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. 1-RES-JB
    6. 23755
    Prev
    Next
    INDEX
    Explanations

    words related to land or territories

    oai_token-act-pair · gpt-3.5-turbo

    instances and discussions related to land

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ĸļ
    -0.85
     tremend
    -0.81
    vernment
    -0.81
    pload
    -0.74
    unction
    -0.73
    inki
    -0.71
     Taste
    -0.69
     Fior
    -0.65
     cumbers
    -0.63
    thodox
    -0.63
    POSITIVE LOGITS
     land
    0.85
    mass
    0.85
    mine
    0.84
    scape
    0.80
    lords
    0.80
    lord
    0.79
     mines
    0.78
    holder
    0.77
    locked
    0.77
    plane
    0.77
    Activations Density 0.015%

    No Known Activations