© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 6356
    Prev
    Next
    INDEX
    Explanations

    positive and negative emotional responses related to difficult situations, such as loss, stress, and resolution

    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
    279
    +0.09
    0.4%
    287
    +0.09
    0.4%
    314
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    718
    +0.09
    0.04
    763
    +0.09
    0.03
    53
    +0.09
    0.03
    Negative Logits
    ahime
    -0.84
     [|
    -0.73
     sometime
    -0.60
    YN
    -0.58
    majority
    -0.58
     summers
    -0.57
     AX
    -0.57
    sea
    -0.55
    auld
    -0.55
    ß
    -0.54
    POSITIVE LOGITS
    except
    1.10
     whatsoever
    1.08
     except
    1.01
     resembling
    0.84
    nor
    0.83
     besides
    0.80
     nor
    0.79
    Except
    0.77
    soever
    0.72
    imilar
    0.69
    Activations Density 0.158%

    No Known Activations