INDEX

Explanations

software commands or instructions related to technical operations

oai_token-act-pair · gpt-3.5-turbo

numerical values and measurements related to products

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 eleph

-0.92

 defending

-0.88

 subpoena

-0.87

 betting

-0.86

 barr

-0.86

 increasingly

-0.85

 relat

-0.84

 responding

-0.83

 unemployed

-0.83

hed

-0.82

POSITIVE LOGITS

Pros

1.89

Features

1.85

Rated

1.83

Conclusion

1.82

Ingredients

1.71

Installation

1.70

Overall

1.69

Materials

1.68

Appearance

1.68

Gallery

1.66

Activations Density 0.256%

software commands or instructions related to technical operations

numerical values and measurements related to products

No Comments

No Known Activations