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. 804
    Prev
    Next
    INDEX
    Explanations

    phrases related to economics and construction

    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
    216
    +0.11
    0.5%
    524
    +0.10
    0.5%
    519
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.11
    0.13
    524
    +0.10
    0.09
    216
    +0.09
    0.09
    Negative Logits
    zynski
    -0.73
    ewitness
    -0.68
    ifi
    -0.67
    ãĥĩãĤ£
    -0.66
    cci
    -0.64
    arine
    -0.63
    pps
    -0.63
    yson
    -0.62
    Ĥİ
    -0.62
    ifier
    -0.62
    POSITIVE LOGITS
     techniques
    1.14
     prowess
    1.14
    redients
    1.07
     spree
    1.04
     habits
    1.00
     methods
    1.00
     capabilities
    0.96
     abilities
    0.94
     capability
    0.94
     ability
    0.94
    Activations Density 1.092%

    No Known Activations