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 1536
    5. 8-RES_FS1536-JB
    6. 115
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences that begin with "According to."

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    679
    +0.12
    0.5%
    169
    +0.11
    0.5%
    739
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    609
    +0.12
    0.07
    169
    +0.11
    0.07
    438
    +0.11
    0.06
    Negative Logits
    eg
    -0.73
    IFT
    -0.70
    obyl
    -0.65
    affe
    -0.65
    oppable
    -0.64
    ql
    -0.63
    estern
    -0.63
    igmat
    -0.62
     realise
    -0.61
    otin
    -0.57
    POSITIVE LOGITS
     sources
    0.81
    SOURCE
    0.76
    Sources
    0.68
    ly
    0.68
    tained
    0.67
    Rank
    0.67
    Ĥİ
    0.67
    tains
    0.66
    translation
    0.66
    edly
    0.65
    Activations Density 0.797%

    No Known Activations