INDEX

Explanations

phrases or instances of negation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

 Niet

-0.69

lined

-0.68

å¥

-0.68

grounds

-0.67

holes

-0.66

quickShipAvailable

-0.65

tions

-0.64

 cleaner

-0.61

LOT

-0.59

Line

-0.59

POSITIVE LOGITS

 necessarily

1.11

 hesitate

1.06

icably

1.05

 underestimate

0.93

 participate

0.93

 bother

0.91

 speculate

0.89

 confuse

0.88

 exceed

0.88

 interfere

0.88

Activations Density 0.039%

phrases or instances of negation

No Comments

No Known Activations