INDEX

Explanations

phrases related to the name "Nada"

oai_token-act-pair · gpt-3.5-turbo

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

icles

-0.77

names

-0.72

minded

-0.69

sheet

-0.68

URES

-0.68

giving

-0.68

urers

-0.67

 clipboard

-0.67

rored

-0.67

icians

-0.66

POSITIVE LOGITS

qua

0.99

illac

0.94

ivari

0.81

ascus

0.81

que

0.78

$$

0.77

uthor

0.77

qi

0.77

0.76

Activations Density 0.018%

phrases related to the name "Nada"

No Comments

No Known Activations