Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 49152
    5. 8-RES_FS49152-JB
    6. 41576
    Prev
    Next
    INDEX
    Explanations

    phrases indicating actions or objectives involving learning or improvement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    608
    +0.13
    0.6%
    665
    +0.13
    0.6%
    718
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.13
    0.03
    718
    +0.13
    0.02
    49
    +0.11
    0.02
    Negative Logits
    lights
    -0.89
    listed
    -0.74
    holder
    -0.72
    teen
    -0.72
     Hartford
    -0.70
    isite
    -0.69
     Coff
    -0.68
     Appears
    -0.68
    quished
    -0.66
    prem
    -0.64
    POSITIVE LOGITS
     differentiate
    1.28
     cope
    1.26
     recognize
    1.24
     behave
    1.23
     manipulate
    1.16
     tolerate
    1.15
     wrestle
    1.14
     imitate
    1.13
     distinguish
    1.13
     identify
    1.10
    Activations Density 0.040%

    No Known Activations