INDEX

Explanations

mentions of a specific word or concept, "NY", potentially referring to New York

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "ny" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

EMP

-0.87

Reviewed

-0.73

rador

-0.72

 FEMA

-0.71

ENDED

-0.70

CVE

-0.69

ModLoader

-0.67

ãĥ¼ãĥĨ

-0.66

PKK

-0.65

Spread

-0.65

POSITIVE LOGITS

ny

0.99

mph

0.99

acht

0.93

heter

0.85

omi

0.81

 Mellon

0.80

ansky

0.79

alty

0.74

akov

0.73

 Giuliani

0.73

Activations Density 0.006%

mentions of a specific word or concept, "NY", potentially referring to New York

occurrences of the term "ny" in various contexts

No Comments

No Known Activations