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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 7968
    Prev
    Next
    INDEX
    Explanations

    references to the sun and sunlight

    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
    603
    +0.12
    0.5%
    420
    +0.11
    0.5%
    27
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    27
    +0.12
    0.02
    603
    +0.11
    0.02
    384
    +0.11
    0.02
    Negative Logits
     Halls
    -0.73
    axter
    -0.70
    ngth
    -0.69
     ABE
    -0.69
    ussen
    -0.69
    ij士
    -0.68
    ourgeois
    -0.68
     Mellon
    -0.67
    USS
    -0.67
     Hodg
    -0.67
    POSITIVE LOGITS
    flower
    1.35
    shine
    1.29
    burst
    1.24
    lit
    1.19
    beam
    1.18
    nah
    1.14
    rises
    1.13
    burn
    1.11
    spot
    1.06
    bat
    1.05
    Activations Density 0.022%

    No Known Activations