INDEX

Explanations

terms related to capturing or maintaining official records

oai_token-act-pair · gpt-3.5-turbo

the term "records" and its variations related to documentation and data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

nen

-0.80

eca

-0.78

bite

-0.76

osi

-0.74

 Galile

-0.73

AFTA

-0.73

anism

-0.72

outher

-0.70

hattan

-0.69

tein

-0.69

POSITIVE LOGITS

 records

1.11

 record

0.93

 holder

0.85

 Records

0.84

 databases

0.83

 Record

0.83

 documenting

0.79

keeping

0.79

record

0.77

 Holder

0.77

Activations Density 0.032%

terms related to capturing or maintaining official records

the term "records" and its variations related to documentation and data

No Comments

No Known Activations