© 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
    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. 22441
    Prev
    Next
    INDEX
    Explanations

    the word "at" near the verb "look."

    oai_token-act-pair · gpt-3.5-turbo

    references to the phrase "look at."

    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
    trap
    -0.71
     campaigner
    -0.69
    widget
    -0.66
    pite
    -0.65
    cell
    -0.65
    hement
    -0.63
    Cry
    -0.63
    SUP
    -0.62
    bender
    -0.62
     subsequ
    -0.62
    POSITIVE LOGITS
    ĸļ
    0.97
    =-=-=-=-
    0.80
    ļéĨĴ
    0.76
     least
    0.75
     angles
    0.71
    umbn
    0.68
    rical
    0.67
     replay
    0.66
    olphins
    0.66
     Hogwarts
    0.65
    Activations Density 0.041%

    No Known Activations