INDEX

Explanations

references to specific organizations or legal entities

oai_token-act-pair · gpt-3.5-turbo

the definite article "the."

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

 âĢº

-0.79

ibaba

-0.77

abi

-0.73

arten

-0.71

ratom

-0.69

itars

-0.68

RGB

-0.67

itto

-0.65

scape

-0.63

amera

-0.63

POSITIVE LOGITS

 aforementioned

1.05

 latter

1.03

ses

0.99

 same

0.98

 slightest

0.95

 likes

0.91

 respective

0.89

 usual

0.89

 afore

0.87

 outset

0.87

Activations Density 0.213%

references to specific organizations or legal entities

the definite article "the."

No Comments

No Known Activations