INDEX

Explanations

specific types or categories of items or concepts

oai_token-act-pair · gpt-3.5-turbo

discussions related to types of tasks and roles 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 @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

cffffcc

-0.81

Interstitial

-0.71

NetMessage

-0.63

Merit

-0.62

nova

-0.61

zar

-0.60

iae

-0.59

RAW

-0.59

nai

-0.57

hig

-0.57

POSITIVE LOGITS

 they

0.90

yip

0.88

soever

0.86

we

0.84

you

0.80

 thou

0.79

 THEY

0.79

 corresponds

0.77

he

0.75

 awaits

0.75

Activations Density 0.227%

specific types or categories of items or concepts

discussions related to types of tasks and roles in various contexts

No Comments

No Known Activations