INDEX

Explanations

technical terms related to software development, such as "modules" and "packages"

oai_token-act-pair · gpt-3.5-turbo

references to software components and packages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 piece

-0.92

 ruler

-0.88

 statue

-0.85

 deed

-0.82

 statement

-0.81

 portion

-0.81

 passage

-0.80

 affair

-0.80

 mechanic

-0.80

 thing

-0.80

POSITIVE LOGITS

 architectures

1.81

 packages

1.73

 frameworks

1.72

 libraries

1.66

 databases

1.64

 formats

1.60

 technologies

1.57

 modules

1.51

 interfaces

1.51

 repositories

1.48

Activations Density 0.089%

technical terms related to software development, such as "modules" and "packages"

references to software components and packages

No Comments

No Known Activations