© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 12-RES-JB
    11. 2944
    Prev
    Next
    INDEX
    Explanations

    past tense verbs (actions that have already happened)

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    instances of the verb "was" indicating past actions or states

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 12-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post
    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_post
    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
     tides
    -0.63
     newsletters
    -0.61
     margins
    -0.59
     tide
    -0.57
     buckets
    -0.56
     zones
    -0.56
     appointments
    -0.56
     Fighters
    -0.56
     equations
    -0.55
     degrees
    -0.55
    POSITIVE LOGITS
    hes
    1.22
    wolves
    1.05
     founded
    1.00
     teasp
    1.00
    hers
    0.98
     born
    0.97
     pione
    0.95
    abi
    0.93
    wolf
    0.92
     invented
    0.87
    Activations Density 0.140%

    No Known Activations