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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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. 1781
    Prev
    Next
    INDEX
    Explanations

    financial investments and business plans mentioned in news articles

    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
    51
    +0.11
    0.5%
    289
    +0.10
    0.5%
    688
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    688
    +0.11
    0.09
    529
    +0.10
    0.07
    554
    +0.10
    -0.03
    Negative Logits
     eloqu
    -0.88
     weird
    -0.82
     innoc
    -0.82
     fame
    -0.81
     truth
    -0.80
    mom
    -0.80
     freak
    -0.80
     hating
    -0.79
     rant
    -0.78
     injustice
    -0.78
    POSITIVE LOGITS
     phased
    1.16
     2021
    1.10
     2019
    1.08
     rollout
    1.04
     operational
    1.03
     2020
    1.03
     outpatient
    1.02
     staffing
    1.01
    ufact
    1.00
     workforce
    0.99
    Activations Density 4.294%

    No Known Activations