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. 12206
    Prev
    Next
    INDEX
    Explanations

    references to "cycles" and related concepts

    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
    288
    +0.13
    0.6%
    326
    +0.12
    0.5%
    546
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.13
    0.01
    546
    +0.12
    0.01
    593
    +0.10
    0.01
    Negative Logits
    inez
    -0.99
    orage
    -0.95
    essors
    -0.84
    avez
    -0.79
    alty
    -0.79
    oaded
    -0.77
    anamo
    -0.77
    amina
    -0.77
    imus
    -0.76
     Aires
    -0.75
    POSITIVE LOGITS
     cycle
    1.08
    cycle
    0.91
     erg
    0.89
     cycles
    0.88
     Cycle
    0.85
    cles
    0.83
     repeats
    0.82
    cycles
    0.82
     alternating
    0.79
    chron
    0.76
    Activations Density 0.013%

    No Known Activations