INDEX

Explanations

comparative terms indicating superiority or increased quantity

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

Ire

-0.69

VICE

-0.67

LAB

-0.64

CTR

-0.63

 largeDownload

-0.62

 stead

-0.61

 valve

-0.60

maid

-0.59

 clearance

-0.58

 centr

-0.57

POSITIVE LOGITS

atos

1.58

assis

1.43

rax

0.86

osaurs

0.85

alan

0.85

ority

0.85

ormal

0.83

agar

0.82

atra

0.82

kees

0.82

Activations Density 0.005%

comparative terms indicating superiority or increased quantity

No Comments

No Known Activations