© 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. 11-RES-JB
    11. 11128
    Prev
    Next
    INDEX
    Explanations

    adjectives or verbs related to cause and effect

    oai_token-act-pair · gpt-3.5-turbo

    references to causation or contributing factors

    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
    adel
    -0.86
    apest
    -0.84
    isode
    -0.83
     sidx
    -0.79
    emn
    -0.77
    onomy
    -0.77
    ura
    -0.76
    aurus
    -0.75
    finals
    -0.75
    daq
    -0.75
    POSITIVE LOGITS
     inexper
    1.41
     inadequate
    1.29
     faulty
    1.28
     factors
    1.22
     insufficient
    1.18
     improper
    1.15
     lack
    1.15
     misunderstanding
    1.15
     incompetence
    1.14
     genetics
    1.14
    Activations Density 0.303%

    No Known Activations