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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 4-RES-JB
    11. 10367
    Prev
    Next
    INDEX
    Explanations

    parts or sections of a text/document indicated by the word "Part" followed by a number

    oai_token-act-pair · gpt-3.5-turbo

    references to sections or parts within a document or article

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

    No Comments

    Negative Logits
    berus
    -0.71
    ãĥīãĥ©ãĤ´ãĥ³
    -0.66
    è¦ļéĨĴ
    -0.66
    anca
    -0.65
    apons
    -0.64
    incinn
    -0.63
     sugg
    -0.62
    ministic
    -0.62
     hawk
    -0.61
     haun
    -0.60
    POSITIVE LOGITS
    ially
    1.18
    icularly
    1.13
    ners
    1.11
    icular
    1.09
    nered
    1.08
    icipated
    0.99
    ition
    0.99
    ner
    0.98
    icle
    0.96
    icles
    0.96
    Activations Density 0.018%

    No Known Activations