© 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
    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. 23966
    Prev
    Next
    INDEX
    Explanations

    words related to proper names and locations, specifically with emphasis on characters that may appear in stories or real-world individuals connected to particular places

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the character string "str" in various contexts

    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
    chwitz
    -0.66
    rice
    -0.63
    doms
    -0.56
     remotely
    -0.55
     Jinn
    -0.55
    rals
    -0.53
    States
    -0.53
     participation
    -0.53
     Samar
    -0.52
     touch
    -0.52
    POSITIVE LOGITS
    glers
    1.05
    ngth
    0.83
    cko
    0.67
    gaard
    0.67
    sung
    0.66
    eways
    0.66
    pta
    0.65
    bg
    0.64
    utsche
    0.64
    gren
    0.64
    Activations Density 0.199%

    No Known Activations