© 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. 10-RES-JB
    11. 19020
    Prev
    Next
    INDEX
    Explanations

    phrases indicating impossibility or denial

    oai_token-act-pair · gpt-3.5-turbo

    negative sentiments or expressions of inability

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rather
    -0.78
    probably
    -0.73
    might
    -0.72
    staking
    -0.66
    arently
    -0.66
     Might
    -0.65
    maybe
    -0.65
     clarify
    -0.63
    not
    -0.63
    prototype
    -0.63
    POSITIVE LOGITS
     able
    1.27
     tolerated
    1.20
     anymore
    1.03
     bothered
    1.00
     anywhere
    0.97
     allowed
    0.96
     permitted
    0.93
     construed
    0.93
     dissu
    0.93
     swayed
    0.91
    Activations Density 0.147%

    No Known Activations