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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 1-RES-JB
    6. 18955
    Prev
    Next
    INDEX
    Explanations

    phrases related to deadlines or expected occurrences

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "due" and related phrases indicating causation or obligation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    anon
    -0.66
    jc
    -0.65
     Maple
    -0.65
     Sheep
    -0.64
    amped
    -0.63
    JC
    -0.63
    assisted
    -0.63
    flower
    -0.63
     Mayo
    -0.61
    amps
    -0.60
    POSITIVE LOGITS
    */(
    0.86
     diligence
    0.79
     dilig
    0.74
    ority
    0.73
    endment
    0.73
    merce
    0.71
    laws
    0.70
    actionDate
    0.69
    oldown
    0.68
    hedral
    0.66
    Activations Density 0.020%

    No Known Activations