© 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. 11-RES-JB
    11. 22573
    Prev
    Next
    INDEX
    Explanations

    references to nature or the environment

    oai_token-act-pair · gpt-3.5-turbo

    nouns related to physical structures and natural elements

    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
    tains
    -0.77
    soever
    -0.71
    Recomm
    -0.65
    PLA
    -0.61
    doms
    -0.60
    Recommend
    -0.60
    gery
    -0.59
     subjects
    -0.56
    nant
    -0.56
    Allows
    -0.56
    POSITIVE LOGITS
     were
    1.14
     are
    1.12
     weren
    1.11
     aren
    1.06
     remain
    0.99
     have
    0.96
     expire
    0.90
     revert
    0.90
    cape
    0.90
     collided
    0.89
    Activations Density 0.247%

    No Known Activations