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 24576
    5. 8-RES_FS24576-JB
    6. 622
    Prev
    Next
    INDEX
    Explanations

    phrases related to time measurements, financial transactions, and legal issues

    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
    492
    +0.13
    0.6%
    327
    +0.11
    0.5%
    338
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    81
    +0.13
    0.03
    104
    +0.11
    0.02
    409
    +0.10
    0.02
    Negative Logits
    ãĥķãĤ©
    -0.93
    raq
    -0.84
    Cu
    -0.80
    Tor
    -0.77
     Slime
    -0.75
     Unic
    -0.73
    Mob
    -0.72
    Pal
    -0.72
    Tele
    -0.70
     Scor
    -0.68
    POSITIVE LOGITS
     year
    1.28
     YEAR
    1.20
     years
    1.17
    year
    1.14
    Year
    1.14
     decade
    1.06
     Year
    1.06
    years
    1.01
     Years
    1.00
     millennium
    0.96
    Activations Density 0.094%

    No Known Activations