INDEX

Explanations

a variety of important nouns, particularly those related to legal, medical, environmental, and community-related 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_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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

erity

-0.69

uberty

-0.63

rison

-0.62

cknow

-0.60

elight

-0.60

sonian

-0.58

 Nightmares

-0.58

inventoryQuantity

-0.57

 History

-0.57

 Expect

-0.56

POSITIVE LOGITS

 apiece

1.14

 totaling

1.13

 ranging

0.99

 worldwide

0.99

 spanning

0.97

 including

0.95

 nationwide

0.90

 across

0.84

 categorized

0.84

 total

0.82

Activations Density 0.182%

a variety of important nouns, particularly those related to legal, medical, environmental, and community-related contexts

No Comments

No Known Activations