INDEX

Explanations

phrases related to real-world locations and contexts

oai_token-act-pair · gpt-3.5-turbo

terms associated with tangible or real-world 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

 Yorkers

-0.78

paio

-0.77

rencies

-0.77

orsi

-0.71

HCR

-0.69

enegger

-0.69

*/(

-0.65

 Converted

-0.64

————

-0.64

arcity

-0.64

POSITIVE LOGITS

 vicinity

1.10

 realm

1.02

 arena

0.98

 universe

0.96

 area

0.96

 context

0.96

 guise

0.96

osphere

0.94

 sphere

0.93

 direction

0.89

Activations Density 0.513%

phrases related to real-world locations and contexts

terms associated with tangible or real-world contexts

No Comments

No Known Activations