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. 8-RES-JB
    6. 17312
    Prev
    Next
    INDEX
    Explanations

    dates and time expressions like "hour", "days after", "30 and 6:30 am", "minutes into"

    oai_token-act-pair · gpt-3.5-turbo

    instances of commas in the text

    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
    vre
    -0.75
    ãĤ´ãĥ³
    -0.71
    urally
    -0.70
    ongevity
    -0.69
    opia
    -0.63
    Really
    -0.63
    agine
    -0.62
    alo
    -0.62
    ãĤ¦ãĤ¹
    -0.61
    far
    -0.60
    POSITIVE LOGITS
     however
    0.84
     CTR
    0.77
     rumors
    0.77
     someone
    0.70
    FTWARE
    0.70
     appellant
    0.68
     assailants
    0.68
    someone
    0.68
     she
    0.68
     TMZ
    0.67
    Activations Density 0.220%

    No Known Activations