INDEX

Explanations

phrases related to machines or tools

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

raints

-0.88

alez

-0.83

omes

-0.78

uates

-0.77

uating

-0.75

si

-0.74

ivating

-0.72

 sqor

-0.72

orb

-0.71

uds

-0.70

POSITIVE LOGITS

gun

1.08

guns

0.99

 readable

0.82

 gunned

0.80

 learning

0.78

 guns

0.77

wreck

0.75

 malfunction

0.73

 Learning

0.73

 machine

0.73

Activations Density 0.022%

phrases related to machines or tools

No Comments

No Known Activations