© 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. 11-RES-JB
    11. 24002
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences containing the word "Each" followed by various descriptions or actions

    oai_token-act-pair · gpt-3.5-turbo

    the word "Each" and its variations, emphasizing individual contributions and perspectives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    models
    -0.78
    amina
    -0.74
    apesh
    -0.73
    olas
    -0.72
    ools
    -0.71
    ables
    -0.70
    afort
    -0.69
    stones
    -0.69
    downs
    -0.69
    abs
    -0.68
    POSITIVE LOGITS
     successive
    1.21
     participant
    1.04
     iteration
    1.01
     year
    1.00
     month
    0.99
     member
    0.98
     person
    0.97
     week
    0.95
    tains
    0.92
     individual
    0.91
    Activations Density 0.037%

    No Known Activations