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. 11-RES-JB
    6. 19116
    Prev
    Next
    INDEX
    Explanations

    phrases related to technological services or company names

    oai_token-act-pair · gpt-3.5-turbo

    references to an "On" feature or product status in services

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     prayers
    -0.61
     Slug
    -0.61
     forces
    -0.59
     sneaky
    -0.58
     forgiven
    -0.58
     compens
    -0.58
     comforting
    -0.57
     partisans
    -0.56
     dug
    -0.55
     discouraging
    -0.54
    POSITIVE LOGITS
    yx
    1.40
    demand
    1.29
    slaught
    1.29
    eness
    1.26
    assis
    1.23
    etime
    1.22
    eworld
    1.20
    shore
    1.16
    Demand
    1.14
    coming
    1.14
    Activations Density 0.060%

    No Known Activations