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. 4-RES-JB
    6. 23588
    Prev
    Next
    INDEX
    Explanations

    mentions of family relationships, particularly references to the role of a grandfather

    oai_token-act-pair · gpt-3.5-turbo

    references to grandfathers and grandparents

    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
    rd
    -0.81
    acus
    -0.79
    AAF
    -0.72
    ylon
    -0.72
    USE
    -0.71
    okers
    -0.70
    kell
    -0.67
    tnc
    -0.66
    yss
    -0.66
    eq
    -0.65
    POSITIVE LOGITS
    father
    1.05
     Takeru
    0.96
    parents
    0.90
     grandfather
    0.85
     Gohan
    0.80
     grandparents
    0.77
     daddy
    0.76
    dad
    0.74
    ships
    0.74
     Odin
    0.73
    Activations Density 0.008%

    No Known Activations