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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 4-RES-JB
    6. 15004
    Prev
    Next
    INDEX
    Explanations

    mentions of emotions related to anger or frustration

    oai_token-act-pair · gpt-3.5-turbo

    the concept of madness or irrational behavior

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •layer 4 - anger

    No Comments

    Negative Logits
    soDeliveryDate
    -0.81
    OPLE
    -0.78
    ãģ¯
    -0.75
    chnology
    -0.75
    etitive
    -0.74
    çĦ
    -0.73
    zzo
    -0.68
    lished
    -0.68
    oldown
    -0.67
    roma
    -0.66
    POSITIVE LOGITS
    agascar
    1.11
    rid
    0.95
    der
    0.84
    bol
    0.81
    cap
    0.81
     mad
    0.77
    ness
    0.75
    mad
    0.74
    iate
    0.74
     dash
    0.73
    Activations Density 0.008%

    No Known Activations