INDEX

Explanations

discussion of policies, political figures, and public affairs

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

hoe

-0.75

owered

-0.74

culus

-0.71

rug

-0.70

 Armored

-0.70

itching

-0.69

cigarettes

-0.69

bees

-0.68

 bowling

-0.64

ynthesis

-0.64

POSITIVE LOGITS

 deserves

1.06

 deserve

0.99

 rightly

0.90

 rightfully

0.90

 deserved

0.88

 therefore

0.88

 respected

0.79

vou

0.78

 justified

0.78

 justifies

0.78

Activations Density 0.210%

discussion of policies, political figures, and public affairs

No Comments

No Known Activations