INDEX

Explanations

the word "split" at different activation levels

oai_token-act-pair · gpt-3.5-turbo

instances of the term "split" and other sections of text that may indicate separation or division

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.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

trak

-0.78

Marginal

-0.69

jamin

-0.69

urai

-0.68

anamo

-0.67

FORE

-0.66

Adv

-0.65

anto

-0.65

enegger

-0.65

sburgh

-0.65

POSITIVE LOGITS

 hairs

1.15

 evenly

0.90

 between

0.89

ters

0.82

creen

0.80

 split

0.75

 unequ

0.75

 apart

0.75

 Split

0.74

Split

0.73

Activations Density 0.047%

the word "split" at different activation levels

instances of the term "split" and other sections of text that may indicate separation or division

No Comments

No Known Activations