© 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 12288
    9. 8-RES_FS12288-JB
    11. 9795
    Prev
    Next
    INDEX
    Explanations

    specific names or abbreviations of organizations

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    232
    +0.13
    0.6%
    447
    +0.12
    0.6%
    145
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.13
    0.12
    145
    +0.12
    0.07
    631
    +0.11
    0.07
    Negative Logits
    etheless
    -0.91
     sight
    -0.74
     shroud
    -0.71
     separatist
    -0.69
    WAYS
    -0.67
     silence
    -0.67
    anwhile
    -0.67
     writ
    -0.66
    ãģį
    -0.65
    ãģ§
    -0.65
    POSITIVE LOGITS
    iman
    1.09
    istar
    1.06
    amon
    1.05
    ima
    1.04
    aq
    1.03
    aci
    1.03
    acia
    1.03
    avia
    1.02
    ona
    1.02
    rington
    1.02
    Activations Density 0.459%

    No Known Activations