INDEX

Explanations

organizations and institutions within 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_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

tumblr

-0.75

 partying

-0.71

 handshake

-0.67

 manipulating

-0.67

 revenge

-0.64

 murdering

-0.63

 emulate

-0.61

 piss

-0.61

 steroids

-0.61

ãĥİ

-0.61

POSITIVE LOGITS

elli

0.96

ijk

0.89

herty

0.86

kson

0.86

(@

0.85

 Chak

0.85

lette

0.83

 Cohen

0.83

elman

0.82

essler

0.82

Activations Density 1.539%

organizations and institutions within news articles

No Comments

No Known Activations