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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 4-RES-JB
    6. 15258
    Prev
    Next
    INDEX
    Explanations

    words related to geographical locations specifically focusing on countries

    oai_token-act-pair · gpt-3.5-turbo

    terms related to specific musical references and composers, particularly Mozart and related keywords

    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
    Not in Any Lists

    No Comments

    Negative Logits
     scant
    -0.78
     Scots
    -0.69
    âĢ¢âĢ¢âĢ¢âĢ¢
    -0.67
     Viking
    -0.67
     Citizens
    -0.65
    aneous
    -0.64
    ¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯
    -0.64
     Celtic
    -0.64
     Vive
    -0.64
    UCT
    -0.63
    POSITIVE LOGITS
    wana
    1.00
     Moz
    0.99
    amb
    0.98
    olean
    0.96
    adish
    0.95
    illian
    0.93
    rolet
    0.90
    ongo
    0.89
    pheus
    0.87
    onga
    0.86
    Activations Density 0.009%

    No Known Activations