© 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. 2154
    Prev
    Next
    INDEX
    Explanations

    instances of a specific word being mentioned or found

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "found."

    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
    idium
    -0.83
     paced
    -0.69
     partic
    -0.66
    weather
    -0.63
    cussion
    -0.61
    cius
    -0.60
     fue
    -0.60
    cised
    -0.59
    oiler
    -0.58
     attendant
    -0.58
    POSITIVE LOGITS
    ãĤ¤ãĥĪ
    0.91
    Ô
    0.85
    使
    0.84
    ãĤ¼
    0.83
    çīĪ
    0.80
    ãģ®å
    0.79
    éļ
    0.78
    herer
    0.73
    Find
    0.72
    DragonMagazine
    0.70
    Activations Density 0.055%

    No Known Activations