INDEX

Explanations

comparisons of quantity or degrees

oai_token-act-pair · gpt-3.5-turbo

variations of the word "more."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Procedure

-0.75

DragonMagazine

-0.71

xtap

-0.67

Origin

-0.67

 Fail

-0.66

boy

-0.66

-0.65

 Passage

-0.65

DER

-0.64

piece

-0.63

POSITIVE LOGITS

 than

1.26

 efficient

1.15

 expensive

1.14

 costly

1.09

 stringent

1.08

 sophisticated

1.07

 manageable

1.06

 complicated

1.01

 advantageous

0.97

 advanced

0.93

Activations Density 0.153%

comparisons of quantity or degrees

variations of the word "more."

No Comments

No Known Activations