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. 21646
    Prev
    Next
    INDEX
    Explanations

    references to the word "Home"

    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
    152
    +0.11
    0.5%
    270
    +0.11
    0.5%
    710
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    152
    +0.11
    0.04
    270
    +0.11
    0.03
    444
    +0.09
    0.03
    Negative Logits
    terday
    -0.80
    andum
    -0.80
    uel
    -0.72
    utical
    -0.68
    uality
    -0.66
     Marketable
    -0.64
    ual
    -0.63
    ãģĨ
    -0.63
    ongyang
    -0.62
    asm
    -0.62
    POSITIVE LOGITS
    opathic
    1.13
    coming
    1.12
    opath
    1.10
    grown
    1.09
    brew
    1.09
    opathy
    1.07
    owners
    1.02
    pages
    0.99
    owner
    0.93
    pun
    0.92
    Activations Density 0.040%

    No Known Activations