Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & 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. 4033
    Prev
    Next
    INDEX
    Explanations

    phrases describing physical actions and body positions, such as raising hands, linking arms, or holding hands up

    oai_token-act-pair · gpt-3.5-turbo

    references to raised hands and body language related to surrender or protest

    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
    ERAL
    -0.80
    >>>
    -0.73
    java
    -0.71
    produ
    -0.69
    haps
    -0.69
    utical
    -0.65
    phabet
    -0.65
    minist
    -0.65
    76561
    -0.64
    asio
    -0.64
    POSITIVE LOGITS
     clenched
    1.14
     pressed
    1.13
     tucked
    1.11
     folded
    1.11
     taped
    1.11
     tied
    1.10
     handcuffed
    1.10
     dangling
    1.09
     locked
    1.08
     draped
    1.08
    Activations Density 0.119%

    No Known Activations