INDEX

Explanations

comparative terms

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

psc

-0.74

OE

-0.69

uti

-0.62

YP

-0.59

otide

-0.59

ims

-0.58

Slow

-0.57

 WARN

-0.57

Ĥİ

-0.57

imil

-0.55

POSITIVE LOGITS

1.16

an

1.15

 someone

0.90

0.86

 something

0.86

 another

0.84

 somebody

0.80

someone

0.76

 somet

0.72

Ãł

0.69

Activations Density 0.291%

comparative terms

No Comments

No Known Activations