© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 8-RES-JB
    11. 9057
    Prev
    Next
    INDEX
    Explanations

    phrases related to playing or interacting with different objects or concepts

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "play with" in various contexts

    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
    hra
    -0.81
    nery
    -0.79
    levard
    -0.75
    itizens
    -0.69
    zynski
    -0.69
    itures
    -0.69
    elman
    -0.66
     gladly
    -0.66
    vt
    -0.66
    mite
    -0.66
    POSITIVE LOGITS
     dolls
    1.01
     impunity
    0.89
     toys
    0.84
    draw
    0.81
     darts
    0.81
    è¦ļéĨĴ
    0.78
     ropes
    0.77
     scissors
    0.77
     gust
    0.76
     sticks
    0.76
    Activations Density 0.111%

    No Known Activations