© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 4980
    Prev
    Next
    INDEX
    Explanations

    positive evaluations or assessments, particularly in terms of performance or quality

    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
    717
    +0.13
    0.6%
    373
    +0.12
    0.6%
    745
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.13
    0.08
    636
    +0.12
    0.04
    250
    +0.11
    0.04
    Negative Logits
     occupied
    -0.63
     menace
    -0.63
     claimant
    -0.62
     accompl
    -0.62
     territory
    -0.61
     intact
    -0.60
     advancing
    -0.59
     achievement
    -0.59
     attacker
    -0.59
     vanquished
    -0.59
    POSITIVE LOGITS
    ggie
    0.79
    ļéĨĴ
    0.78
     AMA
    0.71
    zens
    0.65
    ĸļ
    0.65
    umerable
    0.63
     darn
    0.63
    alle
    0.63
    spective
    0.62
    fitting
    0.62
    Activations Density 0.225%

    No Known Activations