INDEX

Explanations

positive recognition or acknowledgment

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

 depths

-0.68

 rall

-0.60

 specificity

-0.59

rouse

-0.58

 encount

-0.58

 occurrence

-0.58

tnc

-0.57

imes

-0.57

aple

-0.56

 Shack

-0.56

POSITIVE LOGITS

ocobo

0.78

è»

0.75

ãĥĸ

0.72

 from

0.72

 PLUS

0.71

DragonMagazine

0.70

 backing

0.67

âĺ

0.67

 thanks

0.67

RM

0.65

Activations Density 0.240%

positive recognition or acknowledgment

No Comments

No Known Activations