INDEX

Explanations

dates in a specific format

oai_token-act-pair · gpt-3.5-turbo

sequences of letters and characters, particularly patterns in titles or acronyms

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Materials

-0.64

 brim

-0.63

poke

-0.63

caps

-0.62

dos

-0.61

tip

-0.61

Bloomberg

-0.61

 Flavoring

-0.60

Shape

-0.60

picture

-0.59

POSITIVE LOGITS

hyde

0.70

oire

0.69

.:

0.67

culosis

0.67

.,

0.66

.?

0.65

ament

0.65

eteria

0.62

acus

0.61

.—

0.61

Activations Density 0.071%

dates in a specific format

sequences of letters and characters, particularly patterns in titles or acronyms

No Comments

No Known Activations