INDEX

Explanations

phrases related to lists or itemizations

oai_token-act-pair · gpt-3.5-turbo

phrases that emphasize a notable characteristic or classification of subjects

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

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

etheless

-0.70

leneck

-0.67

ledge

-0.66

lene

-0.66

onomy

-0.66

lished

-0.65

perty

-0.63

 Span

-0.63

formance

-0.62

vertising

-0.62

POSITIVE LOGITS

ioxide

0.73

wolf

0.73

eps

0.71

OND

0.69

types

0.69

 indications

0.69

 Marketable

0.68

ä½ľ

0.67

ilk

0.66

 kindred

0.64

Activations Density 0.166%

phrases related to lists or itemizations

phrases that emphasize a notable characteristic or classification of subjects

No Comments

No Known Activations