© 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. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 3710
    Prev
    Next
    INDEX
    Explanations

    words related to social media or interaction with others, such as "blocked," "unblock," and "follow."

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    668
    +0.17
    0.6%
    301
    +0.14
    0.5%
    651
    +0.14
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.17
    0.12
    299
    +0.14
    0.08
    701
    +0.14
    0.06
    Negative Logits
    udeb
    -0.99
    tti
    -0.85
     challeng
    -0.84
     Quadro
    -0.83
    iosyn
    -0.80
     appropri
    -0.80
     incon
    -0.79
     sacrific
    -0.78
     neigh
    -0.76
     Dh
    -0.74
    POSITIVE LOGITS
    ment
    1.68
    able
    1.67
    ABLE
    1.48
    ables
    1.43
    ments
    1.42
    edIn
    1.40
    ings
    1.33
    MENT
    1.32
    enment
    1.32
    ability
    1.28
    Activations Density 2.225%

    No Known Activations