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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 3-RES-JB
    6. 1918
    Prev
    Next
    INDEX
    Explanations

    The neuron is primarily looking for words related to providing comfort, support, or assurance in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    terms related to reassurance or the act of providing comfort or assurance

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    VID
    -0.72
    BLE
    -0.71
    FORE
    -0.70
    Gi
    -0.70
    Interstitial
    -0.67
    ById
    -0.65
     Everywhere
    -0.65
     Horse
    -0.64
    LOD
    -0.64
    KER
    -0.64
    POSITIVE LOGITS
    essment
    1.29
    ignment
    1.28
    essed
    1.19
    igned
    1.19
    ortment
    1.15
    essing
    1.12
    urances
    1.09
    essor
    1.08
    imar
    1.05
    igning
    1.03
    Activations Density 0.005%

    No Known Activations