INDEX

Explanations

phrases indicating inability to perform an action or function

oai_token-act-pair · gpt-3.5-turbo

instances of the word "unable" 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

soType

-0.67

gone

-0.66

ItemImage

-0.65

irie

-0.62

worn

-0.62

 Bullet

-0.61

 Tradition

-0.61

assault

-0.60

inventoryQuantity

-0.60

soDeliveryDate

-0.60

POSITIVE LOGITS

ioned

0.83

 compe

0.81

 adequately

0.78

 afford

0.72

 satisf

0.71

oooooooo

0.71

untarily

0.70

 feas

0.70

 miser

0.70

 muster

0.69

Activations Density 0.051%

phrases indicating inability to perform an action or function

instances of the word "unable" in various contexts

No Comments

No Known Activations