© 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. 3248
    Prev
    Next
    INDEX
    Explanations

    sequences of numbers and alphabets that seem to belong to a structured format or code

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

    technical specifications and model identifiers for electronic products

    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
    estate
    -0.73
    withstanding
    -0.65
     clos
    -0.65
     disproportion
    -0.62
     damned
    -0.59
     spoil
    -0.59
    ,'"
    -0.56
    things
    -0.56
    enance
    -0.55
    idays
    -0.55
    POSITIVE LOGITS
    KT
    1.18
    YR
    0.98
    XT
    0.96
    FH
    0.96
    HM
    0.96
    UV
    0.95
    KK
    0.95
    BN
    0.94
    MF
    0.94
    WN
    0.94
    Activations Density 0.214%

    No Known Activations