© 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. 3584
    Prev
    Next
    INDEX
    Explanations

    numbers related to casualties and attacks in war or conflict situations

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

    negative sentiments or themes related to harm and conflict

    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
     hatch
    -0.83
     microsc
    -0.83
     takedown
    -0.82
     square
    -0.78
     regress
    -0.76
     highs
    -0.74
     pity
    -0.74
     pores
    -0.74
     traps
    -0.73
     wonders
    -0.73
    POSITIVE LOGITS
    controlled
    1.79
    dominated
    1.77
    held
    1.68
    run
    1.65
    led
    1.60
    linked
    1.57
    owned
    1.55
    backed
    1.51
    affiliated
    1.50
    aligned
    1.49
    Activations Density 0.047%

    No Known Activations