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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 18210
    Prev
    Next
    INDEX
    Explanations

    words related to future actions

    oai_token-act-pair · gpt-3.5-turbo

    the word "hasn't" and its variations to indicate lack or absence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     inter
    -0.63
     ext
    -0.58
     Ext
    -0.57
     separate
    -0.54
     absolute
    -0.53
     free
    -0.53
     FIG
    -0.53
    ext
    -0.52
    igate
    -0.52
     enc
    -0.51
    POSITIVE LOGITS
     hasn
    3.01
     hadn
    2.14
     haven
    2.04
     isn
    1.91
     doesn
    1.86
     wasn
    1.81
     didn
    1.64
    doesn
    1.64
     wouldn
    1.50
     doesnt
    1.45
    Activations Density 0.009%

    No Known Activations