INDEX

Explanations

keywords related to quantities or rankings, such as "number," "percentage," or "top."

oai_token-act-pair · gpt-3.5-turbo

data points related to quantitative measurements and significant changes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 guiActiveUn

-0.80

 unfocusedRange

-0.79

ounded

-0.78

kas

-0.74

erey

-0.71

ouston

-0.71

assetsadobe

-0.68

rocal

-0.67

OUN

-0.66

senal

-0.66

POSITIVE LOGITS

 imaginable

1.64

 EVER

1.28

 possible

1.13

 ever

1.13

 conceivable

1.06

ever

1.02

 practicable

0.97

Ever

0.87

 achievable

0.83

 Ever

0.78

Activations Density 0.496%

keywords related to quantities or rankings, such as "number," "percentage," or "top."

data points related to quantitative measurements and significant changes

No Comments

No Known Activations