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. 14735
    Prev
    Next
    INDEX
    Explanations

    links or commands prompting interaction or engagement with multimedia content

    oai_token-act-pair · gpt-3.5-turbo

    commands and prompts related to viewing and sharing media content

    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
    idable
    -0.66
    ãĤ´ãĥ³
    -0.65
     Reincarn
    -0.61
    etheless
    -0.61
    doms
    -0.59
    akra
    -0.59
     pse
    -0.58
    wra
    -0.58
    quartered
    -0.58
    ppings
    -0.57
    POSITIVE LOGITS
    SPONSORED
    0.76
    hent
    0.70
     enlarg
    0.68
     enlarged
    0.65
     VIDEOS
    0.64
     Advertisement
    0.62
     LAR
    0.61
    rect
    0.60
     enlarge
    0.59
    oice
    0.58
    Activations Density 0.111%

    No Known Activations