INDEX

Explanations

phrases or words related to gaps or distances between entities

oai_token-act-pair · gpt-3.5-turbo

phrases indicating gaps or distances between different entities or concepts

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

OGR

-0.92

ober

-0.78

ãĤ¡

-0.76

ONSORED

-0.75

ickr

-0.73

rak

-0.72

di

-0.71

br

-0.71

uddin

-0.71

DI

-0.69

POSITIVE LOGITS

 halves

0.82

 worlds

0.77

 extremes

0.68

 genders

0.68

 thirds

0.67

 sexes

0.67

two

0.66

 them

0.65

 Worlds

0.65

 peoples

0.63

Activations Density 0.042%

phrases or words related to gaps or distances between entities

phrases indicating gaps or distances between different entities or concepts

No Comments

No Known Activations