© 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 24576
    9. 8-RES_FS24576-JB
    11. 19638
    Prev
    Next
    INDEX
    Explanations

    words related to excitement and positivity

    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
    519
    +0.12
    0.6%
    218
    +0.12
    0.6%
    55
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    55
    +0.12
    0.09
    519
    +0.12
    0.07
    566
    +0.10
    0.07
    Negative Logits
    soever
    -0.86
     ILCS
    -0.84
    sequently
    -0.73
    udeau
    -0.73
    rouse
    -0.72
    atform
    -0.71
    ilst
    -0.69
    cific
    -0.69
    istrates
    -0.69
    enture
    -0.68
    POSITIVE LOGITS
     gonna
    1.31
     alright
    1.05
     funny
    1.05
     awesome
    1.03
     awfully
    0.99
     hilarious
    0.99
     cute
    0.98
     fuckin
    0.97
     kinda
    0.97
     fucked
    0.97
    Activations Density 0.331%

    No Known Activations