Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 8-RES-JB
    6. 6651
    Prev
    Next
    INDEX
    Explanations

    the verb "is" followed by various types of content

    oai_token-act-pair · gpt-3.5-turbo

    instances of the verb "is" indicating a description or definition

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.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

    Negative Logits
    uld
    -0.76
    know
    -0.76
    usercontent
    -0.70
     stake
    -0.69
    sburg
    -0.68
    ye
    -0.66
    want
    -0.62
    inav
    -0.61
    ality
    -0.61
    ossom
    -0.61
    POSITIVE LOGITS
     reminiscent
    1.13
     comprised
    1.02
     currently
    1.00
     housed
    0.98
     slated
    0.98
     meant
    0.97
     definitely
    0.97
     supposed
    0.97
     designed
    0.95
     also
    0.94
    Activations Density 0.394%

    No Known Activations