© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 7222
    Prev
    Next
    INDEX
    Explanations

    phrases related to work or tasks

    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
    608
    +0.16
    0.7%
    373
    +0.14
    0.7%
    113
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.16
    0.03
    113
    +0.14
    0.02
    282
    +0.09
    0.02
    Negative Logits
     heartbeat
    -0.65
    audi
    -0.65
     Rahman
    -0.65
    direction
    -0.63
    payment
    -0.61
     payable
    -0.60
     incoming
    -0.60
    RD
    -0.59
    currency
    -0.58
    ials
    -0.58
    POSITIVE LOGITS
     grips
    1.14
    asted
    1.05
    wered
    0.83
    ench
    0.78
    pless
    0.76
    ARC
    0.75
     nominate
    0.75
     choose
    0.74
     settle
    0.74
     pick
    0.74
    Activations Density 0.043%

    No Known Activations