INDEX

Explanations

information related to veterans and military service

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

orie

-0.72

eller

-0.70

maps

-0.69

hran

-0.68

fully

-0.67

Buzz

-0.67

iPhone

-0.66

worms

-0.64

furt

-0.64

alter

-0.63

POSITIVE LOGITS

spective

0.92

 disability

0.84

 Disability

0.84

 Veterans

0.83

umatic

0.78

roleum

0.72

 vets

0.72

vet

0.71

Veter

0.71

 veterans

0.70

Activations Density 1.110%

information related to veterans and military service

No Comments

No Known Activations