,
|
| 1913 |
+
"step": 1185,
|
| 1914 |
+
"token_acc": 0.769907066568505
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.7777777777777778,
|
| 1918 |
+
"grad_norm": 12.843599319458008,
|
| 1919 |
+
"learning_rate": 1.291276635164929e-07,
|
| 1920 |
+
"loss": 0.7847061634063721,
|
| 1921 |
+
"step": 1190,
|
| 1922 |
+
"token_acc": 0.773934673479954
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 0.7810457516339869,
|
| 1926 |
+
"grad_norm": 20.338703155517578,
|
| 1927 |
+
"learning_rate": 1.2552412944389974e-07,
|
| 1928 |
+
"loss": 0.7957106590270996,
|
| 1929 |
+
"step": 1195,
|
| 1930 |
+
"token_acc": 0.7721588604887588
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"epoch": 0.7843137254901961,
|
| 1934 |
+
"grad_norm": 10.63708782196045,
|
| 1935 |
+
"learning_rate": 1.2196436049243581e-07,
|
| 1936 |
+
"loss": 0.7906211853027344,
|
| 1937 |
+
"step": 1200,
|
| 1938 |
+
"token_acc": 0.7707042187518306
|
| 1939 |
+
},
|
| 1940 |
+
{
|
| 1941 |
+
"epoch": 0.7875816993464052,
|
| 1942 |
+
"grad_norm": 9.61670207977295,
|
| 1943 |
+
"learning_rate": 1.1844877269349729e-07,
|
| 1944 |
+
"loss": 0.8031095504760742,
|
| 1945 |
+
"step": 1205,
|
| 1946 |
+
"token_acc": 0.7700618999067395
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 0.7908496732026143,
|
| 1950 |
+
"grad_norm": 10.670196533203125,
|
| 1951 |
+
"learning_rate": 1.1497777691501481e-07,
|
| 1952 |
+
"loss": 0.8217704772949219,
|
| 1953 |
+
"step": 1210,
|
| 1954 |
+
"token_acc": 0.76580138437299
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 0.7941176470588235,
|
| 1958 |
+
"grad_norm": 12.682234764099121,
|
| 1959 |
+
"learning_rate": 1.115517788134338e-07,
|
| 1960 |
+
"loss": 0.7944438457489014,
|
| 1961 |
+
"step": 1215,
|
| 1962 |
+
"token_acc": 0.7717567448118721
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.7973856209150327,
|
| 1966 |
+
"grad_norm": 18.573143005371094,
|
| 1967 |
+
"learning_rate": 1.0817117878630677e-07,
|
| 1968 |
+
"loss": 0.7697268009185791,
|
| 1969 |
+
"step": 1220,
|
| 1970 |
+
"token_acc": 0.7789009541902335
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.8006535947712419,
|
| 1974 |
+
"grad_norm": 16.35167694091797,
|
| 1975 |
+
"learning_rate": 1.0483637192549727e-07,
|
| 1976 |
+
"loss": 0.7881369590759277,
|
| 1977 |
+
"step": 1225,
|
| 1978 |
+
"token_acc": 0.7785905643828424
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 0.803921568627451,
|
| 1982 |
+
"grad_norm": 20.62697982788086,
|
| 1983 |
+
"learning_rate": 1.0154774797100669e-07,
|
| 1984 |
+
"loss": 0.79795241355896,
|
| 1985 |
+
"step": 1230,
|
| 1986 |
+
"token_acc": 0.7751004241062674
|
| 1987 |
+
},
|
| 1988 |
+
{
|
| 1989 |
+
"epoch": 0.8071895424836601,
|
| 1990 |
+
"grad_norm": 21.030271530151367,
|
| 1991 |
+
"learning_rate": 9.830569126542415e-08,
|
| 1992 |
+
"loss": 0.7898718357086182,
|
| 1993 |
+
"step": 1235,
|
| 1994 |
+
"token_acc": 0.7737865370530661
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 0.8104575163398693,
|
| 1998 |
+
"grad_norm": 12.68388557434082,
|
| 1999 |
+
"learning_rate": 9.511058070900884e-08,
|
| 2000 |
+
"loss": 0.7856592178344727,
|
| 2001 |
+
"step": 1240,
|
| 2002 |
+
"token_acc": 0.7744902400832122
|
| 2003 |
+
},
|
| 2004 |
+
{
|
| 2005 |
+
"epoch": 0.8137254901960784,
|
| 2006 |
+
"grad_norm": 10.876585960388184,
|
| 2007 |
+
"learning_rate": 9.196278971540788e-08,
|
| 2008 |
+
"loss": 0.7776141166687012,
|
| 2009 |
+
"step": 1245,
|
| 2010 |
+
"token_acc": 0.7763189305323466
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"epoch": 0.8169934640522876,
|
| 2014 |
+
"grad_norm": 12.351463317871094,
|
| 2015 |
+
"learning_rate": 8.886268616801473e-08,
|
| 2016 |
+
"loss": 0.7695632934570312,
|
| 2017 |
+
"step": 1250,
|
| 2018 |
+
"token_acc": 0.7758939485767702
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.8202614379084967,
|
| 2022 |
+
"grad_norm": 10.065323829650879,
|
| 2023 |
+
"learning_rate": 8.58106323769755e-08,
|
| 2024 |
+
"loss": 0.8167810440063477,
|
| 2025 |
+
"step": 1255,
|
| 2026 |
+
"token_acc": 0.7680006711972481
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 0.8235294117647058,
|
| 2030 |
+
"grad_norm": 12.23916244506836,
|
| 2031 |
+
"learning_rate": 8.280698503684458e-08,
|
| 2032 |
+
"loss": 0.7955585956573487,
|
| 2033 |
+
"step": 1260,
|
| 2034 |
+
"token_acc": 0.7705074202026926
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 0.826797385620915,
|
| 2038 |
+
"grad_norm": 12.86502456665039,
|
| 2039 |
+
"learning_rate": 7.985209518489855e-08,
|
| 2040 |
+
"loss": 0.7748407363891602,
|
| 2041 |
+
"step": 1265,
|
| 2042 |
+
"token_acc": 0.7784362149643144
|
| 2043 |
+
},
|
| 2044 |
+
{
|
| 2045 |
+
"epoch": 0.8300653594771242,
|
| 2046 |
+
"grad_norm": 11.304476737976074,
|
| 2047 |
+
"learning_rate": 7.694630816010955e-08,
|
| 2048 |
+
"loss": 0.7772586822509766,
|
| 2049 |
+
"step": 1270,
|
| 2050 |
+
"token_acc": 0.7760128711423139
|
| 2051 |
+
},
|
| 2052 |
+
{
|
| 2053 |
+
"epoch": 0.8333333333333334,
|
| 2054 |
+
"grad_norm": 13.227031707763672,
|
| 2055 |
+
"learning_rate": 7.40899635627859e-08,
|
| 2056 |
+
"loss": 0.7880292415618897,
|
| 2057 |
+
"step": 1275,
|
| 2058 |
+
"token_acc": 0.7727952000089552
|
| 2059 |
+
},
|
| 2060 |
+
{
|
| 2061 |
+
"epoch": 0.8366013071895425,
|
| 2062 |
+
"grad_norm": 11.008593559265137,
|
| 2063 |
+
"learning_rate": 7.128339521488303e-08,
|
| 2064 |
+
"loss": 0.7893757820129395,
|
| 2065 |
+
"step": 1280,
|
| 2066 |
+
"token_acc": 0.7725503091657516
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 0.8398692810457516,
|
| 2070 |
+
"grad_norm": 12.050971984863281,
|
| 2071 |
+
"learning_rate": 6.852693112098901e-08,
|
| 2072 |
+
"loss": 0.7573928356170654,
|
| 2073 |
+
"step": 1285,
|
| 2074 |
+
"token_acc": 0.7817575406032483
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.8431372549019608,
|
| 2078 |
+
"grad_norm": 13.571142196655273,
|
| 2079 |
+
"learning_rate": 6.582089342999092e-08,
|
| 2080 |
+
"loss": 0.8093362808227539,
|
| 2081 |
+
"step": 1290,
|
| 2082 |
+
"token_acc": 0.7683763996326047
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 0.8464052287581699,
|
| 2086 |
+
"grad_norm": 12.272185325622559,
|
| 2087 |
+
"learning_rate": 6.316559839742514e-08,
|
| 2088 |
+
"loss": 0.7729239463806152,
|
| 2089 |
+
"step": 1295,
|
| 2090 |
+
"token_acc": 0.7779632532268013
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 0.8496732026143791,
|
| 2094 |
+
"grad_norm": 12.929800987243652,
|
| 2095 |
+
"learning_rate": 6.056135634851673e-08,
|
| 2096 |
+
"loss": 0.7964875221252441,
|
| 2097 |
+
"step": 1300,
|
| 2098 |
+
"token_acc": 0.7716209144780574
|
| 2099 |
+
},
|
| 2100 |
+
{
|
| 2101 |
+
"epoch": 0.8529411764705882,
|
| 2102 |
+
"grad_norm": 11.469837188720703,
|
| 2103 |
+
"learning_rate": 5.8008471641910916e-08,
|
| 2104 |
+
"loss": 0.808901596069336,
|
| 2105 |
+
"step": 1305,
|
| 2106 |
+
"token_acc": 0.7661655142565666
|
| 2107 |
+
},
|
| 2108 |
+
{
|
| 2109 |
+
"epoch": 0.8562091503267973,
|
| 2110 |
+
"grad_norm": 42.33794021606445,
|
| 2111 |
+
"learning_rate": 5.55072426341035e-08,
|
| 2112 |
+
"loss": 0.7958111763000488,
|
| 2113 |
+
"step": 1310,
|
| 2114 |
+
"token_acc": 0.7731623407589073
|
| 2115 |
+
},
|
| 2116 |
+
{
|
| 2117 |
+
"epoch": 0.8594771241830066,
|
| 2118 |
+
"grad_norm": 19.68001365661621,
|
| 2119 |
+
"learning_rate": 5.3057961644571044e-08,
|
| 2120 |
+
"loss": 0.7775318145751953,
|
| 2121 |
+
"step": 1315,
|
| 2122 |
+
"token_acc": 0.7772653867904178
|
| 2123 |
+
},
|
| 2124 |
+
{
|
| 2125 |
+
"epoch": 0.8627450980392157,
|
| 2126 |
+
"grad_norm": 14.147223472595215,
|
| 2127 |
+
"learning_rate": 5.066091492160768e-08,
|
| 2128 |
+
"loss": 0.7547184944152832,
|
| 2129 |
+
"step": 1320,
|
| 2130 |
+
"token_acc": 0.7815907531535335
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.8660130718954249,
|
| 2134 |
+
"grad_norm": 12.238420486450195,
|
| 2135 |
+
"learning_rate": 4.831638260887155e-08,
|
| 2136 |
+
"loss": 0.7707562923431397,
|
| 2137 |
+
"step": 1325,
|
| 2138 |
+
"token_acc": 0.776531370150297
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.869281045751634,
|
| 2142 |
+
"grad_norm": 6.894777297973633,
|
| 2143 |
+
"learning_rate": 4.602463871264356e-08,
|
| 2144 |
+
"loss": 0.8057853698730468,
|
| 2145 |
+
"step": 1330,
|
| 2146 |
+
"token_acc": 0.7700311722729691
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 0.8725490196078431,
|
| 2150 |
+
"grad_norm": 11.661809921264648,
|
| 2151 |
+
"learning_rate": 4.3785951069804984e-08,
|
| 2152 |
+
"loss": 0.7872296810150147,
|
| 2153 |
+
"step": 1335,
|
| 2154 |
+
"token_acc": 0.7698627163089369
|
| 2155 |
+
},
|
| 2156 |
+
{
|
| 2157 |
+
"epoch": 0.8758169934640523,
|
| 2158 |
+
"grad_norm": 9.87257194519043,
|
| 2159 |
+
"learning_rate": 4.160058131653449e-08,
|
| 2160 |
+
"loss": 0.811611557006836,
|
| 2161 |
+
"step": 1340,
|
| 2162 |
+
"token_acc": 0.7710193472610412
|
| 2163 |
+
},
|
| 2164 |
+
{
|
| 2165 |
+
"epoch": 0.8790849673202614,
|
| 2166 |
+
"grad_norm": 12.700177192687988,
|
| 2167 |
+
"learning_rate": 3.946878485773153e-08,
|
| 2168 |
+
"loss": 0.7809323787689209,
|
| 2169 |
+
"step": 1345,
|
| 2170 |
+
"token_acc": 0.7797607662542116
|
| 2171 |
+
},
|
| 2172 |
+
{
|
| 2173 |
+
"epoch": 0.8823529411764706,
|
| 2174 |
+
"grad_norm": 11.138504028320312,
|
| 2175 |
+
"learning_rate": 3.739081083716622e-08,
|
| 2176 |
+
"loss": 0.7353385925292969,
|
| 2177 |
+
"step": 1350,
|
| 2178 |
+
"token_acc": 0.7883354786422578
|
| 2179 |
+
},
|
| 2180 |
+
{
|
| 2181 |
+
"epoch": 0.8856209150326797,
|
| 2182 |
+
"grad_norm": 10.677797317504883,
|
| 2183 |
+
"learning_rate": 3.536690210836207e-08,
|
| 2184 |
+
"loss": 0.807730770111084,
|
| 2185 |
+
"step": 1355,
|
| 2186 |
+
"token_acc": 0.7688365282215123
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.8888888888888888,
|
| 2190 |
+
"grad_norm": 13.794832229614258,
|
| 2191 |
+
"learning_rate": 3.339729520621426e-08,
|
| 2192 |
+
"loss": 0.7856162071228028,
|
| 2193 |
+
"step": 1360,
|
| 2194 |
+
"token_acc": 0.7761172240395794
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.8921568627450981,
|
| 2198 |
+
"grad_norm": 10.534261703491211,
|
| 2199 |
+
"learning_rate": 3.148222031934461e-08,
|
| 2200 |
+
"loss": 0.7700923919677735,
|
| 2201 |
+
"step": 1365,
|
| 2202 |
+
"token_acc": 0.7776234567901235
|
| 2203 |
+
},
|
| 2204 |
+
{
|
| 2205 |
+
"epoch": 0.8954248366013072,
|
| 2206 |
+
"grad_norm": 12.092926979064941,
|
| 2207 |
+
"learning_rate": 2.9621901263200467e-08,
|
| 2208 |
+
"loss": 0.7857433319091797,
|
| 2209 |
+
"step": 1370,
|
| 2210 |
+
"token_acc": 0.7747586125644637
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"epoch": 0.8986928104575164,
|
| 2214 |
+
"grad_norm": 12.823472023010254,
|
| 2215 |
+
"learning_rate": 2.7816555453896383e-08,
|
| 2216 |
+
"loss": 0.8035625457763672,
|
| 2217 |
+
"step": 1375,
|
| 2218 |
+
"token_acc": 0.7689068469805668
|
| 2219 |
+
},
|
| 2220 |
+
{
|
| 2221 |
+
"epoch": 0.9019607843137255,
|
| 2222 |
+
"grad_norm": 11.723134994506836,
|
| 2223 |
+
"learning_rate": 2.606639388280524e-08,
|
| 2224 |
+
"loss": 0.7778918266296386,
|
| 2225 |
+
"step": 1380,
|
| 2226 |
+
"token_acc": 0.7751281040599133
|
| 2227 |
+
},
|
| 2228 |
+
{
|
| 2229 |
+
"epoch": 0.9052287581699346,
|
| 2230 |
+
"grad_norm": 9.832569122314453,
|
| 2231 |
+
"learning_rate": 2.437162109189933e-08,
|
| 2232 |
+
"loss": 0.7959203243255615,
|
| 2233 |
+
"step": 1385,
|
| 2234 |
+
"token_acc": 0.7709774050935979
|
| 2235 |
+
},
|
| 2236 |
+
{
|
| 2237 |
+
"epoch": 0.9084967320261438,
|
| 2238 |
+
"grad_norm": 7.625786781311035,
|
| 2239 |
+
"learning_rate": 2.2732435149845685e-08,
|
| 2240 |
+
"loss": 0.7646852493286133,
|
| 2241 |
+
"step": 1390,
|
| 2242 |
+
"token_acc": 0.77947380810111
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 0.9117647058823529,
|
| 2246 |
+
"grad_norm": 18.579334259033203,
|
| 2247 |
+
"learning_rate": 2.1149027628857395e-08,
|
| 2248 |
+
"loss": 0.785280418395996,
|
| 2249 |
+
"step": 1395,
|
| 2250 |
+
"token_acc": 0.7748307161186059
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.9150326797385621,
|
| 2254 |
+
"grad_norm": 21.786314010620117,
|
| 2255 |
+
"learning_rate": 1.962158358230476e-08,
|
| 2256 |
+
"loss": 0.7815417766571044,
|
| 2257 |
+
"step": 1400,
|
| 2258 |
+
"token_acc": 0.7772699876530929
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 0.9183006535947712,
|
| 2262 |
+
"grad_norm": 12.593238830566406,
|
| 2263 |
+
"learning_rate": 1.8150281523088172e-08,
|
| 2264 |
+
"loss": 0.788847541809082,
|
| 2265 |
+
"step": 1405,
|
| 2266 |
+
"token_acc": 0.7735945401336416
|
| 2267 |
+
},
|
| 2268 |
+
{
|
| 2269 |
+
"epoch": 0.9215686274509803,
|
| 2270 |
+
"grad_norm": 12.225467681884766,
|
| 2271 |
+
"learning_rate": 1.6735293402774686e-08,
|
| 2272 |
+
"loss": 0.7762408256530762,
|
| 2273 |
+
"step": 1410,
|
| 2274 |
+
"token_acc": 0.777554104530809
|
| 2275 |
+
},
|
| 2276 |
+
{
|
| 2277 |
+
"epoch": 0.9248366013071896,
|
| 2278 |
+
"grad_norm": 18.23349380493164,
|
| 2279 |
+
"learning_rate": 1.5376784591502655e-08,
|
| 2280 |
+
"loss": 0.7871239662170411,
|
| 2281 |
+
"step": 1415,
|
| 2282 |
+
"token_acc": 0.772668812095275
|
| 2283 |
+
},
|
| 2284 |
+
{
|
| 2285 |
+
"epoch": 0.9281045751633987,
|
| 2286 |
+
"grad_norm": 13.332287788391113,
|
| 2287 |
+
"learning_rate": 1.4074913858654447e-08,
|
| 2288 |
+
"loss": 0.779380989074707,
|
| 2289 |
+
"step": 1420,
|
| 2290 |
+
"token_acc": 0.774930297884829
|
| 2291 |
+
},
|
| 2292 |
+
{
|
| 2293 |
+
"epoch": 0.9313725490196079,
|
| 2294 |
+
"grad_norm": 57.881465911865234,
|
| 2295 |
+
"learning_rate": 1.2829833354301045e-08,
|
| 2296 |
+
"loss": 0.779120922088623,
|
| 2297 |
+
"step": 1425,
|
| 2298 |
+
"token_acc": 0.7779425482551604
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 0.934640522875817,
|
| 2302 |
+
"grad_norm": 10.205779075622559,
|
| 2303 |
+
"learning_rate": 1.1641688591420507e-08,
|
| 2304 |
+
"loss": 0.7665719985961914,
|
| 2305 |
+
"step": 1430,
|
| 2306 |
+
"token_acc": 0.7798292735114926
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.9379084967320261,
|
| 2310 |
+
"grad_norm": 20.218006134033203,
|
| 2311 |
+
"learning_rate": 1.0510618428891394e-08,
|
| 2312 |
+
"loss": 0.7749678611755371,
|
| 2313 |
+
"step": 1435,
|
| 2314 |
+
"token_acc": 0.7785859158615576
|
| 2315 |
+
},
|
| 2316 |
+
{
|
| 2317 |
+
"epoch": 0.9411764705882353,
|
| 2318 |
+
"grad_norm": 15.648624420166016,
|
| 2319 |
+
"learning_rate": 9.436755055264645e-09,
|
| 2320 |
+
"loss": 0.789072847366333,
|
| 2321 |
+
"step": 1440,
|
| 2322 |
+
"token_acc": 0.7732983056851054
|
| 2323 |
+
},
|
| 2324 |
+
{
|
| 2325 |
+
"epoch": 0.9444444444444444,
|
| 2326 |
+
"grad_norm": 14.973705291748047,
|
| 2327 |
+
"learning_rate": 8.420223973314322e-09,
|
| 2328 |
+
"loss": 0.7757076263427735,
|
| 2329 |
+
"step": 1445,
|
| 2330 |
+
"token_acc": 0.778316297232112
|
| 2331 |
+
},
|
| 2332 |
+
{
|
| 2333 |
+
"epoch": 0.9477124183006536,
|
| 2334 |
+
"grad_norm": 15.601005554199219,
|
| 2335 |
+
"learning_rate": 7.461143985370566e-09,
|
| 2336 |
+
"loss": 0.8004892349243165,
|
| 2337 |
+
"step": 1450,
|
| 2338 |
+
"token_acc": 0.7671663615724172
|
| 2339 |
+
},
|
| 2340 |
+
{
|
| 2341 |
+
"epoch": 0.9509803921568627,
|
| 2342 |
+
"grad_norm": 20.841663360595703,
|
| 2343 |
+
"learning_rate": 6.559627179434479e-09,
|
| 2344 |
+
"loss": 0.7896615982055664,
|
| 2345 |
+
"step": 1455,
|
| 2346 |
+
"token_acc": 0.7733638273542005
|
| 2347 |
+
},
|
| 2348 |
+
{
|
| 2349 |
+
"epoch": 0.954248366013072,
|
| 2350 |
+
"grad_norm": 40.20832824707031,
|
| 2351 |
+
"learning_rate": 5.7157789160788835e-09,
|
| 2352 |
+
"loss": 0.7690316200256347,
|
| 2353 |
+
"step": 1460,
|
| 2354 |
+
"token_acc": 0.7803495033766122
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 0.9575163398692811,
|
| 2358 |
+
"grad_norm": 18.089740753173828,
|
| 2359 |
+
"learning_rate": 4.929697816134615e-09,
|
| 2360 |
+
"loss": 0.7929338455200196,
|
| 2361 |
+
"step": 1465,
|
| 2362 |
+
"token_acc": 0.7723885017331871
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 0.9607843137254902,
|
| 2366 |
+
"grad_norm": 9.207868576049805,
|
| 2367 |
+
"learning_rate": 4.2014757491644625e-09,
|
| 2368 |
+
"loss": 0.7696738719940186,
|
| 2369 |
+
"step": 1470,
|
| 2370 |
+
"token_acc": 0.7823429663365727
|
| 2371 |
+
},
|
| 2372 |
+
{
|
| 2373 |
+
"epoch": 0.9640522875816994,
|
| 2374 |
+
"grad_norm": 21.471670150756836,
|
| 2375 |
+
"learning_rate": 3.5311978227268168e-09,
|
| 2376 |
+
"loss": 0.7767305850982666,
|
| 2377 |
+
"step": 1475,
|
| 2378 |
+
"token_acc": 0.778716714159379
|
| 2379 |
+
},
|
| 2380 |
+
{
|
| 2381 |
+
"epoch": 0.9673202614379085,
|
| 2382 |
+
"grad_norm": 10.909425735473633,
|
| 2383 |
+
"learning_rate": 2.9189423724286787e-09,
|
| 2384 |
+
"loss": 0.7795670509338379,
|
| 2385 |
+
"step": 1480,
|
| 2386 |
+
"token_acc": 0.7774368584797041
|
| 2387 |
+
},
|
| 2388 |
+
{
|
| 2389 |
+
"epoch": 0.9705882352941176,
|
| 2390 |
+
"grad_norm": 14.513352394104004,
|
| 2391 |
+
"learning_rate": 2.364780952770762e-09,
|
| 2392 |
+
"loss": 0.7744713306427002,
|
| 2393 |
+
"step": 1485,
|
| 2394 |
+
"token_acc": 0.7770122159743907
|
| 2395 |
+
},
|
| 2396 |
+
{
|
| 2397 |
+
"epoch": 0.9738562091503268,
|
| 2398 |
+
"grad_norm": 10.696837425231934,
|
| 2399 |
+
"learning_rate": 1.8687783287847923e-09,
|
| 2400 |
+
"loss": 0.7719961166381836,
|
| 2401 |
+
"step": 1490,
|
| 2402 |
+
"token_acc": 0.7781668662313896
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"epoch": 0.9771241830065359,
|
| 2406 |
+
"grad_norm": 28.523488998413086,
|
| 2407 |
+
"learning_rate": 1.4309924684645093e-09,
|
| 2408 |
+
"loss": 0.7770801067352295,
|
| 2409 |
+
"step": 1495,
|
| 2410 |
+
"token_acc": 0.7771797441490302
|
| 2411 |
+
},
|
| 2412 |
+
{
|
| 2413 |
+
"epoch": 0.9803921568627451,
|
| 2414 |
+
"grad_norm": 12.750672340393066,
|
| 2415 |
+
"learning_rate": 1.0514745359909706e-09,
|
| 2416 |
+
"loss": 0.7724951267242431,
|
| 2417 |
+
"step": 1500,
|
| 2418 |
+
"token_acc": 0.777609305016609
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.9836601307189542,
|
| 2422 |
+
"grad_norm": 19.97661781311035,
|
| 2423 |
+
"learning_rate": 7.302688857526718e-10,
|
| 2424 |
+
"loss": 0.788493013381958,
|
| 2425 |
+
"step": 1505,
|
| 2426 |
+
"token_acc": 0.7733385032958511
|
| 2427 |
+
},
|
| 2428 |
+
{
|
| 2429 |
+
"epoch": 0.9869281045751634,
|
| 2430 |
+
"grad_norm": 10.919777870178223,
|
| 2431 |
+
"learning_rate": 4.674130571621915e-10,
|
| 2432 |
+
"loss": 0.7954597473144531,
|
| 2433 |
+
"step": 1510,
|
| 2434 |
+
"token_acc": 0.7740816211014887
|
| 2435 |
+
},
|
| 2436 |
+
{
|
| 2437 |
+
"epoch": 0.9901960784313726,
|
| 2438 |
+
"grad_norm": 18.24049186706543,
|
| 2439 |
+
"learning_rate": 2.629377702687563e-10,
|
| 2440 |
+
"loss": 0.8122730255126953,
|
| 2441 |
+
"step": 1515,
|
| 2442 |
+
"token_acc": 0.7664521981464683
|
| 2443 |
+
},
|
| 2444 |
+
{
|
| 2445 |
+
"epoch": 0.9934640522875817,
|
| 2446 |
+
"grad_norm": 28.266395568847656,
|
| 2447 |
+
"learning_rate": 1.1686692216805738e-10,
|
| 2448 |
+
"loss": 0.770718765258789,
|
| 2449 |
+
"step": 1520,
|
| 2450 |
+
"token_acc": 0.7794897588509794
|
| 2451 |
+
},
|
| 2452 |
+
{
|
| 2453 |
+
"epoch": 0.9967320261437909,
|
| 2454 |
+
"grad_norm": 13.187555313110352,
|
| 2455 |
+
"learning_rate": 2.921758420926235e-11,
|
| 2456 |
+
"loss": 0.7856090068817139,
|
| 2457 |
+
"step": 1525,
|
| 2458 |
+
"token_acc": 0.7774288330702203
|
| 2459 |
+
},
|
| 2460 |
+
{
|
| 2461 |
+
"epoch": 1.0,
|
| 2462 |
+
"grad_norm": 15.244508743286133,
|
| 2463 |
+
"learning_rate": 0.0,
|
| 2464 |
+
"loss": 0.7558686256408691,
|
| 2465 |
+
"step": 1530,
|
| 2466 |
+
"token_acc": 0.7827098193732773
|
| 2467 |
+
}
|
| 2468 |
+
],
|
| 2469 |
+
"logging_steps": 5,
|
| 2470 |
+
"max_steps": 1530,
|
| 2471 |
+
"num_input_tokens_seen": 0,
|
| 2472 |
+
"num_train_epochs": 1,
|
| 2473 |
+
"save_steps": 100,
|
| 2474 |
+
"stateful_callbacks": {
|
| 2475 |
+
"TrainerControl": {
|
| 2476 |
+
"args": {
|
| 2477 |
+
"should_epoch_stop": false,
|
| 2478 |
+
"should_evaluate": false,
|
| 2479 |
+
"should_log": false,
|
| 2480 |
+
"should_save": true,
|
| 2481 |
+
"should_training_stop": true
|
| 2482 |
+
},
|
| 2483 |
+
"attributes": {}
|
| 2484 |
+
}
|
| 2485 |
+
},
|
| 2486 |
+
"total_flos": 3.451903791688214e+19,
|
| 2487 |
+
"train_batch_size": 1,
|
| 2488 |
+
"trial_name": null,
|
| 2489 |
+
"trial_params": null
|
| 2490 |
+
}
|