INDEX

Explanations

instances of the word "level" followed by other words

oai_token-act-pair · gpt-3.5-turbo

references to levels and degrees of various attributes or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

maid

-0.79

ttes

-0.79

ãĤ¼ãĤ¦ãĤ¹

-0.77

 Cosponsors

-0.76

orative

-0.73

 Vaugh

-0.72

 Jinn

-0.68

ista

-0.68

ores

-0.67

ists

-0.67

POSITIVE LOGITS

 maturity

1.06

 sophistication

1.05

 professionalism

0.88

 autonomy

0.88

 accomplishment

0.85

 immersion

0.82

 consciousness

0.82

 competence

0.82

 abstraction

0.80

 accommodation

0.80

Activations Density 0.089%

instances of the word "level" followed by other words

references to levels and degrees of various attributes or conditions

No Comments

No Known Activations