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 6144
    5. 8-RES_FS6144-JB
    6. 1822
    Prev
    Next
    INDEX
    Explanations

    comparative phrases highlighting similarities or differences

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs6144-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    6,144
    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
    373
    +0.12
    0.5%
    218
    +0.11
    0.5%
    430
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    59
    +0.12
    0.07
    121
    +0.11
    0.09
    320
    +0.10
    0.06
    Negative Logits
    £ı
    -0.79
    Interstitial
    -0.78
     redes
    -0.76
     obser
    -0.75
    paren
    -0.72
    estic
    -0.69
     arenas
    -0.69
     downtime
    -0.67
    entimes
    -0.66
    pec
    -0.66
    POSITIVE LOGITS
     Stephen
    1.22
     Mr
    1.21
     Mike
    1.21
     Michael
    1.21
     Richard
    1.20
     Kevin
    1.20
     Jon
    1.19
     Donald
    1.19
     John
    1.19
     Juan
    1.18
    Activations Density 1.789%

    No Known Activations