INDEX

Explanations

dates and times mentioned in sentences

oai_token-act-pair · gpt-3.5-turbo

dates and times related to events

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

 careless

-0.81

dad

-0.75

failed

-0.73

 intercepted

-0.67

 explanations

-0.66

 accidental

-0.65

 misinterpret

-0.65

inventoryQuantity

-0.65

 justified

-0.63

 wiret

-0.61

POSITIVE LOGITS

1.09

 tomorrow

1.05

1.01

0.86

uesday

0.85

Tickets

0.84

 evenings

0.83

 Thurs

0.83

 noon

0.83

Activations Density 0.167%

dates and times mentioned in sentences

dates and times related to events

No Comments

No Known Activations