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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 7807
    Prev
    Next
    INDEX
    Explanations

    references to sports match results and player performance

    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
    382
    +0.10
    0.5%
    715
    +0.10
    0.4%
    244
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.10
    0.04
    130
    +0.10
    0.02
    99
    +0.10
    0.03
    Negative Logits
    unk
    -0.78
    unks
    -0.70
     Oprah
    -0.70
    IFT
    -0.69
     OW
    -0.68
    OFF
    -0.65
    yout
    -0.65
     Tycoon
    -0.65
     Unt
    -0.65
    ifted
    -0.65
    POSITIVE LOGITS
    C
    1.29
    c
    1.25
    cs
    1.21
    ci
    1.12
    CI
    1.10
    cb
    1.08
    CB
    1.05
    Cs
    1.02
    cp
    1.01
    cer
    0.98
    Activations Density 0.218%

    No Known Activations