INDEX

Explanations

occurrences of the word "keyword" and its variations related to search engine optimization

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

ardy

-0.79

llah

-0.79

oÄŁ

-0.76

ersen

-0.73

kens

-0.73

ilon

-0.72

arus

-0.72

adra

-0.72

owship

-0.71

 Democr

-0.69

POSITIVE LOGITS

 keywords

1.09

 keyword

1.02

 phrase

0.98

 searches

0.90

 phrases

0.90

 sear

0.88

word

0.86

 prefix

0.81

 adjective

0.79

 stuffing

0.78

Activations Density 0.006%

occurrences of the word "keyword" and its variations related to search engine optimization

No Comments

No Known Activations