© 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. 9950
    Prev
    Next
    INDEX
    Explanations

    instances of information or reported statements that have been revealed or communicated in a document

    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
    393
    +0.12
    0.5%
    533
    +0.11
    0.5%
    679
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    533
    +0.12
    0.03
    655
    +0.11
    0.03
    274
    +0.10
    0.03
    Negative Logits
    izont
    -0.89
    ãĤ´ãĥ³
    -0.83
    estern
    -0.68
    quer
    -0.67
    aukee
    -0.67
    PRESS
    -0.65
    ocaust
    -0.65
    ãĥij
    -0.64
     pione
    -0.64
    istor
    -0.63
    POSITIVE LOGITS
     consequently
    1.08
     they
    0.96
     therefore
    0.91
    soever
    0.91
     furthermore
    0.89
     moreover
    0.86
     there
    0.80
     accordingly
    0.79
     although
    0.75
     THEY
    0.74
    Activations Density 0.093%

    No Known Activations