© 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. 0-RES-JB
    11. 7877
    Prev
    Next
    INDEX
    Explanations

    locations or locations within certain contexts

    oai_token-act-pair · gpt-3.5-turbo

    mentions of locations or geographic names

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ":[{"
    -0.76
    iper
    -0.71
     scr
    -0.69
     Phys
    -0.65
     Jian
    -0.65
     =>
    -0.64
     Healer
    -0.63
    pu
    -0.61
     epid
    -0.60
     Bio
    -0.60
    POSITIVE LOGITS
    LAND
    4.12
    landers
    2.71
    land
    2.47
    lands
    1.89
    lander
    1.71
    LAN
    1.30
    landish
    1.29
    FIELD
    1.15
    Land
    1.15
     Lands
    1.12
    Activations Density 0.008%

    No Known Activations