© 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. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 8-RES_POST_32K-OAI
    6. 5479
    Prev
    Next
    INDEX
    Explanations

    references to different types of rocks and geological features

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    129
    +0.13
    0.6%
    450
    +0.12
    0.6%
    383
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.04
    325
    +0.12
    0.03
    129
    +0.11
    0.02
    Negative Logits
    ufact
    -0.91
     Hots
    -0.76
    uthor
    -0.72
    riot
    -0.71
    ghai
    -0.69
    CTV
    -0.68
    iator
    -0.67
    ersive
    -0.67
    nea
    -0.65
    Interstitial
    -0.64
    POSITIVE LOGITS
     quar
    0.93
     formations
    0.86
     quarry
    0.84
    lake
    0.79
     ledge
    0.72
     slab
    0.71
     bould
    0.71
     fragments
    0.70
     reef
    0.70
     piles
    0.69
    Activations Density 0.099%

    No Known Activations