INDEX

Explanations

elements or aspects of different subjects or objects

oai_token-act-pair · gpt-3.5-turbo

references to specific components or features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

sburgh

-0.76

sett

-0.74

scrib

-0.72

ctors

-0.71

zee

-0.71

spe

-0.69

istani

-0.68

ker

-0.66

claimed

-0.66

ceed

-0.66

POSITIVE LOGITS

 thereof

1.02

ality

0.99

als

0.94

hetical

0.83

ARY

0.82

aries

0.76

of

0.74

 element

0.74

 elements

0.73

alogy

0.73

Activations Density 0.062%

elements or aspects of different subjects or objects

references to specific components or features

No Comments

No Known Activations