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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 5170
    Prev
    Next
    INDEX
    Explanations

    references to additional details or explanations within the text

    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
    163
    +0.11
    0.5%
    157
    +0.10
    0.4%
    47
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.11
    0.08
    131
    +0.10
    0.05
    163
    +0.09
    0.05
    Negative Logits
     effected
    -0.71
    edo
    -0.69
     lasted
    -0.66
     thanking
    -0.65
    icial
    -0.64
    mouth
    -0.64
     behaved
    -0.62
     centered
    -0.62
     inbox
    -0.61
     appealed
    -0.61
    POSITIVE LOGITS
    }}}
    0.82
     ¶
    0.78
     Figures
    0.71
     Protocol
    0.67
    idable
    0.67
    WAYS
    0.65
    lance
    0.65
    Nation
    0.63
     Sok
    0.63
     detail
    0.62
    Activations Density 0.605%

    No Known Activations