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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 12-RES-JB
    11. 3316
    Prev
    Next
    INDEX
    Explanations

    phrases related to a specific TV show or movie, particularly about its production process and characters

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    repeated references to "South" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 12-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post
    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_post
    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
     mishand
    -0.63
     compromise
    -0.61
     stren
    -0.59
     pinch
    -0.58
     conscience
    -0.58
     profit
    -0.58
     negligent
    -0.57
     compromising
    -0.57
     fear
    -0.57
     possible
    -0.57
    POSITIVE LOGITS
    wark
    1.28
    western
    1.27
     Carolina
    1.05
    ampton
    1.04
    west
    1.00
    African
    0.94
     Dakota
    0.94
     Africa
    0.93
    park
    0.93
     Korean
    0.92
    Activations Density 0.043%

    No Known Activations