INDEX

Explanations

phrases that indicate comparisons or analogies

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

#$

-0.66

Posts

-0.66

aldo

-0.65

rous

-0.65

ãģł

-0.64

OUS

-0.63

itton

-0.62

regon

-0.59

ouk

-0.59

OUR

-0.59

POSITIVE LOGITS

pired

1.21

pires

1.20

 portrayed

0.97

 depicted

0.95

ociated

0.95

pects

0.93

 well

0.93

phy

0.91

semb

0.89

pire

0.88

Activations Density 0.084%

phrases that indicate comparisons or analogies

No Comments

No Known Activations