Adding EasyDeL Checkpoints

README.md

@@ -0,0 +1,159 @@
+---
+library_name: easydel
+pipeline_tag: text-generation
+tags:
+  - easydel
+  - jax
+  - "llama"
+  - "CausalLM"
+  - "vanilla"
+---
+
+<p align="center">
+  <img alt="easydel" src="https://raw.githubusercontent.com/erfanzar/easydel/main/images/easydel-logo-with-text.png">
+</p>
+
+<h1 align="center">meta-llama/Llama-3.3-70B-Instruct</h1>
+
+<div align="center">
+  EasyDeL checkpoint converted from meta-llama/Llama-3.3-70B-Instruct.
+</div>
+
+## Overview
+
+This checkpoint is intended to be loaded with EasyDeL on JAX (CPU/GPU/TPU). It supports sharded loading with `auto_shard_model=True` and configurable precision via `dtype`, `param_dtype`, and `precision`.
+
+## Quickstart
+
+```python
+import easydel as ed
+from jax import numpy as jnp, lax
+
+repo_id = "EasyDeL/Llama-3.3-70B-Instruct"
+
+dtype = jnp.bfloat16  # try jnp.float16 on many GPUs
+
+model = ed.AutoEasyDeLModelForCausalLM.from_pretrained(
+    repo_id,
+    dtype=dtype,
+    param_dtype=dtype,
+    precision=lax.Precision("fastest"),
+    sharding_axis_names=("dp", "fsdp", "ep", "tp", "sp"),
+    sharding_axis_dims=(1, -1, 1, 1, 1),
+    config_kwargs=ed.EasyDeLBaseConfigDict(
+        attn_dtype=dtype,
+        attn_mechanism=ed.AttentionMechanisms.VANILLA,
+        fsdp_is_ep_bound=True,
+        sp_is_ep_bound=True,
+        moe_method=ed.MoEMethods.FUSED_MOE,
+    ),
+    auto_shard_model=True,
+    partition_axis=ed.PartitionAxis(),
+)
+```
+
+If the repository only provides PyTorch weights, pass `from_torch=True` to `from_pretrained(...)`.
+
+## Sharding & Parallelism (Multi-Device)
+
+EasyDeL can scale to multiple devices by creating a logical device mesh. Most EasyDeL loaders use a 5D mesh:
+
+- `dp`: data parallel (replicated parameters, different batch shards)
+- `fsdp`: parameter sharding (memory saver; often the biggest axis)
+- `ep`: expert parallel (MoE; keep `1` for non-MoE models)
+- `tp`: tensor parallel (splits large matmuls)
+- `sp`: sequence parallel (splits sequence dimension)
+
+Use `sharding_axis_names=("dp","fsdp","ep","tp","sp")` and choose `sharding_axis_dims` so that their product equals your device count.
+You can use `-1` in `sharding_axis_dims` to let EasyDeL infer the remaining dimension.
+
+<details>
+<summary>Example sharding configs</summary>
+
+```python
+# 8 devices, pure FSDP
+sharding_axis_dims = (1, 8, 1, 1, 1)
+
+# 8 devices, 2-way DP x 4-way FSDP
+sharding_axis_dims = (2, 4, 1, 1, 1)
+
+# 8 devices, 4-way FSDP x 2-way TP
+sharding_axis_dims = (1, 4, 1, 2, 1)
+```
+</details>
+
+## Using via `eLargeModel` (ELM)
+
+`eLargeModel` is a higher-level interface that wires together loading, sharding, training, and eSurge inference from a single config.
+
+```python
+from easydel import eLargeModel
+
+repo_id = "EasyDeL/Llama-3.3-70B-Instruct"
+
+elm = eLargeModel.from_pretrained(repo_id)  # task is auto-detected
+elm.set_dtype("bf16")
+elm.set_sharding(axis_names=("dp", "fsdp", "ep", "tp", "sp"), axis_dims=(1, -1, 1, 1, 1))
+
+model = elm.build_model()
+# Optional: build an inference engine
+# engine = elm.build_esurge()
+```
+
+<details>
+<summary>ELM YAML config example</summary>
+
+```yaml
+model:
+  name_or_path: "EasyDeL/Llama-3.3-70B-Instruct"
+
+loader:
+  dtype: bf16
+  param_dtype: bf16
+
+sharding:
+  axis_dims: [1, -1, 1, 1, 1]
+  auto_shard_model: true
+```
+</details>
+
+## Features
+
+**EasyDeL:**
+- JAX native implementation and sharded execution
+- Configurable attention backends via `AttentionMechanisms.*`
+- Precision control via `dtype`, `param_dtype`, and `precision`
+
+## Installation
+
+```bash
+pip install easydel
+```
+
+## Links
+
+- EasyDeL GitHub: https://github.com/erfanzar/EasyDeL
+- Docs: https://easydel.readthedocs.io/en/latest/
+
+## Supported Tasks
+
+- CausalLM
+
+## Limitations
+
+- Refer to the original model card for training data, evaluation, and intended use.
+
+## License
+
+EasyDeL is released under the Apache-2.0 license. The license for this model's weights may differ; please consult the original repository.
+
+## Citation
+
+```bibtex
+@misc{Zare Chavoshi_2023,
+    title={EasyDeL: An open-source library for enhancing and streamlining the training process of machine learning models},
+    url={https://github.com/erfanzar/EasyDeL},
+    author={Zare Chavoshi, Erfan},
+    year={2023}
+}
+```

chat_template.jinja

@@ -0,0 +1,109 @@
+{{- bos_token }}
+{%- if custom_tools is defined %}
+    {%- set tools = custom_tools %}
+{%- endif %}
+{%- if not tools_in_user_message is defined %}
+    {%- set tools_in_user_message = true %}
+{%- endif %}
+{%- if not date_string is defined %}
+    {%- set date_string = "26 Jul 2024" %}
+{%- endif %}
+{%- if not tools is defined %}
+    {%- set tools = none %}
+{%- endif %}
+
+{#- This block extracts the system message, so we can slot it into the right place. #}
+{%- if messages[0]['role'] == 'system' %}
+    {%- set system_message = messages[0]['content']|trim %}
+    {%- set messages = messages[1:] %}
+{%- else %}
+    {%- set system_message = "" %}
+{%- endif %}
+
+{#- System message + builtin tools #}
+{{- "<|start_header_id|>system<|end_header_id|>\n\n" }}
+{%- if builtin_tools is defined or tools is not none %}
+    {{- "Environment: ipython\n" }}
+{%- endif %}
+{%- if builtin_tools is defined %}
+    {{- "Tools: " + builtin_tools | reject('equalto', 'code_interpreter') | join(", ") + "\n\n"}}
+{%- endif %}
+{{- "Cutting Knowledge Date: December 2023\n" }}
+{{- "Today Date: " + date_string + "\n\n" }}
+{%- if tools is not none and not tools_in_user_message %}
+    {{- "You have access to the following functions. To call a function, please respond with JSON for a function call." }}
+    {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }}
+    {{- "Do not use variables.\n\n" }}
+    {%- for t in tools %}
+        {{- t | tojson(indent=4) }}
+        {{- "\n\n" }}
+    {%- endfor %}
+{%- endif %}
+{{- system_message }}
+{{- "<|eot_id|>" }}
+
+{#- Custom tools are passed in a user message with some extra guidance #}
+{%- if tools_in_user_message and not tools is none %}
+    {#- Extract the first user message so we can plug it in here #}
+    {%- if messages | length != 0 %}
+        {%- set first_user_message = messages[0]['content']|trim %}
+        {%- set messages = messages[1:] %}
+    {%- else %}
+        {{- raise_exception("Cannot put tools in the first user message when there's no first user message!") }}
+{%- endif %}
+    {{- '<|start_header_id|>user<|end_header_id|>\n\n' -}}
+    {{- "Given the following functions, please respond with a JSON for a function call " }}
+    {{- "with its proper arguments that best answers the given prompt.\n\n" }}
+    {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }}
+    {{- "Do not use variables.\n\n" }}
+    {%- for t in tools %}
+        {{- t | tojson(indent=4) }}
+        {{- "\n\n" }}
+    {%- endfor %}
+    {{- first_user_message + "<|eot_id|>"}}
+{%- endif %}
+
+{%- for message in messages %}
+    {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %}
+        {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' }}
+    {%- elif 'tool_calls' in message %}
+        {%- if not message.tool_calls|length == 1 %}
+            {{- raise_exception("This model only supports single tool-calls at once!") }}
+        {%- endif %}
+        {%- set tool_call = message.tool_calls[0].function %}
+        {%- if builtin_tools is defined and tool_call.name in builtin_tools %}
+            {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' -}}
+            {{- "<|python_tag|>" + tool_call.name + ".call(" }}
+            {%- for arg_name, arg_val in tool_call.arguments | items %}
+                {{- arg_name + '="' + arg_val + '"' }}
+                {%- if not loop.last %}
+                    {{- ", " }}
+                {%- endif %}
+                {%- endfor %}
+            {{- ")" }}
+        {%- else  %}
+            {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' -}}
+            {{- '{"name": "' + tool_call.name + '", ' }}
+            {{- '"parameters": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- "}" }}
+        {%- endif %}
+        {%- if builtin_tools is defined %}
+            {#- This means we're in ipython mode #}
+            {{- "<|eom_id|>" }}
+        {%- else %}
+            {{- "<|eot_id|>" }}
+        {%- endif %}
+    {%- elif message.role == "tool" or message.role == "ipython" %}
+        {{- "<|start_header_id|>ipython<|end_header_id|>\n\n" }}
+        {%- if message.content is mapping or message.content is iterable %}
+            {{- message.content | tojson }}
+        {%- else %}
+            {{- message.content }}
+        {%- endif %}
+        {{- "<|eot_id|>" }}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' }}
+{%- endif %}

checkpoint_metadata.json

@@ -0,0 +1,6 @@
+{
+  "timestamp": "2025-12-28T23:06:13.860306",
+  "custom_metadata": {
+    "step": 0
+  }
+}

config.json

@@ -0,0 +1,215 @@
+{
+  "_external_rope_config_kwargs": {},
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "attn_mechanism": "vanilla",
+  "backend": null,
+  "bits": null,
+  "blocksize_b": 1,
+  "blocksize_k": 128,
+  "blocksize_q": 128,
+  "bos_token_id": 128000,
+  "decode_attn_mechanism": null,
+  "dtype": "bfloat16",
+  "easy_method": "train",
+  "embd_pdrop": 0.0,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "fcm_max_ratio": -1,
+  "fcm_min_ratio": -1,
+  "flash_attention_backward_pass_impl": "triton",
+  "fsdp_is_ep_bound": true,
+  "gradient_checkpointing": "",
+  "gradient_checkpointing_targets": null,
+  "hardware_abstraction": false,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 8192,
+  "initializer_range": 0.02,
+  "intermediate_size": 28672,
+  "kv_cache_quantization_config": null,
+  "kv_cache_sharding_sequence_axis_name": "sp",
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 131072,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "moe_force_xla_gmm": false,
+  "moe_method": "fused_moe",
+  "moe_tiling_size_batch": 4,
+  "moe_tiling_size_dim": 128,
+  "moe_tiling_size_seqlen": 128,
+  "num_attention_heads": 64,
+  "num_hidden_layers": 80,
+  "num_key_value_heads": 8,
+  "number_rep_kv": 1,
+  "operation_configs": null,
+  "pallas_k_block_size": 128,
+  "pallas_m_block_size": 128,
+  "pallas_n_block_size": 128,
+  "partition_axis": {
+    "attention_dim_axis": null,
+    "attention_kv_dim_axis": null,
+    "batch_axis": [
+      "fsdp",
+      "dp"
+    ],
+    "bias_head_sequence_axis": null,
+    "bias_key_sequence_axis": null,
+    "data_parallel_axis": "dp",
+    "decode_attention_dim_axis": null,
+    "decode_attention_kv_dim_axis": null,
+    "decode_batch_axis": [
+      "fsdp",
+      "dp"
+    ],
+    "decode_head_axis": "tp",
+    "decode_key_sequence_axis": "sp",
+    "decode_kv_head_axis": "tp",
+    "decode_query_sequence_axis": null,
+    "expert_axis": "ep",
+    "expert_gate_axis": null,
+    "expert_parallel_axis": "ep",
+    "fully_sharded_data_parallel_axis": "fsdp",
+    "head_axis": "tp",
+    "hidden_state_axis": "tp",
+    "key_sequence_axis": "sp",
+    "kv_head_axis": "tp",
+    "mlp_intermediate_axis": "tp",
+    "query_sequence_axis": "sp",
+    "sequence_axis": "sp",
+    "sequence_parallel_axis": "sp",
+    "tensor_parallel_axis": "tp",
+    "vocab_axis": "tp"
+  },
+  "platform": null,
+  "precompute_masks": true,
+  "pretraining_tp": 1,
+  "quantization_config": null,
+  "resid_pdrop": 0.0,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": {
+    "factor": 8.0,
+    "high_freq_factor": 4.0,
+    "low_freq_factor": 1.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "llama3"
+  },
+  "rope_theta": 500000.0,
+  "scan_attention_layers": false,
+  "scan_layers": false,
+  "scan_mlp_chunk_size": 1024,
+  "scan_ring_attention": true,
+  "sequence_axis_name": "sp",
+  "sharding_axis_dims": [
+    1,
+    -1,
+    1,
+    1,
+    1
+  ],
+  "sharding_axis_names": [
+    "dp",
+    "fsdp",
+    "ep",
+    "tp",
+    "sp"
+  ],
+  "sharding_dcn_axis_dims": null,
+  "sp_is_ep_bound": true,
+  "tie_word_embeddings": false,
+  "transformers_version": "4.57.3",
+  "use_cache": true,
+  "use_expert_tensor_mode": false,
+  "use_ring_of_experts": false,
+  "use_scan_mlp": false,
+  "use_sharded_kv_caching": false,
+  "use_sharding_constraint": false,
+  "vocab_size": 128256
+}

generation_config.json

@@ -0,0 +1,13 @@
+{
+  "bos_token_id": 128000,
+  "do_sample": true,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.57.3",
+  "trust_remote_code": false
+}

model/lm_head/kernel/.zarray

@@ -0,0 +1 @@
+{"chunks":[2048,128256],"compressor":{"id":"zstd","level":1},"dimension_separator":".","dtype":"bfloat16","fill_value":null,"filters":null,"order":"C","shape":[8192,128256],"zarr_format":2}

model/lm_head/kernel/0.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e023f98490ad0c18d8dcc3799bf1ef5030e45b0cde092e637a9dae001c8352de
+size 409061244

model/lm_head/kernel/1.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5acfdf8413f6724abc155586ffc7c5069fb8bc4d4a75ee66bddc8283bc8371e
+size 409068416

model/lm_head/kernel/2.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9dded260b2b571006e6e93127f77f183b04a48de490ee274bdf22ccb23cc82d
+size 409030322

model/lm_head/kernel/3.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b708c5e12e425c1b37f12d587354c836c0f62178c285550a1b0272777fb78e7
+size 409114571

model/model/embed_tokens/embedding/.zarray

@@ -0,0 +1 @@
+{"chunks":[32064,8192],"compressor":{"id":"zstd","level":1},"dimension_separator":".","dtype":"bfloat16","fill_value":null,"filters":null,"order":"C","shape":[128256,8192],"zarr_format":2}

model/model/embed_tokens/embedding/0.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:080bb9f2970310ce2a212b63cf186cce35e043ff6d108bb652afcca071ea471f
+size 410847426

model/model/embed_tokens/embedding/1.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b59fa495224bfa92dd3b50eba4c41ba5245b6bdea3b75fc57cc6255e8719ba98
+size 410227838

model/model/embed_tokens/embedding/2.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8069b75450a21dbc80564c2c3f09d19df8ca69e8cfc8224f79dddac9774815ce
+size 410412179

model/model/embed_tokens/embedding/3.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d5e5893d629660bcc3957c964e12fd0a7bdaf688bff01fbb92b452fa56d545b
+size 410781418

model/model/layers/0/input_layernorm/kernel/.zarray

@@ -0,0 +1 @@
+{"chunks":[8192],"compressor":{"id":"zstd","level":1},"dimension_separator":".","dtype":"bfloat16","fill_value":null,"filters":null,"order":"C","shape":[8192],"zarr_format":2}

model/model/layers/0/mlp/down_proj/kernel/.zarray

@@ -0,0 +1 @@
+{"chunks":[28672,2048],"compressor":{"id":"zstd","level":1},"dimension_separator":".","dtype":"bfloat16","fill_value":null,"filters":null,"order":"C","shape":[28672,8192],"zarr_format":2}

model/model/layers/0/mlp/down_proj/kernel/0.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b87f7ca613b8b5d10498be5b7c9a458c2ea5c50c1233b8e2f85bba5f665a7857
+size 92157948

model/model/layers/0/mlp/down_proj/kernel/0.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a75bf55e0e51686488f8c6599a8d8e49c3a177bb53ec03bed743925c1594108
+size 92166225

model/model/layers/0/mlp/down_proj/kernel/0.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3777332e5ace8f8594bf35b44639e3da053b1f579419104533354c36049f420
+size 92147315

model/model/layers/0/mlp/down_proj/kernel/0.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c120b55bfa6a4b973373ed2f5c8a7952c9140e57f6f82c60d963b2d6cea7516e
+size 92154734

model/model/layers/0/mlp/gate_proj/kernel/.zarray

@@ -0,0 +1 @@
+{"chunks":[2048,28672],"compressor":{"id":"zstd","level":1},"dimension_separator":".","dtype":"bfloat16","fill_value":null,"filters":null,"order":"C","shape":[8192,28672],"zarr_format":2}

model/model/layers/0/mlp/gate_proj/kernel/0.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a62d10c42631f75b3997a48f3c71665afc7c4b50fe2788335af4ed593fc93c4
+size 92342325

model/model/layers/0/mlp/gate_proj/kernel/1.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e02fbcdad22b53a4c4c2dc1dbdf88de93849912392628819c70a3e4e77318dd
+size 92335334

model/model/layers/0/mlp/gate_proj/kernel/2.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c3c14aef21060c164931dcb94e81660739e910d3835095088c1b6278b5283f5
+size 92324009

model/model/layers/0/mlp/gate_proj/kernel/3.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:024834f5991235639117b36f05651bf3c1a0c6fc8370228d58c3826fe6a4797a
+size 92318876

model/model/layers/0/mlp/up_proj/kernel/.zarray

@@ -0,0 +1 @@
+{"chunks":[2048,28672],"compressor":{"id":"zstd","level":1},"dimension_separator":".","dtype":"bfloat16","fill_value":null,"filters":null,"order":"C","shape":[8192,28672],"zarr_format":2}

model/model/layers/0/mlp/up_proj/kernel/0.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1a7f150a8f7c056c150f8981756630d600ae3c591b82d57042f15da33479dd0
+size 92327317

model/model/layers/0/mlp/up_proj/kernel/1.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16728bcb4fcc53a08974f26885ab050aa9af5f1d330f68780ecd8a315ca31eb8
+size 92320030

model/model/layers/0/mlp/up_proj/kernel/2.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b141d2eb4ed6d339bff6aaafaae23ac6755931ca8192476984627c2bca5f0912
+size 92308700

model/model/layers/0/mlp/up_proj/kernel/3.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9050589ad66ead6e9f2bb3c85663268179f08d4c47843546c11aa926f396d61
+size 92304207

model/model/layers/0/post_attention_layernorm/kernel/.zarray

@@ -0,0 +1 @@
+{"chunks":[8192],"compressor":{"id":"zstd","level":1},"dimension_separator":".","dtype":"bfloat16","fill_value":null,"filters":null,"order":"C","shape":[8192],"zarr_format":2}

model/model/layers/0/self_attn/k_proj/kernel/.zarray

@@ -0,0 +1 @@
+{"chunks":[2048,1024],"compressor":{"id":"zstd","level":1},"dimension_separator":".","dtype":"bfloat16","fill_value":null,"filters":null,"order":"C","shape":[8192,1024],"zarr_format":2}

model/model/layers/0/self_attn/k_proj/kernel/0.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65ce52ba4ab16c6c3faaed0bad30d53d4de688174c0955ce59003bb4ce7d88c9
+size 3435822

model/model/layers/0/self_attn/k_proj/kernel/1.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43e8b5002b7bf8fbb0c9192cfe7afae9792870e44196f649d724c283723e0fab
+size 3438004

model/model/layers/0/self_attn/k_proj/kernel/2.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:259d4a4bf9591c9a78e388f3e5169b11f5da3de6de2f1b861f2e500291dd3bb9
+size 3435360

model/model/layers/0/self_attn/k_proj/kernel/3.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:99072bfa624a15f794ca179f010a4db0e77059ebb19412f2fe60e77439f0704b
+size 3440675

model/model/layers/0/self_attn/o_proj/kernel/.zarray

@@ -0,0 +1 @@
+{"chunks":[8192,2048],"compressor":{"id":"zstd","level":1},"dimension_separator":".","dtype":"bfloat16","fill_value":null,"filters":null,"order":"C","shape":[8192,8192],"zarr_format":2}

model/model/layers/0/self_attn/o_proj/kernel/0.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b93b26d39cce6f94173ad00f9ce4dec3599ac0d6b70af54e199d113095c0901e
+size 26206461

model/model/layers/0/self_attn/o_proj/kernel/0.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e35cd0562c3f2e6e933714fda7edc8b2b3d75aa23fec60a9878ef39c24c39df8
+size 26209401

model/model/layers/0/self_attn/o_proj/kernel/0.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa15a48e1441579ca3baac0c6da232afc6bf999caf4f80fd785e0f16182d8af9
+size 26208277

model/model/layers/0/self_attn/o_proj/kernel/0.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c27c9c1c17de6607485be5a034accea336a9ab30b0134bd50e91843880d69daa
+size 26205970

model/model/layers/0/self_attn/q_proj/kernel/.zarray

@@ -0,0 +1 @@
+{"chunks":[2048,8192],"compressor":{"id":"zstd","level":1},"dimension_separator":".","dtype":"bfloat16","fill_value":null,"filters":null,"order":"C","shape":[8192,8192],"zarr_format":2}

model/model/layers/0/self_attn/q_proj/kernel/0.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b7bea62181f2b584aab3bec70bdb95141f50476f8ee53eb037c19d63fdeff5a
+size 27411754

model/model/layers/0/self_attn/q_proj/kernel/1.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4de20190a9608167b07a157bf2ac96b80f89d65ef48fcc297e8dec6510bb451b
+size 27421372

model/model/layers/0/self_attn/q_proj/kernel/2.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bec1193a97ff457719cf58636349497799b292ef3074dde56fd3b4c89a408f3c
+size 27407551

model/model/layers/0/self_attn/q_proj/kernel/3.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28f62585273bd3a983361d73c653b41d9b84995c82c994dd65978625ec487c4c
+size 27434725