Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 1146
    Prev
    Next
    INDEX
    Explanations

    words related to historical events and military operations

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    701
    +0.14
    0.7%
    678
    +0.11
    0.5%
    373
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    701
    +0.14
    0.03
    232
    +0.11
    0.02
    582
    +0.09
    0.01
    Negative Logits
     Clash
    -0.62
    ++++++++++++++++
    -0.60
    ITED
    -0.59
    ZA
    -0.59
    INC
    -0.58
     sterling
    -0.58
     rake
    -0.57
     segregated
    -0.57
     mosaic
    -0.57
     cubes
    -0.57
    POSITIVE LOGITS
    ulhu
    1.52
    ttp
    1.43
    orne
    1.19
    istle
    1.11
    umbnail
    1.10
    ieves
    1.06
    ousand
    1.05
    irteen
    1.04
    ousands
    1.03
    irty
    1.03
    Activations Density 0.029%

    No Known Activations