© 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. 5013
    Prev
    Next
    INDEX
    Explanations

    instances of the word "where."

    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
    264
    +0.12
    0.5%
    504
    +0.10
    0.4%
    483
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    300
    +0.12
    0.04
    30
    +0.10
    0.03
    135
    +0.09
    0.04
    Negative Logits
    sense
    -0.80
    ragon
    -0.76
    availability
    -0.74
    termin
    -0.73
    ahime
    -0.70
    wik
    -0.69
    onomic
    -0.69
    mind
    -0.67
    ithering
    -0.67
    erity
    -0.65
    POSITIVE LOGITS
     employees
    0.78
     houses
    0.73
     students
    0.73
     residents
    0.73
     gunmen
    0.70
     tourists
    0.70
     robbers
    0.69
     workers
    0.68
     they
    0.68
     laborers
    0.68
    Activations Density 0.147%

    No Known Activations