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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 37118
    Prev
    Next
    INDEX
    Explanations

    occurrences of the term "mond" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    642
    +0.19
    0.9%
    459
    +0.13
    0.6%
    553
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    642
    +0.19
    0.01
    464
    +0.13
    0.01
    556
    +0.10
    0.01
    Negative Logits
    glers
    -0.79
     abdom
    -0.76
    NRS
    -0.71
    interstitial
    -0.68
    Interstitial
    -0.68
    ASED
    -0.64
    kson
    -0.64
    ngth
    -0.63
    RPG
    -0.62
    cki
    -0.61
    POSITIVE LOGITS
    iac
    1.01
    iate
    0.99
    e
    0.90
    mond
    0.87
    sey
    0.86
    ieth
    0.84
    ian
    0.79
     Beach
    0.79
    ous
    0.77
    iated
    0.77
    Activations Density 0.005%

    No Known Activations