INDEX

Explanations

keywords related to the product name "XL"

oai_token-act-pair · gpt-3.5-turbo

references to XL and NX related concepts or products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

flies

-0.83

keeping

-0.82

 association

-0.74

âĸ¬

-0.71

ãĥĨ

-0.71

ãĥĩãĤ£

-0.70

ãĤ±

-0.70

é¾įå

-0.69

ÑĢ

-0.67

fly

-0.66

POSITIVE LOGITS

XL

1.16

XL

1.05

VII

0.95

xus

0.90

III

0.86

anes

0.86

VI

0.83

iang

0.80

oyd

0.79

II

0.79

Activations Density 0.004%

keywords related to the product name "XL"

references to XL and NX related concepts or products

No Comments

No Known Activations