© 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. 2-RES-JB
    11. 995
    Prev
    Next
    INDEX
    Explanations

    words starting with "cere" or the name "Edmund"

    oai_token-act-pair · gpt-4-turbo

    phrases related to medical conditions or terms, especially those related to the brain or nervous system

    oai_token-act-pair · gpt-3.5-turbo

    references to specific characters or topics with the prefix "Cere" or related terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Magnus
    -0.69
     hashing
    -0.68
    creen
    -0.67
     OPS
    -0.67
    icago
    -0.67
    iage
    -0.64
    minster
    -0.64
     Elm
    -0.63
    heimer
    -0.62
    iamond
    -0.62
    POSITIVE LOGITS
    brate
    0.92
     Cere
    0.92
    milo
    0.88
    brates
    0.87
    unal
    0.84
    mony
    0.83
    ember
    0.78
    rious
    0.77
    ndra
    0.77
    ppo
    0.76
    Activations Density 0.022%

    No Known Activations