© 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. 18943
    Prev
    Next
    INDEX
    Explanations

    runner-ups or second-place finishes in competitions

    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
    216
    +0.11
    0.5%
    445
    +0.10
    0.5%
    155
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    727
    +0.11
    0.01
    326
    +0.10
    0.01
    445
    +0.10
    0.01
    Negative Logits
    olia
    -0.79
    rification
    -0.74
     ILCS
    -0.70
     Debor
    -0.69
     holdings
    -0.68
    ornia
    -0.67
    rox
    -0.66
    inian
    -0.66
    orescent
    -0.65
    illery
    -0.64
    POSITIVE LOGITS
    ners
    1.02
    swick
    0.99
    gaard
    0.94
    wear
    0.85
    ways
    0.84
    nings
    0.84
    runner
    0.81
    bum
    0.79
    aways
    0.77
     stitch
    0.76
    Activations Density 0.012%

    No Known Activations