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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 13786
    Prev
    Next
    INDEX
    Explanations

    phrases related to opinions and evaluations

    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
    640
    +0.11
    0.5%
    74
    +0.10
    0.5%
    44
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.11
    0.07
    6
    +0.10
    0.05
    245
    +0.10
    0.04
    Negative Logits
    urated
    -0.85
    orate
    -0.77
    oled
    -0.67
    urous
    -0.64
     Aval
    -0.64
    oire
    -0.64
    daq
    -0.62
    ashtra
    -0.61
     vict
    -0.59
    orer
    -0.58
    POSITIVE LOGITS
     Firstly
    1.31
    Firstly
    1.24
    âĹı
    1.05
    Reason
    0.97
    %:
    0.86
    First
    0.85
    ·
    0.80
    âĢ¢
    0.79
    Because
    0.77
     First
    0.77
    Activations Density 0.217%

    No Known Activations