© 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 24576
    9. 8-RES_FS24576-JB
    11. 19807
    Prev
    Next
    INDEX
    Explanations

    explicit pleas for financial support or donations

    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
    +0.11
    0.5%
    228
    +0.10
    0.4%
    203
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.11
    0.02
    228
    +0.10
    0.02
    50
    +0.09
    0.02
    Negative Logits
    ãĤ»
    -0.72
    ARK
    -0.64
    aukee
    -0.64
    PRESS
    -0.61
    EMP
    -0.61
    0200
    -0.61
    atis
    -0.60
    cue
    -0.58
    ahime
    -0.58
    Guard
    -0.58
    POSITIVE LOGITS
     they
    0.86
     someday
    0.83
     nobody
    0.82
     whoever
    0.81
     we
    0.76
    uces
    0.74
     everyone
    0.74
    soever
    0.74
     everybody
    0.70
     somebody
    0.70
    Activations Density 0.033%

    No Known Activations