INDEX

Explanations

product reviews and feedback comments

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

 operative

-0.69

 oust

-0.66

 combatants

-0.66

 subter

-0.66

 enshr

-0.65

 outlaw

-0.65

ulent

-0.64

 criminal

-0.63

 embattled

-0.63

 adversaries

-0.63

POSITIVE LOGITS

 myself

1.32

1.29

my

1.21

 haha

1.16

1.11

My

1.08

 Honestly

1.04

Honestly

1.00

 luckily

1.00

My

0.97

Activations Density 3.376%

product reviews and feedback comments

No Comments

No Known Activations