Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 564
    Prev
    Next
    INDEX
    Explanations

    informal expressions and situations

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    447
    +0.16
    0.7%
    46
    +0.11
    0.5%
    9
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    271
    +0.16
    0.08
    288
    +0.11
    0.13
    151
    +0.11
    0.12
    Negative Logits
    ilon
    -0.79
    ATURE
    -0.73
    extra
    -0.68
    CVE
    -0.63
    emp
    -0.62
    =#
    -0.62
    frey
    -0.61
    isite
    -0.61
    elin
    -0.61
    rera
    -0.61
    POSITIVE LOGITS
     plagued
    0.84
     bog
    0.83
     miser
    0.80
     navigating
    0.78
    hooting
    0.76
     owing
    0.75
     grasping
    0.74
    luster
    0.74
     coping
    0.73
     financially
    0.72
    Activations Density 1.731%

    No Known Activations