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. 4-RES-JB
    6. 3387
    Prev
    Next
    INDEX
    Explanations

    terms related to size or magnitude

    oai_token-act-pair · gpt-3.5-turbo

    terms indicating significant quantities or large-scale concepts

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

    No Comments

    Negative Logits
    gently
    -0.82
    pered
    -0.62
     Keller
    -0.62
     fluent
    -0.61
    xit
    -0.61
    gent
    -0.59
     Orient
    -0.58
     Jinn
    -0.58
    İĭ
    -0.57
     taps
    -0.56
    POSITIVE LOGITS
     (>
    0.97
    ãĤ¼ãĤ¦ãĤ¹
    0.78
    occup
    0.74
     conglomer
    0.72
    catentry
    0.69
    ortment
    0.66
    ————
    0.63
    WER
    0.63
    andowski
    0.63
     undertaking
    0.63
    Activations Density 0.492%

    No Known Activations