© 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 49152
    9. 8-RES_FS49152-JB
    11. 37183
    Prev
    Next
    INDEX
    Explanations

    assertive language related to firmness or strong convictions

    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
    718
    +0.12
    0.6%
    624
    +0.10
    0.4%
    4
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    718
    +0.12
    0.01
    624
    +0.10
    0.01
    4
    +0.10
    0.01
    Negative Logits
    ;;;;;;;;;;;;
    -0.83
    querade
    -0.78
    ̶
    -0.76
    vernment
    -0.71
     Emin
    -0.70
    NF
    -0.70
    FactoryReloaded
    -0.69
     livest
    -0.69
    Lv
    -0.66
    ;;;;;;;;
    -0.66
    POSITIVE LOGITS
    ament
    1.25
    ness
    1.14
    nesses
    1.01
     believer
    1.00
     footing
    0.97
     grounding
    0.91
    hearted
    0.89
    ures
    0.87
    ty
    0.87
    est
    0.86
    Activations Density 0.014%

    No Known Activations