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. 0-RES-JB
    6. 14046
    Prev
    Next
    INDEX
    Explanations

    words related to painting

    oai_token-act-pair · gpt-3.5-turbo

    references to types of paint and artistic techniques

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     obedient
    -0.78
    Mos
    -0.66
    ONSORED
    -0.65
     readiness
    -0.63
     GOODMAN
    -0.62
    responsible
    -0.61
     rehearsal
    -0.60
     slut
    -0.59
     reinforced
    -0.59
    euro
    -0.58
    POSITIVE LOGITS
     Paint
    2.55
     Draw
    1.10
    acular
    0.92
     reign
    0.83
    bid
    0.83
    SourceFile
    0.81
     Pavilion
    0.76
     Bid
    0.76
     Pyth
    0.73
     Golf
    0.71
    Activations Density 0.007%

    No Known Activations