© 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. 10-RES-JB
    11. 12718
    Prev
    Next
    INDEX
    Explanations

    words related to stillness, physical surroundings, and attributes like wetness and sourness

    oai_token-act-pair · gpt-3.5-turbo

    terms and phrases related to silence and auditory concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ortium
    -0.66
    ofi
    -0.63
    ³³³³³³³³³³³³³³³³
    -0.62
     Pharm
    -0.62
    lished
    -0.60
    perial
    -0.60
    γ
    -0.60
    ULTS
    -0.60
    emi
    -0.58
     Undead
    -0.58
    POSITIVE LOGITS
     etiquette
    0.74
     habits
    0.65
     pals
    0.63
    ocalypse
    0.61
    pedia
    0.61
     assignment
    0.60
     designation
    0.59
     obsession
    0.58
     cra
    0.58
     woes
    0.58
    Activations Density 0.739%

    No Known Activations