Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 8-RES-JB
    6. 19212
    Prev
    Next
    INDEX
    Explanations

    words related to the abbreviation "Ep" and medical conditions with "hemer"

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the term "Ep" and related terminology

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    FUL
    -0.85
    hips
    -0.81
    LESS
    -0.76
     Pru
    -0.75
    folk
    -0.68
     unauthorized
    -0.67
    holding
    -0.65
     humming
    -0.64
    WARE
    -0.61
     clearance
    -0.60
    POSITIVE LOGITS
    hemer
    1.52
    iphany
    1.51
    iscopal
    1.34
    isodes
    1.25
    istle
    1.25
    onymous
    1.24
    ipel
    1.10
    igraph
    1.06
    idem
    1.04
    otrop
    1.03
    Activations Density 0.029%

    No Known Activations