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 · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 10192
    Prev
    Next
    INDEX
    Explanations

    instances of the word 'provide'

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to offering or supplying services or resources

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    mat
    -0.76
    Boss
    -0.75
    SEE
    -0.68
     Nanto
    -0.67
    Rated
    -0.67
    phase
    -0.64
    FactoryReloaded
    -0.64
    dar
    -0.64
    boss
    -0.64
    beat
    -0.64
    POSITIVE LOGITS
    utical
    0.92
     sust
    0.84
    atives
    0.83
     condolences
    0.79
     refunds
    0.79
    atively
    0.78
     showc
    0.75
    ative
    0.71
    ocative
    0.70
    cially
    0.69
    Activations Density 0.038%

    No Known Activations