Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 2-RES-JB
    6. 36
    Prev
    Next
    INDEX
    Explanations

    It appears that Neuron 4 does not activate for any of the tokens provided, suggesting that it may not have been trained properly or that the given inputs do not include the features that Neuron 4 is designed to detect

    oai_token-act-pair · gpt-4-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Layer
    -0.81
    Maps
    -0.71
    STER
    -0.70
    SOURCE
    -0.70
    DOWN
    -0.70
    bling
    -0.70
     Offline
    -0.68
    grounds
    -0.68
     PAGE
    -0.68
    ãĥ¼ãĥ«
    -0.67
    POSITIVE LOGITS
     suffice
    0.77
     concess
    0.76
    erva
    0.76
    umph
    0.74
    ugal
    0.73
     unexpectedly
    0.72
    acio
    0.72
    arcity
    0.69
     disappoint
    0.69
     qualified
    0.68
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.