© 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 49152
    9. 8-RES_FS49152-JB
    11. 45620
    Prev
    Next
    INDEX
    Explanations

    words that indicate professions or roles ending in 'ist' or 'ists'

    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
    216
    +0.13
    0.6%
    679
    +0.13
    0.6%
    97
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    32
    +0.13
    0.02
    510
    +0.13
    0.02
    235
    +0.12
    0.02
    Negative Logits
    FACE
    -0.85
    MENTS
    -0.75
    ç«
    -0.75
    é¾
    -0.74
    shown
    -0.73
    ãģ®éŃĶ
    -0.72
    ãĥīãĥ©
    -0.71
    zens
    -0.66
    ghazi
    -0.65
    APTER
    -0.64
    POSITIVE LOGITS
     extraord
    0.86
    otle
    0.84
    ische
    0.81
    ophical
    0.79
    ischer
    0.79
    ophe
    0.77
    emi
    0.77
    ribution
    0.75
    atics
    0.75
     tendencies
    0.74
    Activations Density 0.032%

    No Known Activations