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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 4-RES-JB
    6. 8037
    Prev
    Next
    INDEX
    Explanations

    instances of the word "return" followed by a location or a personal pronoun

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "return to" in various contexts

    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
    anooga
    -0.75
    erning
    -0.68
    kered
    -0.68
     precursor
    -0.64
    anners
    -0.63
    erity
    -0.62
    sel
    -0.60
    ussen
    -0.59
    unning
    -0.59
    okane
    -0.59
    POSITIVE LOGITS
     haunt
    0.85
     blows
    0.72
    ãĤ¼
    0.70
     Ashes
    0.69
    owship
    0.68
     fray
    0.68
     shore
    0.67
    å£
    0.67
    abama
    0.67
    ITED
    0.66
    Activations Density 0.078%

    No Known Activations