© 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. 5506
    Prev
    Next
    INDEX
    Explanations

    references to "horns" and "whistle" in various contexts

    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
    553
    +0.12
    0.5%
    385
    +0.10
    0.5%
    309
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.12
    0.01
    288
    +0.10
    0.02
    289
    +0.09
    0.01
    Negative Logits
    izable
    -0.81
    ization
    -0.74
    igate
    -0.72
    arians
    -0.72
    igating
    -0.72
    isation
    -0.71
    igation
    -0.70
    izations
    -0.69
    ivia
    -0.69
    icals
    -0.68
    POSITIVE LOGITS
    pipe
    0.99
    pots
    0.96
    hovah
    0.95
    bone
    0.82
    dropping
    0.82
    sticks
    0.81
    wash
    0.80
    leading
    0.79
    plate
    0.79
    bones
    0.79
    Activations Density 0.054%

    No Known Activations