INDEX

Explanations

terms related to visual or symbolic representations within different contexts

oai_token-act-pair · gpt-3.5-turbo

mentions of the concept of representation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

imb

-0.76

ivot

-0.70

ero

-0.69

frey

-0.69

issy

-0.69

launch

-0.68

kers

-0.68

strap

-0.67

si

-0.64

sis

-0.64

POSITIVE LOGITS

 Represent

1.23

 representation

1.20

 representations

1.18

eering

0.92

 representing

0.89

Represent

0.87

ational

0.83

 represented

0.81

represent

0.78

lations

0.78

Activations Density 0.019%

terms related to visual or symbolic representations within different contexts

mentions of the concept of representation

No Comments

No Known Activations