© 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. 1194
    Prev
    Next
    INDEX
    Explanations

    instances of words related to retraction or retractions in texts relating to research, publications, and controversies

    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
    503
    +0.13
    0.6%
    651
    +0.12
    0.6%
    721
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    651
    +0.13
    0.03
    503
    +0.12
    0.02
    232
    +0.10
    0.02
    Negative Logits
    tips
    -0.76
    SHIP
    -0.70
    STD
    -0.69
    STEM
    -0.66
    latest
    -0.64
     Passion
    -0.63
    xual
    -0.62
     hearts
    -0.61
    ership
    -0.61
    INESS
    -0.61
    POSITIVE LOGITS
    ribut
    1.17
    raction
    1.12
    ainer
    1.08
    rieve
    1.06
    rans
    1.05
    rieving
    1.05
    itled
    1.04
    ribution
    1.03
    reating
    1.03
    ention
    1.00
    Activations Density 0.058%

    No Known Activations