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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 3-RES-JB
    6. 9953
    Prev
    Next
    INDEX
    Explanations

    text related to directions or spatial positions, specifically focusing on the left side

    oai_token-act-pair · gpt-3.5-turbo

    references to the left side or direction

    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
    alist
    -0.74
    ASED
    -0.71
    MAT
    -0.71
     conduc
    -0.68
    enegger
    -0.67
     Cosponsors
    -0.66
    inarily
    -0.65
    ITNESS
    -0.64
    yrinth
    -0.64
    REC
    -0.63
    POSITIVE LOGITS
    overs
    1.00
    wing
    0.98
     untreated
    0.86
     wing
    0.84
     handed
    0.84
    ward
    0.83
     hemisphere
    0.78
     hand
    0.78
    wich
    0.74
     undone
    0.73
    Activations Density 0.036%

    No Known Activations