INDEX

Explanations

terms related to different types of environments, such as natural, built, operational, etc

oai_token-act-pair · gpt-3.5-turbo

references to different types of environments

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

 count

-0.66

ilee

-0.64

 Forbes

-0.63

 counted

-0.63

ichi

-0.63

law

-0.63

head

-0.62

 deductible

-0.61

hur

-0.60

ittee

-0.60

POSITIVE LOGITS

 environments

3.54

vironments

2.43

 climates

2.13

 ecosystems

1.89

 contexts

1.85

 atmosp

1.82

 landscapes

1.81

 habitats

1.78

 environment

1.76

 situations

1.68

Activations Density 0.013%

terms related to different types of environments, such as natural, built, operational, etc

references to different types of environments

No Comments

No Known Activations