INDEX

Explanations

references to physical edges or boundaries in a variety of contexts

oai_token-act-pair · gpt-3.5-turbo

references to edges or boundaries in various contexts

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

isson

-0.71

odder

-0.70

asses

-0.70

atorium

-0.67

onte

-0.67

atoon

-0.65

oka

-0.65

alogue

-0.65

asting

-0.65

leaders

-0.64

POSITIVE LOGITS

 edges

0.79

bare

0.75

pace

0.73

 thereof

0.71

layer

0.71

 Detail

0.70

pring

0.69

paces

0.68

 Catalyst

0.67

cape

0.66

Activations Density 0.021%

references to physical edges or boundaries in a variety of contexts

references to edges or boundaries in various contexts

No Comments

No Known Activations