© 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 3072
    9. 8-RES_FS3072-JB
    11. 3025
    Prev
    Next
    INDEX
    Explanations

    terms related to technical specifications and product features

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    99
    +0.14
    0.6%
    527
    +0.13
    0.6%
    57
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.14
    0.07
    99
    +0.13
    0.06
    232
    +0.11
    0.06
    Negative Logits
    loo
    -0.99
    hips
    -0.96
    ienced
    -0.83
    DOWN
    -0.81
     bearer
    -0.79
    trap
    -0.71
    allery
    -0.71
    acies
    -0.68
    REAM
    -0.66
    gerald
    -0.66
    POSITIVE LOGITS
    arest
    1.11
    verend
    1.08
    braska
    1.07
    olithic
    0.97
    jad
    0.92
    cker
    0.91
    cht
    0.88
    uve
    0.83
    arthed
    0.83
    gger
    0.81
    Activations Density 0.731%

    No Known Activations