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. 0-RES-JB
    6. 3073
    Prev
    Next
    INDEX
    Explanations

    species names of animals

    oai_token-act-pair · gpt-3.5-turbo

    terms related to scientific classification, particularly of organisms and biological taxa

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     PW
    -0.68
     rate
    -0.64
     dive
    -0.62
     Jets
    -0.60
     elim
    -0.60
     rates
    -0.59
     Rockets
    -0.59
     dump
    -0.59
     downhill
    -0.59
    RT
    -0.59
    POSITIVE LOGITS
    icus
    4.71
    ici
    1.96
    icum
    1.91
    alus
    1.52
    ica
    1.38
    ican
    1.35
    ensis
    1.34
    icans
    1.33
    icia
    1.33
    icas
    1.28
    Activations Density 0.019%

    No Known Activations