© 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. 9-RES-JB
    11. 11402
    Prev
    Next
    INDEX
    Explanations

    verbs indicating actions or goals

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "take" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     accompanies
    -0.74
     agre
    -0.70
    tions
    -0.67
    ese
    -0.67
    hess
    -0.63
    ateg
    -0.60
    Gear
    -0.60
    rehens
    -0.58
    ominated
    -0.58
    vine
    -0.58
    POSITIVE LOGITS
     advantage
    1.42
     care
    1.02
    aways
    1.02
     aback
    0.94
     liberties
    0.94
     away
    0.93
    advant
    0.90
     cues
    0.88
     pains
    0.88
     responsibility
    0.87
    Activations Density 0.092%

    No Known Activations