INDEX

Explanations

instances of specific types of concepts or entities within a context

oai_token-act-pair · gpt-3.5-turbo

various types of opportunities and options in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

-0.70

ãĤ´ãĥ³

-0.68

staking

-0.65

comings

-0.61

 conclud

-0.60

 exting

-0.60

WAYS

-0.60

ãĤĴ

-0.58

ãĥ¡

-0.58

hig

-0.58

POSITIVE LOGITS

 they

1.14

we

1.01

 THEY

0.94

he

0.93

you

0.93

soever

0.89

she

0.86

they

0.86

 thou

0.85

it

0.83

Activations Density 0.155%

instances of specific types of concepts or entities within a context

various types of opportunities and options in different contexts

No Comments

No Known Activations