© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 8-RES-JB
    11. 14334
    Prev
    Next
    INDEX
    Explanations

    specifically identifies the word "been" in various phrases or sentences

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "been."

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    rouse
    -0.77
    âĨij
    -0.65
    Nich
    -0.64
     Niet
    -0.62
    ierrez
    -0.61
     MacArthur
    -0.61
    Dialogue
    -0.61
    inks
    -0.61
     Brach
    -0.59
    ï
    -0.59
    POSITIVE LOGITS
     awhile
    0.92
     deprecated
    0.91
     released
    0.87
     replaced
    0.87
    cffffcc
    0.84
     seen
    0.83
     updated
    0.83
     eaten
    0.81
     notified
    0.80
     able
    0.80
    Activations Density 0.046%

    No Known Activations