INDEX

Explanations

terms related to power dynamics and hierarchies, including words related to government structures and brand power

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

Von

-0.75

eday

-0.69

eryl

-0.69

algia

-0.67

auder

-0.65

eret

-0.65

ALK

-0.64

akeru

-0.64

Bei

-0.63

roit

-0.63

POSITIVE LOGITS

houses

1.16

stroke

1.01

lifting

1.00

lessness

0.97

 outage

0.91

 wielded

0.90

puff

0.89

plant

0.88

FUL

0.86

 vested

0.85

Activations Density 0.584%

terms related to power dynamics and hierarchies, including words related to government structures and brand power

No Comments

No Known Activations