© 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 24576
    9. 8-RES_FS24576-JB
    11. 18806
    Prev
    Next
    INDEX
    Explanations

    emotional and societal terms related to personal experiences and societal issues

    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
    420
    +0.11
    0.5%
    338
    +0.11
    0.5%
    737
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.11
    0.07
    737
    +0.11
    0.05
    325
    +0.10
    0.06
    Negative Logits
    Rated
    -0.70
    SF
    -0.65
    odore
    -0.64
     âĢº
    -0.62
    TF
    -0.61
     Lod
    -0.59
    byter
    -0.58
     Uz
    -0.58
    uilt
    -0.58
    esides
    -0.58
    POSITIVE LOGITS
    ousel
    0.85
    hole
    0.83
     mentality
    0.82
    chairs
    0.78
     politic
    0.76
     fallacy
    0.76
    elight
    0.74
    osphere
    0.74
    arity
    0.73
    belt
    0.71
    Activations Density 0.307%

    No Known Activations