© 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 12288
    9. 8-RES_FS12288-JB
    11. 774
    Prev
    Next
    INDEX
    Explanations

    information on the availability of products or services

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    496
    +0.13
    0.6%
    558
    +0.12
    0.5%
    288
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    496
    +0.13
    0.04
    677
    +0.12
    0.03
    106
    +0.10
    0.03
    Negative Logits
    ²¾
    -0.77
    kos
    -0.74
    mot
    -0.73
    cig
    -0.69
    bey
    -0.68
    asus
    -0.67
    dream
    -0.67
    pson
    -0.66
    mother
    -0.65
    cher
    -0.63
    POSITIVE LOGITS
    available
    0.76
     availability
    0.74
     unrestricted
    0.72
     online
    0.71
    isable
    0.69
     downloadable
    0.68
    Choice
    0.67
     simultane
    0.66
    ibaba
    0.65
    oin
    0.64
    Activations Density 0.082%

    No Known Activations