INDEX

Explanations

phrases related to personal development and self-improvement

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

atl

-0.68

 Penny

-0.63

 Seah

-0.60

anka

-0.58

UFF

-0.56

 Ashton

-0.56

 Bild

-0.56

 Pierce

-0.55

OTOS

-0.54

 Sandy

-0.54

POSITIVE LOGITS

 surely

1.23

 chances

1.20

 inevitably

1.04

 nevertheless

0.96

 likely

0.96

 automatically

0.94

 logically

0.91

 probably

0.89

 understandably

0.84

 invariably

0.84

Activations Density 6.281%

phrases related to personal development and self-improvement

No Comments

No Known Activations