INDEX

Explanations

locations mentioned in news articles

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

inarily

-0.75

isoft

-0.71

natureconservancy

-0.69

 guiActiveUnfocused

-0.69

 guiName

-0.68

INAL

-0.67

 href

-0.66

BIL

-0.66

Cause

-0.63

-----------

-0.62

POSITIVE LOGITS

DC

0.97

bourg

0.76

 Lumpur

0.73

 Janeiro

0.72

clinton

0.72

 Jefferson

0.69

 Kong

0.68

 Nevada

0.67

DC

0.65

Activations Density 0.067%

locations mentioned in news articles

No Comments

No Known Activations