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. 3-RES-JB
    6. 7569
    Prev
    Next
    INDEX
    Explanations

    words related to the planet Mars or things associated with Mars

    oai_token-act-pair · gpt-3.5-turbo

    references to the term "Martian."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ulet
    -0.87
    reme
    -0.84
    uer
    -0.84
    reply
    -0.83
    irs
    -0.80
    hr
    -0.80
    reen
    -0.79
    unal
    -0.78
    forward
    -0.78
    committee
    -0.78
    POSITIVE LOGITS
     Martian
    0.92
     Worlds
    0.72
     Curiosity
    0.69
     Giant
    0.67
     Beasts
    0.67
     stripes
    0.66
     Commando
    0.66
     Dimensions
    0.65
     Dynamics
    0.64
     Conditions
    0.63
    Activations Density 0.011%

    No Known Activations