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. 12088
    Prev
    Next
    INDEX
    Explanations

    statistics and survey results in text

    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
    481
    +0.15
    0.7%
    607
    +0.09
    0.4%
    488
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    546
    +0.15
    0.04
    607
    +0.09
    0.04
    393
    +0.08
    0.05
    Negative Logits
    ¬¼
    -0.73
    onial
    -0.66
    argo
    -0.64
    hid
    -0.64
    quished
    -0.63
     arrang
    -0.63
    displayText
    -0.63
    NetMessage
    -0.61
     fronts
    -0.61
    DragonMagazine
    -0.61
    POSITIVE LOGITS
     71
    1.05
     61
    1.04
     51
    1.03
     53
    1.03
     46
    1.03
     41
    1.03
     62
    1.01
     47
    1.01
     63
    1.00
     77
    1.00
    Activations Density 0.328%

    No Known Activations