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. 8-RES-JB
    6. 36
    Prev
    Next
    INDEX
    Explanations

    references to a person named "Tipp"

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the term "ipp," suggesting a focus on specific concepts or terms related to a context that involves 'ipp.'

    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
    netflix
    -0.67
    vae
    -0.65
    ModLoader
    -0.64
    CVE
    -0.63
    ãĥ¥
    -0.63
    BuyableInstoreAndOnline
    -0.62
    #$
    -0.62
    nces
    -0.62
    ========
    -0.62
    meal
    -0.60
    POSITIVE LOGITS
    ocamp
    1.34
    sburgh
    1.19
    erton
    1.01
    etts
    1.00
    erman
    0.99
    erella
    0.98
    enger
    0.97
    eh
    0.92
    ortun
    0.91
    onent
    0.90
    Activations Density 0.034%

    No Known Activations