INDEX

Explanations

prepositional phrases indicating location or specialization

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lement

-0.68

bol

-0.60

=]

-0.60

een

-0.60

cade

-0.59

ison

-0.59

lash

-0.58

catentry

-0.58

ily

-0.58

ieth

-0.57

POSITIVE LOGITS

tow

1.08

ked

1.06

 lieu

0.97

 abundance

0.95

built

0.95

 mind

0.94

 front

0.94

clusions

0.93

 regards

0.91

alien

0.89

Activations Density 0.130%

prepositional phrases indicating location or specialization

No Comments

No Known Activations