© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 3993
    Prev
    Next
    INDEX
    Explanations

    legal and political references, including specific names of politicians and their actions

    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
    340
    +0.11
    0.5%
    456
    +0.10
    0.5%
    481
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    547
    +0.11
    0.04
    25
    +0.10
    0.04
    430
    +0.10
    0.05
    Negative Logits
    Rated
    -0.66
    ''.
    -0.65
    "}
    -0.59
     };
    -0.57
    channelAvailability
    -0.55
    malink
    -0.55
     Directive
    -0.55
     penetrate
    -0.53
     Chau
    -0.53
    }}}
    -0.52
    POSITIVE LOGITS
    that
    0.77
    whether
    0.70
    ello
    0.67
     whether
    0.64
    iety
    0.63
    tics
    0.63
    ouncing
    0.62
     that
    0.61
    iasm
    0.60
    ãĤ¤ãĥĪ
    0.59
    Activations Density 0.306%

    No Known Activations