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. 3522
    Prev
    Next
    INDEX
    Explanations

    medical conditions and treatment words, such as types of dermatitis and medical procedures

    oai_token-act-pair · gpt-3.5-turbo

    indicators of plant-related topics and plant health issues

    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
    )!
    -0.64
    }}}
    -0.61
     accordingly
    -0.56
     others
    -0.55
     embold
    -0.54
     afterwards
    -0.53
     later
    -0.53
     sadly
    -0.53
     secondly
    -0.52
     endif
    -0.52
    POSITIVE LOGITS
    particip
    0.65
    FTWARE
    0.63
    SHIP
    0.59
    unte
    0.58
    oscopic
    0.56
    tele
    0.55
    perty
    0.53
    houn
    0.51
    monitor
    0.51
    odcast
    0.51
    Activations Density 1.364%

    No Known Activations