© 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. 9630
    Prev
    Next
    INDEX
    Explanations

    personal names

    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
    167
    +0.13
    0.6%
    151
    +0.10
    0.5%
    421
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    167
    +0.13
    0.01
    232
    +0.10
    0.02
    326
    +0.10
    0.02
    Negative Logits
    اÙĦ
    -0.68
    ãĥĨãĤ£
    -0.65
     Afric
    -0.64
    fields
    -0.63
    xual
    -0.62
    CONCLUS
    -0.59
     Alz
    -0.58
    adium
    -0.57
    addons
    -0.57
    Effective
    -0.57
    POSITIVE LOGITS
    gement
    1.62
    gments
    1.44
    gment
    1.37
    iciary
    1.37
    icial
    1.34
    gements
    1.20
    ging
    1.19
    ges
    1.10
    icator
    1.07
    ged
    1.00
    Activations Density 0.075%

    No Known Activations