INDEX

Explanations

differences or discrepancies in various contexts

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

rollers

-0.79

der

-0.69

trak

-0.67

ATA

-0.65

ivery

-0.64

ãĤ®

-0.63

psc

-0.63

ODE

-0.63

vez

-0.63

leeve

-0.61

POSITIVE LOGITS

 between

1.45

between

1.30

 Between

1.30

yip

0.99

 separating

0.92

iveness

0.91

Between

0.84

 amongst

0.83

 favoring

0.80

ially

0.80

Activations Density 0.921%

differences or discrepancies in various contexts

No Comments

No Known Activations