© 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. 19996
    Prev
    Next
    INDEX
    Explanations

    financial and legal terms and activities

    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
    165
    +0.10
    0.5%
    594
    +0.10
    0.5%
    111
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.10
    0.07
    162
    +0.10
    0.03
    594
    +0.10
    0.03
    Negative Logits
    xual
    -0.88
    yip
    -0.76
    AppData
    -0.74
    arij
    -0.73
    namese
    -0.71
    etheless
    -0.69
    genre
    -0.68
    çĶ°
    -0.68
    llah
    -0.67
    archy
    -0.67
    POSITIVE LOGITS
     upkeep
    1.11
     expenses
    1.01
     renovations
    0.87
     services
    0.85
     repairs
    0.85
     subscriptions
    0.85
     goods
    0.84
     fees
    0.84
     equipment
    0.83
     necessities
    0.82
    Activations Density 0.511%

    No Known Activations