© 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. 4-RES-JB
    11. 10957
    Prev
    Next
    INDEX
    Explanations

    text captions or descriptions in a document

    oai_token-act-pair · gpt-3.5-turbo

    references to captions and subtitles in media content

    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
    prus
    -0.75
    ces
    -0.66
    cit
    -0.65
    kus
    -0.65
     dime
    -0.65
    atra
    -0.64
     Grail
    -0.63
    robat
    -0.62
     pistols
    -0.62
     Cups
    -0.61
    POSITIVE LOGITS
     caption
    0.93
    ed
    0.88
    TextColor
    0.82
    deck
    0.79
    xual
    0.79
    acters
    0.78
    edin
    0.77
    escription
    0.76
    named
    0.75
    writer
    0.74
    Activations Density 0.005%

    No Known Activations